Experimental Neurosurgery in Animal Models

Neuromethods 116
Miroslaw Janowski Editor
Experimental
Neurosurgery
in Animal Models
NEUROMETHODS
Series Editor
Wolfgang Walz
University of Saskatchewan
Saskatoon, SK, Canada
For further volumes:

http://www.springer.com/series/7657
Experimental Neurosurgery
in Animal Models
Edited by
Miroslaw Janowski
The Russell H. Morgan Department of Radiology and Radiological Science, Division of MR Research
Institute for Cell Engineering, Johns Hopkins University School of Medicine, Baltimore, MD, USA;
NeuroRepair Department, Mossakowski Medical Research Centre, Warsaw, Poland;
Department of Neurosurgery, Mossakowski Medical Research Centre, Warsaw, Poland
Editor
Miroslaw Janowski
The Russell H. Morgan Department of Radiology
and Radiological Science
Division of MR Research Institute for Cell Engineering
Johns Hopkins University School of Medicine
Baltimore, MD, USA
NeuroRepair Department
Mossakowski Medical Research Centre
Warsaw, Poland
Department of Neurosurgery
Mossakowski Medical Research Centre
Warsaw, Poland
ISSN 0893-2336 ISSN 1940-6045 (electronic)

Neuromethods
ISBN 978-1-4939-3728-8 ISBN 978-1-4939-3730-1 (eBook)
DOI 10.1007/978-1-4939-3730-1
Library of Congress Control Number: 2016939947
© Springer Science+Business Media New York 2016

This work is subject to copyright. All rights are reserved by the Publisher, whether the whole or part of the material is
concerned, specifically the rights of translation, reprinting, reuse of illustrations, recitation, broadcasting, reproduction
on microfilms or in any other physical way, and transmission or information storage and retrieval, electronic adaptation,
computer software, or by similar or dissimilar methodology now known or hereafter developed.
The use of general descriptive names, registered names, trademarks, service marks, etc. in this publication does not
imply, even in the absence of a specific statement, that such names are exempt from the relevant protective laws and
regulations and therefore free for general use.
The publisher, the authors and the editors are safe to assume that the advice and information in this book are believed to
be true and accurate at the date of publication. Neither the publisher nor the authors or the editors give a warranty,
express or implied, with respect to the material contained herein or for any errors or omissions that may have been made.
Printed on acid-free paper
This Humana Press imprint is published by Springer Nature

The registered company is Springer Science+Business Media LLC New York
Preface to the Series
Experimental life sciences have two basic foundations: concepts and tools. The Neuromethods
series focuses on the tools and techniques unique to the investigation of the nervous system
and excitable cells. It will not, however, shortchange the concept side of things as care has
been taken to integrate these tools within the context of the concepts and questions under
investigation. In this way, the series is unique in that it not only collects protocols but also
includes theoretical background information and critiques which led to the methods and
their development. Thus it gives the reader a better understanding of the origin of the
techniques and their potential future development. The Neuromethods publishing program
strikes a balance between recent and exciting developments like those concerning new ani-
mal models of disease, imaging, in vivo methods, and more established techniques, includ-
ing, for example, immunocytochemistry and electrophysiological technologies. New
trainees in neurosciences still need a sound footing in these older methods in order to apply
a critical approach to their results.
Under the guidance of its founders, Alan Boulton and Glen Baker, the Neuromethods
series has been a success since its first volume published through Humana Press in 1985.
The series continues to flourish through many changes over the years. It is now published
under the umbrella of Springer Protocols. While methods involving brain research have
changed a lot since the series started, the publishing environment and technology have
changed even more radically. Neuromethods has the distinct layout and style of the Springer
Protocols program, designed specifically for readability and ease of reference in a laboratory
setting.
The careful application of methods is potentially the most important step in the process
of scientific inquiry. In the past, new methodologies led the way in developing new disci-
plines in the biological and medical sciences. For example, Physiology emerged out of
Anatomy in the nineteenth century by harnessing new methods based on the newly discov-
ered phenomenon of electricity. Nowadays, the relationships between disciplines and meth-
ods are more complex. Methods are now widely shared between disciplines and research
areas. New developments in electronic publishing make it possible for scientists that
encounter new methods to quickly find sources of information electronically. The design of
individual volumes and chapters in this series takes this new access technology into account.
Springer Protocols makes it possible to download single protocols separately. In addition,
Springer makes its print-on-demand technology available globally. A print copy can there-
fore be acquired quickly and for a competitive price anywhere in the world.
Saskatoon, Canada Wolfgang Walz
v
Preface
In the field of neuroscience, animal surgeries are often performed both to develop animal
models and to test the application of various therapies. Surgery has always been considered
an art and individual variability is high. However, proper surgical preparation is directly
related to the achieved results. While there has been progress in the automation of tissue
sampling and image analysis, surgery is still a manual procedure. Thus, animal surgery is
somewhat of a bottleneck because the great variation in individual skill cannot ensure that
the experimental results will always be of the highest quality. Despite the importance of the
surgical procedure, the technical description in journal articles is usually very brief, which
may also cause difficulties when someone attempts to reproduce the experiment. Moreover,
the technical obstacles that a researcher might encounter, as well as tips and tricks about
how to overcome them, are usually not mentioned in the literature. In addition, surgical
techniques are not sold as kits with instructions included. Thus, a reference book, in which
procedures in the form of a corpus of instructions, is highly desired. Considering that surgi-
cal technique is meticulous and deserves a full explanation of the technical details to per-
form procedures properly, the book Experimental Neurosurgery in Animal Models has been
prepared to address these challenges.
For many years, small-animal models were favored mostly due to low cost, ease of care,
and the possibilities for high throughput. While they are still valuable for answering some
basic research questions, the translation of therapeutic approaches from bench to bed is
usually unsuccessful. Thus, there is a growing awareness that therapies should be tested in
large-animal models prior to clinical application. Although, currently, very few laboratories
perform neurosurgical procedures on large animals, there is a growing interest in using
these animals. Therefore, it would be of great value to have access to the operative expertise
of leaders in the field of large-animal surgery. This book answers that need and also high-
lights the experienced laboratories that could serve as a reference for newcomers. Thus, the
part of the book devoted to large animals is especially compelling and sets the standard for
state-of-the-art translational research. While the initial chapters of the book present the
standard small-animal models now used in neuroscience, these are later followed by a
description of procedures in large-animal models.
While manual precision is equally important in all models, the complexity of surgical
dissection varies. The first six chapters focus primarily on the brain, while the next six chap-
ters concern the spinal cord in rodents. The last four chapters provide a description of
operative procedures in large animals. The book begins with three chapters that describe
rapid procedures that do not always require the use of a scalpel or in which the use of a
scalpel is very limited, but all these chapters are related to the major neurosurgical disci-
plines, such as neurotrauma, radiosurgery, and stereotaxy. The next two chapters describe
the very complicated craniotomies in small animals. The sixth chapter deals with the
advances in the use of robotics, which is expected to have growing role in animal models.
The next two chapters are devoted to the presentation of models of spine injury, and the
following chapter describes microsurgical access to the spinal cord. Chapters 10 and 11
vii
viii Preface
present various methods of injection to the spine and CSF through the cisterna magna.
Chapter 12 focuses on cranial and peripheral nerve dissection using a very advanced water-
jet dissection method.
The large-animal section begins by detailing the performance of a craniotomy in swine.
This procedure is universal and can be used for wide brain access for various purposes, as
well as for surgical training. Thus, this would be of interest not only to researchers but also
to neurosurgical residents and neurosurgeons. The next chapter presents stereotaxy, which
is far more complex in a large-animal setting, but, due to advances in sophisticated technol-
ogy, is a very powerful method for the translation of animal research to the clinical scenario,
particularly as monkeys are often used as the experimental species. The sheep model of
stroke was long-awaited after many unsuccessful attempts to translate the positive small-
animal data to the clinical setting. There is a good possibility that this model will have wide
preclinical utility, particularly in the current climate that mandates that clinical tests be
preceded by relevant animal studies. The last chapter is devoted to the extremely important
neurosurgical disease, subarachnoid hemorrhage. The chapter introduces a reader to the
complexity of pathological sequelae that can be directly related to the neurosurgical tech-
nique and provides both the successes and failures related to the use of various techniques,
which allows researchers to build on the enormous experience of this group in studying this
disease in a primate model.
This book is expected to gather the interest of various readerships. It will be very useful
for basic researchers, who need to establish animal models in the field of neuroscience,
especially in neurosurgery. The vast group of neurosurgical residents can treat it as a reposi-
tory of research and training opportunities for the use of animal models. The book may also
facilitate the selection of an appropriate animal model as well as serve as a basis for further
technical improvements and refinements of these models for academic neurosurgeons run-
ning their own labs simultaneously with clinical practice. Thus, it is expected that the book
will be warmly received and will serve frequently as a handbook during the planning and
performance of surgical procedures on the central nervous system.
Baltimore, MD, USA Miroslaw Janowski

Contents
Preface to the Series. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . v

Preface. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . vii
Contributors. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . xi
1 Animal Models of Traumatic Brain Injury . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1

Fredrik Clausen
2 Experimental Radiosurgery in Animal Models . . . . . . . . . . . . . . . . . . . . . . . . . 13
Ajay Niranjan, Wendy Fellows-Mayle, Douglas Kondziolka,
and L. Dade Lunsford
3 Stereotactic Surgery in Rats. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 31
Jaroslaw Maciaczyk, Ulf D. Kahlert, Máté Döbrössy,
and Guido Nikkhah
4 Rat Middle Cerebral Artery (MCA) Occlusion Models
Which Involve a Frontotemporal Craniectomy . . . . . . . . . . . . . . . . . . . . . . . . 55
Hideaki Imai, Nobuhito Saito, and I. Mhairi Macrae
5 Inferior Colliculus Approach in a Rat . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 73
Dennis T.T. Plachta
6 Why Robots Entered Neurosurgery . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 85
Jason W. Motkoski and Garnette R. Sutherland
7 Impact Model of Spinal Cord Injury . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 107
Dorothée Cantinieaux, Rachelle Franzen, and Jean Schoenen
8 Acute Clip Compression Model of SCI . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 111
Jared T. Wilcox and Michael G. Fehlings
9 Microsurgical Approach to Spinal Canal in Rats. . . . . . . . . . . . . . . . . . . . . . . . 119
Mortimer Gierthmuehlen and Jan Kaminsky
10 Stereotaxic Injection into the Rat Spinal Cord. . . . . . . . . . . . . . . . . . . . . . . . . 133
Charla C. Engels and Piotr Walczak
11 Surgical Access to Cisterna Magna Using Concorde-Like Position
for Cell Transplantation in Mice and CNS Dissection within Intact
Dura for Evaluation of Cell Distribution . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 141
Miroslaw Janowski
12 Animal Models for Experimental Neurosurgery of Peripheral
and Cranial Nerves . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 151
Joachim Oertel, Christoph A. Tschan, and Doerther Keiner
13 Surgery of the Brain and Spinal Cord in a Porcine Model . . . . . . . . . . . . . . . . 165
Jan Regelsberger
ix
x Contents
14 Real-Time Convection Delivery of Therapeutics to the Primate Brain . . . . . . . 175

Dali Yin, Massimo S. Fiandaca, John Forsayeth,
and Krystof S. Bankiewicz
15 Focal Cerebral Ischemia by Permanent Middle Cerebral Artery
Occlusion in Sheep: Surgical Technique, Clinical Imaging,
and Histopathological Results . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 195
Björn Nitzsche, Henryk Barthel, Donald Lobsien, Johannes Boltze,
Vilia Zeisig, and Antje Y. Dreyer
16 A Nonhuman Primate Model of Delayed Cerebral Vasospasm
After Aneurismal Subarachnoid Hemorrhage. . . . . . . . . . . . . . . . . . . . . . . . . . 227
Ryszard M. Pluta, John Bacher, Boris Skopets,
and Victoria Hoffmann
Index . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 251
Contributors
JOHN BACHER • Division of Veterinary Resources, Office of Research Services,

National Institutes of Health, Bethesda, MD, USA
KRYSTOF S. BANKIEWICZ • Department of Neurosurgery, University of California
San Francisco, San Francisco, CA, USA
HENRYK BARTHEL • Department of Nuclear Medicine, University of Leipzig, Leipzig,
Germany
JOHANNES BOLTZE • Department of Medical Cell Technology, Fraunhofer Research
Institution for Marine Biotechnology, Lubeck, Germany; Neuroscience Center,
Massachusetts General Hospital and Harvard Medical School, Charlestown, MA, USA
DOROTHÉE CANTINIEAUX • GIGA-Neuroscience, University of Liège, Liège, Belgium
FREDRIK CLAUSEN • Section of Neurosurgery, Department of Neuroscience,
Uppsala University, Uppsala, Sweden
MÁTÉ DÖBRÖSSY • Department of Stereotactic and Functional Neurosurgery,
University of Freiburg Medical Center – Neurocentre, Freiburg, Germany
ANTJE Y. DREYER • Department of Cell Therapy, Fraunhofer Institute for Cell Therapy
and Immunology, Leipzig, Germany
CHARLA C. ENGELS • Division of MR Research, Russell H. Morgan Department of
Radiology and Radiological Science, Cellular Imaging Section, Institute for Cell
Engineering, The Johns Hopkins University School of Medicine, Baltimore, MD, USA;
Department of Radiology, Faculty of Medical Sciences, University of Warmia and
Mazury, Olsztyn, Poland
MICHAEL G. FEHLINGS • Spine Program, University Health Network, Toronto Western
Hospital, University of Toronto, Toronto, ON, Canada
WENDY FELLOWS-MAYLE • The Department of Neurological Surgery, The University of
Pittsburgh, Pittsburgh, PA, USA; The Center for Image-Guided Neurosurgery, University
of Pittsburgh Medical Center, Pittsburgh, PA, USA
MASSIMO S. FIANDACA • Department of Neurosurgery, University of California
San Francisco, San Francisco, CA, USA
JOHN FORSAYETH • Department of Neurosurgery, University of California San Francisco,
San Francisco, CA, USA
RACHELLE FRANZEN • GIGA-Neuroscience, University of Liège, Liège, Belgium
MORTIMER GIERTHMUEHLEN • Department of Neurosurgery, University of Freiburg,
Freiburg, Germany
VICTORIA HOFFMANN • Division of Veterinary Resources, Office of Research Services,
HIDEAKI IMAI • Department of Neurosurgery, Faculty of Medicine, The University of Tokyo,
Tokyo, Japan
xi
xii Contributors
MIROSLAW JANOWSKI • The Russell H. Morgan Department of Radiology and Radiological

Science, Division of MR Research Institute for Cell Engineering, Johns Hopkins
University School of Medicine, Baltimore, MD, USA; NeuroRepair Department,
Mossakowski Medical Research Centre, Warsaw, Poland; Department of Neurosurgery,
Mossakowski Medical Research Centre, Warsaw, Poland
ULF D. KAHLERT • Department of Neurosurgery, Heinrich-Heine University Duesseldorf,
Duesseldorf, Germany
JAN KAMINSKY • Department of Neurosurgery, University of Freiburg, Freiburg, Germany
DOERTHER KEINER • Neurochirurgische Klinik, Universitaetsklinikum des Saarlandes
Homburg/Saar, Homburg/Saar, Germany
DOUGLAS KONDZIOLKA • The Department of Neurological Surgery, The University of
Pittsburgh, Pittsburgh, PA, USA; The Center for Image-Guided Neurosurgery,
University of Pittsburgh Medical Center, Pittsburgh, PA, USA
DONALD LOBSIEN • Department of Neuroradiology, University of Leipzig, Leipzig,
Germany
L. DADE LUNSFORD • Department of Neurological Surgery, The University of Pittsburgh,
Pittsburgh, PA, USA; The Center for Image-Guided Neurosurgery, University of
Pittsburgh Medical Center, Pittsburgh, PA, USA
JAROSLAW MACIACZYK • Department of Neurosurgery, Heinrich-Heine University
Duesseldorf, Duesseldorf, Germany
I. MHAIRI MACRAE • Institute of Neuroscience and Psychology, College of Medical Veterinary
and Life Sciences, University of Glasgow, Glasgow, Scotland, UK
JASON W. MOTKOSKI • Division of Neurosurgery, Seaman Family MR Research Centre,
Foothills Medical Centre, Calgary, AB, Canada
GUIDO NIKKHAH • Department of Stereotactic and Functional Neurosurgery, University of
Freiburg Medical Center – Neurocentre, Freiburg, Germany
AJAY NIRANJAN • The Department of Neurological Surgery, The University of Pittsburgh,
Pittsburgh, PA, USA; The Center for Image-Guided Neurosurgery, University of
Pittsburgh Medical Center, Pittsburgh, PA, USA
BJÖRN NITZSCHE • Department of Cell Therapy, Fraunhofer Institute for Cell Therapy and
Immunology, Leipzig, Germany; Department of Nuclear Medicine, University of Leipzig,
Leipzig, Germany
JOACHIM OERTEL • Neurochirurgische Klinik, Universitaetsklinikum des Saarlandes
Homburg/Saar, Homburg/Saar, Germany
DENNIS T.T. PLACHTA • Laboratory for Biomedical Microtechnology, Department of
Microsystems Engineering, University of Freiburg – IMTEK, Freiburg, Germany
RYSZARD M. PLUTA • Surgical Neurology Branch, National Institute of Neurological
Disorders and Stroke, National Institutes of Health, Bethesda, MD, USA;
Fishbein Fellow, JAMA, Chicago, IL, USA
JAN REGELSBERGER • Department of Neurosurgery, University Medical Center Hamburg
Eppendorf, Hamburg, Germany
NOBUHITO SAITO • Department of Neurosurgery, Faculty of Medicine, The University of
Tokyo, Tokyo, Japan
JEAN SCHOENEN • GIGA-Neuroscience, University of Liège, Liège, Belgium
BORIS SKOPETS • Division of Veterinary Resources, Office of Research Services,
Contributors xiii
GARNETTE R. SUTHERLAND • Division of Neurosurgery, Seaman Family MR Research

Centre, Foothills Medical Centre, Calgary, AB, Canada; Department of Clinical
Neurosciences, University of Calgary, AB, Canada
CHRISTOPH A. TSCHAN • Klinik für Neurochirurgie, Ludmillenstift Meppenr, Germany
PIOTR WALCZAK • Division of MR Research, Russell H. Morgan Department of Radiology
and Radiological Science, Cellular Imaging Section, Institute for Cell Engineering, The
Johns Hopkins University School of Medicine, Baltimore, MD, USA
JARED T. WILCOX • Spine Program, University Health Network, Toronto Western Hospital,
University of Toronto, Toronto, ON, Canada
DALI YIN • Department of Neurosurgery, University of California San Francisco,
San Francisco, CA, USA
VILIA ZEISIG • Department of Nuclear Medicine, University of Leipzig, Leipzig, Germany
Chapter 1
Animal Models of Traumatic Brain Injury

Fredrik Clausen
Abstract
Animal models of traumatic brain injury (TBI) have been the core of the research on the molecular, cellular,
and functional effects of the disease. To be able to simulate the heterogeneous aspects of TBI several
models have been designed. This chapter aims to describe the three most commonly used experimental
models of TBI in rodents.
Key words Traumatic brain injury, Mice, Rats, Controlled cortical impact, Fluid percussion injury,
Weight drop injury, Stereotaxy
1 Introduction
1.1 The Need Basic traumatic brain injury (TBI) research is reliant on animal
for Animal Models models to study the multitude of effects the event has on the brain.
in Basic Research The brain is simply too complex to simulate in vitro or in silico.
on Traumatic Brain Add to this, the interaction with the blood stream and immune
Injury system and the complexity grows another magnitude.
The first animal models of TBI were setup in larger species such
as dogs and cats, though presently most experiments are done in
rodents. Porcine models of TBI have become more common and
they are more clinically relevant as the pig brain is closer to the
human in regard to size and anatomy, but the differences in cost and
effort between using pigs compared to rodents are substantial.
Most studies are done in young male adult rats and mice,
which actually is quite relevant as young males are over represented
in TBI, as they are more prone to experience vehicle accidents,
sports injuries, and violence. Interestingly, also the elderly suffer a
higher risk of TBI due to falls, though very few studies in aged
rodents have been performed. One large difference between
human TBI and most animal models is that while human TBI is
heterogenous, experimental TBI is most often homogenous. In
human patients there are often secondary complications such as
multitrauma, intoxication, and preexisting disease.
Miroslaw Janowski (ed.), Experimental Neurosurgery in Animal Models, Neuromethods, vol. 116,
DOI 10.1007/978-1-4939-3730-1_1, © Springer Science+Business Media New York 2016
1
2 Fredrik Clausen
Animal welfare is a big issue in several countries and it is of

course recommended that the experimenters abide to the current
laws and regulation regarding animal experiments. It should be
noted that even though experimental TBI may sound very brutal,
even in rodent models at a severe setting most animals recuperate
quickly. The loss of motor function and cognitive acuity is hard to
spot in a standard animal cage and requires special tests to assess
properly.
1.2 Most Commonly Controlled cortical impact (CCI) was developed by Dixon et al.
Used Models of TBI [1] to use as a rat model of TBI and was later adapted to be used
in mice [2]. It is one of the most widely used TBI models. The model
1.2.1 Controlled
relies on a pneumatically driven piston to compress the exposed
Cortical Impact
brain, and the severity of the injury is determined by the depth of
compression and speed of the piston (Fig. 1). This results in a pre-
dominantly focal contusion at the site of impact, but have effects
throughout the brain and gives rise to contralateral changes in the
hippocampus and thalamus.
There are several manufacturers of CCI devices, but the origi-
nal model made at Virginia Commonwealth University (Richmond,
Virginia) is still in production by Amscien Inc [3].
Fig. 1 Schematic drawing of the controlled cortical injury device. Piston (1), micrometer gauge (2), oscilloscope
(3), pressure hoses (4), control unit (5), signal conditioner (6), automatic average rod speed measurement unit
(7), and computer (8). The stereotactic frame is placed in the CCI device. The piston (1) is extended to its maxi-
mum and placed perpendicularly against the exposed dura mater. The piston is then retracted and lowered to
the desired compression depth, measured by the micrometer gauge (2). The piston is firmly locked in position
and the trauma induced. The acceleration of the piston is registered by the oscilloscope (3), amplified by the
signal conditioner (6), measured by the rod speed measurement unit (7), and the recorded by the computer (8)
Models of Brain Trauma 3
Fig. 2 Schematic drawing of the fluid percussion injury device. Pendulum with hammer head (1), piston (2),
fluid-filled cylinder (3), pressure transducer (4), nozzle (5), pressure monitor (6), and computer (7). The injury
is induced by attaching the animal to the nozzle (5) and releasing the pendulum (1) from the desired height.
The hammer on the pendulum will hit the piston (2) that causes a rise in pressure in the cylinder (3) which
sends a pressure pulse into the skull of the animal. The pressure is registered by the pressure transducer (4),
measured by the pressure monitor (6), and recorded on the computer (7)
1.2.2 Fluid Fluid percussion injury (FPI) was originally developed in cats [4]
Percussion Injury and was later adapted to be used in rat [5, 6] and mice [7]. The
model is based on the rapid injection of a fluid pulse on the exposed
dura mater of the brain with a subsequent displacement of the
brain. The fluid pulse is generated as a hammer head mounted on
a pendulum pushes a piston into a fluid-filled cylinder and is
directed into a nozzle coupled either directly or via a hose to the
skull cavity (Fig. 2).
The injury severity depends on the pressure of the fluid
pulse, which can be changed by adjusting the height of the fall
of the hammer head. The pressure is recorded close to the end
of the nozzle spout and has been found to correlate well to the
pressure inside the skull in rats [8]. The position of the craniot-
omy and coupling to the FPI device majorly affects the outcome
[9]. To that end two variants of FPI are referred to. In lateral
FPI (LFPI), the injury is centered over the parietal cortex, resulting
in a more focal injury than central FPI (CFPI) that is centered
over the midline.
FPI results in more global brain effects than CCI and if set at
moderate or severe injury it elicits apnea and can shock the brain
stem fatally. In the most severe injury settings of LFPI, it is expected
that around 25 % of the animals do not survive the trauma. CFPI
has an even greater global effect and is in general used with lower
pressure than LFPI as it easily shocks the brain stem causing
4 Fredrik Clausen
extended apnea or death. CFPI is used when a more diffuse axonal

injury (DAI) is wanted without a defined contusion.
There are currently two commercially available FPI devices
that differ in the design. The first model is produced by Amscien
[3] and has a continuous water pillar through the device into the
brain allowing the pressure wave to travel in a straight line into the
skull cavity. Dragonfly R&D Inc. [10] manufactures a model with
a flexible hose attached to the nozzle, making it possible to deliver
the pressure pulse into the skull cavity with the rodent in the ste-
reotaxic frame. Frey et al. have developed a novel design where the
pendulum and piston are replaced by a picospritzer [11], though
this model is not yet commercially available.
1.2.3 Weight Drop Injury Weight drop injury (WDI), also called closed head injury (CHI),
differs from CCI and FPI as it is an acceleration/deceleration
model of TBI. Out of the three TBI models described in this chap-
ter it is the simplest equipment wise. The basic concept with a
weight falling through a guide tube, hitting the head of the rodent,
and accelerating it into a foam bead underneath the animal (Fig. 3)
was first described by Shapira et al. [12]. The model can be used
with or without protecting the skull bone. Foda and Marmarou
developed the model by attaching a steel disk to the skull of the
rodent to reduce fracturing of the skull [13, 14].
Fig. 3 Schematic drawing of the weight drop injury device. Weight (1), guide tube
(2), foam bed (3), and impact (4). The weight (1) is released into the guide tube
(2) and hits the head of the animal. The impact accelerates the head of the ani-
mal into the foam bed (3)
This is the most heterogeneous model of the three discussed in

this chapter, especially if there are skull fractures involved. However,
this is not necessarily a bad thing, as human TBI is very heteroge-
neous. The severity of the injury is determined by the size of the
weight, the distance it falls, and the flexibility of the material under-
neath the animal. WDI causes a predominantly diffuse injury, but
many studies also show visible lesions and cavities.
1.3 General To be able to perform the craniotomy and other procedures on

Considerations the rodent skull a stereotaxic frame is needed to secure the head of
the animal properly. There are several manufacturers of stereotaxic
1.3.1 Stereotaxy
frames for rodents, but it is advisable to consider the needs for
the laboratory when it comes to species as some frames are more
easily converted from use for rats to mice. A need for microma-
nipulators attached to the frame can also influence the choice of
manufacturer.
1.3.2 Craniotomy CCI and FPI both require that a craniotomy is made to deliver the
force that causes the trauma. In CCI, the craniotomy is typically
made slightly larger than the tip of the piston used. For FPI, the
craniotomy is preferably performed with a trephine of a size that
offers the best possible fit for the coupling to the device.
The placement of the craniotomy, and subsequently the
place of the injury, is of great importance. Studies have shown
that moving the craniotomy for FPI results in different outcome
in regards to lesion size and functional deficits [9]. It is also one
of the largest factors when it comes to inter-operator and inter-
laboratory differences.
After the injury is made, the bone piece removed during the
craniotomy is replaced. To more accurately model a CHI, the bone
piece can be fastened using tissue glue and/or bone cement. This
will result in a larger injury as the contused brain won’t swell out
of the cavity and a higher intracranial pressure (i.c.p.) will be
higher, causing a reduction in blood flow to the injured area [15].
1.3.3 Anesthesia There are many different ways to keep the rodent sedated and
and Temperature anesthetized during surgery, and the rules and regulation varies
between different countries as to what constitutes satisfactory
anesthesia in small animals. All forms of anesthesia have strengths
and weaknesses.
Gas anesthesia with isoflurane with nitrous oxide and oxygen
has been shown to be neuroprotective in itself [16], which can
mask smaller treatment effects. Though isoflurane so far is deemed
harmless to humans, proper ventilation of the operating table
should be uses to protect the experimenter. Halothane should not
be used as it is carcinogenous to humans.
Pentobarbital and chloral hydrate offer adequate sedation, but
does not offer pain relief making them illegal to use in some
6 Fredrik Clausen
countries due to animal welfare concerns. Interestingly, pentobar-

bital decreases the metabolism of the brain, whereas chloral hydrate
increases it [17].
Hypnorm/dormicum offers both sedation and pain relief but
needs frequent injections to keep the animal sedated.
Local anesthesia can be injected into the scalp before it is
opened to reduce the discomfort and stress in the animal. It is rec-
ommended to use local anesthesia in any other surgical wound
made during the operation to diminish the postoperative pain as
much as possible.
In most experimental setups, normothermia is desired and
since several forms of anesthesia decreases metabolism and body
temperature a way to keep the animal heated is necessary. This can
be achieved by a heating pad and/or a heating lamp. For longer
experiments a combination of the two is recommended. As cere-
bral hypothermia lowers the metabolism of the brain and is consid-
ered neuroprotective, it is useful to monitor the brain temperature
as well as the core body temperature and keep it above
36.5 °C. There are thin probes available that can be placed between
the brain and the skull bone or alternatively between the temporal
muscle and the skull bone if there is no room for the probe in the
craniotomy.
As the animal is unconscious from the anesthesia and trauma,
after the surgery is finished it is essential to setup a heated recovery
cage where the animal can regain consciousness in a warm and
calm environment to reduce stress. It is necessary to heat the cage
with a heating lamp to avoid postoperative hypothermia that could
influence the outcome.
1.3.4 Physiological In shorter experiments (less than 30 min anesthesia), physiological

Monitoring monitoring could be restricted to core temperature or blood pres-
sure if there is noninvasive measuring available.
In longer experiments (i.e., microdialysis, imaging, and com-
plicated treatments), it is advisable to monitor blood pressure,
blood gases, and brain temperature to make sure that the animal is
within normal physiological parameters (or nonphysiological if
that is part of the design of the experiment).
1.3.5 Selecting Naturally the primary concern when choosing TBI model is which
a TBI Model outcome measures that is being studied. If DAI is to be studied,
the choice is between WDI and MFPI. WDI is easier and less costly
to setup and the experimental procedure is quicker than
MFPI. However, the inherent heterogeneity of the model makes it
necessary to do larger groups of animals. If a more focal injury is of
interest then the choice is between CCI and LFPI. Once again,
FPI is the more labor intensive and costly method, but if the labo-
ratory also is interested in DAI the same FPI device can be used for
both applications.
2 Materials
2.1 Controlled Surgical tools: Dumont forceps, delicate scissors, flat small forceps
Cortical Impact (8–9 cm), hemostatic forceps (10–12 cm), scalpel, dental drill or
trephine, stereotaxic frame (some models of CCI come with a basic
2.1.1 Materials Needed
stereotaxic setup, but it is recommended to use a free-standing
for CCI
stereotaxic frame, especially if you do other surgical procedures on
the rat, i.e., micro dialysis or stereotaxic injections), sutures, and
needle. Tissue adhesive to reattach the bone piece if so desired.
2.2 Fluid Surgical tools: Dumont forceps, flat small forceps (8–9 cm), hemo-
Percussion Injury static forceps (10–12 cm), scalpel, dental drill or trephine (recom-
mended), stereotaxic frame, sutures, and needle.
2.2.1 Materials Needed
Trauma coupling: Luer lok injection needle adapted to fit the
for FPI
craniotomy (note that the needle part is entirely removed), tissue
adhesive, bone/dental cement, 2 mm screw and appropriate screw-
driver, and 2 mm drill.
2.3 Weight The equipment is relatively easy to make in house as all that is
Drop Injury needed is an appropriate weight, a guide tube that fits the weight
and a flexible material to rest the animal on.
2.3.1 Materials Surgery (if a protection is attached to the skull bone): scalpel, for-
ceps, hemostatic forceps, protective disk, tissue adhesive, and ste-
reotaxic frame.
3 Method
3.1 Controlled 1. Sedate the animal and attach it to the stereotaxic frame.
Cortical Impact 2. Trim the fur on the head if wanted.
3. Inject local anesthesia under the scalp and open up the scalp
along the midline using a scalpel or scissors. Retract the skin
and expose the skull bone. Keep the scalp retracted using
hemostatic forceps.
4. Use Dumont forceps to clear the skull bone from periost. If
necessary retract the muscle lateral to the lateral ridges to
achieve more space for the craniotomy.
5. Use the midline and bregma sutures on the skull bone to posi-
tion the craniotomy (Fig. 4).
6. Use a dental drill or trephine to perform the craniotomy
without causing a rift to the dura mater. Remove the bone
piece and place it in sterile, isotonic saline if it is to be
replaced later (Fig. 5).
7. Move the sterotaxic frame to the CCI device.
8 Fredrik Clausen
Fig. 4 Schematic drawing of the exposed rodent skull
Fig. 5 Most common placement of the craniotomy for FPI or CCI
8. Find the null position and retract the piston. Lower the piston
the desired distance. Perform the CCI and time the length of
apnea (mostly applicable in mice).
9. Move the stereotaxic frame back to the operating table.
10. Replace the bone piece and use tissue adhesive to secure it in
its former place.
11. Suture the scalp and move the animal to a recovery cage.
3.2 Fluid 1. Sedate the animal and place it in the sterotaxic frame.
Percussion Injury 2. Trim the fur on the head and inject the scalp with local anes-
thesia. Open the scalp along the midline and retract the skin to
expose the skull bone. Keep the scalp retracted using hemo-
static forceps.
3. Use Dumont forceps to clean the skull from connective tissue
and periosteum.
Fig. 6 Possible locations for the anchoring screws
4. Use the midline and bregma sutures as reference points to

place the craniotomy in the desired place. Make the craniot-
omy using a trephine, be sure not to rupture the dura mater
(Fig. 5).
5. Use a drill to make a hole for the anchor screw, without going
all the way through the bone. Attach the screw to the skull
without penetrating the skull bone (Fig. 6).
6. Place the trauma coupling in the craniotomy and secure it with
tissue adhesive.
7. Pour dental cement around the trauma coupling. Make sure to
include the anchor screw in the cast. Allow the dental cement
to set properly.
8. Wean the animal off anesthesia enough for it to regain the
pinch reflex in the paws.
9. Attach the animal to the FPI device and perform the injury.
10. Time the length of apnea if at a moderate or severe setting.
11. Return the animal to the heated pad and measure the time till
it regains the pinch reflex in the contralateral paws.
12. Replace the bone piece and attach it with tissue adhesive and
suture the scalp.
13. Move the animal to a recovery cage when the pinch reflex has
returned.
3.3 Weight 1. Sedate the animal.

Drop Injury 2. If a protective disk is used, place the animal in a stereotaxic
frame.
3. Inject local anesthesia under the skin and expose the skull bone
by opening the scalp along the midline, retract the skin to
expose the skull bone. Keep the scalp retracted using hemo-
static forceps.
10 Fredrik Clausen
4. Remove connective tissue and periost from the skull bone.

5. Make sure the bone is dry before applying the tissue adhesive
to attach the disk in the preferred position, use the midline and
bregma sutures as reference points.
6. Place the animal on the flexible bed and position the guide
tube so the weight will hit the disk or the head of the animal
according to the decided protocol.
7. Release the weight from the decided height.
8. Move the rat immediately after the weight hits to avoid a sec-
ond impact as the weight rebounds.
9. Remove the disk and suture the scalp.
10. Move the animal to a recovery cage.
4 Notes
4.1 Controlled Firstly, it is important to remember that even though most experi-
Cortical Impact ments are made in age matched, inbred male subjects, these are
biological experiments and there may be differences between
individuals, both physiological and behavioral.
Aside from individual differences in the lab animals, there are
two factors that can cause a great variability between operators and
laboratories in CCI. The first is the placement of the trauma, a few
mm difference in where the piston strikes can cause a different
outcome. The second is the null position before the impact. This
can be made depressing the cortex slightly or just touching it with
the piston. The difference between those two positions is about
0.5 mm, which is the difference between a moderate and severe
injury both in mice and rats.
Another possible source of variability is the angle of the piston
to the exposed brain. It should be perpendicular and the angle is
easily adjusted on the CCI device if needed.
If a severe injury is desired, a rift on the dura mater is expected
after impact, whereas on moderate or mild setting ruptures should
be avoided.
4.2 Fluid The FPI device itself can cause variability if there are air bubbles
Percussion Injury present in the cylinder or nozzle. Before starting the experiment
a few test hits should be made and the pressure curve checked for
irregularities. If the pressure curve is full of spikes it is a sign of air
bubbles in the system and they should be removed before start-
ing the experiment. Make sure that the pressure peak is at the
desired value.
When applying tissue adhesive to the trauma coupling after
attaching it to the craniotomy it is possible for the tissue adhesive
to leak onto the brain if the trauma coupling does not fit well. The
trauma coupling then has to be removed and cleaned or discarded,

since the tissue adhesive will set on the brain surface and decrease
the pressure of the trauma. When doing FPI in mice, the tissue
adhesive can attach hard enough to the dura mater to cause a rift
when the trauma coupling is removed. In mice, it may be advisable
to use a very fast acting adhesive on just one or two points of
the trauma coupling to avoid this problem.
4.3 Weight To make the experiments as reproducible as possible it is important

Drop Injury that the weight hits the animals as similarly as possible. This can to
some extent be achieved by forming the flexible bed so that the
animals rests in the same position. Keeping the guide tube properly
lubricated to allow the weight to accelerate as similarly as possible
from experiment to experiment is also of importance. If a protec-
tive disk is used it should be mounted as similar as possible in all
animals. To avoid a second impact from the rebound of the weight
the animal can be moved or a catch for the weight built into the
device.
References
1. Dixon CE, Clifton GL, Lighthall JW, Yaghmai injury in the mouse. Acta Neuropathol
AA, Hayes RL (1991) A controlled cortical 98:396–406
impact model of traumatic brain injury in the 8. Clausen F, Hillered L (2005) Intracranial pres-
rat. J Neurosci Methods 39:253–262 sure changes during fluid percussion, con-
2. Smith DH, Soares HD, Pierce JS, Perlman trolled cortical impact and weight drop injury
KG, Saatman KE, Meaney DF et al (1995) A in rats. Acta Neurochir (Wien) 147:775–780
model of parasagittal controlled cortical impact 9. Floyd CL, Golden KM, Black RT, Hamm RJ,
in the mouse: cognitive and histopathologic Lyeth BG (2002) Craniectomy position affects
effects. J Neurotrauma 12:169–178 morris water maze performance and hippo-
3. www.amscien.com (2004) AmScien Instruments, campal cell loss after parasagittal fluid percus-
Richmond, VA sion. J Neurotrauma 19:303–316
4. Stalhammar D, Galinat BJ, Allen AM, Becker 10. www.dragonflyinc.com (2010) Dragonfly
DP, Stonnington HH, Hayes RL (1987) A Research & Development Incorporated,
new model of concussive brain injury in the cat Ridgeley, WV
produced by extradural fluid volume loading: 11. Frey LC, Hellier J, Unkart C, Lepkin A,
I. Biomechanical properties. Brain Inj Howard A, Hasebroock K et al (2009) A novel
1:73–91 apparatus for lateral fluid percussion injury in
5. Dixon CE, Lyeth BG, Povlishock JT, Findling the rat. J Neurosci Methods 177:267–272
RL, Hamm RJ, Marmarou A et al (1987) A 12. Shapira Y, Shohami E, Sidi A, Soffer D,
fluid percussion model of experimental brain Freeman S, Cotev S (1988) Experimental
injury in the rat. J Neurosurg 67:110–119 closed head injury in rats: mechanical, patho-
6. McIntosh TK, Vink R, Noble L, Yamakami I, physiologic, and neurologic properties. Crit
Fernyak S, Soares H et al (1989) Traumatic Care Med 16:258–265
brain injury in the rat: characterization of a lat- 13. Foda MA, Marmarou A (1994) A new model
eral fluid-percussion model. Neuroscience of diffuse brain injury in rats. Part II:
28:233–244 Morphological characterization. J Neurosurg
7. Carbonell WS, Grady MS (1999) Regional 80:301–313
and temporal characterization of neuronal, 14. Marmarou A, Foda MA, van den Brink W,
glial, and axonal response after traumatic brain Campbell J, Kita H, Demetriadou K (1994)
12 Fredrik Clausen
A new model of diffuse brain injury in rats. 16. Goren S, Kahveci N, Alkan T, Goren B, Korfali
Part I: Pathophysiology and biomechanics. E (2001) The effects of sevoflurane and isoflu-
J Neurosurg 80:291–300 rane on intracranial pressure and cerebral per-
15. Zweckberger K, Eros C, Zimmermann R, Kim fusion pressure after diffuse brain injury in rats.
SW, Engel D, Plesnila N (2006) Effect of early J Neurosurg Anesthesiol 13:113–119
and delayed decompressive craniectomy on 17. Uematsu M, Takasawa M, Hosoi R, Inoue O
secondary brain damage after controlled corti- (2009) Uncoupling of flow and metabolism by
cal impact in mice. J Neurotrauma 23: chloral hydrate: a rat in-vivo autoradiographic
1083–1093 study. Neuroreport 20:219–222
Chapter 2
Experimental Radiosurgery in Animal Models

Ajay Niranjan, Wendy Fellows-Mayle, Douglas Kondziolka,
and L. Dade Lunsford
Abstract
Lars Leksell described stereotactic radiosurgery as a method to destroy intracranial targets using a single,
high dose of focused, ionizing radiation administered using stereotactic guidance. Radiosurgery is an
impressive blend of minimally invasive technologies guided by a multidisciplinary team of surgeons, oncol-
ogists, medical physicists, and engineers. The long-term results of radiosurgery are now available and have
established it as an effective noninvasive management modality for intracranial vascular malformations and
many tumors. A variety of experimental models have been used to study the effect of radiosurgery in brain.
The results of experimental radiosurgery have enhanced our understanding of the biological impact of
radiosurgery on different tissues. Additional applications of radiosurgery in the management of malignant
tumors and functional disorders are being assessed.
Key words Experimental, Animal models, Epilepsy, Radiosurgery, Functional disorders, Tumors,
Vascular malformations
1 Introduction
Radiosurgery is a surgical technique which is designed to produce

a specific radiobiological effect within a sharply defined target vol-
ume using a single, high dose of focused, ionizing radiation admin-
istered using stereotactic guidance. Normal tissue effects are
limited by the highly focused nature of the radiosurgical beams.
Whereas conventional fractionated radiotherapy is generally most
effective in killing rapidly dividing cells, radiosurgery induces bio-
logical responses irrespective of the mitotic activity, oxygenation,
and inherent radiosensitivity of target cells. The field of stereotactic
radiosurgery represents one of the fundamental shifts in the neuro-
logical surgery over the last two decades. Compared to conven-
tional neurosurgery techniques, stereotactic radiosurgery is
minimally invasive and relies on biological response of tissues in
order to eradicate or inactivate them. Considering the unique
biological response of tissues to radiosurgery, it is important to
13
14 Ajay Niranjan et al.
study the biological effects of radiosurgery in both normal and

pathological nervous system tissues in animal models. Information
gained from radiosurgical research studies is useful in devising
future strategies to prevent damage to normal tissue without com-
promising treatment efficacy.
1.1 History Initial radiosurgery experiments were performed using the rabbit
of Experimental and goat central nervous system (CNS) models. These experiments
Radiosurgery Using were designed to investigate the use of focused radiation as neuro-
Animal Models surgical tool. The early histological results (third to eighth day)
using proton beam radiation on rabbit spinal cord tissue showed
complete transection of the spinal cord with 400 Gy using 1.5 mm
beam diameter and with 200 Gy using 10 mm beam diameter [1].
The goat brain model was also used to document sharply defined
lesions in deep parts of the brain, 4–7 weeks after 200 Gy of stereo-
tactic multiple port proton beam radiation. Rexed et al. preformed
proton beam radiosurgery on rabbit brain with 200 Gy using a
1.5 mm collimator [2]. These investigators documented a well-
demarcated lesion at the target. In a similar model, Leksell et al.
used cross-fired irradiation with a narrow beam of high-energy to
create well-circumscribed intracerebral lesions of appropriate size
and shape [2]. Andersson et al. in a study of long-term effect of
proton radiosurgery on goat brain documented that there were no
adverse effects in or around the lesion after 1.5–4 years after
200 Gy radiosurgery [3]. Nilsson et al. irradiated (100–300 Gy)
the basilar artery of cats by stereotactic technique using 179-source
cobalt-60 prototype gamma unit [4]. Histology demonstrated
endothelial wall injury, and hyalinization and necrosis of the mus-
cular layers. These investigations demonstrated that radiosurgery
could be potentially used to create sharply defined lesions in deep
parts of the brain.
1.2 Radiosurgery Lunsford et al. [5] and Kondziolka et al. [6] studied the radiobio-
as a Lesioning logical effects of stereotactic radiosurgery using a baboon model.
Technique A central dose of 150 Gy using an 8 mm collimator was delivered
to the caudate, thalamus, or pons regions using the gamma knife.
No imaging changes were noted at 4 weeks after irradiation. MR
imaging documented a circumscribed contrast enhanced lesion by
6–8 weeks and frank necrosis at the irradiated target by 24 weeks.
Kondziolka et al. [7] irradiated the frontal lobe of rats with
maximal doses of 30–200 Gy using a 4 mm collimator and studied
histologic changes 90 days after radiosurgery. While detectable his-
tologic alterations were noted with doses of more than 70 Gy,
necrosis was seen only in tissues irradiated with more than 100 Gy.
Blatt et al. [8] evaluated serial tissue changes after 125 Gy using
linear accelerator (LINAC) radiosurgery of internal capsule of cats.
Serial imaging and histopathological evaluations showed tissue
necrosis accompanied by vascular proliferation and edema by
Radiosurgery in Animals 15
3.5 weeks. Kamiryo et al. studied the radiosensitivity of brain by

evaluating the effects of radiosurgery dose and time rat brain. The
rat brain was irradiated using the maximum doses of 50, 75, or
120 Gy and analyzed for histologic changes and blood–brain bar-
rier integrity up to 12 months [9]. Whereas tissue irradiated with
higher doses (120 Gy) showed alterations in astrocytic morphol-
ogy by 3 days, such changes were not observed until 3 months
with lower doses (50 Gy). Blood–brain barrier breakdown was
noted within 3 weeks of 120 Gy irradiation but was not seen even
up to 12 months after 50 Gy. These findings suggested that the
latent period between irradiation and detection of pathologic alter-
ations was dependent on both the dose and the biological end
point used. The impact of dose and biological end points on latency
was also reported by Karger et al. [10], who evaluated the rat brain
using MRI at 15, 17, or 20 months after treatment with 26–50 Gy
of LINAC-based radiosurgery. No radiation-induced effect on
MRI was noted at any time point for doses less than 30 Gy. After
40 Gy radiosurgery, the latency of detectable MRI changes was
19–20 weeks, whereas the latency after 50 Gy was 15–16 weeks. In
a similar study focusing on vascular changes after a maximal dose
of 75 Gy delivered to the rat brain using a gamma knife, it was
noted that vascular changes preceded necrosis [11]. This finding
suggests that the vascular response is also an important component
in the biologic effect of radiosurgery (Table 1).
1.3 Radiation A few strategies for radioprotection of normal tissue have already
Protection Studies been explored. The initial strategies included use of cerebral protec-
in Animal Models tive agents while delivering a high dose to tumor cells. Oldfield et al.
[12] documented protection from radiation-induced brain injury
using pentobarbital. Buatti et al. [13] found that 21-aminosteroids
(21-AS) protected the cat brain from injury due to radiosurgery and
was significantly more effective than corticosteroids [13]. Kondziolka
et al. showed that 15 mg/kg but not 5 mg/kg of U-74389G (a
21-AS) was effective at reducing brain injury in the rat when admin-
istered 1 h prior to radiosurgery. U74389G ameliorated vasculopa-
thy and regional edema and delayed the onset of necrosis, while
gliosis remained unaffected [14]. Preliminary evidence suggests that
this agent may be acting through reduction of the cytokines induced
by brain irradiation.
1.4 Enhancing Although benign tumor radiosurgery is associated with high tumor
the Effect control rates malignant glial tumors often recur. Additional strate-
of Radiosurgery gies to improve cell kill of malignant brain tumors are needed.
in Animal Models Niranjan et al. studied the synergistic effect of tumor necrosis factor
alpha (TNF-α) on enhancing the tumor response to radiosurgery.
TNF-α can act as a tumoricidal agent with direct cytotoxicity medi-
ated through binding to its cognate cell-surface receptors and a vari-
ety of activities triggering a multifaceted immune attack on tumors
Table 1
Central nervous system response to radiosurgery
Year Central
of Animal Radiosurgery dose Collimator Radiosurgery
Author study model target (Gy) size (mm) technique Results
Lunsford 1990 Baboon Caudate, 150 8 Gamma knife MR imaging and
thalamus, histology
pons documented lesion
45–60 days
posttreatment
Kondziolka 1992 Rat R. Frontal 30– 4 Gamma knife Histology at 90 day
lobe 200 showed tissue
changes at lower
doses (60 Gy) and
necrosis at higher
doses (100 Gy)
Blatt 1994 Cat Internal 149 10 LINAC MRI and serial
capsule histopathology
indicated mass effect
and neurologic
deficits at 3.5–4.5
weeks, some necrosis
12–29 weeks, and
late resorption of
necrosis
Kamiryo 2001 Rat Parietal 75 4 Gamma knife Electron microscopy at
cortex 3.5 months showed
decreased vascularity
and increased
capillary diameter in
irradiated regions;
basement membrane
changes precede
vascular damage
Karger 2002 Rat Parietal 26–50 3 LINAC MR imaging
cortex documented contrast
enhancement at 15
weeks after 50 Gy
and 19 weeks after
40 Gy radiosurgery
[15–20]. In addition, locally produced TNF-α has been reported to

enhance the sensitivity of tumors to radiation in nude mice [15]. We
employed a replication defective herpes simplex virus (HSV), as a
vector to deliver thymidine kinase (TK) and/or tumor necrosis fac-
tor (TNF-α) genes to U-87 MG tumors in nude mice. Radiosurgery
was performed 48 h after gene transfer using 15 Gy to the tumor
margin (21.4 Gy to the center). Daily ganciclovir therapy (GCV)
was started after gene transfer and continued for 10 days. The com-
bination of radiosurgery with TNF-α or with HSV-TK-GCV (sui-
cide gene therapy) and TNF-α significantly improved median
survival of animals [21]. In additional experiments, the connexin-43
gene was added to enhance the formation of gap junctions between
tumor cells, which should facilitate the intercellular dissemination of
TK-activated GCV from virus-infected cells to noninfected sur-
rounding cells. This creates a bystander effect that can improve
tumor cell killing [22]. Addition of connexin-43 gene to this para-
digm (TK-GCV + TNF-α + radiosurgery) further improved survival
(90 % survival in tumor-bearing mice). We also studied this strategy
in a 9 L rat glioma model and found that addition of radiosurgery to
suicide gene therapy (SGT) significantly improved animal survival
compared to SGT alone. The combination of HSV-based SGT
(TK-GCV), TNF-α gene transfer, and radiosurgery was more effec-
tive than SGT or radiosurgery alone. The combination of SGT with
radiosurgery was also more effective than SGT or radiosurgery
alone. Although, the exact mechanism of this effect is unclear and
remains the subject of future investigations, these experiments indi-
cate that gene therapy could be an effective strategy for enhancing
the radiobiological impact of radiosurgery. In other studies, tumor
sensitization to radiation was apparently mediated by extracellular
TNF-α promoting the destruction of tumor vessels, whereas HSV
vector mediated TNF-α enhanced killing of malignant glioma cell
cultures is presumably a consequence of an intracellular TNF-α
activity [20, 23] (Table 2).
1.5 Functional Radiosurgery is rapidly expanding beyond its use as a treatment of

Radiosurgery brain tumors and AV malformations. It has been found effective
in Animal Models for other neurologic disorders, such as epilepsy, movement disor-
ders, and trigeminal neuralgia. The promise of “functional” radio-
surgery has led to a need to investigate its efficacy, limitations, and
potential drawbacks.
The potential efficacy of radiosurgery for the treatment of epi-
lepsy has been evaluated using rat models. Kainic acid reproducibly
induces epilepsy in rats when injected into the hippocampus. Mori
et al. [24] treated kainic acid-induced epilepsy in rats with doses of
20–100 Gy radiosurgery using gamma knife. The efficacy of the
treatment on epilepsy was evaluated by direct observation and
scalp EEG for 42 days. Even 20 Gy significantly reduced the num-
ber of seizures, and the efficacy improved with increasing dose.
Only doses higher than 60 Gy induced histologic changes. Maesawa
et al. [25] irradiated epileptic rats with a single dose of 30 or 60 Gy.
Both doses significantly reduced EEG-defined seizures. The latency
to this effect was less after the higher dose (5–9 weeks for 60 Gy
versus 7–9 Gy for 30 Gy). While kainic acid injection alone reduced
performance of rats on the water maze task, the performance of
rats that had radiosurgery after kainic acid administration was not
Table 2
Experimental radiosurgery for malignant brain tumors
Year
First of Animal Maximum Tumor Collimator Experimental
author study model dose (Gy) model size (mm) treatment Results
Kondziolka 1992 Rat 30–100 C6 4 Radiosurgery Treated animals
Glioma survived 39
days (control
29 days).
Treated
tumors had
hypocellular
appearance
with cellular
edema
Niranjan 2000 Nude 21.4 U 87 MG 4 Radiosurgery + The combination
mouse HSV-based gene treatment
therapy enhanced
median
survival
(75 days) with
89 % animal
surviving
Nakahara 2002 Rat 32 MADB 4 Radiosurgery + The combination
106 cytokine treatment
cells transduced tumor significantly
cell vaccine prolonged
animal survival
and protected
animals from a
subsequent
challenge by
parental
tumor cells
placed in the
CNS
Niranjan 2003 Rat 21.4 9L 4 Radiosurgery + The combination
Glioma HSV-based gene of
therapy radiosurgery
and multigene
therapy
enhanced
median animal
survival
(150 days)
with 75 %
animal
surviving
different from controls. Liscak et al. [26] evaluated the effects of

radiosurgery on normal hippocampus in an effort to identify
potential normal tissue complications and determine dose limits
for hippocampal radiosurgery. This study employed four separate
4 mm isocenters to irradiate the entire hippocampus with
25–100 Gy. Doses <50 Gy did not cause any perceptible changes
based on histology, MRI, and Morris water maze testing. In con-
trast, the performance on the Morris water maze was significantly
worse for animals who were treated with >50 Gy. These investiga-
tions support the concept that radiosurgery may be an effective
method for treating epilepsy, but they also suggest that doses to
the hippocampus should be limited to reduce potential effects on
learning and memory.
The effect of radiosurgery on potential targets for the treat-
ment of movement disorders has been evaluated. De Salles and
colleagues [27] used a LINAC and 3 mm collimator to deliver a
maximal dose of 150 Gy to the subthalamic nucleus of one vervet
monkey and to the substantia nigra of another. Follow-up MRI
detected a 3 mm lesion that did not increase in size throughout the
course of the study. Kondziolka et al. [28] examined the effects of
thalamic radiosurgery in a baboon model, and reported that a dose
of 100 Gy (central dose using 4-mm collimator) was sufficient to
induce contrast enhancement of MR images and coagulative necro-
sis as evaluated by histology.
Radiosurgery has significant potential as an effective, noninva-
sive method for treatment of trigeminal neuralgia. Kondziolka
et al. investigated the effect of gamma knife irradiation on the tri-
geminal nerve in the baboon [29]. A central dose of 80 or 100 Gy
using a 4 mm collimator was delivered to the normal proximal
trigeminal nerve. Follow-up MRI at 6 months showed a 4 mm
region of contrast enhancement on the nerve. Histology showed
that both large and small fibers were affected with axonal degen-
eration occurring after 80 Gy and necrosis after 100 Gy. Neither
dose was effective at selectively damaging fibers responsible for
transmission of pain while maintaining those responsible for other
sensations, which would be optimal for effective treatment of tri-
geminal neuralgia. This study demonstrated that it was possible to
noninvasively and precisely affect specific nerves using the gamma
knife. In a recent study Zhao et al. investigated the effect of dose
and single versus two isocenters on trigeminal nerve radiosurgery
in rhesus monkeys [30]. These authors delivered a central radiation
dose of 60, 70, 80, or 100 Gy using 4-mm collimator at trigeminal
nerve root. One side of the nerve was exposed to single-target-
point irradiation, and the contralateral side was exposed to double-
target-point irradiation. Histological examination at 6 months
revealed that the target doses of 80 Gy resulted in partial degenera-
tion and loss of axons and demyelination. The extent of histologi-
cal changes was identical with the single-target-point and the
double-target-point irradiation (Table 3).
Table 3
Experimental functional radiosurgery
Year
of Animal Maximum Region(s) Irradiation Collimator
First author study model dose (Gy) irradiated technique size (mm) Results
Ishikawa 1999 Rat 200 Medial Gamma 4 Sequential MRI and
temporal knife histopathology
lobe showed consistent
necrosis at 2
weeks after
200 Gy
radiosurgery.
Mori 2000 Rat 20–100 Hippocampus Gamma 4 Reduction in
knife seizure
frequency after
≥20 Gy
radiosurgery.
Maesawa 2000 Rat 30–60 Hippocampus Gamma 4 Reduction in
knife seizure
frequency after
30–60 Gy
radiosurgery,
shorter latency
after higher
dose, learning
and memory
unaffected.
Kondziolka 2000 Baboon 80–100 Trigeminal Gamma 4 Axonal
nerve knife degeneration on
electron
microscopy 6
months after
radiosurgery at
all doses
Chen 2001 Rat 20–40 Hippocampus Gamma 4 Substantially
knife reduction in
seizure
frequency and
duration by
subnecrotic
(20–40 Gy)
radiosurgery
De Salles 2001 Monkey 150 Subthalamic LINAC 3 MRI and histology
nucleus, showed that
substantia necrotic lesion
nigra remained at
<3 mm size after
LINAC
radiosurgery.
(continued)
Table 3
(continued)
Year
of Animal Maximum Region(s) Irradiation Collimator
First author study model dose (Gy) irradiated technique size (mm) Results
Kondziolka 2002 Baboon 100 Thalamus Gamma 4 MRI, histology
knife showed necrosis
at 6 months
Liscak 2002 Rat 25–150 Hippocampus Gamma 4 Altered memory
knife performance
after >50 Gy
radiosurgery
Zerris 2002 Rat 140 Caudate– Gamma 4 Radiosurgery
putamen knife significantly
complex reduced
6-OHDA-
induced
hemiparkinsonian
behavior. Areas
surrounding
necrotic lesions
were highly
positive for
GDNF
Brisman 2003 Rat 5–130 Hippocampus Proton NA Proton radiosurgery
CGE beam with doses 90
CGE or higher
resulted in adverse
behavioral effects
and necrosis in 3
months. 30 or 60
CGE radiosurgery
led to marked
increase in
HSP-72 staining
but no necrosis
Zhao 2011 Monkey 60–100 Trigeminal Gamma 4 Irradiation at 80 Gy
nerve knife can cause partial
degeneration and
loss of axons and
demyelination. A
100-Gy dose can
cause some
necrosis of
neurons. No
additional effect
of double-target-
point irradiation
was seen.
Vincent et al. performed hypothalamic radiosurgery on geneti-

cally obese Zucker rats and studied the effect of subnecrotic hypo-
thalamic radiosurgery on body weight set point [31]. These
investigators performed radiosurgery using a total dose of 40 Gy
delivered to two nearby targets in the medial hypothalamus. These
investigators noted significant and sustained reductions in weight
set point for animals that received radiosurgery compared to sham-
treated animals after a latency of 7 weeks. No gross behavioral
abnormalities were noted. Histopathological analysis showed no
abnormalities except a small area of necrosis in one animal. At the
University of Pittsburgh, the authors are investigated the feasibility
of hypothalamic radiosurgery using a primate model. The results
showed that hypothalamic radiosurgery is feasible and safe.
2 Materials and Methods
2.1 Small Animal 1. The rats/mice are anesthetized with Ketamine and Acepromazine
Radiosurgery Models administered intramuscularly.
(Mice Model/Rat Model) 2. Anesthetized animals are placed in a stereotactic head frame
2.1.1 Animal Preparation (David Kopf Instruments, Tujunga, CA).
for Radiosurgery 3. A small craniotomy is drilled 2 mm to the right of midline and
1 mm anterior to the coronal suture. Dura was not opened.
4. A predetermined number of cells (1 × 105 U-87MG glioblas-
toma cells in a 3-μl volume) is implanted stereotactically in the
right frontal lobe region 3 mm below the dura mater. This area
corresponds to the lateral portion of the right striatum of
the mouse.
5. A drug or viral vector can also be injected using the above
technique.
6. The injection needle is removed.
7. A 2 mm section of a 25-gauge needle is placed in the craniot-
omy site over the dura for later stereotactic targeting.
8. The craniotomy is then sealed with bone wax and the scalp is
closed with a 3-0 silk suture.
2.1.2 Radiosurgery 1. Animals are anesthetized and placed on a small animal specially
Technique for Small modified platform which is attached to stereotactic frame.
Animals 2. Animals are secured in place using transparent adhesive tape.
3. Angiography fiducial box is attached to the stereotactic frame.
4. Lateral and posteroanterior plain X-rays are taken. It is impor-
tant to make sure that all nine fiducial markers as well as a metal
marker place on animal skull are visible on X-ray films (Fig. 1).
5. X-ray films are scanned into a Gamma Knife planning computer.
Fig. 1 Figure showing anteroposterior (a) and lateral (b) views of plain X-ray films for stereotactic radiosurgery
of small animals. Three anesthetized rats are placed on a small animal specially modified platform which is
attached to stereotactic frame. Angiography fiducial box is attached to the stereotactic frame. Lateral and
anteroposterior plain X-rays are taken. Note that the fiducial markers from Angiography fiducial box as well as
a metal marker placed on animal skull are visible on X-ray films
6. Target is defined based on its predetermined distance from the

metal skull marker. We routinely used a point 3 mm perpendicu-
lar to an extradural metal marker in the right frontal brain region,
which corresponded to the center of the tumor cell injection.
7. Radiosurgery planning is performed using Leksell Gamma Plan®.
8. For small animals, a plan is achieved using one 4-mm radiation
isocenter.
9. Depending upon the goal of radiosurgery a central or a margin
dose is selected. For our experiments involving tumor radio-
surgery, we selected a margin dose of 15 Gy (center dose,
21.4 Gy) and delivered it to the 70 % isodose line using a 4-mm
collimator (Fig. 2).
10. The final plan is printed and exported to Leksell Gamma Unit.
The print out shows the x, y, and z coordinates of the planned
isocenter as well as time required at that position in order to
deliver the desired dose.
11. Radiosurgery is performed by positioning the small animal
frame at the x, y, and z coordinates of the isocenter in the
Leksell gamma knife unit (Elekta Instruments, Atlanta, GA)
and setting the time obtained from dose plan.
12. At the end of the treatment, animal platform with stereotactic
frame is removed from the Gamma Unit. Animals are removed
from the frame and put in their respective cages.
13. Animals are observed till the regain consciousness and are full
awake.
Fig. 2 Figure showing radiosurgery dose plan for rat model of demyelination. For
radiosurgery planning, stereotactic X-ray films are scanned into a Gamma Knife
planning computer. Target is defined based on its predetermined distance from
the metal skull marker or skull suture. Radiosurgery planning is performed using
Leksell Gamma Plan® using 4-mm radiation isocenter
2.1.3 Animal Observation 1. All animals are observed twice daily to monitor external appear-
Protocol ance, feeding behavior, and locomotion (ability to walk to a
distance of 50 cm in 10 s).
2. The contralateral limbs are observed daily for the development
of paresis both passively and actively.
3. Animals are sacrificed at the first sign of an adverse event (pare-
sis, inability to feed) and brains are removed for histological
examination.
4. Animals surviving through the 75-day observation period are
euthanized and the brains removed for histological examination.
2.2 Large Animal 1. Large animals (monkeys) are individually housed in stainless
Radiosurgery Models steel cages in air-conditioned and temperature and light-cycle-
(Baboon Model, controlled rooms.
Monkey Model) 2. Animals are anesthetized using intravenous Propofol (2,6-diiso-
2.2.1 Animal Preparation propylphenol) infusion and were intubated and maintained on
and Radiosurgery inhalation anesthetic agents.
3. The animal is brought in the laboratory adjacent to MR unit.
The standard Model-G Leksell Head frame is used for large
Fig. 3 Diagrammatic representation showing monkey with standard Model-G Leksell head frame anchored to
his head. Large animals are anesthetized using intravenous Propofol (2,6-diisopropylphenol) infusion and the
Leksell head frame is anchored to their head using two pins on the forehead and two pins on the back of head.
Two front posts are attached to cheek (maxilla) using 60–70 mm long pins. The two back posts are fixed on
the occipital ridge using long 80–90 mm pins. The MR fiducial box is then attached on top of the head ring and
stereotactic MRI is performed. These stereotactic MR images are imported into the dose planning computer for
radiosurgery dose planning
animals. This frame is widely used clinically for stereotactic

applications of brain.
4. In humans, the frame is anchored to head using two pins on the
forehead and two pins on the back of head. Because monkeys do
not have a convex forehead and their head size is much smaller
compared to human, special technique was used to anchor
Leksell head frame. Two front posts were attached to cheek
(maxilla) using 60–70 mm long pins. The two back posts were
fixed on the occipital ridge using long 80–90 mm pins (Fig. 3).
5. After securing the head ring, the fiducial box is attached on top
of the head ring.
6. The animal is taken to MR unit for the second set of images
(stereotactic images).
7. Head frame is stabilized using a modified head holder that fits
MR table and head coil.
8. The stereotactic MR images using an FOV (field of view) of
250 × 250 mm are obtained using standard head coil.
9. The images are transferred to Gammaplan computer via ether-

net system.
10. Stereoatctic images are defined using GammaPlan software.
11. The target is selected based on the goals of the experiment.
The target could be thalamus, trigeminal nerve, caudate
nucleus, hypothalamus, etc.
12. An optimum margin and central dose is selected.
13. The plan is printed and exported to Leksell Gamma unit.
14. The animal is observed and maintained under anesthesia
throughout the planning period.
15. Once the dose plan is ready, the animal is secured to Gamma
unit by attaching his head frame to the unit.
16. x, y, and z coordinates obtained from the dose planning are set
on the Gamma Unit.
17. The time obtained for desired dose delivery is set on the unit
and the treatment is initiated.
18. At the completion of treatment animal is removed from the unit.
19. Head frame is removed and head is dressed with antiseptic and
antibiotic crèmes.
20. Anesthesia reversal is given and endotracheal tube is removed
once animal starts breathing on her own.
21. Animal is placed in the cage and monitored till she is fully
awake and active.
2.2.2 Animal Observation 1. All animals are observed daily to monitor external appearance,
Protocol feeding behavior, and locomotion (ability to walk to a distance
of 50 cm in 10 s).
2. The contralateral limbs are observed daily for the development
of paresis both passively and actively.
3. Depending upon the goals of research follow-up MR imaging
is performed.
4. Animals are sacrificed at the first sign of an adverse event (pare-
sis, inability to feed) and brains are removed for histological
examination.
5. Depending upon the protocol Animals are euthanized and the
brains removed for histological examination.
3 Notes
1. Anesthesia is a critical component of small animal radiosurgery.

Anesthesia needs to be titrated judiciously in order to keep
mice or rats immobile during the whole experiment which can
last up to 3 h. While higher doses of anesthetic agents can kill

the animals, the lower doses can make experiment ineffective if
animal moves during the radiosurgery treatment. Every animal
is weighed prior to experiment and doses of anesthetics are
titrated accordingly. Rats were usually anesthetized with an
intramuscular or Intraperitoneal injection of Ketamine and
Acepromazine (9:1) at a dose of 44 mg/kg. Animals are given
additional doses at the slightest sign of movement of any body
parts or wakefulness.
2. While placing the animals on stereotactic frame care is taken to
make sure that when angiography fiducial box is attached to
the frame, the animal is located in the middle of the frame.
Special precaution is taken to immobilize the animal by gently
taping there head to the frame platform.
3. One plain X-rays are taken, we always ensure that all nine fidu-
cial markers (pluses and crosses) are visible on both anteropos-
terior and lateral view. Visualization of these markers is critical
to scanning these films into gamma plan computer. Once these
films are digitized these can be used for dose planning.
4. Dose planning is performed using the gamma plan software.
For small animal radiosurgery, a 4 mm beam diameter collima-
tor is selected for radiation delivery. If a larger lesion is desired
(such as for demyelination experiment), two shot dose plan
can be designed. The margin dose is prescribed to 50 % isodose
line. For most tumor radiosurgery, however, target is much
smaller than 4 mm. In such cases, the volume covered by deliv-
ered radiation can be decreased by prescribing the margin dose
to a higher isodose line (such as 70 %).
5. Proper technique for target selection is important in order to
treat normal or tumor brain. If the goal is to treat the normal
brain (to study the effect of radiation) then a target can be
selected based on distances from the coronal suture, which can
be visualized easily in plain X-rays. However, if the goals is to
treat a previously implanted tumor, then the exact coordinates
of the tumor center are needed. To identify the location of the
implanted tumor we use a metal marker on the dura mater of
the animals. We usually implant the tumor cells 3 mm below
the dura. On the day of tumor implantation the anesthetized
animal is placed in a stereotactic head frame (David Kopf
Instruments, Tujunga, CA) and a small craniotomy is drilled,
2 mm to the right of midline and 1 mm anterior to the coronal
suture. The dura is not opened. Tumor cells in a 3-μl volume
are implanted stereotactically in right frontal lobe region 3 mm
below the dura mater. This area corresponds to the lateral por-
tion of the right corpus striatum of the mouse. After removal
of the injection needle, a 2 mm section of a 25-gauge needle is
placed in the craniotomy site over the dura for later stereotactic
targeting. The craniotomy was then sealed with bone wax and
the scalp is closed with 3-0 silk suture. On the day of radiosur-
gery this marker is seen on the X-ray films. Because we know
that the implanted tumor is 3 mm below the marker we can
center the radiosurgery target 3 mm below the metal marker.
4 Future Experimental Radiosurgery
Rapid developments in computer hardware and software, imaging

technologies, and stereotactic techniques have contributed to
improvement in radiosurgery technology. As a result, the role of
radiosurgery has expanded beyond its initial application for func-
tional neurosurgery, pain management, arteriovenous malforma-
tions, and selected skull base tumors. The clinical spectrum of
radiosurgery now includes a wide variety of primary and secondary
brain tumors, vascular malformations, and functional brain disor-
ders. Although radiosurgery provides survival benefits in diffuse
malignant brain tumors, cure is still not possible. Additional strate-
gies are needed to specifically target tumor cells while sparing nor-
mal CNS tissue. Further animal model-based research is needed to
develop new treatment strategies that would maximize the effec-
tiveness of radiosurgery on target tissue and minimize injury to
other areas.
References
1. Larsson B, Leksell L, Rexex B, Sourander P, Biological effects of stereotactic radiosurgery

Mair W, Anderson B (1958) The high-energy in the normal primate brainstem. In: Lunsford
proton beam as neurosurgical tool. Nature LD (ed) Stereotactic radiosurgery update.
182(6):1222–1223 Elsevier, New York, pp 291–294
2. Leksell L, Larsson B, Anderson B, Rexed B, 7. Kondziolka D, Lunsford LD, Claassen D,
Sourander P, Mair W (1960) Lesions in the Maitz AH, Flickinger JC (1992) Radiobiology
depth of the brain produced by a beam of of radiosurgery: part I. The normal rat brain
high-energy protons. Acta Radiol Ther Phys model. Neurosurgery 31(2):271–279
Biol 53:251–264 8. Blatt DR, Friedman WA, Bova FJ, Theele DP,
3. Andersson B, Larsson B, Leksell L et al (1970) Mickle JP (1994) Temporal characteristics of
Histopathology of late local radiolesions in the radiosurgical lesions in an animal model.
goat brain. Acta Radiol Ther Phys Biol 9(5): J Neurosurg 80(6):1046–1055
385–394 9. Kamiryo T, Kassell NF, Thai QA, Lopes MB,
4. Nilsson A, Wennerstrand J, Leksell D, Lee KS, Steiner L (1996) Histological changes
Backlund EO (1978) Stereotactic gamma irra- in the normal rat brain after gamma irradia-
diation of basilar artery in cat. Preliminary tion. Acta Neurochir (Wien) 138(4):451–459
experiences. Acta Radiol Oncol Radiat Phys 10. Karger CP, Munter MW, Heiland S, Peschke P,
Biol 17(2):150–160 Debus J, Hartmann GH (2002) Dose-
5. Lunsford LD, Altschuler EM, Flickinger JC, response curves and tolerance doses for late
Wu A, Martinez AJ (1990) In vivo biological functional changes in the normal rat brain after
effects of stereotactic radiosurgery: a primate stereotactic radiosurgery evaluated by mag-
model. Neurosurgery 27(3):373–382 netic resonance imaging: influence of end
6. Kondziolka D, Lunsford LD, Altschuler EM, points and follow-up time. Radiat Res 157(6):
Martinez AJ, Wu A, Flickinger JC (1992) 617–625
11. Kamiryo T, Lopes MB, Kassell NF, Steiner L, 21. Niranjan A, Moriuchi S, Lunsford LD et al
Lee KS (2001) Radiosurgery-induced micro- (2000) Effective treatment of experimental
vascular alterations precede necrosis of the glioblastoma by HSV vector-mediated TNF
brain neuropil. Neurosurgery 49(2):409–414, alpha and HSV-tk gene transfer in combina-
discussion 14–15 tion with radiosurgery and ganciclovir admin-
12. Oldfield EH, Friedman R, Kinsella T et al istration. Mol Ther 2(2):114–120
(1990) Reduction in radiation-induced brain 22. Marconi P, Tamura M, Moriuchi S et al (2000)
injury by use of pentobarbital or lidocaine pro- Connexin 43-enhanced suicide gene therapy
tection. J Neurosurg 72(5):737–744 using herpesviral vectors. Mol Ther 1(1):71–81
13. Buatti JM, Friedman WA, Theele DP, Bova 23. Moriuchi S, Oligino T, Krisky D et al (1998)
FJ, Mendenhall WM (1996) The lazaroid Enhanced tumor cell killing in the presence of
U74389G protects normal brain from ste- ganciclovir by herpes simplex virus type 1 vector-
reotactic radiosurgery-induced radiation directed coexpression of human tumor necrosis
injury. Int J Radiat Oncol Biol Phys 34(3): factor-alpha and herpes simplex virus thymidine
591–597 kinase. Cancer Res 58(24):5731–5737
14. Kondziolka D, Somaza S, Martinez AJ et al 24. Mori Y, Kondziolka D, Balzer J et al (2000)
(1997) Radioprotective effects of the Effects of stereotactic radiosurgery on an animal
21-aminosteroid U-74389G for stereotactic model of hippocampal epilepsy. Neurosurgery
radiosurgery. Neurosurgery 41(1):203–208 46(1):157–165, discussion 65–68
15. Staba MJ, Mauceri HJ, Kufe DW, Hallahan 25. Maesawa S, Kondziolka D, Dixon CE, Balzer
DE, Weichselbaum RR (1998) Adenoviral J, Fellows W, Lunsford LD (2000) Subnecrotic
TNF-alpha gene therapy and radiation damage stereotactic radiosurgery controlling epilepsy
tumor vasculature in a human malignant gli- produced by kainic acid injection in rats.
oma xenograft. Gene Ther 5(3):293–300 J Neurosurg 93(6):1033–1040
16. Cao G, Kuriyama S, Du P et al (1997) 26. Liscak R, Vladyka V, Novotny J Jr et al (2002)
Complete regression of established murine Leksell gamma knife lesioning of the rat hip-
hepatocellular carcinoma by in vivo tumor pocampus: the relationship between radiation
necrosis factor alpha gene transfer [comment]. dose and functional and structural damage.
Gastroenterology 112(2):501–510 J Neurosurg 97(5 Suppl):666–673
17. Han SK, Brody SL, Crystal RG (1994) 27. De Salles AA, Melega WP, Lacan G, Steele LJ,
Suppression of in vivo tumorigenicity of human Solberg TD (2001) Radiosurgery performed
lung cancer cells by retrovirus-mediated transfer of with the aid of a 3-mm collimator in the sub-
the human tumor necrosis factor-alpha cDNA. Am thalamic nucleus and substantia nigra of the
J Respir Cell Mol Biol 11(3):270–278 vervet monkey. J Neurosurg 95(6):990–997
18. Ostensen ME, Thiele DL, Lipsky PE (1989) 28. Kondziolka D, Conce M, Niranjan A, Maesawa
Enhancement of human natural killer cell S, Fellows W (2002) Histology of the 100 Gy
function by the combined effects of tumor thalomotomy in the baboon. Radiosurgery
necrosis factor alpha or interleukin-1 and 4(4):279–284
interferon-alpha or interleukin-2. J Biol 29. Kondziolka D, Lacomis D, Niranjan A et al
Response Mod 8(1):53–61 (2000) Histological effects of trigeminal nerve
19. Owen-Schaub LB, Gutterman JU, Grimm EA radiosurgery in a primate model: implications for
(1988) Synergy of tumor necrosis factor and trigeminal neuralgia radiosurgery. Neurosurgery
interleukin 2 in the activation of human cyto- 46(4):971–976, discussion 6–7
toxic lymphocytes: effect of tumor necrosis 30. Zhao ZF, Yang LZ, Jiang CL, Zheng YR,
factor alpha and interleukin 2 in the generation Zhang JW (2010) Gamma Knife irradiation-
of human lymphokine-activated killer cell induced histopathological changes in the tri-
cytotoxicity. Cancer Res 48(4):788–792 geminal nerves of rhesus monkeys. J Neurosurg
20. Gridley DS, Archambeau JO, Andres MA, 113(1):39–44
Mao XW, Wright K, Slater JM (1997) Tumor 31. Vincent DA, Alden TD, Kamiryo T et al (2005)
necrosis factor-alpha enhances antitumor The baromodulatory effect of gamma knife irra-
effects of radiation against glioma xenografts. diation of the hypothalamus in the obese Zucker
Oncol Res 9(5):217–227 rat. Stereotact Funct Neurosurg 83(1):6–11
Chapter 3
Stereotactic Surgery in Rats

Jaroslaw Maciaczyk, Ulf D. Kahlert, Máté Döbrössy, and Guido Nikkhah
Abstract
Animal models represent the final step to complete preclinical investigations. Here, we describe in detail
the principles and procedures for the surgical, toxin-induced animal models for Parkinson’s disease (PD),
and Huntington’s disease (HD). Using highly precise stereotactic intracerebral injections of toxins into the
nigrostriatal pathway and basal ganglia, we are able to target specific neural circuits in different regions of
the dopaminergic and GABAergic system. In addition, validated protocols for adult and neonatal cell
transplantation to reconstruct the destructed neuronal circuits as models for neural repair are described.
Key words Stereotactic neurosurgery, Rat, Cell transplantation, Parkinson’s disease model,
Huntington’s disease model, Neural stem cells, Neurorepair
1 Introduction
Stereotactic surgery has been an invaluable tool applied since early

20th century in both experimental and clinical neuroscience to
precisely target deep located brain regions for lesioning, injection
of anatomical tracers, implantation of electrodes, neurotransplan-
tation, sampling of various brain pathologies, gene delivery, etc. Its
history begins with the development of the first stereotactic appa-
ratus by Horsley and Clarke in 1908 [1], who also coined the term
“stereotaxis” deriving from Greek words “stereos” meaning “three
dimensional” and “taxis” meaning “to touch or to move.” Despite
further technical development and refinement of stereotactic
equipment the principles of the surgical procedure enabling mini-
mal invasive targeting of the deep brain structures based on the
three-dimensional Cartesian coordinate system remained constant
over the last hundred years. Whereas in humans the accuracy of
this minimal invasive method relies on the preoperative imaging, in
rodents the precision of the stereotactic targeting is based on the
relation between bony landmarks of the cranium with deep located
structures, which can be easily determined from stereotactic atlases
such as The Mouse Brain in Stereotaxic Coordinates [2] and The
31
32 Jaroslaw Maciaczyk et al.
Rat Brain in Stereotaxic Coordinates [3]. The coordinates describe

the distance of the target in three dimensions (x, y, and z) from
bregma—an intersection of the coronal and sagittal sutures on the
surface of the skull, where the x plane represents a mediolateral
(ML), the y plane rostrocaudal distance from the reference point,
and z plane the dorsoventral (DV) distance from dura level that can
be easily calculated upon the immobilization of the head of experi-
mental animal in the stereotactic apparatus.
In this chapter, we are going to focus on the stereotactic surgi-
cal procedures in both adult and neonatal rats routinely performed
in our laboratory to create rodent models of neurodegenerative
disorders, i.e., Parkinson’s disease (PD), Huntington’s disease
(HD), and for experimental neurotransplantation approaches.
2 PD Rat Models
Human idiopathic PD is a progressive neurodegenerative disorder,

primarily characterized by degeneration of the dopaminergic neu-
rons of the substantia nigra (SN) and subsequent loss of dopami-
nergic innervation of the striatum, leading to deregulation of the
complex neurotransmitter system of the basal ganglia. Therefore,
modeling of PD in rodents requires the depletion of the dopamine
(DA) releasing terminals in their projection areas. In rodents, a
standardized DA depletion of the basal ganglia can be reached by
axotomy of nigral afferents (traditional model), by systemic appli-
cation of 1-methyl-4-phenyl-1,2,5,6-tetrahydropyridine (MPTP)
in mice, or by unilateral intracerebral stereotactic injection of
6-hydroxydopamine [4]—being a most common rodent model of
PD. The neurotoxicity of 6-OHDA is based on its potent inhibi-
tory effect on the mitochondrial respiratory enzymes (chain com-
plexes I and IV, 2) that consequently leads to loss of dopaminergic
neurons [5, 6]. Based on different injection sites of the 6-OHDA
within the nigrostriatal pathway various grades of severity of lesion-
induced PD can be achieved (Fig. 1):
1. The caudate–putamen unit (CPu)—referred as to “partial” or
“terminal lesion,” which also leads to relatively mild, local DA
depletion resembling an early and intermediate stage of PD.
2. The medial forebrain bundle (MFB) which leads to extensive
DA depletion recapitulating a later PD status, referred as to
“complete lesion.”
3. The substantia nigra pars compacta (SNc), which leads to more
specific and moderate DA depletions for the representation of
earlier PD status.
Independently from the model chosen the lesion coordinates
are set according to bregma as a reference point for the anteroposterior
Stereotaxy in Rats 33
A B C
AC
PF
NS
MFB
NA
OT
A
VTA SN
Fig. 1 Sagittal schematic diagram of the rodent brain indicating the three main targets for the 6-OHDA induced
dopamine (DA) depletion. Partial DA loss can be achieved by infusion of the toxin in the striatum at the presyn-
aptic level of the nigrostriatal projections (A), whereas complete DA depletion is produced if the toxin is injected
into the medium forebrain bundle (B). Infusion of toxin into the substantia nigra has also been used (C) but this
option gives less consistent striatal DA pathology. SN substantia nigra, VTA ventral tegmental area, A amyg-
dala, OT olfactory tubercle, NA nucleus accumbens, MFB medium forebrain bundle, S septum, NS neostriatum
(striatum), PF prefrontal cortex, AC anterior cingulated cortex
(AP) and ML coordinates and the dura as reference for the DV

coordinate, using the rat brain atlas Paxinos and Watson [7]. The
6-OHDA is used in a concentration of 3.6 μg/μl in saline contain-
ing 0.2 % (w/v) ascorbic acid.
3 Injection of 6-OHDA into the CPu
6-OHDA injection into the CPu, referred to as “partial” or “ter-

minal lesion” results in more selective damage of the nigrostriatal
dopaminergic system (Fig. 2c, d). Following the injection of the
neurotoxin, due to its retrograde transport to SN pars compacta
(SNc), exclusively nigral dopaminergic neurons projecting to the
injection area undergo degeneration and cell death. In this animal
model, usually ventrolateral and dorsomedial striatum have been
targeted. In rodents, the ventrolateral portion of the CPu receives
input from motor and sensorimotor areas of the neocortex whereas
its DA innervations exclusively projects from the SNc. In contrast,
Fig. 2 (a–f) Following 6-OHDA lesion, sections are stained for DA afferents and cell body using tyrosine hydroxylase
as the marker. Control animals with an intact nigrostriatal projections show complete DA innervation throughout the
caudate putamen unit (CPU, a), and full complement of cell body staining in the midbrain both in the substantia nigra
and the ventral tegmental area (b). Following partial/terminal 6-OHDA lesion, striatal DA depletion will occur selec-
tively in the areas targeted but will remain intact in other striatal regions (c); and the partial lesion will be reflected at
the midbrain level as well (d). Medial forebrain bundle lesions result in the complete loss of DA in the striatum (e) and
at the midbrain level affecting both the substantia nigra and the ventral tegmental area (f). V ventricle, Ctx cortex, Str
striatum, VTA ventral tegmental area, SN substantia nigra. Scale bar = 1 mm (a, c, e), and 500 μm (b, d, f)
the dorsal part of the CPu is characterized by a mixed DA innerva-

tion from both SN and ventral tegmental area (VTA), receiving
inputs from frontal cortical areas as well as from the limbic system
being an equivalent of the nucleus caudatus in humans.
Interestingly, more pronounced effects on locomotion and
drug-induced rotation behavior can be achieved by lesioning the
dorsomedial part of the CPu, whereas injection of the neurotoxin
into the ventrolateral parts of the CPu provokes predominantly
difficulties with movement initiation, sensorimotor orientation,
and skilled motor behavior. Therefore, lesions in rodent ventrolat-
eral CPu resemble most closely the depletion of DAnergic innerva-
tion in putamen of patients with PD [8]. However, to achieve
long-lasting contralateral behavior deficits the neurotoxin has to be
distributed over multiple injection sites along the rostrocaudal axis
of the ventrolateral CPu with a bilateral CPu lesion as a closest
parallel to the human disease [9, 10].
4 Injection of 6-OHDA in the MFB
Unilateral 6-OHDA injection into the MFB leads to an almost

total destruction of the DAnergic neurons of the SNc projecting to
striatum as well as of the VTA projecting to the nucleus accumbens
[52] (Fig. 2e, f), causing a postsynaptic denervation supersensitiv-
ity of DA receptors. As a result of the lesion-induced imbalance
between the nigrostriatal systems of both hemispheres, the animals
show unilateral sensorimotor deficits enabling the evaluation of the
lesion by behavioral analysis. The most robust one is a spontaneous
postural motor asymmetry, causing the animals to rotate toward
their impaired hemisphere. This can be enhanced by stress and in
particular during drug-induced rotation using D1/D2 receptor
agonist apomorphine and/or DA reuptake inhibitor d-amphet-
amine (for further details, see [11–13]). In contrast, the bilateral
MFB lesion of the dopaminergic nigrostriatal system in adult ani-
mals leads to a severe sensorimotor impairment, rapid deteriora-
tion of the general condition with aphagia and adipsia, necessitating
intensive care, and total parenteral nutrition [14, 15]. However, an
advantage of bilateral injection is the avoidance of sprouting of
axons from an intact side of the brain accounting for reported par-
tial compensation of neurological deficits after the unilateral lesion.
Nevertheless, the standard 6-OHDA rat model, caused by unilat-
eral injection of 6-OHDA into the MFB is not only more prag-
matic, but also permits a direct comparison of lesion effects and
therapeutic results within one animal by the comparison of both
hemispheres.
5 Injection of 6-OHDA into the SNc
Injection of the neurotoxin into SNc results in a more selective deple-

tion of DAnergic neurons and less dramatic damage in the DA sys-
tem. Animals receive unilaterally either medial and/or a single lateral
injection into the SNc. The loss of dopaminergic neurons within the
SNc, as measured with tyrosine hydroxylase (TH) immunohisto-
chemistry reaches approximately 90 %. Interestingly, also VTA seems
to be affected by the injection of neurotoxin showing only about
70 % of surviving DAnergic neurons as compared to the unlesioned
side. The single lateral injection of the 6-OHDA spares the dopami-
nergic cells in the medial SNc and reflects a neuropathological find-
ing of PD patients with the DAnergic cell loss present mainly within
the lateral aspect of the SN [16]. In consequence, at the lesion side
the density of DAnergic fibers within the lateral CPu diminishes dra-
matically as compared to medial parts of the caudate–putamen unit
[17]. Moreover, the remaining DAnergic CPu innervation corre-
sponds clearly to the extent of DAnergic cell depletion in SNc and
correlates with the number of rotations in response to apomorphine,
but not to d-amphetamine. Some authors reported bilateral SNc
neurotoxin injections better resembling a clinical picture of PD with
both hemispheres affected by the pathological process [18]. One of
the major difficulties of this model is the small size of the target struc-
ture making the injection of the toxin into the SNc without lesioning
adjacent structures, i.e., VTA [17] a very challenging task and restrict-
ing its application to very rare experimental designs.
6 Neonatal 6-OHDA Injections
To produce a bilateral degeneration of the DAnergic nigrostriatal

pathway in neonatal rats, 6-OHDA solution is injected transcuta-
neously into both lateral ventricles on postnatal day 1 (P1) [19–
21]. Anatomical landmark to define the correct coordinates is the
bregma, in this developmental stage still visible through the skin.
Other than in the adult MFB-lesion model, this bilateral lesion
surgery does not result in severe akinesia and severe sensorimotor
deficits. In most cases, the lesioned neonatal rats can be raised by
their mother without additional support; however, it has also been
observed that the mother can reject lesioned pups in which case
the affected pups need to be hand feed until they gain autonomy.
7 Rodent Model of HD
HD is a hereditary progressive neurodegenerative disorder charac-

terized by the development of emotional, behavioral, and psychiat-
ric abnormalities; loss of previously acquired intellectual or
cognitive functions; and motor disturbances. The gene involved in

HD, called huntingtin (Htt) or Interesting Transcript 15 (IT15)
gene (historically known as HD gene), is located on the short arm
of chromosome 4 (4p16.3) [22] The sequence of three DNA
bases, cytosine–adenine–guanine (CAG) coding for amino acid
glutamate [23] located at the 5′end of the HD gene show increased
number of repeats corresponding to the onset of the clinical HD
symptoms [24]. For CAG repeats equal to or greater than 35, the
HD gene is thought to have 100 % penetrance [25]. The neuro-
pathological hallmark of HD is the loss of DARPP-32 positive,
GABA-ergic striatal medium spiny projection neurons leading to
the degeneration of the basal ganglia, and development of the clin-
ical symptoms of the disease [26–28]. There are numerous
rodent—mainly mice—transgenic models of HD but these models
are best to study molecular pathways and mechanisms that are
affected by the mutant huntingtin as they do not reliable repro-
duce the striatal pathology. Therefore, experimental modeling of
the HD requires a selective depletion of this striatal cell subpopula-
tion achieved by stereotactic, uni-, or bilateral injection of a neuro-
toxin into the CPu. The most widely used is an excitotoxic
quinolinic acid (QA), an agonist of the N-methyl-d-aspartate
(NMDA) receptor and an endogenous metabolite of tryptophan,
selectively affecting striatal medium spiny neurons containing
GABA, dynorphin, and enkephalin [29] causing striatal degenera-
tion in rats [30, 31] similar to the one seen in human HD. For a
variety of reasons detailed below, QA became the preferred excito-
toxin for use in HD studies. Several experimental data favor an
excitotoxic hypothesis as a possible pathogenesis of HD. It has
been reported that in brains of HD patients, the activity of
3-hydroxyanthranilate oxygenase, which is the biosynthetic enzyme
for QA, is increased [32] in parallel with reduced levels of kyn-
urenic acid, an antagonist of NMDA receptors that may modulate
QA-induced neurotoxicity [33]. Moreover, there are obvious simi-
larities between pathology observed in the HD brain and in the
QA model. In QA-induced striatal lesions, the levels of GABA and
substance P, both of them localized in the medium spiny neurons,
are reduced and the content of somatostatin present in the medium
aspiny neurons is preserved [29]. In accordance with these neuro-
chemical findings, some morphological studies in rats have shown
that intrastriatal injections of QA selectively depletes medium spiny
neurons with relative sparing of medium aspiny neurons and large
neurons [29, 32]. Another reason for the widespread use of QA in
HD research is that the cell death caused by the excitotoxin may
closely resemble oxidative damage-related mechanism of neuronal
death seen in HD brains [34–37]. QA injections in rodents pro-
duce the most reliable and morphologically reproducible lesions as
well as motor, cognitive, and motivational impairments manifested
by HD patients in early (but not later) stages of HD. Motor-related
behavioral deficits in the rat model include: spontaneous and

apomorphine-induced ipsilateral rotation behavior upon unilateral
QA injection [38] due to the loss of the ipsilateral striatal projec-
tion neurons and the modulatory effect that the dopamine signal-
ing has on the output neurones; animals also show hyperlocomotion,
impairments in skilled paw use, and lateralized sensory motor defi-
cits [39–42]. Cognitive functions, such as discriminative capability
and response selection [43–45] as well as the water maze and the
T-maze based tests [46–48], indicating impaired visuospatial skills
and memory recall, are also impaired in rats with bilateral striatal
lesion. At the morphological level, QA-induced neuronal loss leads
to a dose-dependant progressive atrophy of the striatum, starting 4
weeks after the toxin injection. Depending on the lesion parame-
ters, by 3–4 months post-lesioning, the volume of the entire stria-
tum is reduced by around 30–70 % with concomitant enlargements
of the ipsilateral ventricle [40].
Taken together, the QA-induced lesions of the striatum in
rodents resemble some crucial neuropathological features of HD
(Fig. 3a–f). The model is therefore suitable for exploring the feasi-
bility of intrastriatal grafts to replace for lost striatal medium spiny
neurons.
8 Other Stereotactic Procedures in Rats
Beside the stereotactic injections of the neurotoxic substances to

selectively lesion certain brain structures, as described above; the ste-
reotactic procedure may also be applied for intracerebral cell implan-
tation. This technique is an important part of an methodological
armamentarium of experimental neurooncology aiming at the
implantation of malignant cells, derived either directly from the
postoperative tumor specimen or from in vitro culture of the tumor-
derived cell line, and regenerative neurosurgery, enabling the precise
intracranial implantation of stem cells of different origin or predif-
ferentiated precursors in order to reconstitute the pathologically
changed cytoarchitecture of the recipients’ brain or to deliver cells
ectopically producing neurotransmitters and growth factors. With
the development of new experimental fields the emergence of novel
applications of the stereotactic procedures could be observed. This
includes viral and short-hairpin RNA implantations for in vivo gene
manipulation, microdialysis for studies of brain metabolism with ste-
reotactically implanted probes, interstitial radiation, or implantation
of electrodes for both registering of the electric activity of specific
brain regions and their stimulation.
The general principles of the stereotactic procedure do not sig-
nificantly differ in all of the above mentioned applications; therefore
further description of the technique will be based on stereotactic
lesioning and neurotransplantation in adult and neonatal animals.
Fig. 3 (a–f) Following quinolinic acid (QA) lesion, sections from the control (a–c) and the lesion (d–f) groups
are typically stained with a selection of markers specific for neurones (NeuN; a, d), dopaminergic afferents (TH,
tyrosine-hydroxylase; b, e), and medium spiny neurones (DARPP-32; c, f). The control sections show regular
staining, and no anatomical deformation. However, lesioned sections have enlarged ventricles, collapsed
axons of passage, shrunk striatal tissue, a necrotic core, and exhibit a general reduction of staining in the stria-
tum. Ctx cortex, CC corpus collasum, V ventricle, AC anterior commisure. Scale bar = 1 mm
9 General Stereotactic Surgical Procedure in Adult and Neonatal Rats
In the following paragraphs, the general stereotactic surgical

procedure in adult and neonatal rats are explained.
10 Materials
The choice of experimental animal, its gender and age, will depend
entirely on the experimental objective of the investigator. Typically,
within a given investigation, surgical procedures are performed on
young adult rats (200–250 g at the start of the study) housed
under 12-h light/12-h dark conditions, temperature controlled
facilities (22–24 °C), with food and water ad libitum, and up to
five rats per cage. We advise, following the fixation of the experi-
mental animal in the stereotactic apparatus, in order to optimize
the precision of procedures to perform all steps under an operating
microscope such as the SMED-Studer Medical, Engineering-AG,
Switzerland Yasargil System, VM-900 as used in our laboratory.
All experiments must be carried out according to guidelines

and regulations of the relevant local and national authorities, and
it is the direct responsibility of the PI to be aware of these
responsibilities.
10.1 Equipment Disposable scalpel No. 10 (Feather Safety Razor, Japan), Wullstein
retractors (No. 17018-11), adson forceps (No. 91106-12), MORIA
forceps (Straight, No. 11370-40) MORIA forceps (Curved, No.
11370-42), Micro-Mosquito (Straight, serrated, No. 13010-12),
Hartman Hemostatic forceps (No. 13002-10), Michel suture clips
(No. 12040-02), applying forceps for Michel suture clips (No.
12018-12), Ear punch for animal identification (No. 24210-02)
from FST (Fine Science Tools GmbH, Germany) catalog, a bone
scraper and “surgical hooks” made from clipped and bent needles.
small animal stereotactic apparatus, i.e., Stoelting stereotactic frame
no. 51600 and Cunningham neonatal rat adaptor no 51625
(Stoelting, USA) [49], high speed microdrill: Proxxon Micromot
40 with small dental drill bits (Proxxon, Germany), 5 μl-calibrated
borosilicate glass capillaries (i.e., BF100-50-7.5, Sutter, USA),
micropipette puller (Sutter P-97, Sutter, USA), Hamilton microliter
syringes 2 and 10 μl (Hamilton Europe, Switzerland), operating
microscope (SMED-Studer Medical, Engineering-AG, Switzerland
Yasargil System, VM-900). Isoflurane-vaporizer for neonatal surger-
ies, micro pump (World Precision Instruments Inc., UK), cotton
swabs, a microdrill holder attached to the arm of the frame.
10.2 Reagents Ethanol 70 % vol/vol, anesthetics and analgesics (isofluran, ketamine,

xylazine, lidocaine, buprenorphine), lubricant eye ointment (i.e.,
Dexpanthenol), sterile PBS, neuotoxins (6-OHDA, QA) for lesion-
ing or cell suspension for neurotransplantation approaches, Borgal
anibiotic solution 24 % (Sulfadoxin + Trimethoprim) (Intervet,
Germany). When choosing the anesthetic, the investigator needs to
be aware that some products might interfere with the action of the
neurotoxin. For example, ketamine, a noncompetitive NMDA antag-
onist, has shown to mitigate the lesion induced by QA, a glutamate
analog acting on the NMDA receptor, as the two products have the
same principle target [50].
10.3 Surgical Prior to procedure the surgical area has to be cleaned and disinfected
Procedure with 70 % ethanol and the tools used either for lesioning or stereo-
tactic intracerebral cell implantation should be sterilized by autoclav-
ing or immersion in the ethanol solution (and then air dried).
10.3.1 Anesthesia Using a vaporizer (Fig. 4a), the isoflurane solution is converted into
its gaseous form and delivered to the animal by O2 into an induction
box (Fig. 4b) with isoflurane at a gas flow rate of approximately
5.0 vol% with 1.5 l O2/min. It is essential that once the animals are
induced—but still waiting for the surgery—the percentage of
isoflurane is reduced from 5 vol% to around 3 vol% as keeping the

animals too long on the higher percentage will result in possible
severe respiratory impairment and significantly increase a periopera-
tive risk of. In case of adult animals, standard intraperitoneal injec-
tion of 10 mg/kg i.p. Ketamine hydrochloride (Ketamine® 10 %
Essex Pharma GmbH, Germany) and 5 mg/kg i.p. Xylazine hydro-
chloride (Rompun®, Bayer AG, Germany) have been chosen. In our
experience, it has never been necessary to neither intubate the rats
nor to control the blood gases and body temperatures. For perform-
ing the QA intrastriatal injections, it is important to use inhalation
anesthesia, i.e., with isofluran, usually induced at 5.0 vol% with 1.5 l
O2/min, and maintained at approximately 2.0 vol% with 0.8–1.0 l
O2/min, due to significant reduction of QA excitotoxic effect on the
rat striatum when using ketamine, as described earlier [50]. During
the surgeries performed on neonatal rodents, hypothermia has been
the method of choice for reliably anesthetizing animals up to the
eighth day of age. For this purpose, neonatal rats have been covered
with crushed ice for approximately 5–7 min depending on size of
the animal (usually 1 min/g body weight suffices). Following the
induction the hypothermic anesthesia can be safely maintained for
up to 30 min by adding 10 g of dry ice every 10 min to a 50 % etha-
nol bath into the reservoir of the Cunningham neonatal rat adaptor
no. 51625 (Stoelting, USA) [49]. This will maintain the tempera-
ture of the instrument at approximately 5 °C. For longer surgical
procedures, the animal should be removed from the stereotactic
apparatus after 20–30 min and warmed up to the point of slight
responsiveness to nociceptive stimuli (i.e., a pinch of the tail or paw).
Afterwards, the hypothermia can be induced again as described
above and the animal repositioned in the apparatus for the next
phased of the surgery. For surgery in older neonates (beyond P10),
the animals have been anesthetized with isoflurane/oxygen-ventila-
tion: 2.5 vol% of isoflurane given with 3 l O2/min. Some authors
advocate for administration of 0.02 mg/kg body weight of atropine
given subcutaneously prior the anesthesia to reduce bronchial secre-
tion and improve breathing.
The effectiveness of the induction and depth of anesthesia can
be monitored by the responsiveness of the animals to nociceptive
stimuli, as described above.
Fixation In anesthetized animals, the fur on the skull has to be shaved and
of the Experimental Animal the skin disinfect with 70 % ethanol solution. Afterwards, one ear
in the Stereotactic bar of the apparatus should be fixed in the stereotactic frame
Apparatus (Stoelting stereotactic frame no. 51600) and the animal’s head
should be gently positioned so, that the ear canal is lead onto the
fixed ear bar. Keeping the head of the animal without changing
position the second ear bar should be introduced into the ear canal
to complete the fixation (Fig. 4c, d). It is important to apply only
moderate pressure and nonrupture ear bars with wide angle-tip in
Fig. 4 (a–d) Liquid Isoflurane is converted by O2 to its gaseous form in the vaporizer (a) and carried to the
induction box (b) by plastic tubing. The principal components of a stereotactic frame are two ear bars that
move laterally and a tooth bar that is moveable backwards and forwards; the adjustable arm to which the drill,
the lesioning, or the transplantation equipment can be attached is not depicted here (c). The head of the anes-
thetized animal is fixed in the frame using the ear bars and the tooth bar with the nose clamp gently fastened
(d). More detailed description can be found in the text
order to avoid injuring of the tympanic membrane. In case of con-

tinuation of the gas anesthesia, as soon as the head is fixed in the ear
bars, the gas inlet needs to be secured close to the animal’s nose to
ensure that the required level of anesthesia is maintained. Correct
fixation positions the head of the animal horizontal and symmetrical
to the ear bars as shown in Fig. 4d and enables its free vertical
movement precluding at the same time movements lateral to the
ear bar axis. Further, the lower jaw of the animals should be gently
pull down with small forceps to allow the incisor adapter (tooth
bar) be introduced into the animal’s mouth deep enough to place
the incisors in the opening of the adapter. After that, we suggest to
pull the adapter slightly backward to exert traction on the animal’s
head, which significantly improves the stability of fixation. For tar-
geting striatum either for neurotoxin injections (QA HD model or
terminal 6-OHDA lesion) or for cell implantation tooth bar is usu-
ally set at 0.0 mm. In case of MFB lesion, a “flat skull position” with
tooth bar set at +3.4/−2.3 mm for the first and second trajectory,
respectively, is necessary. The final step of mounting the animal into
the stereotactic apparatus is the fixation with the nose clamp applied
with a very low pressure on the animal’s nostrils. In some cases,
especially when working with very young and small animals the
nose clamp could be completely omitted. Finally, in order to pre-
vent the obstruction of the upper respiratory tract of the animal and
secure unproblematic breathing throughout the procedure the
tongue should be pulled out and aside.
10.3.2 Craniectomy, After proper fixation of the animal in the stereotactic apparatus, we
Coordinates, routinely apply the operating microscope for further steps of the
and Stereotactic Injection procedure to maximize precision. Using scalpel a midline incision of
1–2 cm exposing the bregma and lambda as anatomical landmarks
should be made (Fig. 5a). The subcutaneous tissue should be care-
fully removed using a small bone scraper and margins of the wound
should be retracted leaving the skull exposed. It is important to keep
the skull moist with sterile PBS throughout the surgery, as men-
tioned above. The horizontal position of the skull depends on the
position of the tooth bar and differs according to performed proce-
dure. The most critical step for calculating the coordinates is the
proper measuring of the x and y coordinates of the bregma. For this
purpose, the tip of the Hamilton syringe or the tip of the drill bit,
mounted to the holder arm of the stereotactic frame has to be low-
ered to the level of the skull pointing the intersection of the coronal
and sagital sutures (Fig. 5b and inset). The coordinates of this point
can be read from the x (anterior–posterior, AP) and y (mediolateral,
ML) arms the frame. To calculate the coordinates of the cannula
entry point for further craniotomy, the coordinates of skull entry
point, as determined from a stereotactic brain atlas, has to be added
to the coordinates of the bregma. The standard coordinates for ste-
reotactic targets used in our laboratory are listed in Sect. 11. In the
next step, the skull over the target area is going to be thinned using
the high speed drill, usually leaving a thin bonny lamella through
which blood vessels and the dura are visible. It is important not to
drill through the bone, as it would probably cause an injury to the
surface of the brain. To remove the last part of the skull, we use self-
made “surgical hooks” from clipped and bent 27G needles enabling
the elevation of the carefully perforated edges of the craniotomy and
their removal with fine forceps. Afterwards, applying the same “sur-
gical hooks” very careful perforation of the dura is going to be per-
formed. An alternative to holding the drill in the hand is to have it
attached with an adaptor to a stereotactic arm using the tip of the
drill bit to locate the bregma, measure out the appropriate
Fig. 5 (a, b) The exposed skull reveals the skull plates that join up at the bregma (at the intersection of the dots,
a). The bregma is used as the point of reference for the anterior–posterior and the medial–lateral coordinates.
If using a fixed drill with a fine drill bit, the burr holes at the required coordinates can be measured out with the
drill (b, and inset)
coordinates, and make the burr holes (Fig. 6a). Further steps of the
stereotactic procedure depend on the type of surgery. In case of the
injection of neurotoxins for either PD or HD model, a 10-μl
Hamilton syringe with a 30-gauge steel cannula, mounted to the
holder of the stereotactic apparatus is going to be applied. After fill-
ing the Hamilton syringe with the neurotoxin solution, the tip of the
attached needle is lowered to the level of the dura, which is a refer-
ence for the z-axis, i.e., the DV coordinate of the target. Afterwards,
the needle is slowly introduced into the brain parenchyma and a
deposit of the neurotoxin is injected with an injection rate of approx-
imately 2 μl/min, although this is a parameter that depends on the
discretion of the investigator. Using a minipump system for the
toxin injection is an option that can improve consistency (Fig. 6b–f).
The cannula should be then held in place for 3 min before retraction
to prevent a retrograde flux of the neurotoxin along the trajectory
canal. Preparation of 6-OHDA is described in detail in Sect. 11 of
the chapter. To prevent oxidation, 6-OHDA solution needs to be
kept in dark on ice being made up fresh from powder after every 3 h
of surgery. The cannula needs to be reloaded with fresh toxin after
each animal. Similar to the 6-OHDA, details concerning the prepa-
ration and handling of QA is described in Sect. 11. QA is more sta-
ble then the 6-OHDA solution, nevertheless similar precautions are
taken such as protecting it from light and keeping it on ice. QA can
be made up and aliquoted in units of 50 μl up to 12 months in
advance if stored at −20 °C. If kept on ice, a single aliquot can be
used for an entire lesioning session but then must be disposed of. To
ensure consistent toxin quality throughout the day, the lesion can-
nula needs to be reloaded between each animal.
10.3.3 Transplantation Implantation of the cell suspension differs in some steps signifi-
cantly from described above, standard lesioning procedure. One of
the most critical phases is the preparation of the tissue for grafting.
Depending on the experimental paradigm graft can be composed of
pieces of the tissue of interest or be prepared as a cell suspension
[51]. The latter requires usually enzymatic and mechanical dissocia-
tion of the tissue/cell culture. The types of enzymes, length of
incubation, and subsequent mechanical separation of cells depend
strictly on the cell type and usually have to be determined empiri-
cally prior to implantation, and the reader needs to refer to key
publications (for example, [15]). Due to the relatively low rate of
cell survival following the stereotactic implantation, especially in
case of dopaminergic precursors it is important to monitor the via-
bility of the single cell suspension, i.e., according to standard Trypan
blue exclusion method or using automatic cell counters. This
parameter seems to be critical for the survival of grafted cells, so
that the viability of the sample amenable for transplantation in our
laboratory must not be lower then 90–95 %. After the counting, the
cells are resuspended in a desired volume of the transplantation
Fig. 6 (a–f) The adjustable arm attached to the stereotactic frame can accommodate the drill (a), as well as
other instrument. In the case of QA lesions, a micropump (b) is used to exert precise pressure onto the plunger
of a 10 μl Hamilton syringe (c) which has a 280 μm thick (internal diameter) polythene tube filled with saline.
The 30 gauge lesioning cannula that penetrates the brain is attached to the adjustable arm (d, e). To ensure
precision during the all surgical procedure, the use of a microscope is recommended (f)
solution (usually cell culture medium in case of serum-free growth

conditions or Hank’s balanced salt solution—HBSS—for cells cul-
tivated in serum supplemented medium) to the final density. In
order to prevent a reaggregation of the cells that may cause a plug-
ging of the syringe precluding the reproducible implantation of cell
deposits 0.05 % DNase is routinely added to the transplantation
solution in our laboratory. Standard cell implantation procedure
requires a 2 μl Hamilton microsyringe with 26-gauge steel cannula.

After performing a craniectomy as described above, the cell suspen-
sion is slowly drawn up to fill the syringe with the desired volume
plus 10 % of the total syringe capacity. Following that the syringe is
lowered to the level of dura and the DV coordinates of the target
are calculated. After perforating the dura with the bevel of the small
needle syringe, cannula is slowly introduced into the recipient’s
brain to reach the required vertical coordinate and left at this posi-
tion about 1 min before starting injection. In the next step, the
desired volume of the cell suspension is injected at a rate of approxi-
mately 0.5 μl/min and the needle is left thereafter in place for addi-
tional 2 min prior the careful withdrawal. If more than one cell
deposits is to be placed at different depths along the same trajec-
tory, the deepest one should be injected first, followed by the next
deepest, etc. [52].
In order to minimize the brain trauma inevitably caused by the
cell implantation and to allow the injection of small graft deposits
ranging 50–500 nl in a reproducible manner a microtransplantation
approach can be applied [53, 54]. The most crucial modification
represents the introduction of the glass capillary connected to the
end of blunt-end of the steel Hamilton syringe cannula using a cuff
made of the polyethylene tubing (Fig. 7a, b [52, 55]). The glass
micropipette of a desired diameter is prepared from the borosilicate
glass capillary using a pipet puller (Sutter P-97). The temperature
and time settings are usually empirically determined to obtain a
micropipette having a long (8–10 mm) slowly tapering shank with
final tip diameter of 50–75 μm. The tip of the pipette must be bro-
ken square at the level of the desired inner diameter, which is done
Fig. 7 (a, b) Cells can be introduced into the brain using either regular or microtransplantation method. The tips
of the regular metal Hamilton cannula (left side, 500 μm outer diameter) and the glass capillary (right side,
50–70 μm) are depicted with 1.0 μl of medium being extruded (a). The microtransplantation instrument con-
sists of a 2 μl Hamilton microsyringe fitted with the glass capillary using a cuff of polyethylene tubing as an
adapter (b). Scale bar = 500 μm
easily under dissecting microscope. Therefore, the tip of the capil-

lary can be tailored to accommodate various types of cell suspen-
sions. After assembling of the whole system, it has to be completely
filled with fluid (e.g., sterile saline or implantation medium) and
devoid of air collections by removing the plunger and backfilling
the syringe to eliminate dead space. A special care must be taken no
to damage a very fragile tip of the glass capillary and to keep its
lumen open. For the latter, any contact of the tip with the blood
should be avoided. Following retraction from the brain the capillary
should be immediately rinsed with the implantation medium to
counteract the aggregation of the cell in the tip. This microtrans-
plantation approach has significantly improved the grafting proce-
dure in the iso- (e.g., [55, 56]), allo- (e.g., [57]), and xenografting
[58] paradigms in the rat model.
At the end of the session, the steel cannula or the glass capillary
is gently removed, the exposed area carefully rinsed with the sterile
saline solution. While still in the earbars, the scalp wound is closed
either with the wound clips (adult animals) or 7-0 nylon suture
(newborn rats). Antibiotic ointment on the wound can be added
to prevent postoperative infection.
10.3.4 Neonatal Implantation of cells into a neonatal recipient is routinely per-

Transplantation formed using the microtransplantation technique, but requires
some modifications of the procedure. As already mentioned, neo-
natal animals are to be operated under hypothermic anesthesia
using a Cunningham neonatal rat adaptor (Stoelting, USA)
(Fig. 8). Like as for adult rats, the midline skin incision should be
made before fixation in the stereotactic apparatus and the skin,
together with connective tissue should be pull downward at both
sides of the skull to expose the premature external acoustic meatuses
consisting of delicate tube-like membraneous part connecting the
earbud to its cartilaginous part located vetral and anterior to
the transverse and occipital sinuses easily seen through the skull.
The tips of the ear bars should be gently inserted into the cartilagi-
nous external acoustic meatus until the resistance is felt. Please
note that applying an excessive force may distort the animal’s head
indicated by the disappearance of the blood, particularly form the
transverse sinus, due to the sinus constriction. Next, the head is
mounted in the apparatus by inserting the mouthpiece and tight-
ening the nose clamp. Finally, the head positioning such as the
bregma and lambda have the same vertical coordinate, and the
points 3 mm on either side of the lambda are localized on the same
horizontal plane, has to be accomplished. The calculation of the
coordinates of the site of interest does not differ significantly from
the one described for adult recipients. However, due to the fact
that the cranial sutures used to determine the bregma and lambda
point are less distinct in newborns than those in adults, the
Fig. 8 The Cunningham adapter shown with a neonatal rat. The adapter is used
to allow the stereotactic intervention on neonatal rodents too small to operate on
with the adult setup
consistence in choosing the reference point is absolutely essential

and needs practice. Further steps of cell implantation are per-
formed similarly to the procedure described for adult rats.
Key Recommendation Independently of the applied stereotactic procedure, i.e., either

Relating to Lesion neurotoxic lesion or transplantation using the standard or micro-
and Grafting Surgery transplantation approach in neonatal or adult animals, before per-
forming an experimental intervention the investigator should carry
out the following preparative steps:
1. Determine the lesion or transplantation coordinates of the tar-
get using an appropriate standard stereotaxic atlas.
2. Carry out pilot runs of the lesion, and or transplantation to ensure
correct preparation of the toxin, the cells, and to validate the ste-
reotactic coordinates used. This is particularly important when
working with neonatal animals. Lesion coordinates can also be
practiced by injecting a dye, ink, or Trypan blue. After sacrificing
of the animal, the brain can be sectioned to visualize the lesion
sites, the dyes, or the implanted cell deposits. Any misplacements
of the lesion or graft can be then easily measured and corrected.
10.3.5 Postoperative In general, the postoperative complications occur rarely after ste-
Care reotatic procedures in adult experimental animals, particularly in
rodents. After the surgery rats have to be kept warm. In our insti-
tution, we apply routinely a heat lamp during the recovery from
the anesthesia. Furthermore, the breathing pattern of the animal
should be carefully observed. By respiratory arrest in many cases, a
successful resuscitation can be performed, though this problem
appears to be more common in neonatal rats during the rewarming
period rather than in adults. Another important issue of the neona-
tal surgery is maternal neglect that can be prevented to certain
extent by proper preoperative handling of the animals. Special
attention has to be paid to adequate analgesic treatment after the
surgery. We use routinely 0.05 mg/kg body weight of buprenor-
phine (Temgesic) applied subcutaneously with the first injection
prior to regaining consciousness. Additionally, to avoid postopera-
tive dehydration 30 ml/kg body weight of sterile saline should be
injected as a subcutaneous deposit. The sufficient food intake dur-
ing the first phase of the recovery should be facilitated using moist
food pellets put on the dishes inside the animal’s cage for easy food
access. The clinical status of the operated animals should be closely
monitored with special attention to any signs of distress.
11 Notes
1. Typical parameters for the 6-OHDA lesion: Preparation, doses,

coordinates
The lesion coordinates are set according to bregma as a refer-
ence point for the AP and ML coordinates and the dura as refer-
ence for the DV coordinate, using the rat brain atlas Paxinos
and Watson 1998.
Powder form of 6-OHDA is weighed out and stored in
1.5 ml Eppendorfs in the fridge prior to use. When required for
the lesion, appropriate amount of the toxin is freshly made up
with 0.2 % ascorbic acid to a working dilution of 3.6 μg/μl. The
solution is protected from light and kept on ice.
2. Medial Forebrain Bundle:
Typical dose : 3.6 μg/μl 6-OHDA in saline containing 0.2 %
(w/v) ascorbic acid
Track # TB (mm) AP (mm) ML (mm) DV (mm) Volume (μl)

1 −2.3 −4.4 −1.2 −7.8 2.5
2 +3.4 −4.0 −0.8 −8.0 3
The injection rate should be 1.0 μl/min and the cannula is kept in
place for an additional 4 min before it is slowly retracted.
3. Terminal/partial DA lesions, injected into the striatum:
Typical dose: 3 × 7 μg/μl (3.6 μg/μl 6-OHDA in saline con-
taining 0.2 % (w/v) ascorbic acid)

1 0.0 +1.0 −3.0 −5.0 2
2 0.0 −0.1 −3.7 −5.0 2
3 0.0 −1.2 −4.5 −50 2
The injection rate should be 1.0 μl/min and the cannula is kept in
place for an additional 4 min before it is slowly retracted.
4. Typical parameters for the QA lesions: Preparation, doses,
coordinates
Preparing the toxin at the appropriate pH is essential, as this
ensures the complete resolution of the toxin which is generally
purchased in powder form. Measuring the pH is done using
litmus paper, and if the agent is prepared in a too small volume,
one can dip a needle tip into the toxin and spot the needle
onto the litmus paper directly.
Under typical circumstances a stock solution of 0.12 M QA
(molecular weight = 167.12) is prepared. The aim is to prepare
6.25 ml of stock solution using 125 mg of research grade QA.
Dissolve 125 mg of QA in 750 μl PBS (pH 7.4), add 50 μl of
10 M sodium hydroxide.
Sonicate the above solution for 15 min.
Add 3200 μl PBS. The total volume at this stage will be 4 ml,
and this is will permit the use of a pH meter.
Add 50 μl of 10 M sodium hydroxide to bring the solution to
pH 7.4; if pH needs to be adjusted use sodium hydroxide or
concentrated hydrochloric acid.
Add 2200 μl of PBS to obtain the required concentration of
0.12 M QA.
Check pH again, and if needed adjust to pH 7.4.
Aliquot 50 μl into Eppendorfs, label and store in freezer at
−20 °C.
If the required concentration is 0.09 M QA, then add 16.7 μl
of PBS to a 50 μl aliquot of 0.12 M QA. The stock can be
stored at −20 °C safely for 12 months; beyond this time point
a new batch should be made up.
The amount of QA injected is typically expressed either as “X”
number of deposits of “Y” μl each of “Z” M (molarity): for
example, four deposits of 0.2 μl of 0.12 M QA; or as “X” nmol
(molality): for example, 96 nmol QA. Each deposit is infused
with the micropump over 90 s, with 1 min between different
vertical deposits, and a 3 min wait prior to removal of the can-
nula from the brain to eliminate/ reduce lesion damage due to
toxin reflux.
Typical coordinates of QA striatal lesion:

1 0.0 +1.0 +2.9 −5.0/−4.0 0.2
2 0.0 −0.4 +3.3 −5.07−4.0 0.2
Under this two track and two deposits/track protocol a total of

four deposits of 0.20 μl (total of 0.8 μl) of QA is released into
the striatum.
References
1. Horsley V, Clarke RH (1908) The structure 12. Marshall JF, Ungerstedt U (1977)
and functions of the cerebellum examined by a Supersensitivity to apomorphine following
new method. Brain 31:45–124 destruction of the ascending dopamine neu-
2. Paxinos G, Franklin K (2012) The mouse rons: quantification using the rotational
brain in stereotaxic coordinates. Academic model. Eur J Pharmacol 41:361–367
Press, San Diego 13. Dunnett SB, Robbins TW (1992) The func-
3. Paxinos G, Franklin K (2006) The rat brain in tional role of mesotelencephalic dopamine sys-
stereotaxic coordinates. Academic Press, San tems. Biol Rev Camb Philos Soc 67:491–518
Diego 14. Dunnett SB, Björklund A, Schmidt RH, Stenevi
4. Cenci MA, Whishaw IQ, Schallert T (2002) U, Iversen SD (1983) Intracerebral grafting of
Animal models of neurological deficits: how rel- neuronal cell suspensions. V. Behavioural recov-
evant is the rat? Nat Rev Neurosci 3:574–579 ery in rats with bilateral 6-OHDA lesions fol-
5. Olanow CW (1993) A radical hypothesis for neu- lowing implantation of nigral cell suspensions.
rodegeneration. Trends Neurosci 16:439–444 Acta Physiol Scand Suppl 522:39–47
6. Blum D, Torch S, Lambeng N, Nissou M, 15. Dunnett SB (2010) Chapter 55: neural trans-
Benabid AL, Sadoul R, Verna JM (2001) plantation. Handb Clin Neurol 95:885–912
Molecular pathways involved in the neurotoxicity 16. Goto S, Hirano A, Matsumoto S (1989)
of 6-OHDA, dopamine and MPTP: contribu- Subdivisional involvement of nigrostriatal loop
tion to the apoptotic theory in Parkinson’s disin idiopathic Parkinson’s disease and striatoni-
ease. Prog Neurobiol 65:135–172 gral degeneration. Ann Neurol 26:766–770
7. Paxinos G, Watson C (2006) The rat brain in ste- 17. Carman LS, Gage FH, Shults CW (1991)
reotaxic coordinates. Academic Press, San Diego Partial lesion of the substantia nigra: relation
8. McGeer PL, Itagaki S, Akiyama H, McGeer between extent of lesion and rotational behav-
EG (1988) Rate of cell death in parkinsonism ior. Brain Res 553:275–283
indicates active neuropathological process. 18. van Oosten RV, Cools AR (2002)
Ann Neurol 24:574–576 Differential effects of a small, unilateral,
9. Amalric M, Moukhles H, Nieoullon A, 6-hydroxydopamine-induced nigral lesion on
Daszuta A (1995) Complex deficits on reac- behavior in high and low responders to nov-
tion time performance following bilateral elty. Exp Neurol 173:245–255
intrastriatal 6-OHDA infusion in the rat. Eur 19. Bentlage C, Nikkhah G, Cunningham MG,
J Neurosci 7:972–980 Björklund A (1999) Reformation of the
10. Kirik D, Rosenblad C, Björklund A (1998) nigrostriatal pathway by fetal dopaminergic
Characterization of behavioral and neurode- micrografts into the substantia nigra is criti-
generative changes following partial lesions of cally dependent on the age of the host. Exp
the nigrostriatal dopamine system induced by Neurol 159:177–190
intrastriatal 6-hydroxydopamine in the rat. 20. Nikkhah G, Cunningham MG, Cenci MA,
Exp Neurol 152:259–277 McKay RD, Björklund A (1995) Dopaminergic
11. Creese I, Burt DR, Snyder SH (1977) microtransplants into the substantia nigra of neo-
Dopamine receptor binding enhancement natal rats with bilateral 6-OHDA lesions.
accompanies lesion-induced behavioral super- I. Evidence for anatomical reconstruction of the
sensitivity. Science 197:596–598 nigrostriatal pathway. J Neurosci 15:3548–3561
21. Nikkhah G, Eberhard J, Olsson M, Björklund A Huntington disease victims. Proc Natl Acad
(1995) Preservation of fetal ventral mesence- Sci U S A 85:4079–4081
phalic cells by cool storage: in-vitro viability and 33. Beal MF, Matson WR, Swartz KJ, Gamache
TH-positive neuron survival after microtrans- PH, Bird ED (1990) Kynurenine pathway
plantation to the striatum. Brain Res 687:22–34 measurements in Huntington’s disease stria-
22. Gusella JF, Wexler NS, Conneally PM, Naylor tum: evidence for reduced formation of kyn-
SL, Anderson MA, Tanzi RE, Watkins PC, urenic acid. J Neurochem 55:1327–1339
Ottina K, Wallace MR, Sakaguchi AY (1983) A 34. Beal MF, Ferrante RJ, Swartz KJ, Kowall NW
polymorphic DNA marker genetically linked to (1991) Chronic quinolinic acid lesions in rats
Huntington’s disease. Nature 306:234–238 closely resemble Huntington’s disease. J Neurosci
23. (1993) A novel gene containing a trinucleo- 11:1649–1659
tide repeat that is expanded and unstable on 35. Beal MF, Matson WR, Storey E, Milbury P,
Huntington’s disease chromosomes. The Ryan EA, Ogawa T, Bird ED (1992) Kynurenic
Huntington’s Disease Collaborative Research acid concentrations are reduced in Huntington’s
Group. Cell 72:971–983. disease cerebral cortex. J Neurol Sci 108:80–87
24. Kieburtz K, MacDonald M, Shih C, Feigin A, 36. Ferrante RJ, Kowall NW, Cipolloni PB, Storey E,
Steinberg K, Bordwell K, Zimmerman C, Beal MF (1993) Excitotoxin lesions in primates
Srinidhi J, Sotack J, Gusella J (1994) as a model for Huntington’s disease: histopatho-
Trinucleotide repeat length and progression of logic and neurochemical characterization. Exp
illness in Huntington’s disease. J Med Genet Neurol 119:46–71
31:872–874 37. Roberts RC, Ahn A, Swartz KJ, Beal MF,
25. Brinkman RR, Mezei MM, Theilmann J, DiFiglia M (1993) Intrastriatal injections of
Almqvist E, Hayden MR (1997) The likeli- quinolinic acid or kainic acid: differential pat-
hood of being affected with Huntington dis- terns of cell survival and the effects of data analy-
ease by a particular age, for a specific CAG size. sis on outcome. Exp Neurol 124:274–282
Am J Hum Genet 60:1202–1210 38. Vazey EM, Chen K, Hughes SM, Connor B
26. Borrell-Pagès M, Zala D, Humbert S, Saudou (2006) Transplanted adult neural progenitor cells
F (2006) Huntington’s disease: from hunting- survive, differentiate and reduce motor function
tin function and dysfunction to therapeutic impairment in a rodent model of Huntington’s
strategies. Cell Mol Life Sci 63:2642–2660 disease. Exp Neurol 199:384–396
27. Deng YP, Albin RL, Penney JB, Young AB, 39. Döbrössy MD, Dunnett SB (2003) Motor
Anderson KD, Reiner A (2004) Differential loss training effects on recovery of function after
of striatal projection systems in Huntington’s striatal lesions and striatal grafts. Exp Neurol
disease: a quantitative immunohistochemical 184:274–284
study. J Chem Neuroanat 27:143–164 40. Döbrössy MD, Dunnett SB (2006) The effects
28. Nakamura K, Aminoff MJ (2007) of lateralized training on spontaneous forelimb
Huntington’s disease: clinical characteristics, preference, lesion deficits, and graft-mediated
pathogenesis and therapies. Drugs Today functional recovery after unilateral striatal
(Barc) 43:97–116 lesions in rats. Exp Neurol 199:373–383
29. Beal MF, Kowall NW, Ellison DW, Mazurek 41. Döbrössy MD, Dunnett SB (2006)
MF, Swartz KJ, Martin JB (1986) Replication Morphological and cellular changes within
of the neurochemical characteristics of embryonic striatal grafts associated with
Huntington’s disease by quinolinic acid. enriched environment and involuntary exer-
Nature 321:168–171 cise. Eur J Neurosci 24:3223–3233
30. Bordelon YM, Chesselet MF, Nelson D, Welsh 42. Döbrössy MD, Dunnett SB (2007) The corri-
F, Erecińska M (1997) Energetic dysfunction dor task: striatal lesion effects and graft-
in quinolinic acid-lesioned rat striatum. mediated recovery in a model of Huntington’s
J Neurochem 69:1629–1639 disease. Behav Brain Res 179:326–330
31. Ribeiro CAJ, Grando V, Dutra Filho CS, 43. Döbrössy MD, Svendsen CN, Dunnett SB
Wannmacher CMD, Wajner M (2006) (1996) Bilateral striatal lesions impair reten-
Evidence that quinolinic acid severely impairs tion of an operant test of short-term memory.
energy metabolism through activation of Brain Res Bull 41:159–165
NMDA receptors in striatum from developing 44. Brasted PJ, Döbrössy MD, Robbins TW,
rats. J Neurochem 99:1531–1542 Dunnett SB (1998) Striatal lesions produce
32. Schwarcz R, Okuno E, White RJ, Bird ED, distinctive impairments in reaction time per-
Whetsell WO (1988) 3-Hydroxyanthranilate formance in two different operant chambers.
oxygenase activity is increased in the brains of Brain Res Bull 46:487–493
45. Döbrössy MD, Svendsen CN, Dunnett SB tional coupling of the two nigrostriatal path-
(1995) The effects of bilateral striatal lesions ways. J Comp Neurol 432:217–229
on the acquisition of an operant test of short 53. Nikkhah G, Cunningham MG, Jödicke A,
term memory. Neuroreport 6:2049–2053 Knappe U, Björklund A (1994) Improved graft
46. Furtado JC, Mazurek MF (1996) Behavioral survival and striatal reinnervation by microtrans-
characterization of quinolinate-induced lesions plantation of fetal nigral cell suspensions in the
of the medial striatum: relevance for Huntington’s rat Parkinson model. Brain Res 633:133–143
disease. Exp Neurol 138:158–168 54. Nikkhah G, Olsson M, Eberhard J, Bentlage C,
47. Isacson O, Brundin P, Kelly PA, Gage FH, Cunningham MG, Björklund A (1994) A micro-
Björklund A (1984) Functional neuronal transplantation approach for cell suspension graft-
replacement by grafted striatal neurones in the ing in the rat Parkinson model: a detailed account
ibotenic acid-lesioned rat striatum. Nature of the methodology. Neuroscience 63:57–72
311:458–460 55. Nikkhah G, Rosenthal C, Falkenstein G,
48. Shear DA, Dong J, Gundy CD, Haik-Creguer Roedter A, Papazoglou A, Brandis A (2009)
KL, Dunbar GL (1998) Comparison of intrastria- Microtransplantation of dopaminergic cell sus-
tal injections of quinolinic acid and 3-nitropropionic pensions: further characterization and optimi-
acid for use in animal models of Huntington’s dis- zation of grafting parameters. Cell Transplant
ease. Prog Neuropsychopharmacol Biol Psychiatry 18:119–133
22:1217–1240 56. Hahn M, Timmer M, Nikkhah G (2009)
49. Cunningham MG, McKay RD (1993) A hypo- Survival and early functional integration of
thermic miniaturized stereotaxic instrument dopaminergic progenitor cells following trans-
for surgery in newborn rats. J Neurosci plantation in a rat model of Parkinson’s dis-
Methods 47:105–114 ease. J Neurosci Res 87:2006–2019
50. Jiang W, Büchele F, Papazoglou A, Döbrössy 57. Brandis A, Kuder H, Knappe U, Jödicke A,
M, Nikkhah G (2009) Ketamine anaesthesia Schönmayr R, Samii M, Walter GF, Nikkhah G
interferes with the quinolinic acid-induced (1998) Time-dependent expression of donor-
lesion in a rat model of Huntington’s disease. and host-specific major histocompatibility
J Neurosci Methods 179:219–223 complex class I and II antigens in allogeneic
51. Pruszak J, Just L, Isacson O, Nikkhah G dopamine-rich macro- and micrografts: com-
(2009) Isolation and culture of ventral mesen- parison of two different grafting protocols.
cephalic precursor cells and dopaminergic neu- Acta Neuropathol (Berl) 95:85–97
rons from rodent brains. Curr Protoc Stem 58. Steiner B, Winter C, Blumensath S, Paul G,
Cell Biol; Chapter 2:Unit 2D.5 Harnack D, Nikkhah G, Kupsch A (2008)
52. Roedter A, Winkler C, Samii M, Walter GF, Survival and functional recovery of trans-
Brandis A, Nikkhah G (2001) Comparison planted human dopaminergic neurons into
of unilateral and bilateral intrastriatal hemiparkinsonian rats depend on the cannula
6-hydroxydopamine-induced axon terminal size of the implantation instrument. J Neurosci
lesions: evidence for interhemispheric func- Methods 169:128–134
Chapter 4
Rat Middle Cerebral Artery (MCA) Occlusion Models Which

Involve a Frontotemporal Craniectomy
Hideaki Imai, Nobuhito Saito, and I. Mhairi Macrae
Abstract
In this chapter, we describe the technical approach for exposure and occlusion of the rat middle cerebral
artery (MCA), focusing mainly on proximal electrocoagulation of the MCA, the Tamura model. This
model requires training and expertise in microsurgical techniques so that the artery can be exposed and
occluded without damaging the underlying brain tissue. However, once the required skills are acquired, a
very reproducible ischemic insult can be produced with good recovery and low mortality.
Through extensive experience in the use of this model, we have modified the original Tamura model
to make the surgery more straightforward and less invasive. In this chapter, we describe the MCAO
procedure step by step, comprehensively noting the surgical preparation, body position, skin incision,
craniotomy, dural incision, diathermy of the MCA, and the prevention of infection. We have also included
a series of photographs of the surgical site at each step to facilitate training in the model.
Key words Rat, Middle cerebral artery occlusion, Focal cerebral ischemia, Rodent, Tamura model,
Diathermy, Electrocoagulation, Endothelin-1, MCAO
1 Introduction
Animal stroke models are indispensable for both the investigation

of the pathophysiology of cerebral ischemia and the evaluation of
preclinical pharmacological intervention [1]. Most rodent models
involve occlusion of the middle cerebral artery (MCAO) using
either an intraluminal approach (e.g., Koizumi model) [2] or by
exposure and direct surgical occlusion of the blood vessel (e.g.,
Tamura model) [3]. The most suitable experimental stroke model
to use in a given study depends on the scientific question being
investigated and each model has inherent advantages and disadvan-
tages. The Tamura model, for example, has a robust advantage in
terms of the reliability and reproducibility in the production of
ischemic lesions in the ipsilateral middle cerebral artery (MCA)
territory including the frontal cortex, dorsal parietal cortex, and
the lateral part of the caudate–putamen. However, presumably due
Miroslaw Janowski (ed.), Experimental Neurosurgery in Animal Models, Neuromethods 116 vol. 116,
55
56 Hideaki Imai et al.
to the requirement for substantial microsurgical expertise, the use

of this model, and others that require surgical exposure of the
MCA, have become less prevalent compared to the intraluminal
filament model. The Tamura model induces permanent MCAO
while the intraluminal filament model can be adapted for either
permanent or transient MCAO. However, because of the craniec-
tomy, the Tamura model is associated with lower mortality than
the intraluminal filament and other intact skull models.
In this chapter, we provide a full description of the neurosurgi-
cal procedures for direct exposure of the MCA in the rat including
anesthesia, physiological monitoring, positioning, and neurovascular
anatomy. Details are provided for MCA occlusion using electroco-
agulation (proximal and distal MCA occlusion), endothelin-1
(ET-1), and mechanical devices (Fig. 1).
2 Animals
Sprague-Dawley rats are the strain of choice in the authors’ labora-

tories. Adult male rats (e.g., from Charles River, Tsukuba, Japan)
are maintained in controlled temperature (24 ± 1 °C) and humidity
(55 ± 5 %) under a 12-h light: 12-h dark cycle with free access to
rat chow and water (see Note 1).
3 Methods
3.1 General Surgical For all surgical procedures, anesthesia is induced with isoflurane
Preparation (4–5 %) and then subsequently maintained with 1.5–2 % isoflurane
in nitrous oxide:oxygen (70:30) on a facemask. The rats are then
intubated transorally (16 gauge intubation tube is suitable for rats
of 250–350 g) and artificially ventilated by a small animal respira-
tor pump with a tidal volume of 3–4 ml and respiration rate from
45 (surgical tracheostomy) to 60 (nonsurgical, oral intubation)
breaths per minute. The right femoral artery is cannulated for con-
tinuous physiologic monitoring (Fig. 2). Arterial pressure is moni-
tored throughout the experiment and arterial blood samples are
taken at regular intervals for assessment of respiratory status using
a direct-reading electrode system (Bayer, Newbury, Berkshire,
UK). Rats are maintained normotensive (MABP > 80 mmHg),
normocapnic (36 < PaCO2 < 44 mmHg), and adequately oxygen-
ated (PO2 > 100 mmHg) while under anesthesia. Rectal tempera-
ture is maintained at 37 °C with a heating lamp or homeothermic
blanket during the operation.
3.2 Surgical Good aseptic technique is essential in experimental stroke surgery

Exposure of the MCA and efforts should be made to ensure the operating area is as clean
and sterile as possible. All surgical instruments should be sterilized
Permanent Middle Cerebral Artery Occlusion in Rats 57
Fig. 1 There are several ways to occlude the MCA—by using neurosurgical pro-
cedures such as electrocoagulation (proximal (a) and distal (b) MCA occlusion),
endothelin-1 (c), and mechanical devices (d)
Fig. 2 The right femoral artery is cannulated for continuous physiologic monitoring.
If needed, the femoral vein is also available for intravenous injection
prior to use and laid out on a sterile drape beside the animal. Sterile
swabs, sutures, saline, etc., are employed. The use of a surgical
operating microscope is recommended for the entire surgical
procedure and the craniectomy site should be irrigated frequently
with sterile saline or artificial CSF (see Note 2).
3.3 Position of the The animal is placed in the right lateral position (Fig. 3), raising
rat on the operating the head 20° from the surgery table to allow a left frontotemporal
table approach (see Note 3). In order to protect the left eye, the eyelid
should be closed by a suture or tape (Fig. 4). Fur around the
planned skin incision should be shaved using an electric hair clipper
and then a skin antiseptic solution applied (Fig. 4). Administration
of a local anesthetic (1–2 mg/kg ropivicaine, bupivacaine) subcu-
taneously (line block) to the wound site prior to skin incision is
recommended.
Fig. 3 The animal is placed in the right lateral position, raising the head 20° from
the surgery table to allow a left frontotemporal approach
3.4 Skin Incision A 1.5 cm vertical skin incision is performed between the left eye-
and the Way ball and ear auricule, using electrocoagulation to stem any bleed-
to Approach the Skull ing (Fig. 5). The temporal fascia and muscle are incised just under
Base the skin incision from the zygomatic arch to the linear tempolaris
(top of the temporal muscle) using bipolar forceps for cutting with
coagulation. If the surgical approach is correct, the coronoid pro-
cess of the mandible (Fig. 6) is a good landmark and should emerge
surrounded by the temporal muscle. After dissecting the temporal
muscle, the coronoid process is taken away. The zygoma is not
rongeured away. After splitting the temporal muscle and reflecting
it to the rostral and caudal side, the surgical field is open to observe
the temporal skull, root of zygoma, and zygomatic arch.
Fig. 4 The eyelid should be closed by a suture or tape in order to protect the left
eye. Fur around the planned skin incision should be shaved using an electric hair
clipper and then a skin antiseptic solution applied
3.5 Approach The thin membrane is penetrated and dissected from the temporal
to the Middle Fossa: skull base. Then just along the temporal muscle under the zygo-
Expose the matic arch, the mandibular nerve is visible from the foramen ovale.
Frontotemporal Skull The craniotomy point of the temporal skull base between the
foramen ovale and orbital fissure can now be identified (Fig. 7).
3.6 Craniectomy A single entry burr hole is made with a dental drill. The temperature
is controlled by irrigation with sterile saline solution to keep the
dura matter intact. Micro forceps are used to expand the burr hole
and to perform the frontotemporal craniectomy (Fig. 8). The lat-
eral part of the temporal bone and temporal skull are removed
with rongeurs. Complete removal of the bone ridge facilitates
access from the proximal end of the MCA in the basal cistern to
Fig. 5 A 1.5-cm vertical skin incision is performed between the left eyeball and
ear auricule, using electrocoagulation to stem any bleeding. Then, the temporal
fascia and muscle are incised just under the skin incision from the zygomatic
arch to the linear tempolaris (top of the temporal muscle) using bipolar forceps
for cutting with coagulation
the distal end of the MCA where it crosses the inferior cerebral
vein (ICV).
The dura matter and arachnoid membrane are opened by
perforating with a fine needle and retraction, exposing the full
visualized brain within the craniectomy. CSF is released, thereby
producing further brain exposure.
3.7 Exposure of MCA The olfactory tract and ICV must be exposed as landmarks to
from Proximal definitively identify the MCA. Distally, the ICV crosses the MCA
to Distal Extent [at an angle of ~90°] and proximally the branching arteries of the
MCA such as lenticulo striate artery (LSA) (Fig. 9).
Fig. 6 After splitting the temporal muscle, the coronoid process of the mandible
emerges surrounded by the temporal muscle and is a good landmark. After dis-
secting the temporal muscle, the coronoid process is taken away
3.8 Permanent Focal Permanent focal cerebral ischemia is accomplished by occlusion of

Cerebral Ischemia the MCA (see Note 4), as introduced by Tamura et al. [3] with
some modification [4] as follows: For electrocoagulation of the
3.8.1 Electrocoagulation main trunk of the MCA, the forceps are delicately inserted under
of the MCA: Proximal the MCA at the olfactory tract. To facilitate this, you should first
MCAO (Fig. 1a) electrocoagulate the proximal perforating arteries and lateral stri-
ate arteries of the MCA with short bursts of current using the
diathermy forceps. This will provide space to then place one side of
the diathermy forceps between the MCA stem and brain surface,
the other side on the surface of the MCA stem (see Notes 5 and 6).
After gently clamping the MCA, to reduce the blood flow
through the blood vessel, electrocoagulation is exerted for a few
seconds. This procedure is repeated several times until the blood
flow is completely stopped. Then, using a segmental approach, the
Fig. 7 To approach the middle fossa, the thin membrane is penetrated and
dissected from the temporal skull base. The craniotomy point of the temporal
skull base between the foramen ovale and orbital fissure can now be identified
main trunk of the MCA is electrocoagulated starting from the

proximal MCA and working your way distally till you reach the
intersection of the MCA with the ICV (Figs. 9, 10, and 11). The
MCA is then transected at the level of the olfactory tract to ensure
the completeness of the occlusion. Sham-operated controls
undergo the same procedure to expose the MCA but the artery
and its branches are not occluded.
3.8.2 Electrocoagulation The model can be modified to reduce the size and location of isch-
of the MCA: Distal MCAO emic tissue by applying electrocoagulation to a small, more distal
(Fig. 1b) portion of the MCA. For example, a short (2 mm) occlusion and
transection, just distal to the ICV, will spare the caudate–putamen
and confine ischemia to the cortex [5] (Fig. 1b). This variation is
Fig. 8 For the craniectomy, firstly, a single entry burr hole is made with a dental
drill. Secondly, microforceps are used to expand the burr hole and to perform the
frontotemporal craniectomy. At this stage, the MCA, ICV, and olfactory tract can
be visualized through the intact dura matter
less technically demanding and induces a milder neurological

deficit than proximal MCAO.
3.9 Transient Focal The peptide ET-1 is a potent vasoconstrictor of cerebral blood
Cerebral Ischemia vessels with a prolonged duration of action [6] capable of blocking
flow and inducing downstream ischemia. The model requires the
3.9.1 ET-1-Induced MCA same surgical approach as is used for the electrocoagulation mod-
Occlusion (Fig. 1c) els, and the exposed MCA can be transiently occluded by topical
application of ET-1 (Fig. 1c). Once the dura has been opened and
the MCA exposed, a fine (30 gauge) sterile needle is used to
puncture the arachnoid membrane at several points on either side
of the blood vessel to improve peptide access. ET-1 (25 μl of 10−7
to 10−4 M) is then topically applied to constrict the artery
Fig. 9 After the dura matter and arachnoid membrane are opened, the MCA,
inferior cerebral vein (ICV), which crosses the MCA, the olfactory tract and the
branching arteries of the MCA such as lenticulo striate artery (LSA) are exposed
sufficiently to block blood flow (see Note 7). This can be confirmed
visually using the operating microscope. The higher the concentra-
tion of ET-1 applied, the more severe and prolonged the ischemia
and the larger the infarct, which has both a cortical and subcortical
component, similar to proximal MCAO [7]. As the effect of the
peptide wears off, the MCA diameter returns to normal and blood
flow is gradually reestablished.
3.9.2 Mechanical Mechanical occlusion of the MCA provides the flexibility to induce
Occlusion of the MCA permanent or transient occlusion of the main trunk of the MCA or
(Fig. 1d) its branches. Using mechanical devices such as microaneurysm
clips [8, 9] (Fig. 1d), hooks [10], and ligature snares [11], models
have been developed to induce focal ischemia (30 min to 2 h), fol-
lowed by reperfusion (see Note 8). Mechanical occlusion at a
Fig. 10 Electrocoagulation of the MCA is performed over the olfactory tract. Then,
using a segmental approach, the main trunk of the MCA is electrocoagulated
starting from the proximal MCA and working your way distally till you reach the
intersection of the MCA with the ICV
single point on the MCA can result in unacceptable variability in

lesion size. In order to limit this, MCAO is often combined with
uni- or bilateral common carotid artery occlusion or employed
specifically in strains such as the spontaneously hypertensive and
spontaneously hypertensive stroke-prone rat which have impaired
cortical collateral blood flow. Microaneurysm clips (e.g., Codman,
AVM micro clips with 10 g closing pressure) are small enough for
use in the rat and are loaded into a special applicator for attach-
ment to the MCA.
Mechanical occlusion can cause damage to the MCA with
resultant impairment in reperfusion when the occluding device is
removed. Recent modifications of the suture model have been
published [12], which limit direct damage to the occluded vessels
with the use of an occluding suture.
Fig. 11 Lower magnification microscopic view of Fig. 10 provides a good orienta-

tion of the surgical anatomy of the MCA. The transected MCA on the olfactory
tract can be identified in the center of the craniectomy
3.10 Closure Before closing the wound, the surgical field should be washed with
of the Surgical Wound quantities of sterile saline to prevent infection. Muscle and skin lay-
ers are sutured with 4-0 Vicryl (Johnson & Johnson, New
Brunswick, NJ) (Fig. 12). After surgery, a subcutaneous injection
of sterile saline (2.5 ml into each of two sites) is administered to
prevent post-anesthetic dehydration and should be repeated twice
a day until the animal is drinking normally Analgesia should also
be administered for the first 2–3 days after stroke surgery to
limit postoperative pain (e.g., carprieve, buprenorphine, and
paracetamol. Follow the recommendation of your local vet).
The eyelid suture is removed and anesthesia withdrawn to allow
the animal to recovery. Body temperature is maintained until the
rat is fully conscious and when spontaneous respiration returns, the
intubation tube is withdrawn and the rat returned to a clean cage
with softened rat chow and water.
Fig. 12 Muscle and skin layers are sutured with 4-0 Vicryl. The eyelid suture is
removed
3.11 Assessment Quantitative histopathology, employing light microscopic

of Ischemic Damage in examination of neuronal perikarya on multiple hematoxylin- and
the Rat eosin-stained sections (Fig. 13a), is an established technique for
the volumetric assessment of ischemic damage and drug effi-
cacy [13]. Areas of infarct are transcribed onto line diagrams of
coronal sections throughout the MCA territory (Fig. 13b)
Alternatively, 2,3,5-triphenyltetrazolium chloride (TTC) staining
can be used to detect ischemic damage [14]. TTC reacts with
intact mitochondrial oxidative enzyme systems to produce a red
colored formazan. The region where ischemia has damaged the
tissue, mitochondria remain uncolored and easily distinguishable
(Fig. 13c). More recently, magnetic resonance imaging is increas-
ingly used for the assessment of ischemia and ischemic damage.
Diffusion-weighted imaging is a sensitive and reliable modality for
detection of the ischemic injury in the acute phase (Fig. 13d) and
T2 scans provide images of the final infarct.
Fig. 13 Assessment of ischemic damage after MCA occlusion can be achieved on hematoxylin and eosin (H.E.)
stained sections (a), 2,3,5-triphenyltetrazolium chloride (TTC) brain slices (c), and MR imaging (d). Line
diagrams from a stereotaxic atlas of the rat brain can be used for the volumetric assessment of ischemic
damage (b). Infarct is represented by black shading
4 Notes
1. Selection of the strain of rat: MCA occlusion can be carried out

on any strain of rat but reproducibility in outcome measures
such as infarct size varies in different strains [15]. The authors
prefer to use Sprague-Dawley rats of 300–350 g and aged
10–12 weeks. In animals below 300 g the surgical field is rela-
tively small, making the approach more difficult and in animals
above 350 g the skull is noticeably thicker making the exposure
of the MCA more challenging. Reproducibility can also be
improved by using animals of a defined age and bodyweight.
2. To avoid any postsurgical infection, the surgical field should

be washed with copious amounts of sterile saline or artificial
cerebral spinal fluid. In our experience, no cases of infection
have occurred when this procedure has been followed.
3. The complexity of the Tamura electrocoagulation model is
almost exclusively attributed to achieving the correct anatomi-
cal orientation so that the craniectomy is made to reveal the
MCA with minimum invasiveness. Achieving the correct orien-
tation of the head for the surgical approach is crucial because of
the deep surgical field needed to expose the proximal MCA at
the skull base. It takes time and practice for the researcher to
learn the correct position of the rat, angle of the microscope,
and surgical approach.
4. Occlude the MCA on the same side of the brain in each animal
to limit variability in outcome measures.
5. For the Tamura model, the most critical point in the procedure
is learning how to electrocoagulate the MCA without rupturing
the blood vessel or injuring the underlying brain tissue. This
requires a fine and controlled technique that is more difficult
than electrocoagulation techniques used in human surgery
where an equivalent size of microvessel is easily coagulated. The
blood vessel walls of the rat MCA are much thinner than those
of the human and are electrocoagulated under mean arterial
pressures of approximately 80–90 mmHg.
6. Electrocoagulation of the rat MCA: (a) To ensure reproducible
ischemic lesions, the perforating artery from the MCA and the
LSA must be coagulated first when carrying out proximal MCA
occlusion. This ensures that the ischemic lesion includes the
dorsolateral caudate nucleus. Moreover, once this procedure is
complete, it enables the diathermy forceps to be inserted
between the MCA and cerebral surface to clamp the main trunk
of the MCA thereby facilitating a successful MCA occlusion by
reducing the blood flow through the artery; (b) keep the tips of
the diathermy forceps clean and polished so that the forceps do
not adhere to the MCA when the blood vessel is being electro-
coagulated; (c) once electrocoagulation has stopped the blood
flow through the MCA at the level of the olfactory tract, the
subsequent proximal to distal electrocoagulation of the main
MRC trunk is much easier to achieve.
7. For ET-1 induced transient MCAO, consistency in the potency
of ET-1 is very important to control variability. The peptide is
purchased in lyophilized form and should be made up to the
required concentration, aliquoted out into single use vials and
frozen at −80 °C. A fresh aliquot should be thawed for each
experiment and not refrozen.
8. Microaneurysm clips, ligatures, and sutures are applied to the

MCA under the magnification of an operating microscope. This
also facilitates visual inspection of the artery and/or its branches
after removal of the occluding device to ensure the return of
blood flow.
References
1. Macrae IM (1992) New models of focal cere- the rat middle cerebral arter y. Acta
bral ischaemia. Br J Clin Pharmacol Neuropathol 78:605–614
34:302–308 9. Buchan AM, Xue D, Slivka A (1992) A new
2. Koizumi J, Yoshida Y, Nakazawa T, Ooneda G model of temporary focal neocortical ischemia
(1986) Experimental studies of ischemic brain in the rat. Stroke 23:273–279
edema, I: a new experimental model of cere- 10. Kaplan B, Brint S, Tanabe J, Jacewicz M, Wang
bral embolism in rats in which recirculation X-J, Pulsinelli W (1991) Temporal thresholds
can be introduced in the ischemic area. Jpn for neocortical infarction in rats subjected to
J Stroke 8:1–8 reversible focal cerebral ischemia. Stroke
3. Tamura A, Graham DI, McCulloch J, Teasdale 22:1032–1039
GM (1981) Focal ischemia in the rat. Part I: 11. Shigeno T, Teasdale GM, McCulloch J,
description of technique and early neuro- Graham DI (1985) Recirculation model fol-
pathological consequences following middle lowing MCA occlusion in rats. Cerebral blood
cerebral artery occlusion. J Cereb Blood Flow flow, cerebrovascular permeability and brain
Metab 1:53–60 edema. J Neurosurg 63:272–277
4. Imai H, McCulloch J, Graham DI, Masayasu 12. Luo W, Wang Z, Li P, Zeng S, Luo Q (2008)
H, Macrae IM (2002) New method for the A modified mini-stroke model with region-
quantitative assessment of axonal damage in directed reperfusion in rat cortex. J Cereb
focal cerebral ischemia. J Cereb Blood Flow Blood Flow Metab 28:973–983
Metab 22:1080–1089 13. Bederson JB, Pitts LH, Germano SM,
5. Shigeno T, McCulloch J, Graham DI, Nishimura MC, Davis RL, Bartkowski HM
Mendelow AD, Teasdale GM (1985) Pure (1986) Evaluation of
cortical ischemia versus striatal ischaemia. Surg 2,3,5-triphenyltetrazolium chloride as a stain
Neurol 24:47–51 for detection and quantification of experimen-
6. Robinson MJ, McCulloch J (1990) tal cerebral infarction in rats. Stroke
Contractile responses to endothelin in feline 17:1304–1308
cortical vessels in situ. J Cereb Blood Flow 14. Osborne KA, Shigeno T, Balarsky AM, Ford I,
Metab 10:285–289 McCulloch J, Teasdale GM, Graham DI
7. Macrae IM, Robinson MJ, Graham DI, Reid (1987) Quantitative assessment of early brain
JL, McCulloch J (1993) Endothelin induced damage in a rat model of focal cerebral isch-
reductions in cerebral blood flow: dose- aemia. J Neurol Neurosurg Psychiatry
dependency, time course and neuropathologi- 50:402–410
cal consequences. J Cereb Blood Flow Metab 15. Duverger D, MacKenzie ET (1988) The
13:276–284 quantification of cerebral infarction following
8. Dietrich WD, Nakayama H, Watson BD, focal ischemia in the rat: influence of strain,
Kanemitsu H (1989) Morphological con- arterial pressure, blood glucose concentration,
sequences of early reperfusion following and age. J Cereb Blood Flow Metab
thrombotic or mechanical occlusion of 8:449–461
Chapter 5
Inferior Colliculus Approach in a Rat

Dennis T.T. Plachta
Abstract
The inferior colliculus (IC) of the rat is a well-investigated and understood model for topographical map-
ping of the frequency domain (Clopton et al., Exp Neurol 42(3):532–540, 1974; Kelly and Masterton, J
Comp Physiol Psychol 91(4): 930–936, 1977; Borg, Hear Res 8(2) 101–115, 1982; Ryan et al., Hear Res
36(2–3): 181–189, 1988; Zhang et al., Hear Res 117(1–2):1–12, 1998). As a central hub for binaural audi-
tory processing in the midbrain (Du et al., Eur J Neurosci 30(9): 1779–1789, 2009) it shows a variety of
response patterns to a given complex auditory stimulation (Kelly et al., Hear Res 56(1–2):273–280, 1991;
Kelly and Li, Hearing Res 104:112–126, 1997). It is therefore a major target for neuroscientific approaches
of the ascending and descending auditory pathway. Approaching the IC is, however, not only valuable for
scientists interested in auditory processing, but also for students learning the proceedings of standard elec-
trophysiological experimentation. In addition, engineers of biomedical devices (e.g., flexible penetrating
electrodes) can take benefit from the IC approach (Kisban et al., Conference proceedings: annual interna-
tional conference of the IEEE Engineering in Medicine and Biology Society IEEE Engineering in Medicine
and Biology Society Conference, 2007:175–178). A critical test of the suitability of shaft electrodes is their
successful implantation in vivo. The steady tonotopic structure of the IC and its three subdivisions provides
an almost perfect anatomic testing ground (Saldana and Merchan, J Comp Neurol 319(3):417–437, 1992).
Additionally, the anatomical procedure to access the IC requires only a medium level of surgical skills and
the testing apparatus can be kept relatively small and manageable. The current study describes the necessary
anatomical steps and materials needed for the aforementioned scenarios.
Key words Inferior colliculus, IC approach, Auditory, Mid-brain, Rat, Tonotopy
The inferior colliculus (IC) of the rat is a well-investigated and

understood model for topographical mapping of the frequency
domain [1–5]. As a central hub for binaural auditory processing in
the midbrain [6] it shows a variety of response patterns to a given
complex auditory stimulation [7, 8]. It is therefore a major target
for neuroscientific approaches of the ascending and descending
auditory pathway. Approaching the IC is, however, not only valu-
able for scientists interested in auditory processing, but also for
students learning the proceedings of standard electrophysiological
experimentation. In addition, engineers of biomedical devices
(e.g., flexible penetrating electrodes) can take benefit from the IC
approach [9]. A critical test of the suitability of shaft electrodes is
their successful implantation in vivo. The steady tonotopic
73
74 Dennis T.T. Plachta
structure of the IC and its three subdivisions provides an almost

perfect anatomic testing ground [10]. Additionally, the anatomical
procedure to access the IC requires only a medium level of surgical
skills and the testing apparatus can be kept relatively small and
manageable. The current study describes the necessary anatomical
steps and materials needed for the aforementioned scenarios.
1 Materials
Female Sprague-Dawley rats weighing between 250 and 300 g are

used. For acute experiments, Ketamine (10 %, 0.6 ml/kg body-
weight (BW)) and Medetomidine (0.3 mg/kg) are used for initial
anesthesia. If the experiments last more than 4 h, additional anesthe-
sia must be applied: use the same concentrations but at 1/5th of the
initial dose. Four hours after the initial application of the anesthetics,
a refresh has to be applied at least every second hour. Both medica-
tions can be applied subcutaneousely (s.c.) using a single syringe.
Especially for students and beginners in animal experimenta-
tion, a plexiglass tube (diameter: 8 cm, height: 20 cm, one end
closed and perforated for air flow) can be helpful to safely apply the
initial anesthetics. The rats like to crawl into the tube. The experi-
menter can then hold the tail and apply the anesthetic while the rat
is safely immobilized in the tube.
The anesthetized rat will cool down and dehydrate. To prevent
this and to maintain stable physiological parameters, ringer solution
(Braun, Ringerlösung 0.9 % NaCl) has to be applied every other hour
(1 ml s.c.). Additionally the body temperature has to be supported
using either a heat-lamp (e.g., Thermolux Pet-Mat 10 W, for experi-
ments <4 h) or a closed-loop heat pad (e.g., Animal Temperature
Controller ATC, WPI). For the opening of the skull a drill is neces-
sary (Proxxon, Mini-Mot 40). For optimal results, use a drilling head
Fig. 1 REM image of the head of a flexible polyimide electrode with a large application tip. This tip is used to
retract the insertion guide. Note the electrode contacts on the surface. Those are ideal for recording from dif-
ferent recording depth and in this approach will deliver different best frequencies. The bar indicates 100 μm
Inferior Colliculus Approach 75
Fig. 2 Image of the rat attached to the stereotactic device. Note that no metal is close to the opening wound
and the head is fixed to prevent respiration artifacts
with a ball (0.5 mm ball head). Opening of the skull with the drill and
positioning of the microelectrode should be performed using a dis-
secting microscope (e.g., Leica M60). For postsurgical fixation of the
head and for precise positioning of electrodes a stereotactic device is
strongly recommended (e.g., Stoelting, AnyAngle Stereotaxic
Instrument). Finally, the penetration of the electrode requires a
micromanipulator (e.g., WPI, KITE-R micromanipulator).
Our setup including the auditory stimulation and electrophysi-
ological recording instrumentation is shown in Figs. 1 and 2. The
complete list of devices for surgery and experimentation is pro-
vided in Table 1.
2 Methods
2.1 General Anatomy The inferior colliculus (IC) is a mesencephalic nucleus, primarily part
and Considerations of the ascending auditory pathway. It is roughly oval in shape with a
diameter of around 3 mm and is situated some 2–4 mm beneath the
lambda point (see Fig. 3 for saggital section). The IC is part of the
corpora quadrigemina and separated into three subdivisions: the dor-
sal cortex of the inferior nuclei (DCIC), the external cortex of the
inferior nuclei (ECIC), and the core, the central nucleus of the IC
(CNIC; DCIC and ECIC are not separated in Fig. 3). The CNIC in
particular shows a highly linear tonotopy with low frequency respon-
siveness in the dorsal area and high frequency responses in ventral
regions. The hearing range of a rat extends far into the ultrasound
Table 1
Instruments for surgery and devices for electrophysiology
Surgical equipment
Scalpel, scalpel blade #10, WPI
Noyes scissors, stainless steel, 14 cm, straight, WPI
Vannas scissors superfine, stainless steel, 8 cm, straight, WPI
Dissecting scissors, 10 cm, straight, WPI
2× Dumont #5, 12 cm, straight, WPI
Dumont #5, 11 cm, curved, WPI
Gillies dissecting forceps, 15 cm, 1 × 2 teeth
ALM retractor, 7 cm, 2 × 2 or 4 × 4 prongs, WPI
Cutting needles, ½ circle, size 0, WPI
Probe 14 cm, round tip, 0.25 mm, WPI
Drill, Mini-Mot 40, Proxxon
Drill head, 0.5 mm, round, Proxxon
Longhair shaver, Exact Power EP50, Braun
Cauterizer, Cautery Kit, FST
Q-tip, local dealer
Syringe, 0.1 ml, short needle
Syringe, 1 ml, short needle
Syringe, 5 ml, short needle
Accessory surgical equipment
Animal heat pad, animal temperature controller ATC, WPI
Dissecting microscope, M60, Leica
Anesthetic tube for rat, self-made
Stereotactic device, AnyAngle, Stoelting
Surgical consumables
Ringer, 0.9 % NaCl, Braun
Ketamine, 10 %, Dormitor
Medetomidine, Dormitor
Analgetics, Ketoprofen, Sigma
Dermal glue, Nexaband, Abbot
Dental cement, Promedica
Collagenase, Sigma
Vaseline, local dealer
(continued)
Table 1
(continued)
Agarose, Sigma
Nembutal, Dormitor
Skin disinfectant, Kodan, local CVS
Electrophysiological equipment
Micromanipulator, Kite-R, WPI
Micro stepper, SM325, WPI
Faraday cage, self-made
Preamplifier, Medusa, TDT
AD-Board, Medusa, TDT
DA-Board, RP2, TDT
Oscilloscope, TDS1012, Tektronix
Software, brainware, TDT
Audio amplifier, 7600, KronHite
Loudspeaker, MT22, Morel
Sound-level meter, 2238 Mediator, Bruel&Kjaer
domain. Frequencies from a few Hz up to 38 kHz can be detected

[11]. A comprehensive tonotopical mapping of the IC requires
therefore quite sophisticated stimulation hardware.
2.2 Preparing Depending on whether the planned approach will be an acute one
the Skull or not, the operation place and the instruments have to be steril-
ized and prepared. By default the following steps in this chapter
describe acute implantation.
The head of the anesthetized rat is first shaved using an electric
shaver (e.g., Braun, long hair shaver). Even though the approach
targets only the IC contralateral to the stimulation side, the coat
between the neck and both eyes should be removed above both
hemispheres. This is a matter of precaution. If the remaining side
has to be opened up later, this additional shaving avoids hair par-
ticles from contaminating the already open wound. If the experi-
ment is not acute, the skin has to be disinfected locally prior to
using a scalpel (Kodan®). The next step is a 1.5 cm midline incision
between the eyes and the ears (see Fig. 2). The skin is clamped
back with a retractor (Alm self-retaining retractor, 7 cm, WPI).
Alternatively add two transversal terminal incisions (T-shaped inci-
sions) to both ends of the midline cut; this allows the two skin flaps
to be folded inwards. The attached subcutaneous connective tissue
has to be removed from the scull using a pair of fine forceps, fine
scissors, and a scalpel. Do not hesitate to make extensive use of the
scalpel to scratch away the remaining and strongly adhesive
Fig. 3 Drawing of a parasagittal section of the rat brain. Attached are the scale
bars (in mm) given by the stereotactic device. (1) skull, (2) telencephalon, (3)
cerebellum, (4) brainstem, (5) DCIC and ECIC, (6) CNIC, and (7) “lambda” point.
The red bar indicates the position of the opening of the skull the arrow repre-
sents an electrode penetrating through the brain in an angle of 30°
periosteum; otherwise the identification of the sagittal and the

lambdoidal suture is difficult (see Fig. 4). There will be only very
mild bleeding caused by this latter procedure.
2.3 Opening Now the skull should be exposed and the sagittal and the lambdoidal
of the skull suture should be identifiable (see Fig. 4), respectively. Even though the
target nucleus is directly underneath the sagittal suture, it is not wise to
penetrate the skull at this spot. The superior sagittal sinus is just below
the suture and penetrating this major blood vessel will bring any exper-
iment to a swift end. Instead take the drill and force a 4–5 mm diam-
eter whole between the sagittal and lamboidal suture as shown in
Figs. 5a, b and 6. This drilling is one of two crucial steps in this approach
and requires a steady hand and a dissection binocular. Take the head of
the drill and gently drill a circular channel in the bone until the blood
vessels on top of the brain become visible through the thinned out
bone structure. The channel the drill carves into the bone should be
cleaned with ringer solution every now and then to wash away the
bone particles and facilitate optically control of the depth. If the drill is
used with too much force it will penetrate through the remaining thin
layer of bone and damage the brain tissue beneath.
A raw egg presents a very suitable exercise to find the right
moment to stop drilling. Just let the drill do the job and dig in
Fig. 4 Drawing of a rat skull with attached opening for IC approach. (1) sutura
coronalis, (2) bregma point, (3) sutura sagittalis, (4) sutura lamboidea, and (5)
lambda point. The large wire in the right insert is the reference electrode. This
wire is formed like a hook to be fixed underneath the bone
deeper and deeper to the shell with every new circle. Once the
resistance drops, one is generally close to inserting the drill into the
egg and the right moment to stop drilling has been reached.
As soon as the blood vessels are visible through the thinned out
bone in the channel (Fig. 5c) and there are no major bony bridges
left which connect the area within the circle with the remaining
skull, the encircled bone material can be removed using a pair of
thin but strong forceps (see Fig. 5d). Use one end of the forceps to
penetrate between the encircled bone and the dura mater on top of
the brain in a steep angle. Then carefully liftoff the encircled mate-
rial. This step is the second critical one since the penetration must
be done very precisely in order to not damage the dura and the
underlying brain surface as well. The liftoff step might require quite
some force depending on how much bone structure is left after the
drilling. Again this liftoff can be best practiced using a raw egg.
2.4 Preparing After the liftoff step the opening has to be prepared for the pene-
the Penetration tration of the electrode. If this approach is not acute, it is reason-
able to keep the removed bone lid in a cooled ringer dish until the
craniotomy is subsequently closed.
Due to the drilling the fringe of the opening may possess sharp
edges, which even under magnified vision might be difficult to iden-
tify. As a matter of precaution it is conducive to use a tiny round probe
Fig. 5 The two critical steps of the IC approach are shown in a sketch. The bone
of the skull is shown in black, the dura mater in blue, the drill in red, and the tips
of the forceps and the rounded tip of the probe in gray. (a) Positioning of the drill,
(b) movement of the drill in circles in order to carve a circular channel (c) stop-
ping of drilling before penetration, (d) use of forceps to liftoff the circular bone
piece, (e) use of round-tipped probe to remove sharp edges, and (f) removal of
all debris and blood from the wound
to detect possible sharp edges and break them away from the dura
(Fig. 5e, f). This way, the risk of electrodes being bent and damaged
while coming too close to these edges is reduced to a minimum.
Modern electrode designs are quite flexible and therefore
tricky to penetrate through intact dura such as the one generally
found in the rat (e.g., flexible array electrode Fig. 7). Even stiff
glass electrodes tend to buckle, bend, and finally break if
Fig. 6 Two images showing the opening in the skull. The dotted line in the left
image shows a recommended area of penetration for a successful approach
toward the IC
penetrated straight through the dura. Except where stiff metal

electrodes will be used it is therefore preferable in most cases to
remove the dura above this round window before penetration.
This can be realized in two ways: either remove the dura using fine
forceps and micro scissors (WPI) or use collagenase (Sigma) to
degrade the collagen fibers and “digest” the dura. The first
approach might result in some bleeding, which can be stopped
using a fine cauterizer (FST Cautery Kit). The latter approach is
“safer” in terms of excess wounds, but might not weaken the dura
enough to allow very sensitive electrodes to penetrate without
buckling (see Fig. 6 for final view of the prepared recording site).
In addition, the digestion of the dura requires some 20–30 min
incubation time.
Fig. 7 Setup used for acute IC-recordings. (1) Animal with heat pillow underneath, (2) stereotactic device, (3)
dissection microscope for electrode positioning and general control of skull opening, (4) preamplifier, (5) loud-
speaker, (6) oscilloscope, (7) closed-loop thermo element, and (8) Faraday cage. Note that parts of the record-
ing and stimulation chain are not shown (like PC, attenuator, AD-board, headphone)
2.5 Recording Track Once the opening window for the electrodes is prepared the ani-
Through the IC mal can be placed in the stereotactic device. Since the first three
steps of the IC approach require full access to and flexibility of the
head of the rat it is not conducive to have the rat fixed in the ste-
reotactic device prior to step 4. The primary functions of the ste-
reotactic device are to allow the fixation of the animal and to reduce
respiration artifacts. If cortical pulsation becomes an issue during
recording, 3 % agar or Vaseline can be used to fill the hole, damp-
ening movement of the cortex and diminishing this artifact.
Since the IC has an offset toward the opening of the skull, the
penetration has to be applied at an angle of 30° (see Fig. 3). For
reproducible tracks, a micromanipulator is crucial. If the penetra-
tion depth is of importance an adjustable or programmable micro
stepper is a welcome but expensive auxiliary piece of equipment,
which delivers precise mapping coordinates.
For a frequency dependent recording, the following equip-
ment is necessary on the recording and the stimulation side of the
setup. The recording requires an electrode and an appropriate
head stage. The signal is then fed into a bio-amplifier and finally
into an AD-board (analog-digital-converter). If the software does
not present the recording in a window it is convenient to have an
extra oscilloscope at hand to monitor the activity at the electrode.
For auditory stimulation, a voltage controlled function genera-
tor of a programmable DA-board is necessary to generate analog
signals. These signals have to be adjusted in amplitude using an
attenuator and fed into a loudspeaker using an audio amplifier. If
calibrated signals are desired, use a hand calibration tool, e.g., from
Bruel&Kjaer.
Details on the setup devices we used can be found in Table 1.
There are three major approaches to record frequency dependent
signals from the IC:
1. Use sinusoidal signals of defined duration and amplitude.
2. Use sweeps of different frequencies and defined amplitude.
3. Use white noise and perform a RevCor (reverse correlation)
analysis [12].
A broadband search stimulus can be used to control the activity
at the electrodes. In the same context is it most convenient to have at
least one channel of the recording streamed to a headphone to listen
for the background “hash,” which typically precedes or accompanies
the presence of clear spiking activity at a given recording site.
2.6 Termination After the experiment is conducted, there are now two possible
of the Experiment courses of action, depending on whether the experiment was an
or Closing acute one, or whether for example a tracer was applied and the ani-
of the Wound mal must survive for a period of days or weeks in order for the tracer
to be distributed throughout the brain. To terminate an experi-
ment, use sodium pentobarbital (Nembutal®, 40–50 ml/kg) and
apply it i.p. If the animal is subject to a tracer study, pick up the lid
of bone material lifted off during trepanation and place it back so
the craniotomy is closed again. Now use dental cement (Promedica)
to fixate it. Take care that the cement covers the lid as well as the
surrounding area of the skull. The wound can be closed using a
suture. Try to place the stitches underneath the skin since rats are
extremely adept at ripping out surgical suture material during
recover. If the use of a suture is not applicable the wound can also
be closed using dermal glue like Nexaband s/c®.
For the post surgery phase, the animal should be administered
with analgesics (Ketoprofen, 5 mg/kg s.c.).
3 Notes
● Especially during the surgical process, ensure that the mouth
of the rat is open and the nose as well as the mouth are not
blocked by the tongue or a surgical drape.
● Monitor the respiration of the animal every half hour. The fre-
quency and depth of the respiration is an important vital indi-
cator regarding longer experimental sessions.
● Use a syringe and the 0.9 % NaCl solution to frequently clean the
wound. This prevents it from drying. It is especially important that
the exposed surface of the brain does not dry out during experi-
mentation. This will result in recording artifacts and eventually, by
means of clogging, lead to damage of the recording electrode.
● Keep an eye out for micturition of the experimental animal. If

this does not happen after the first 2–3 h the animal might
suffer from dehydration or renal failure. Immediately apply
additional 1 or 2 ml 0.9 % NaCl.
● With ongoing experimental duration the animal will fall into
deeper relaxation, i.e., the animal will “flat out” and the eyes
will pop out of the skull. Therefore, use Vaseline to prevent the
eyes of the animal from drying out.
● Do not leave any metal containing parts attached to the wound
(e.g., retractor and forceps). This will result in increased levels
of noise, since the metal will pick up electrical fields like an
antenna.
● A Faraday cage is not mandatory but generally helps to get
stable results and a decent level of background noise during
the recording. In the same context, do not forget to turn off
cell phones and to ground yourself while positioning the elec-
trode with open amplifiers.
● If you purchase a heating mat for an electrophysiological
experiment, make sure the device is appropriate for this since
some heat mats are strong sources of pulsed electromagnetic
radiation, which then will spit in your recording.
References
1. Clopton BM, Winfield JA (1974) Unit 8. Kelly JB, Li L (1997) Hearing Research: two
responses in the inferior colliculus of rat to sources of inhibition affecting binaural
temporal auditory patterns of tone sweeps and evoked responses in the rat’s inferior collicu-
noise bursts. Exp Neurol 42(3):532–540 lus: the dorsal nucleus of the lateral lemniscus
2. Kelly JB, Masterton B (1977) Auditory sensi- and the superior olivary complex. Hear Res
tivity of the albino rat. J Comp Physiol Psychol 104:112–126
91(4):930–936. doi:10.1037/h0077356 9. Kisban S, Herwik S, Seidl K, Rubehn B, Jezzini
3. Borg E (1982) Auditory thresholds in rats of A, Umiltà MA, Fogassi L, et al (2007)
different age and strain. A behavioral and elec- Microprobe array with low impedance elec-
trophysiological study. Hear Res 8(2):101–115 trodes and highly flexible polyimide cables for
4. Ryan AF, Furlow Z, Woolf NK, Keithley EM acute neural recording. Conference proceed-
(1988) The spatial representation of frequency ings: annual international conference of the
in the rat dorsal cochlear nucleus and inferior IEEE Engineering in Medicine and Biology
colliculus. Hear Res 36(2–3):181–189 Society IEEE Engineering in Medicine and
Biology Society Conference, 2007, 175–178.
5. Zhang DX, Li L, Kelly JB, Wu SH (1998) doi:10.1109/IEMBS.2007.4352251
GABAergic projections from the lateral lem-
niscus to the inferior colliculus of the rat. Hear 10. Saldana E, Merchan M (1992) Intrinsic and
Res 117(1–2):1–12 commissural connections of the rat inferior
colliculus. J Comp Neurol 319(3):417–437
6. Du Y, Ma T, Wang Q, Wu X, Li L (2009) Two
crossed axonal projections contribute to binaural 11. Heffner HE, Heffner RS, Contos C, Ott T
unmasking of frequency-following responses in (1994) Audiogram of the hooded Norway rat.
rat inferior colliculus. Eur J Neurosci 30(9):1779– Hear Res 73(2):244–247
1789. doi:10.1111/j.1460-9568.2009.06947.x 12. Klein DJ, Depireux DA, Simon JZ, Shamma
7. Kelly JB, Glenn SL, Beaver CJ (1991) Sound SA (2000) Robust spectrotemporal reverse
frequency and binaural response properties of correlation for the auditory system: optimizing
single neurons in rat inferior colliculus. Hear stimulus design. J Comput Neurosci 9(1):85–
Res 56(1–2):273–280 111. doi:10.1023/A:1008990412183
Chapter 6
Why Robots Entered Neurosurgery

Abstract
Progress in neurosurgery has paralleled technological innovation. Image-guided surgical robotic systems
have emerged as a potential hub for integration of the complex sensory, pathologic, and imaging data sets
that are available to contemporary neurosurgeons. These systems couple the executive capacity of surgeons
with the technical capabilities of machines and have the potential to improve surgical care as neurosurgery
progresses towards the cellular level. Surgery is often performed in animal models prior to clinical applica-
tion, representing a very important safety step in regulatory approval. As the capital investment for surgical
robotic systems decreases, robotic systems may be specifically designed for animal application. In this chap-
ter, we review neurosurgical robotic systems used in humans and animals; present the development, pre-
clinical testing, and early clinical use of a unique image guided MR-compatible neurosurgical robot called
neuroArm; and review the strengths and limitations of using surgical robotic systems in animal models.
Key words neuroArm, Image guidance, Robotics, Clinical integration, Stereotaxy, Microsurgery,
Neurosurgery
1 Introduction
Historically, progress in neurosurgery has paralleled technological

innovation. This trend began with improved neurosurgical instru-
mentation. In 1927, Bovie and Cushing revolutionized neurosur-
gery with the introduction of electrocautery [1]. For the first time,
neurosurgeons were provided a technology that allowed control of
bleeding, resulting in a substantial reduction of operative morbid-
ity and mortality. The ability to achieve hemostasis also allowed
surgeons to begin surgically managing lesions that were previously
considered inoperable.
During the same timeframe, visualization of the surgical site
evolved from the incandescent headlamp, progressing into
today’s counterbalanced operating microscope [2]. Magnification
of the surgical field initiated the microsurgical paradigm, with
narrower surgical corridors and greatly enhanced visual differen-
tiation between normal and pathological tissue. The resulting
85
86 Jason W. Motkoski and Garnette R. Sutherland
microsurgical revolution necessitated the creation of increasingly

small, precise surgical instrumentation [3].
Prior to the 20th century, surgical localization was based on the
principles of Paul Broca and contemporaries, who established con-
cepts of cortical compartmentalization of function based on clinical
pathological correlation [4]. In 1895, Wilhelm Conrad Roentgen
revolutionized diagnostic medicine with the discovery of X-rays [5].
This was improved upon by the serendipitous discovery of pneumo-
encephalography, or air injection, in 1917, which visualized brain
shift through displacement of ventricles [6]. Pneumoencephalography
remained the primary neurological imaging modality until the mid
1950s, when contrast angiography became widespread because of
decreased toxicity of contrast agents. The invention of the micro-
processor allowed for the explosive growth of computer technology,
which when coupled with X-ray imaging, created computerized
tomographic (CT) imaging in the early 1970s [7]. Characterization
of electron spin in the hydrogen atom was also coupled to computer
technology to create magnetic resonance (MR) imaging [8, 9].
These technologies allowed serial 2D imaging of the human body,
and accurate lesion localization within a particular 2D plane.
Volumetric reconstruction followed, presenting interactive 3D vir-
tual models from which additive or destructive lesions on the brain
could be observed. In addition, MR technology evolved to allow
imaging of brain function and metabolism [10, 11].
The explosive growth of imaging technology contributed to a
new paradigm in surgery: one of minimalistic technique. Technologies
rapidly emerged in response, equipping the surgeon with an array of
new instrumentation including endoscopy, high-definition cameras,
computer displays, and elongated tools capable of accessing body
compartments through small portals. It was very quickly shown that
minimalistic surgical technique resulted in decreased length of hos-
pital stay, lower rates of surgical complication, improved patient out-
come, and increased patient satisfaction [12, 13].
In addition, as intracranial operations involve fixed anatomy in
a contained volume, investigators began to exploit triangular
geometry to link preoperative images with intraoperative surgical
navigation [14, 15]. As a result, cranial openings became smaller
and surgeons were able to accurately target deep brain pathology
using patient-specific imaging. Unfortunately, the act of surgical
dissection results in brain shift, whether through anesthetic man-
agement, patient positioning for surgery, drainage of cerebral spi-
nal fluid, progressive excision of pathology, or brain edema. In
response to this challenge, several investigators began to integrate
various imaging technologies into surgical procedure [16–19].
Surgeons were soon provided with exquisite lesion localization
before and during each operation. Neurosurgical corridors became
smaller, pushing surgeons toward their physical limits of precision,
accuracy, and coordination. Investigators from around the world
began to integrate robotic technology into the increasingly complex
Robots in Neurosurgery 87
surgical environment. Robotics allowed the executive capacity of

the human brain to be coupled with the increasingly precise and
accurate technology of machines. The modern neurosurgical para-
digm of informatic surgery was thus born [20].
2 Neurosurgical Robotics
The initial concept of integrating a robotic system into neurosur-

gery was a daunting proposition. In addition to the technical
requirements of the system and its surgical objectives, investigators
needed to resolve challenges associated with sterility, patient safety,
ethical and regulatory approval, considerable financial cost, inte-
gration with imaging technology, and introduction into established
surgical procedure and processes. Multidisciplinary cooperation
between science, medicine, engineering, and industry was crucial
in overcoming the complexity of these challenges. Global aware-
ness and networking allowed incorporation of diverse technologies
as they were developed and became available (Fig. 1).
Neurosurgical robotics (Table 1) began with the introduction
of an industrial robot, Programmable Universal Machine for
Fig. 1 (a) Timeline showing the chronological introduction of technologies into clinical neurosurgery. Over the
past 20 years, robotic systems have been developed to couple the executive decision-making capacity of the
surgeon with the accuracy of imaging technology and the precision of advanced robotics. (b) A screenshot of
the neuroArm human–machine interface showing the integration of 3D magnetic resonance images with
surgical planning (blue cone) and robotic tools (blue and off-white) to target patient pathology
88
Table 1
The evolution of neurosurgical robotics
Year Project name DOF # EE Navigation Imaging Purpose Documented use Accomplishments Limitations
1988 PUMA1,2 6 1 BRW frame CT Retractor Glioma surgery First neurosurgical No FDA approval
robot
Demonstration of Safety concerns with
safety industrial robot in
OR
1994 Minerva3–5 5 1 BRW frame CT Image guided Stereotactic biopsy Mounted inside CT Only 5 DOF
surgery machine
Stereotactic Requires dedicated CT
implantation scanner
Obstructive within CT
gantry
1997 CyberKnife6 6 1 Frameless X-ray Radiosurgery Neurosurgical Focused radiosurgery Expensive
radiation
FDA approved
7,8
1999 RAMS 6 1 N/A N/A Microsurgery Rat carotid Microsurgical ability No haptic feedback
endarterectomy
Increased length of
procedure
1999 Steady Hand9 6 1 N/A N/A Microsurgery Tremor filter and Increased surgical No image guidance
motion stabilizer precision
Haptic feedback No clinical application
2001 Harvard MRI 5 1 Frameless MRI Navigation and Position needle Nonmagnetic Pointing device only
Robot10,11 tool placement holder actuators
2002 Evolution 112–14 6 1 Frameless MRI Endoscopy Neuroendoscopy Endoscopic Narrow working
application envelope
Endoscopic Single arm
ventriculostomy
Transphenoidal No haptic feedback
skull base surgery
2003 NeuroMate15–18 5 1 Frame CT Stereotaxy and Stereotaxy Commercially No microsurgical
based or lesion available ability
Frameless localization
Functional First FDA-approved No tool actuation
neurosurgery neurosurg robot
Drilling at the skull Diverse clinical
base application
2003 NeuRobot19–21 3 3 Frameless N/A Microsurgery Tumor resection Partial resection of Limited to 3 DOF for
meningioma each arm
Telesurgery on rat Low payload
from 40 km away
2005 Georgetown22,23 6 1 N/A Fluoroscopy Stereotaxy Percutaneous facet Accuracy comparable Movement occurs in 1
blocks to manual DOF at a time
technique
Requires fluoroscopy
suite
2006 Pathfinder24 6 1 Frameless CT Stereotaxy Epilepsy surgery Highly accurate CT required for
navigation feducial placement
Surgical ergonomics
(continued)
Robots in Neurosurgery
89
Table 1
90
(continued)
2006 SpineAssist25–27 6 1 Frame CT Spinal Guide for tool FDA approval for Limited range of
based instrumentation positioning spinal surgery application
Guide for screw
placement
2009 NISS28 5 1 CT CT CT-guided Image-guided In vivo and in vitro Ionizing radiation
navigation implantation surgical
implantation
2009 neuroArm29–33 7 2 Frameless MRI Presurgical Various intracranial Microsurgery and Expensive
planning pathology stereotaxy
Microsurgery and Intracranial tumor MRI-compatible
stereotaxy resection robot and tools
Haptic feedback
DOF degrees of freedom, #EE number of end effectors, PUMA programmable universal machine for assembly, BRW Brown–Roberts–Well, CT computer tomographic, FDA
Food and Drug Administration, OR operating room, MRI magnetic resonance imaging, km kilometer, NISS Neuroscience Institute Surgical System
References: 1Kwoh YS, et al. (1988) IEEE Trans Biomed Engg 35(2):153-160. 2Drake JM, et al. (1991) Neurosurgery 29(1):27-33. 3Glauser D, et al. (1995) J Image Guid
Surg 1(5):266-72. 4Hefti J-L, et al. (1998) Comp Aid Surg 3:1-10. 5Frankhauser H, et al. (1994) Stereotact Funct Neurosurg 63(1-4):93-8. 6Adler JR, et al. (1997) Stereotact
Funct Neurosurg 69(1-4 Pt 2):124-8. 7Das H et al. (1999) Comp Aided Surg 4:15-25. 8Le Roux PD, et al. (2001) Neurosurgery 48(3):584-9. 9Taylor R, et al. (1999) Int J
Rob Res 18:1201-1210. 10Chinzei K, et al. (2001) Med Sci Monit 7(1):153-63. 11Chinzei K, et al. (2003) Min Invas Ther & Allied Technol 12(1-2):59-64. 12Zimmerman M,
et al. (2002) Neurosurgery 51:1446-52. 13Zimmerman M, et al. (2004) Acta Neurochir (Wein) 146:697-704. 14Nimsky CH, et al. (2004) Minim Invasive Neurosurg 47(1):41-
6. 15Li QH, et al. (2002) Comp Aided Surg 7(2):90-98. 16Varma TRK, et al. (2006) Int J Med Rob Comp Assist Surg 2(2):107-113. 17Varma TRK, et al. (2003) Stereotact
Funct Neurosurg 80(1-4):132-5. 18Xia T, et al. (2008) Int J Med Robot 4(4):321-330. 19Hongo K, et al. (2002) Neurosurgery 51(4)985-988. 20Goto T, et al. (2003) J
Neurosurg 99(6),1082-4. 21Hongo K, et al. (2006) Acta Neurochir Suppl (Wien) 98:63-66. 22Cleary K, et al. (2002) Acad Radiol 9(7):821-5. 23Cleary K, et al. (2005) Int J
Med Robot 1(2):40-47. 24Eljamel MS, et al. (2006) Int J Med Rob Comp Assist Surg 2:233-7. 25Lieberman IH, et al. (2006) Neurosurgery 59(3):641-50. 26Sukovich W, et al.
(2006) Int J Med Robot 2(2):114-122. 27Barzilay Y, et al. (2006) Int J Med Robot 2(2): 181-193. 28Chan F, et al. (2009) Surg Neurol 71:640-8. 29Louw DF, et al. (2004)
Neurosurgery 54(3):525-36. 30Sutherland GR, et al. (2008) Neurosurgery 62(2):286-93. 31Pandya S, et al. (2009) Neurosurgery 111(6):1141-9. 32Sutherland GR, et al. (2008)
IEEE Eng Med Biol Mag 27(3):59-65. 33Greer AD, et al. (2008) IEEE/ASME Trans Mech 13(3):306-315
Assembly (PUMA), into the operating theater in 1985 [21]. In

1987, Benabid et al. coupled the PUMA 200 robot to a Brown–
Roberts–Wells (BRW) head frame for frame-based stereotaxy [22].
The robotic arm had six degrees of freedom (DOF), and each joint
was equipped with spring-applied, solenoid-release brakes that
would immediately stop motion should any system defect arise.
Using CT imaging for navigation, the system was able to orient a
cannula for needle insertion. The robot was modified to hold a
retractor for the resection of multiple pediatric thalamic astrocyto-
mas in 1991 [23]. These developments provided proof-of-concept
of robotic neurosurgery, allowing multidisciplinary research teams
to begin development of robotic systems designed for specific neu-
rosurgical applications.
In 1994, Frankenhauser et al. introduced the Minerva robot
[24]. The system was coupled to the same BRW headframe used by
Benabid, and mounted inside a CT scanner for image-guided biopsy
and implantation. Unfortunately, the requirement of a dedicated
CT scanner limited applicability of Minerva to the neurosurgical
community at large. Two systems, the Robot Assisted Microsurgical
System (RAMS) and Steady Hand system, were developed in 1999
to enhance microsurgery [25, 26]. While these projects were shown
to improve microsurgical precision and provided haptic feedback,
respectively, they have not yet been clinically applied.
By the year 2000, advances in computer technology, the ubiq-
uity of neurological imaging, and increasing adaptation of robotic
technology across the manufacturing and aerospace industries pro-
vided the potential for development of an image-guided neurosur-
gical robotic system. The Harvard MRI robot was developed and
integrated with intraoperative MR imaging (iMRI) [27]. The
robot manipulator was MR compatible, and used for intraoperative
tool orientation. NeuRobot emerged as a microsurgical robot in
2003 [28]. It was telecapable, and tested in an experimental model
with the surgeon located 40 km away from the surgical site.
In addition to developments in microneurosurgery, robotic
systems were designed for endoscopic and stereotactic application.
The Evolution I robot was developed for endoscopic applications,
and used in 2002 for the transphenoidal removal of pituitary ade-
noma [29]. The CT-based NeuroMate system became the first
neurosurgical robot to receive FDA approval for clinical use [30].
The Pathfinder system followed in 2006, with highly accurate
CT-guided stereotactic application [31]. Finally, the NISS collabo-
ration was published in 2009 with direct application in CT-guided
implantation procedures [32].
Two important robotic systems have been developed for spinal
surgery. In 2005, the Georgetown robot was used for fluoroscopy-
guided percutaneous facet blocks in cadaveric studies and clinical
patients [33]. Accuracy was deemed comparable to manual tech-
nique, but movement was only available in 1 DOF at a time. The
SpineAssist robot has been used for tool positioning and pedicle
screw placement [34]. It received FDA approval and is commer-
cially available to this day.
3 neuroArm
Advances in technology provided the opportunity to develop a

robotic system capable of both microsurgery and stereotaxy [35].
Furthermore, due to the increasing acceptance of intraoperative
MR imaging, it was also desirable to construct a system that could
operate within this imaging environment [36]. While this pre-
sented a significant challenge, it would resolve the problem of dis-
rupting surgical rhythm for intraoperative image acquisition. In
2002, investigators at the University of Calgary (Calgary, Alberta,
Canada), in collaboration with MacDonald, Dettweiller and
Associates, began the development of such a robot.
4 neuroArm: Design and Manufacture
The initial requirements document for neuroArm included the

ability to perform both microsurgery and stereotaxy within the
bore of a 1.5 T magnet. At the time of preliminary design review,
it became evident that stereoscopic vision, a necessary requirement
for microsurgery, could not practically be captured within the
magnet bore. Furthermore, due to requirements of payload
(750 g) and speed (200 mm/s), as well as the size of then existing
position encoders, the manipulators needed to be relatively large.
Both could not be placed within the 70-cm working aperture of
the intraoperative 1.5 T magnet. For these reasons, scope was
changed such that image-guided microsurgery would be per-
formed outside of the bore of the magnet, and stereotaxy, using a
single arm, within the bore of the magnet.
4.1 Manipulators neuroArm consists of two anthropomorphic manipulators, each

with a total of seven DOF: six spatial and one degree of tool actua-
tion [37]. Each manipulator was designed to reflect the limbs and
joints of a human surgeon: shoulder joint to allow for rotation in
both the horizontal and vertical plane, elbow joint to allow for
flexion/extension of the arm, and a wrist joint to allow adjustment
of rotation and pitch. Tool actuation is accomplished through
motion of one tool holder relative to the other. The arms are
mounted on a mobile base, which allows height adjustment to
accommodate the position of the operating table.
To achieve MR compatibility, the manipulators were manu-
factured from titanium, polyetheretherketone (PEEK), and
polyoxymethylene (Delrin; Dupont, Wilmington, DE). Motion

of the arms is accomplished using ultrasonic piezoelectric actua-
tors (Nanomotion, Yokneam, Israel) that have a 20,000-h life-
time, 1-nm resolution, and inherent braking characteristics if
power is lost. Absolute 16-bit sine/cosine encoders provided
0.01-degree accuracy at each joint and retained positional infor-
mation when powered off for imaging. Haptic feedback was pro-
vided in three translational DOF from six-axis force/torque
sensors (ATI Industrial Automation, Apex, NC) that were specifi-
cally manufactured for neuroArm. The end effectors were
designed to hold a variety of tools using a standardized interface
that allows for tool roll and tool actuation (Fig. 2).
4.2 Mobile Base The manipulators are moved in and out of the operating room on
a height-adjustable mobile base. A digitizing arm, mounted on the
mobile base, allows registration of the manipulators to the radio
frequency (RF) coil. The information, transferred to the comput-
erized human–machine interface, allows 3D MR image display
with tool-overlay. The field camera, mounted on the mobile base,
provides overall visual feed of the surgical field. For stereotactic
procedures, the mobile base is used to transport the manipulators
to a platform inserted into the MR magnet (Fig. 3).
4.3 Main System The main system controller consists of four main software applica-
Controller tions, each operating on an individual computer: (1) The command
and status display provides the main graphical control interface for
the neuroArm end effectors. (2) The MRI display provides 2D and
3D volumetric images of patient pathology with tool overlay. (3)
The hand controller interfaces to the left and right human–machine
interface hand controllers process kinematic motion at the human–
machine interface. (4) The controller interfaces to the manipulator
arms and other hardware.
4.4 Human–Machine The human–machine interface recreates the sight, sound, and touch
Interface of surgery, while facilitating integration of advanced imaging and
surgical planning technologies [38] (Fig. 4). Two high-definition
cameras (Ikegami Tsushinki Co, Tokyo, Japan) mounted on the
surgical microscope provide a stereoscopic image at 1000 TV lines
horizontal resolution to a 3D computer monitor (Alienware, USA).
The MRI display can be manipulated by touch with on-screen con-
trols to view patient-specific MR images in 2D or 3D, with real-
time tool overlay. The surgeon is thus able to see the tools as they
are manipulated down the surgical corridor, and their spatial rela-
tionship to the pathology. During stereotaxy, the command status
display provides real-time feedback of end effector orientation rela-
tive to the RF coil and magnet bore.
Fig. 2 (a) The neuroArm end effector uses two standardized connectors (blue) to
hold a surgical tool. The upper connection includes a gear to control tool roll,
while the lower connection moves upward and downward to allow for tool actua-
tion. (b) During surgery, the neuroArm manipulators are draped for sterility, while
the two standardized tool connectors penetrate the drape. The scrub nurse is
able to exchange all the neuroArm tools with the standardized tool connectors
Fig. 3 (a) During stereotaxy, one neuroArm manipulator is placed on a specialized board within the iMRI mag-
net bore, opposite the patient. (b) The iMRI machine moves to the operating table, so the patient and neuroArm
end effector meet at the magnet isocenter. Stereotaxy near the magnet isocenter allows for simplified registra-
tion and optimized image quality throughout the procedure
Fig. 4 The neuroArm human–machine interface recreates the sight, sound, and touch of the surgical site for the
surgeon. The surgeon is provided with a 3D stereoscopic view of the surgical corridor. The command status dis-
play (right of the stereoscopic display) shows the position of the neuroArm manipulators in relation to the radio
frequency coil. The surgeon controls neuroArm using two modified PHANTOM hand controllers, providing 7 DOF
control with 3 DOF force feedback. The surgeon communicates with the surgical team using a wireless headset
The human–machine interface is height adjustable, and two mod-

ified PHANTOM hand controllers (SensAble Technologies, Inc.,
Woburn, MA), which provide haptic feedback in three translational
DOF, are mounted at 45° to optimize ergonomics. The surgeon acti-
vates each of the manipulators by continuously depressing a corre-
sponding foot pedal. Each pedal acts as a live-man switch: if disengaged,
all motion input to the corresponding manipulator is immediately
stopped, bypassing the main system controller. This design is not
dependent on the computational capacity of the main system control-
ler, providing the surgeon ultimate control over any adverse motion of
the manipulators. Motion scaling and tremor filtration can be applied
from the command status display.
As the neuroArm human–machine interface is telecapable, pre-
cise and reliable audio communication with surgical staff is of criti-
cal importance. Each member of the surgical team wears a wireless
headset with a microphone to maintain smooth surgical rhythm.
This alone seems to enhance communication between all members
of the surgical team, empowering each member of the surgical
team to optimize their role in the procedure
4.5 Safety The creation of a telecapable human–machine interface provides an

ergonomic platform from which the surgeon can interact with com-
plex imaging and sensory data sets. For the surgeon to interact
within this complex data set, and at the surgical site, increasingly
sophisticated robotic manipulators have been developed and pro-
vided. This creates some unique challenges relative to patient safety,
as the surgeon is no longer mechanically engaged with the surgical
site. Thus, motion of the robotic manipulators passes through the
main system controller, where all of the input and output from the
surgical site is being analyzed and interpreted. Computational com-
plexity is further increased by redundancy in all manipulator posi-
tion encoders, force measurements, computer power supplies, and
redundant hardware wiring. This introduces the potential for
uncontrolled motion of the robotic manipulators at the surgical
site, should any individual element within the computational pro-
cessing malfunction. While the software is designed to localize any
malfunction in the system, this can take time due to the complex
nature of the computation. To overcome this, the live-man switch
was introduced to provide the surgeon with the absolute ability to
stop robotic motion at any time of the robotic procedure. In addi-
tion, all members of the surgical team are trained on robotic safety,
in particular, on the potential for unexpected collision with object
or patient. As stereotaxy with neuroArm occurs within the confines
of the MR magnet bore, the surgeon is provided with visual feed of
the tool by field cameras within the bore. Prior to executing the
procedure, the surgeon must perform a verification intraoperative
scan containing the tool and destination pathology to assess accu-
racy of graphical MR tool overlay.
5 neuroArm: Preclinical Studies
neuroArm and its components arrived and were installed in the

intraoperative MR Operating Theatre at the Seaman Family MR
Research Centre (Calgary, Alberta, Canada) in 2008. Regulatory
and ethics approval were received in the same year. While such
approval is required for clinical application, introduction, and accep-
tance into the international neurosurgical community requires suit-
able preclinical studies to demonstrate performance and capability of
the system within the time-sensitive environment of surgery. A two-
stage study was conceived to sequentially test (1) the microsurgical
capabilities of neuroArm in a rodent model and (2) the stereotactic
capabilities of neuroArm in a cadaveric model [39].
5.1 Animal Studies Animal studies allowed for in situ testing of neuroArm and accli-
matization to the novel neuroArm interface. Compared to conven-
tional neurosurgery, when using neuroArm for microsurgery,
surgeons were sitting rather than standing, viewing the surgical site
through miniature cameras rather than an operating microscope,
manipulating hand controllers rather than the tools directly,
requesting for tool exchange from the human–machine interface,
communicating via wireless headsets, and relying on the assistant
surgeon for manipulations requiring increased dexterity at the sur-
gical site. Surgeons adjusted to these changes over the course of
the study, indicating a short learning curve.
A Sprague-Dawley rat model was selected to evaluate neuroArm
in microsurgery mode. Each procedure involved three objectives
(bilateral nephrectomy, splenectomy, and bilateral submandibular
gland excision), selected to provide reasonable models of varying
microsurgical landscapes. Procedures were completed using either
neuroArm or conventional hand techniques, and with a common
assistant surgeon. Total surgical time, blood loss, thermal injury, and
vascular injury were recorded for each procedure, then weighted
and combined into a surgical performance score to compare all mea-
sures with a single variable. Using neuroArm and hand techniques,
each surgeon was allowed one complete procedure for familiariza-
tion with the equipment and surgical objectives. Results of the sub-
sequent four procedures were recorded for evaluation [39].
For each procedure, the abdominal cavity and neck were
opened with midline incisions. The renal vessels were exposed,
coagulated with bipolar electrocautery and the kidneys removed
from the abdominal cavity. Splenectomy required hemostasis of
the splenic vessels, then dissection from abdominal contents and
removal from the abdomen. Submandibular gland excision
required more cutting and less hemostasis than the previous objec-
tives given the fibrous nature of surrounding neck tissue. For all
robotic trials, neuroArm was equipped with bipolar forceps in the
right end effector and tissue forceps in the left. For nonrobotic trials,
the surgeons were provided bipolar forceps and a standard selec-
tion of microsurgical instruments.
Results from procedures using neuroArm were compared to
those of procedures using hand techniques. While the use of neu-
roArm increased the total surgical time, there was decreased blood
loss compared to hand trials, resulting in equal overall surgical per-
formance. Increased surgical time had been, in part, expected as
the introduction of intraoperative technology has previously been
shown to increase surgical time. However, surgical time is not the
only predictor of surgical outcome, which is why blood loss and
other performance measures had been recorded. The decreased
blood loss when using neuroArm may have been a result of
increased caution from the surgeon, who was no longer directly
present at the surgical site for a rapid response to a vascular event
should one occur. While decreased blood loss did not reach statisti-
cal significance due to small sample size, it remains an important
measure of surgical performance.
5.2 Cadaveric Following animal studies, the neuroArm navigation system was
Studies tested by image-guided implantation of ferrous oxide coated
nanoparticles in a cadaveric model. The head of the caudate nucleus
and globus pallidus were selected as target implantation sites and
identified on all trials by a senior resident neurosurgeon. This was
an important preclinical study to evaluate the accuracy of the fra-
meless neuroArm navigation software as compared to an already
established navigation system.
Following bilateral frontal craniotomy with a pneumatic drill,
the cadaveric specimens were placed in a head clamp. T1-weighted
MR images were acquired at 2 mm slice thickness. For neuroArm
trials, one neuroArm end effector was placed inside the MR magnet
bore and registered to the head clamp and images. The senior resi-
dent neurosurgeon identified the targets, then planned implantation
trajectory using the tool tip extension feature and the Z-lock feature,
which restricts end effector motion to only the direction of the tool
axis. These features, coupled with 3D tool overlay at the neuroArm
human–machine interface, greatly simplified the implantation pro-
cedure. Nonrobotic implantation was completed on the contralat-
eral side using the VectoVision Sky system (BrainLAB). The
specimens remained in the same head clamp, and the same preopera-
tive T1-weighted MR images were loaded onto VectorVision. For
this implantation, the surgeon was presented with sagittal, axial, and
coronal images at the tool tip. Following all implantation proce-
dures, the specimens were imaged using the same acquisition
sequences to determine the final position of nanoparticles relative to
the desired targets. The neuroArm system was more accurate than
the VectorVision system, but did not reach statistical significance
due to small sample size (n = 4 targets for each modality).
Fig. 5 (a) For microsurgical procedures, neuroArm is positioned at the operating table in the position of the
primary surgeon. The assistant surgeon is able to operate in an ergonomic position relative to neuroArm and
the operating microscope. (b) The neuroArm bipolar forceps can be used to coagulate, as well as remove
pathological tissue
6 neuroArm: Clinical Studies
neuroArm represents a novel paradigm for neurosurgery, which

created a number of practical considerations for implementation
into established neurosurgical procedure. Wireless headsets were
implemented for all members of the surgical team to allow com-
munication with the surgeon at the human–machine interface. The
scrub nurse became responsible for exchanging the neuroArm
tools. Draping of the robot, and the time of draping relative to
preoperative MR imaging and registration to patient anatomy
required careful planning into existing nursing and MR technician
protocols (Fig. 5). For these reasons, clinical integration of neu-
roArm was accomplished in a step-wise fashion.
Among the first 22 cases, 10 were meningioma, 9 glioma, 2
acoustic schwannoma, and 1 brain abscess. Each of these proce-
dures required general anesthetic and craniotomy. For four cases,
neuroArm was registered to the intraoperatively acquired surgical
planning MR images, and used to target the pathology and deter-
mine craniotomy placement (Fig. 6). In all cases, neuroArm was
draped during craniotomy, and brought into the surgical field after
partial dissection of the pathology. Working at the human–machine
Fig. 6 (a) Conventional presurgical planning involves marking of the surgical site following anesthetic and fixa-
tion with pins in a head clamp. (b) Prior to craniotomy, neuroArm navigation software can be used to confirm
and refine craniotomy placement based on intracranial pathology. (c) Patient-specific MR images are loaded
into the MRI display at the neuroArm human–machine interface
interface, the surgeon was able to manipulate tools within the sur-
gical corridors, coagulate vessels to control bleeding, and aspirate
(Fig. 7). For the brain abscess case, the bipolar forceps, mounted
in the right arm, was used to open the tumor capsule and allow
drainage of pus.
There was a disruption in the ongoing integration of neu-
roArm into neurosurgery in 2009, as the 1.5 T iMRI environment
with local RF shielding was upgraded to a 3.0 T iMRI suite that
included whole room RF shielding. This upgrade required a
10-month interval, during which the operating room was shut
down to patient cases. The upgrade to whole room shielding
allowed dramatic improvements in practical aspects for stereotactic
procedures. The manipulator is now able to be attached directly to
the magnet bore, rather than being mounted on an extension
board from the OR table. Cables are now run through the back-
side of the magnet, rather than along the OR table, which was
previously required to prevent penetration of the RF shielding.
Finally, the registration procedure is much simpler as the location
of the manipulator is always constant relative to the magnet’s iso-
center, and thus the patient’s pathology (Fig. 3).
Fig. 7 (a) Positioning neuroArm into surgical procedure is important so that ergonomics of the surgical assis-
tant and scrub nurse are not compromised. (b) At the surgical site, both neuroArm and the assistant surgeon
are able to manipulate tools within the surgical corridor. (c) Sterile drapes are placed over the neuroArm
manipulators, while the tool holders (blue) are able to penetrate the sterile drapes and hold the tools
7 Use of Surgical Robotics in Animal Models
Robotic surgery in animal models occurs primarily as preclinical test-

ing of systems intended for clinical use. Animal studies provide opti-
mal simulation of the surgical environment, recreating the dynamic
landscape of the surgical field and time-sensitive events such as hem-
orrhage. The surgeon becomes puppeteer of the surgical corridor;
manipulating its characteristics to achieve the surgical objective. To
date, there is no superior simulation of clinical surgery than surgery
in animal models. This fact has driven regulatory requirements for
surgical technology across North America, and all surgical robotic
systems require testing in animal surgery. The primary investigator
and colleagues often complete these validation studies, but once
regulatory approval has been achieved, it is no longer necessary to
sacrifice animal life to validate robotic performance.
There does, however, remain a need to train individual sur-

geons on the use of any surgical robotic system. For neuroArm, a
detailed training paradigm has been developed. It begins with
familiarization of equipment using an interactive, internet-based
computerized tutorial. The surgeon manipulates a virtual neu-
roArm through space to accomplish very simplistic objectives
including positioning of equipment, stacking of objects, and tracing
of convoluted tool tip trajectories. The next step involves the use of
the real neuroArm robot in simple tasks: moving rings on pegs or
sponges between containers. These stages provide excellent famil-
iarization of the neuroArm equipment, but do not replicate the
dynamic complexities of the surgical site. For this reason, it remains
customary for new surgeons to complete minor procedures in a
rodent model prior to using neuroArm in clinical application. There
is simply no adequate substitution for this experience at the present
time. It is hoped, that through patient-specific virtual surgery simu-
lators, dynamic environments will be created that allow surgical
rehearsal without the expense of animal life.
When neuroArm is used in the animal setting, it offers the
surgeon certain advantages over conventional techniques. These
advantages are similar to the benefits of using neuroArm in the
clinical setting, and include:
– Increased kinetic precision of the distal tool tip to the order of
micrometers
– Increased tool tip accuracy from sine/cosine absolute position
encoders
– Improved human–machine interface including real-time 3D
tool overlay, visualization of the surgical site and the intuitive
Z-lock feature for stereotaxy-biopsy
The present challenges of robotic surgery in animals are being
rapidly achieved through advances in robotics and human–machine
interface technology. Robotic systems of the future will have
increased dexterity and push the limits of precision and accuracy
toward the cellular level. Human–machine interfaces will become
more intuitive to operate and increasingly integrated with intraop-
erative imaging. Finally, developments in low and high fidelity hap-
tic technology will improve the sense of touch that is recreated at
the human–machine interface. This factor may be the most impor-
tant in driving international adaptation of robotic systems. Present
limitations of robotic surgery with neuroArm include:
– The dexterity of a human surgeon is greater than 6 DOF
because a tool may be held in a single position with multiple
positions of proximal joints in the arms. Due to present com-
puter algorithms, neuroArm dexterity is limited to 6 DOF: any
specific tool position can only be achieved with a unique com-
bination of proximal joint positions. The greater robotics
industry has solved this challenge already, and algorithms are

presently being evaluated for suitability with neuroArm.
– While neuroArm offers low frequency haptic feedback in 3
DOF, it is not able to completely recreate a surgeon’s sense of
kinesthetic touch at the surgical site. Ongoing international
research in haptics is advancing this science and will improve
future haptic interfaces.
– Since the preclinical animal studies, the human–machine inter-
face visualization system was upgraded from Ikegami cameras
inside a microscope-style viewfinder to a single high-definition
3D display. This has overcome limitations at the time of pre-
clinical studies of decreased picture quality compared to the
microscope view at the surgical site.
It is very likely that the future of animal surgery will involve an
increasing number of robotic systems. Perhaps the greatest issue
preventing the widespread development of robotic systems for ani-
mal applications is the high cost of surgical robotics. Robotics
intended for clinical use have benefited from the greater funding
that has been required to initiate their development. However, as
robotic technology becomes more universally applied in animal
procedures, costs of development and implementation will drop,
allowing creation of systems for specific therapeutic purposes in
animal applications.
8 Conclusion
The future of neurosurgery lies in the realm of multidisciplinary

teamwork. As surgical staff gather increasing clinical experience with
neurosurgical robotic systems, robotic technology will continue to
be advanced by teams of engineers, scientists, and technologists
around the world. Mechanically, advances in MR-compatible robot-
ics will allow miniaturization without sacrifices in surgical perfor-
mance. This will overcome present spatial limitations and allow
movement toward real-time, image-guided microsurgery within the
physical constraints of intraoperative imaging devices.
Perhaps of more impact to the surgeon will be the rapid
upgrades in human–machine interface technology. In the future,
the surgeon will be presented with pre- and intraoperative images in
a manner that is surgically relevant and intuitive to manipulate.
Improved computer processing will provide relevant data related to
anatomy, function, and metabolism. The surgeon will be provided
with realistic recreation of touch through ongoing developments in
haptic feedback and hand controller design. Measurement of surgi-
cal forces and their relationship to tissue deformation will open new
areas of research in basic science toward the understanding of tactile
perception and its relation to surgical decision-making.
Neurosurgical robotics will become the hub of technology in

the operating room, allowing an interface for imaging and surgical
management, advanced tool design, image-guided biopsy and
implantation, and realization of individualized therapy through
cell-specific intervention. From this, patients will receive better,
more accurate, and increasingly precise neurosurgical care.
Acknowledgments
Supported by grants from the Canada Foundation for Innovation,

Western Economic Diversification Canada, Alberta Advanced
Education and Technology, Alberta Heritage Foundation for
Medical Research, and the Canadian Institute for Health Research.
References
1. Cushing H, Bovie WT (1928) Electrosurgery versus open colectomy for treatment of non-
as an aid to the removal of intracranial tumors, metastatic colon cancer: a randomized trial.
with a preliminary note on a new surgical- Lancet 359(9325):2224–2229
current generator by W.T. Bovie. Surg Gynecol 13. Liem MS, van der Graaf Y, van Steensel CJ et al
Obstet 27:751–785 (1997) Comparison of conventional anterior sur-
2. Kriss TC, Kriss VM (1998) History of the operat- gery and laparoscopic surgery for inguinal-hernia
ing microscope: from magnifying glass to micro- repair. N Engl J Med 336(22):1541–1547
neurosurgery. Neurosurgery 42(4):899–907 14. Kelly PJ, Kall B, Goerss S (1983) Stereotactic
3. Yasargil MG (1969) Microsurgery applied to CT scanning for the biopsy of intracranial
neurosurgery. Academic, New York lesions and functional neurosurgery. Appl
4. Broca P (1861) Nouvelle observation d’aphe´mie Neurophysiol 46:193–199
produite par une le´sion de la troisie`me cir- 15. Kanner AA, Vogelbaum MA, Mayberg MR
convolution frontale. Bulletins de la Socie´te´ et al (2002) Intracranial navigation by using
d’anatomie 2e serie 6:398–407 low-field intraoperative magnetic resonance
5. Roentgen WC (1895) On a new kind of rays. imaging: preliminary experience. J Neurosurg
Proc Phys-Med Soc, Wurzburg 97:1115–1124
6. Dandy WE (1919) Roentgenography of the 16. Chandler WF, Knake JE, McGillicuddy JE
brain after injection of air into the spinal canal. et al (1982) Intraoperative use of real-time
Ann Surg 70:397 ultrasonography in neurosurgery. J Neurosurg
7. Hounsfield GN (1973) Computerized trans- 57(2):157–163
verse axial scanning (tomography): part 1. 17. Lunsford LD (1982) A dedicated CT system
Description of system. Br J Radiol 46:1016 for the stereotactic operating room. Appl
8. Lauterbur PC (1980) Progress in n.m.r. zeug- Neurophysiol 45(4–5):374–378
matogrpahy imaging. Philos Trans R Soc Lond 18. Black PM, Moriarty T, Alexander E III et al
B Biol Sci 289(1037):483–487 (1997) Development and implementation of
9. Mansfield P, Maudsley AA (1977) Medical intraoperative magnetic resonance imaging and
imaging by NMR. Br J Radiol 50:188–194 its neurosurgical applications. Neurosurgery
41(4):831–835
10. Ogawa S, Tso-Ming L, Nayak AS et al (1990)
Oxygenation-sensitive contrast in magnetic 19. Sutherland GR, Kaibara T, Louw D et al
resonance image of rodent brain at high mag- (1999) A mobile high-field magnetic reso-
netic fields. Magn Reson Med 14:68–78 nance system for neurosurgery. J Neurosurg
91(5):804–813
11. Peeling J, Sutherland GR (1992) High-
resolution 1H NMR spectroscopy studies of 20. Lang MJ, Sutherland GR (2010) Informatic
extracts of human cerebral neoplasms. Magn surgery: the union of surgeon and machine.
Reson Med 24:123–136 World Neurosurg 74(1):118–120
12. Lacy AM, Garcia-Valdecasas JC, Delgado S 21. Kwoh YS, Hou J, Jonckheere EA et al (1988) A
et al (2002) Laparoscopy-assisted colectomy robot with improved absolute positioning
accuracy for CT guided stereotactic brain sur- 31. Eljamel MS (2006) Robotic application in epi-
gery. IEEE Trans Biomed Eng 35(2):153–160 lepsy surgery. Int J Med Robot 2:233–237
22. Benabid AL, Cinquin P, Lavaile S et al (1987) 32. Chan F, Kassim I, Lo C et al (2009) Image-
Computer-driven robot for stereotactic surgery guided robotic neurosurgery—an in vitro and
connected to CT scan and magnetic resonance in vivo point accuracy evaluation experimental
imaging: technological design and preliminary study. Surg Neurol 71(6):640–647
results. Appl Neurophysiol 50(1–6):153–154 33. Cleary K, Watson V, Lindisch D et al (2005)
23. Drake JM, Joy M, Goldenberg A et al (1991) Precision placement of instruments for
Computer- and robot-assisted resection of tha- minimally invasive procedures using a “needle
lamic astrocytomas in children. Neurosurgery driver” robot. Int J Med Robot 1(2):40–47
29(1):27–33 34. Lieberman IH, Togawa D, Kayanja MM et al
24. Fankhauser H, Glauser D, Flury P et al (1994) (2006) Bone-mounted miniature robotic guid-
Robot for CT-guided stereotactic neurosurgery. ance for pedicle screw and translaminar facet
Stereotact Funct Neurosurg 63(1–4):93–98 screw placement: part I—technical develop-
25. Le Roux PD, Das H, Esquenazi S et al (2001) ment and a test case result. Neurosurgery
Robot-assisted microsurgery: a feasibility study 59(3):641–650
in the rat. Neurosurgery 48(3):584–589 35. Louw DF, Fielding T, McBeth PB et al (2004)
26. Taylor R, Jensen P, Whitcomb L et al (1999) A Surgical robotics: a review and neurosurgical pro-
steady-hand robotic system for microsurgical aug- totype development. Neurosurgery 54(3):525–
mentation. Int J Robot Res 18(12):1201–1210 536, discussion 536–537
27. Chinzei K, Miller K (2001) Towards MRI 36. Sutherland GR, Latour I, Greer AD (2008)
guided surgical manipulator. Med Sci Monit Integrating an image-guided robot with intra-
7(1):153–163 operative MRI: a review of the design and con-
28. Hongo K, Kobayashi S, Kakizawa Y et al (2002) struction of neuroArm. IEEE Eng Med Biol
NeuRobot: telecontrolled micromanipulator sys- 27(3):59–65
tem for minimally invasive microneurosurgery- 37. Sutherland GR, Latour I, Greer AD et al (2008)
preliminary results. Neurosurgery 51(4):985–988 An image-guided magnetic resonance-compatible
29. Zimmermann M, Krishnan R, Raabe A et al surgical robot. Neurosurgery 62(2):286–292,
(2004) Robot-assisted navigated endoscopic discussion 292–293
ventriculostomy: implementation of a new 38. Greer AD, Newhook P, Sutherland GR (2008)
technology and first clinical results. Acta Human-machine interface for robotic surgery
Neurochir (Wien) 146(7):697–704 and stereotaxy. IEEE/ASME Trans Mech
30. Varma TR, Eldridge PR, Forster A et al (2003) 13(3):355–361
Use of the NeuroMate stereotactic robot in fra- 39. Pandya S, Motkoski JW, Serrano-Almeida C et al
meless mode for movement disorder surgery. (2009) Advancing neurosurgery with image-
Stereotact Funct Neurosurg 80(1-4):132–135 guided robotics. J Neurosurg 111(6):1141–1149
Chapter 7
Impact Model of Spinal Cord Injury

Dorothée Cantinieaux, Rachelle Franzen, and Jean Schoenen
Abstract
Spinal cord injury (SCI) is a frequent disorder with effective treatment still to be developed. Amongst the
various experimental models of SCI, the impact model at the thoracic level is one of the most useful as it
mimics the contusion injury, which represents 33 % of all spinal cord injuries encountered in the clinic.
Key words Spinal cord injury, Laminectomy
1 Material
Lesions are realized on adult female Wistar rats weighing approxi-

mately 200 g. The whole surgical procedure is performed under
general anaesthesia, with a continuous delivery of a mixture of O2
(N25) with 5 % isoflurane for induction and 2–3 % isoflurane for
maintenance.
« Infinite Horizons Spinal Cord Impactor » was acquired from
Precision Systems and Instrumentation, LLC. Version 5.0.
Mini-osmotic pumps are from Alzet, model 1007D with flow
rate of 0.5 μl/h, during 7 days.
1.1 Preparation Lesion is realized at the T10 level. This level is chosen because it is
of Rat for the Lesion rostrally far enough from the central pattern generator (situated at
the T13-L1 levels and whose lesion would prevent any locomotor
recovery). The T10 level can be located by feeling the thorax of the
rat to identify the final floating ribs: T13 vertebra is at this level, so
T10 is three vertebras rostrally further. It is easier to count verte-
bras by feeling them and by feeling spaces between them with a
scalpel. Moreover, the T10 vertebra can be spotted by the fact that
(1) T9-10-11 vertebras are very squeezed, (2) T9 spinous process
is directed backwards (caudal side), T10 spinous process is almost
vertical and T11 spinous process is oriented towards rostral part,
(3) the T8 vertebra is situated just underneath neck fat tissue, and
107
108 Dorothée Cantinieaux et al.
(4) there is a larger space lightly sloping between T8 and T9 com-

pared to other vertebras.
After careful laminectomy at the T10 level, bared spinal cord is
protected with sponge material. Spinous process of the following
caudal vertebra (11th) is smoothed away by mechanical abrasion, up
to obtain a plane surface or even slightly bended, to create a gutter
directed towards the bared spinal cord (for placing catheter later).
The sponge material can then be removed.
1.2 Positioning Rat is placed on the impactor table (see Fig. 1). Position of the
of Rat on Impactor forceps on rachis determines the success of the lesion. They must
and Lesion be fixed to the vertebras directly adjacent to the laminectomy (9th
and 11th), and must penetrate the muscular tissue deeply enough
on both sides (3–4 mm) in order to firmly immobilize the rachis
and stabilize it during impact (see Fig. 2). Forceps must pinch ver-
tebras on both sides between accessory and lateral processes (see
Fig. 3). Jointed arms carrying forceps must be also tightly screwed.
It’s important the bared spinal cord to be horizontal for the impac-
tor tip, so that all of the spinal cord surface to be injured is reached
with the same impact force.
With micrometric screws allowing to adjust impactor table, rat
is placed in such a way that the center of the laminectomized region
is just below impactor tip (to check the exact position of the tip,
bring it down next to the spinal cord).
Fig. 1 Positioning of the rat on the impactor table

Impact on Spinal Cord 109
Fig. 2 Position of the forceps on the rachis
Dorsal spinous process
Articular process
Forceps
Accessory process
Lateral process
Vertebral foram en (containing spinal
cord)
Vertebral body Rib
Fig. 3 Outline of position of the forceps on the thoracic vertebras (transversal view)
Impactor tip must not touch any bone or muscle pieces at the
time of impact. Before performing the impact, the impactor tip is
carefully brought against spinal cord, and then it is raised up by four
micrometric screw turns (this is for the rod to reach a sufficient speed).
The system is now ready, and the lesion is realized at 250 kilodynes.
110 Dorothée Cantinieaux et al.
On adult female Wistar rats weighting 200 g, this impact force

allows to reach a displacement in the spinal cord of approximately
1400–1500 μm.
Following the impact, the tip is raised up, forceps are removed,
and the rat is replaced on a hot carpet maintained at 37 °C.
1.3 Treatment In order to allow the treatment solution to penetrate the injured
Administration spinal cord tissue, dura must be opened. Under the microscope,
and using a very thin needle, a hole is made in the dura (without
touching the spinal cord tissue), allowing a yellowish clear liquid to
escape. The hole is enlarged by gently pulling dura on 4–6 mm2
out. With a Hamilton syringe, 10 μl of treatment or control solu-
tion is delivered on the top of the lesion. Then, the reservoir of a
mini-osmotic pump linked to a catheter and containing the treat-
ment or control solution is placed subcutaneously in the back, cau-
dally to the lesion. A first stitch with a Vycril 5-0 thread is realized
around tank with adjacent muscles, to fix the tank to the muscles.
The next stitches are made all along the catheter, always with mus-
cles, in order to fasten and direct the catheter towards the lesion
(to place it into the gutter made at the beginning) (see Fig. 4).
Catheter is next cut just above the lesion to deliver solution at the
correct place (see Fig. 4). Muscles between stitches are sutured
above the catheter and the skin is sutured with a Vycril 3-0 thread.
Fig. 4 Placement of the mini-osmotic pump and the catheter in the back and suture with muscles to direct
catheter towards lesioned spinal cord. Grey arrow shows a stitch around the tank of the minipump, black
arrows show stitches made along the catheter, and white arrow shows the bared spinal cord above which
catheter must be cut
Chapter 8
Acute Clip Compression Model of SCI

Jared T. Wilcox and Michael G. Fehlings
Abstract
Developing effective therapeutics to treat spinal cord injury (SCI) requires robust preclinical animal mod-
els with substantive clinical relevance. To extrapolate preclinical studies of SCI to human medicine, the
animal model must exhibit the proper pathophysiology processes, including hypoxia-ischemia, neuroin-
flammation, cell death, excitotoxicity, myelin disruption, axonal degradation, astrogliosis, and glial scar-
ring. The modified aneurysm clip compression SCI model has been established and characterized over
three decades of use (Dolan and Tator, J Neurosurg 51:229–233, 1979; Rivlin and Tator, Surg Neurol
10:38–43, 1978; Joshi and Fehlings, J Neurotrauma 19:175–203, 2002). We present the cervical clip
compression SCI model that delivers a bilateral, dorsoventral lesion. The clip compression model is one of
the first non-transection models of SCI (Dolan and Tator, J Neurosurg 51:229–233, 1979), and remains
the only well-characterized SCI model incorporating dorsoventral compression. This approach requires
concentric access to the dura, observant avoidance of spinal roots, and consistent application of force.
The cervical clip compression SCI model has high translatability, and is considered to be one of the most
highly relevant models of experimental SCI from a translational perspective (Kwon et al., J Neurotrauma
28:1525–1543, 2010). The clip compression model provides a robust and highly translational lesion for
evaluation and assessment of cutting-edge and combinatorial therapeutics (Karimi-Abdolrezaee et al.,
J Neurosci 30:1657–1676, 2010).
Key words Spinal cord injury, Rat, Modified aneurysm clip, Compression, Cervical
1 Materials
Rivlin-Tator modified Kerr-Lougheed 1/4 slightly curved aneurysm

clip and custom clip applicator (Rivlin-Tator rat clip and FEJOTA™
mouse clip available from the Fehlings Laboratory) are required
(Fig. 1) [1, 2]. Lesions are realized on Wistar rats of 290 ± 20 g
body weight anesthetized with isoflurane at 5 % induction and 2 %
maintenance delivered with 1:1 NO2:O2 mixture. Standard surgi-
cal instrumentation is employed including aim screw retractor,
angled offset bone nippers, and dull spinal hook (Fine Science
Tools). Procedure is performed with visualization through surgical
microscope with built-in illumination and zoom/focus capability.
111
112 Jared T. Wilcox and Michael G. Fehlings
Fig. 1 Calibrated clip and schematic for force measurement. (a) Schematic
drawing of the 1/4 slightly curved Fehlings modified aneurysm clip for rat models
of SCI. (b) The rat clip is 1.5 × 1.0 × 0.1 cm with four components assembled. (c)
The clip is applied using an applicator modified with a triggered quick-release
mechanism. (d) Clips are calibrated using physical measurements of the clip and
forces at a 45° tangential line of force at 1-mm intervals. Computer-generated
graphics created by Nikolai Goncharenko and used with permission
2 Methods
2.1 Landmarks Standard care for animals, instrumentation, and safety are observed
and Laminectomy throughout the procedure. We recommend gaseous anaesthesia
with a mixture of isoflurane and nitrous oxide to allow rapid induc-
tion, smooth anaesthesia, rapid awakening postoperatively, and
avoidance of intraoperative hypotension. To realize a lesion with
involvement of the muscles of the paw and not the shoulder, the
calibrated clip (Fig. 1) is applied to the cord at the C7 vertebral
level while avoiding the corresponding C8 nerve roots. The
Clip Contusion of Spinal Cord 113
Fig. 2 Landmarking and laminectomy. (a) A dorsal approach is taken to access the cervical spinal column,
retracting superficial and intermediate muscles, with the spinalis muscle layer incised along the midline
(arrow) and retracted. (b, c) With the clear exposure of vertebral laminae immediately medial to facet joints,
angled bone nippers are used to remove the laminae at the most lateral aspect without underlying tissue dam-
age. (d) The location of tendon-like polygonal tissue (arrows) demarcates the interradicular space for approach
prominent spinal process of T2 is evident with palpation, and the

splenius capitis muscle completely overlies the C4 vertebrae.
Counting rostrally or caudally from these positions clarifies verte-
bral levels. The procedure is initiated by a skin incision made from
occipital protuberance to T2. The cervical spinal column is accessed
(Fig. 2): (1) latissimus and rhomboid muscles are cut along the
midline raphe from the nuchal ligament caudally to the muscular
origins at T2, (2) superficial muscles are retracted with a spring
retractor, (3) incision is made at the midline raphe of spinalis mus-
cle extended rostrally to end before the splenius capitis and resid-
ing artery, (4) attachments of spinalis and deep muscles are removed
from the vertebrae to be laminectomized (C7, T1) and partially
from one additional lamina rostral and caudal (C6, T2), (5) mus-
cles are removed laterally to the articular facets by scraping these
vertebrae with a scalpel, and (6) spinalis is then retracted with an
aim screw retractor. The ligamenta flava of the junctions corre-
sponding to the injury level (i.e., C6/C7 and C7/T1) are then
removed with fine forceps. Angled offset bone nippers elicit the
laminectomy by cutting the laminae as close to the left and right
articular facets as possible. Very little bleeding occurs with proper
avoidance of vessels residing adjacent to the T2 spinal process, and
within the splenius capitis, ligamenta flava, and lateral posterior
vertebral processes.
2.2 Extradural Ease of entry to the cervical cord—with clarity provided by C7/T1
Microdissection laminectomy and muscle retraction—determines the success of clip
of the Cervical Spinal application (Fig. 3). Posterior vertebral processes are laminecto-
Cord mized to an extent allowing the surgeon to visualize the C8 and
T1 dorsal spinal roots at the C7/T1 intervertebral bony junction.
Extension of laminectomy is made using Friedman-Pearson ron-
geurs with 0.5 mm cup. Entrance to the ventral aspect of the cord
is made at, or immediately rostral to, the C7/T1 bony junction to
preserve the dura and ventral spinal vessels. Explicitly, the guiding
Fig. 3 Extradural microdissection and applying the clip. (a) Guiding hook is inserted into the C7/T1 bony junc-
tion to form a patent canal for the clip along the ventral spinal cord. (b) Once the clip is inserted between the
C8 and T1 nerve roots until the distal tip is visible, the lesion is generated as with the contusive impact of the
top clip blade and subsequent 60-s compression by clip-specified closing force. (c) Bruising of the cord occurs
within seconds, resultant from the contusive force of the clip closing (arrowheads) indicating proper level of
tissue damage. (d) Lesion epicenter stained with LFB/H&E displays a central cavitation and scarring (arrow-
heads), significant grey matter loss, and subpial white matter sparing (arrows)
hook and clip are inserted bilaterally in the space between the C8
and T1 spinal roots. Visualizing the triangular tendon-like appear-
ance on the dorsolateral surface of the cord reveals the interradicu-
lar space. Landmarking this space for dural entry is imperative to
avoid distraction, contusion, and avulsion of the spinal roots, which
greatly impair the paw and upper limb. The guiding hook is (1)
inserted into the space caudal to the C8 roots, (2) gently rotated
around the ventral aspect of the cord using (3) small rostrocaudal
angular sweeping motions to loosen the natural adhesion of the
dural to the posterior face of the C6 vertebral body (4) until the
hook tip is seen on the contralateral space.
2.3 Applying the Clip The modified aneurysm clip applies a bilateral contusion followed
by sustained dorsoventral compression (Fig. 3b) [3]. Clips are
inexpensive and are calibrated using common instrumentation
with highly reliable closing strengths (Fz, Fig. 1d) determined by
mathematical reduction of empirical measurement. Stated strengths
refer to the closure of blades at 5° displacement regressed linearly
using sine law under assumptions of Hooke’s law (calculations can
be found in Ref. 1 using Fig. 1d schematic). Meticulous care of the
instrumentation is required. Prior to application, the clip is opened
using an applicator with locking and quick-release mechanism
(Figs. 1d and 3a). The hook is used to guide the bottom blade of
the clip through the space between C8 and T1 spinal roots until
the blade tip is visualized in the contralateral space. Any restriction
in movement while advancing the clip blade is overcome with gen-
tle rostrocaudal pivoting. Continued resistance cannot be over-
come by advancing the clip, and thorough ventral dissection with
a fine nerve hook is instead performed slowly to avoid cord manip-
ulation. Following smooth insertion of the clip around the cord,
the applicator is held so the arm contacting the bottom clip blade
is immobile and the other arm able to move freely. It is imperative
that the bottom blade of the clip does not roam upon clip release,
and the tips of the blades do not contact any bone or muscle.
The release mechanism is then triggered so the clip snaps closed
with a standardized contusive impact [4, 5]. The clip is fully released
without any contact of the applicator for 60 s. Duration of com-
pression can be varied; however, we recommend a standardized
1-min period of compression to reliably produce posttraumatic
ischemia [6]. The applicator is then used to fully open the clip and
remove it from the cord. The contusive impact initiates the primary
physical insult while the compressive force realizes the secondary
pathological processes essential to modelling the clinical presenta-
tion of traumatic SCI. Consistent application of the clip provides a
resultant lesion exhibiting robust and highly reproducible central
cavitation, astrogliosis, neuroinflammation, and spared subpial
axonal rim (Fig. 3d) [3, 5].
2.4 Wound Closure Following application and removal of the clip, retractors are
and Monitoring released and the surgical field is monitored until bleeding and CSF
leakage cease. A rectangular piece of surgifoam is placed over the
exposed dura to provide a transient barrier. Muscles superficial to
the spinalis muscle are closed in layers using 5-0 silk semicontinu-
ous sutures. Skin incision is closed using Michaels wound clips or
discontinuous absorbable sutures. Animals are administered
0.05 mg/kg buprenorphine and 5 cc saline sq, removed from iso-
flurane, and followed until verification of shoulder abduction and
elbow extension. Animals are fed until competent of self-care, and
bladders are expressed until urinary continence returns. Contracture
and weakness of the forepaw should be evident, but shoulder
involvement, cervical kyphosis, and spastic rigidity are unfavorable
outcomes.
2.5 Quality Grey matter destruction, central cavitation, and white matter pres-
Assurance ervation in the subpial rim are expected and consistent (Fig. 3d).
and Outcome Confirmation of reproducibility is performed using lesional dimen-
Measures sion analysis with Abercrombie equation, Cavalieri method, or
StereoInvestigator software (MBF Bioscience). The advantage of
cervical clip compression model (C7 and rostral) is the utility of
neurobehavioral outcomes to assess return of function to the fore-
limb. These include, but are not limited to, the use of grip strength
meters, staircase reaching/grasping test, ladder/grid walk, inclined
plane, catwalk gait assessment, IBB Forelimb Scale, and electro-
physiological measure of motor-, sensory-, and spinal cord-evoked
potentials, H reflex, and H/M Spasticity Index.
3 Notes
● While clip-spring combinations are reliable for over 900
open-close cycles, routine cleaning and calibration should be
performed.
● Meticulous instrument care is required, including periodic clip
re-calibration before and after each set of experiments or 40
applications.
● Care should be taken to avoid the large subcutaneous vein div-
ing into the thoracic muscles at T4 with tributaries adjacent to
T2 spinal process.
● If surgical field needs widening, spinalis can be cut just caudal
to C5.
● Excessive distraction of the spinal roots is evident by a reduced
grip strength or increased paw contracture on the side of clip/
hook insertion.
● Radiculopathy due to the hook/clip is easily reduced by using
a narrow (<0.8 mm) and dulled hook, and carefully visualizing
the ventral roots.
● Fine forceps can be used to carefully move the lateral edge of

the cord (padded with gauze) towards the midline to visualize
nerve roots.
● Damage to the dorsal nerve roots will cause rats to chew their
forepaws.
● Changes in the depth of the clip around the cord are evident as
this causes cord tissue to be “pinched” between the blades
close to the roller.
● Placement of the clip around the cord is the cornerstone of the
model, as this determines all impact and forces and the result-
ing lesion.
● To eliminate all movement of the clip, insert the hook into the
contralateral side and place it firmly underneath the bottom
blade of the clip.
References
1. Dolan EJ, Tator CH (1979) A new method for 4. Kwon BK, Okon EB, Tsai E et al (2010) A
testing the force of clips for aneurysms or exper- grading system to evaluate objectively the
imental spinal cord compression. J Neurosurg strength of pre-clinical data of acute neuropro-
51(2):229–233 tective therapies for clinical translation in spinal
2. Rivlin AS, Tator CH (1978) Effect of duration cord injury. J Neurotrauma 28:1525–1543
of acute spinal cord compression in a new acute 5. Karimi-Abdolrezaee S, Eftekharpour E, Wang
cord injury model in the rat. Surg Neurol J et al (2010) Synergistic effects of transplanted
10(1):38–43 adult neural stem/progenitor cells, chondroiti-
3. Joshi M, Fehlings MG (2002) Development and nase, and growth factors promote functional
characterization of a novel, graded model of clip repair and plasticity of the chronically injured
compressive spinal cord injury in the mouse: Part spinal cord. J Neurosci 30(5):1657–1676
1. Clip design, behavioral outcomes, and histopa- 6. Fehlings MG, Tator CH, Linden RD (1989)
thology, and; Part 2. Quantitative neuroanatomi- The relationships among the severity of spinal
cal assessment and analysis of the relationships cord injury, motor and somatosensory evoked
between axonal tracts, residual tissue, and loco- potentials and spinal cord blood flow. EEG Clin
motor recovery. J Neurotrauma 19(2):175–203 Neurophysiol 74(4):241–259
Chapter 9
Microsurgical Approach to Spinal Canal in Rats

Mortimer Gierthmuehlen and Jan Kaminsky
Abstract
The rodent spine is used for a variety of models, including spinal instability (de Medinaceli and Wyatt, J
Neural Transplant Plast 4:39–52, 1993), neuronal regeneration (Kwon et al., Spine 27:1504–1510,
2002), infection studies (Ofluoglu et al., Arch Orthop Trauma Surg 127:391–396, 2007), and studies
about the cauda-equina-syndrome (Kobayashi et al., J Orthop Res 22:180–188, 2004). It is an interdisci-
plinary target for urologic (Hoang et al., J Neurosci 26:8672–8679, 2006), orthopedic (Iwamoto et al.,
Spine 20:2750–2757, 1995; Spine 22:2636–2640, 1997), neurologic (Takenobu et al., J Neurosci
Methods 104:191–198, 2001), and neurosurgical (Xiao and Godec, Paraplegia 32:300–307, 1994) ques-
tions. However, no standard procedure to approach the spinal cord in rats has been published in detail. We
present a description of a dorsal approach to the spine, spinal canal and myelon of the rat. This approach
provides sufficient exposure of the neural structures to perform extended microsurgery at the spinal nerve
roots, the lateral and dorsal myelon and vertebral structures under a surgical microscope. Perioperative
management, anesthesia, and anatomical landmarks are discussed and common pitfalls are described.
Key words Dorsal approach, Spine, Spinal cord, Rat, Nerve root
1 Introduction
The rodent spine is used for a variety of models, including spinal

instability [1], neuronal regeneration [2], infection studies [3],
and studies about the cauda-equina-syndrome [4]. It is an interdis-
ciplinary target for urologic [5], orthopedic [6, 7], neurologic [8],
and neurosurgical [9] questions. However, no standard procedure
to approach the spinal cord in rats has been published in detail. We
present a description of a dorsal approach to the spine, spinal canal
and myelon of the rat. This approach provides sufficient exposure
of the neural structures to perform extended microsurgery at the
spinal nerve roots, the lateral and dorsal myelon and vertebral
structures under a surgical microscope. Perioperative management,
anesthesia, and anatomical landmarks are discussed and common
pitfalls are described.
119
120 Mortimer Gierthmuehlen and Jan Kaminsky
2 Materials
The procedure is performed in female Wistar rats weighing between

250 and 300 g. Anesthesia is done with intramuscular administra-
tion of ketamine 10 % (0.75 ml/kg bodyweight (BW)) and medeto-
midine (0.15 mg/kg). The instruments we used for surgery are
listed in Table 1. An operating microscope is necessary once the
laminectomy is done.
A self-designed OR table (Fig. 1a) allows stable positioning of
the rat with moderate kyphosis for optimal exposure of the spine.
It also allows the surgeon to position the hands comfortably placed
on each side of the animal. A heat lamp (standard infrared lamp for
human use) and a warming pad (Thermolux Pet-Mat 10 W
attached to a continuously variable dimmer switch) provide con-
stant temperature of the animal during surgery. Sufficient hydra-
tion of the animal is ensured by intermittent subcutaneous injection
of 0.9 % NaCl (1 ml/h/100 g BW).
3 Methods
3.1 Anatomy The lumbar spine of the rat consists of six vertebrae. The landmark
for the caudal lumbar spine is the pelvis; the sixth lumbar vertebra
can be identified between the two cristae iliacae (Fig. 1). This is the
most caudal level of the spinal canal that can be opened safely since
the sacral spinal roots leave the spinal canal dorsally and preparation
in that area is extremely difficult.
Table 1
Instruments we used during surgery
Instrument Manufacturer Comments

OR table Self-made Providing kyphosis and warming
Scissors Pfeilring
Rongeur Niegeloh
Forceps (anat/surg.) Aesculap
Needle-holder Aesculap
Micro-forceps Aesculap
Micro-scissors Aesculap
Micro-needle holder Aesculap
Dura hook (sharp) Aesculap
Nerve dissector Self-made 2-0 suture on top of a dental
instrument (Fig. 15)
Microsurgery of Spinal Cord 121
Fig. 1 (a) The operating table with a continuously adjustable dimmer (A) controls the heat of the warming pad
(B). This pad is formed to provide a kyphosis. The hand rests on both sides (C) allow stable positioning of the
hands. Small clamps (D) hold retracting sutures as necessary. (b) CT scan of the lumbar spine of a rat. The sixth
lumbar vertebra is located between the cristae iliacae (arrow)
3.2 Approach The coat at the operation field is removed with an electric shaver
to the Lumbar Spinal (a professional long-hair shaver is better suited than a cheap stan-
Cord dard shaver). After palpating the cristae iliacae, the L6 spineous
process is identified marking the most caudal process. After disin-
3.2.1 Skin Incision
fection of the skin (Kodan®), a midline incision is done from 2 cm
and Preparation
cranial to 1 cm caudal to the cristae iliacae. The skin is retracted
of the Laminae and Spinal
with 2-0 suture, followed by blunt subcutaneous. After the subcu-
Processes
taneous connective tissue is carefully lifted with forceps and cut
away (Fig. 2)—otherwise it might get entangled in the drill and
cause trouble—the fascia of the paravertebral muscles becomes vis-
ible. The paravertebral tendons connecting to the L6 spineous
process can be identified as the last white stripe before the muscles
attach directly to the sacral bone (Fig. 3). The fascia is incised
superficially and bilaterally to the spineous processes from L3 to
L6 (Fig. 3), while a cranio-caudal direction allows cutting the
paravertebral tendons. Of course, the fascia could also be incised in
a caudo-cranial way, but the anatomy of the tendons can easily
misguide the blade of the scalpel laterally. Again, blunt preparation
is needed to separate the paravertebral muscles from the spineous
processes; the remaining paravertebral tendons are dissected with
scissors. Great care should be taken not to go too lateral but to
dissect closely to the spineous processes. A lateral and deep prepa-
ration may damage the spinal roots emerging from the spinal canal.
A small retractor is inserted to hold back the paravertebral muscles,
and the interspineous tendons are dissected to make the spineous
processes visible (Fig. 4).
Fig. 2 The rat is positioned on the OR table - right side points cranially, left side
caudally. The edges of the wound are retracted with 2-0 suture attached to small
clamps on the operating table. The subcutaneous connective tissue (containing
the blood vessels seen in the photo) should be removed as it may get entangled
with the drill. The paravertebral muscles become visible. The spineous process of
L6 can be localized as the most caudal insertion point for paravertebral tendons
Fig. 3 The sacrum (1), the spineous processes L6 (2), L5 (3), and L4 (2) and the
paravertebral tendons (lines) shine through the muscular and subcutaneous tis-
sue. The tissue should be incised bilaterally in a cranio-caudal direction (arrows).
This prevents the scalpel from being misguided laterally by the paravertebral
tendons
3.2.2 Identification By holding the spinal process of L6 with forceps and moving the
of the Correct Spinal Level sacrum backward and forward with two fingers, a movement
between the L6 and the S1 spineous process becomes obvious.
This again ensures the correct level, as the spineous processes of S1
and S2 do not show any mobility against each other. After laminec-
tomy, the level of L1/L2 can be confirmed if the caudal cone of
Fig. 4 The paravertebral and interprocessous tendons are dissected with scis-
sors and the processes L6 (2), L5 (3), L4 (4), and L3 (5) and the sacrum (1) can
be identified. It is helpful to hold the L6 process with the forceps and move the
sacral bone to identify the correct level. Mobility is seen between L6 and S1, but
absent between S1 and S2
the myelon is seen intradurally. It is easily identified by a small

venous plexus on its dorsal surface.
The identification of the correct level is already a critical step in
the lumbar preparation, but it becomes even more difficult when
surgery at a specific thoracic level is necessary. In this case it is
advisable to identify the lumbosacral junction and count the spine-
ous processes cranially until the target region is found. As the tho-
racic spine consists of 13 vertebrae and anatomical variations are
known, other ways to identify the correct region have been
described (LITERATUR de Medinaceli).
3.2.3 Laminectomy The spineous processes are now removed with a small rongeur.
From this point on, an operation microscope is used.
A large drilling head is chosen to clean the operating site
(Fig. 5). A combined irrigation-suction device (Hydroflow®) is
helpful in providing good vision. Bleeding mostly occurs in the
space between the facet joints and can be coagulated with the
bipolar forceps, but since the spinal nerves emerge only a few mil-
limeters below this area, the coagulation power is adjusted to the
minimal possible. A cranial-to-caudal direction is chosen to open
the spinal canal as its diameter decreases caudally. The drill is now
held perpendicular to the spine; otherwise it could get stuck in
osseous structures and damage the area lateral to the vertebrae.
The lateral edges of the vertebrae and the facet joints are also
reduced, making it more comfortable to cut them in a later stage
of surgery (Fig. 6). Remaining tendons attached to the facet joints
can also be cut now.
Fig. 5 The laminae L6 (dashed line 1), L5 (dashed line 2), L4 (dashed line 3), and
L3 (dashed line 4) can be identified, also their respective facet joints (bold lines
1–4). At this stage, bleeding almost always occurs in the space between the
facet joints and can easily be terminated with bipolar forceps. This is the last
stage of surgery where the bipolar forceps can be used. Once the dura is visible,
the bipolar should be avoided, as heat and electricity may damage the nerves or
cause uncontrollable movements of the rat
Fig. 6 With a smaller drilling head and much irrigation, the laminae are thinned
out and the lateral edges, including the facet joints, are reduced. The spinal canal
(dashed line) and the nerve roots (arrow) become visible. The spinal parts of the
facet joint capsule appear as white stripes in the intervertebral region (white
dots). Here, the bone layer is much thinner. The area framed by a white rectangle
is shown in Fig. 7
The laminae are not entirely removed with the drill, but a small
bone shell is left for safer manual removal. This layer is identified
by small cracks reflecting the light of the OR microscope (Fig. 7).
Drilling is stopped when these small cracks become visible over the
entire length of the spinal canal.
3.2.4 Entering At this step, the rat is positioned with the lower legs extended and
the Spinal Canal hanging downwards, since the following manipulation at the spinal
canal may provoke neural and muscular activity. The small cracks in
the thin osseous layer are inspected with a sharp hook. The hook
should therefore be slipped under the lateral edge of the osseous
layer which is then carefully lifted (illustrated in Fig. 7). Starting
this procedure at the level of a facet joint again reduces the risk of
inadvertently penetrating the dura. Small fragments are removed
with micro-forceps. Bleeding is stopped with warm water and
Gelita®. The intraspinal, fibrous capsules of the facet joints (see
Fig. 7) may sometimes mimic dissected nerve roots, but they are
thicker and harder to remove than nerves. The rongeur is used to
carefully cut away the remaining lateral edges of the laminae
(Fig. 8), and the dural sack is prepared (Fig. 9).
3.2.5 Intra- By carefully resecting the vertebrae’s pedicles it is possible to

and Extradural Approach achieve a far lateral approach to the dural sack and the vertebral
discs. In the lumbar spine this is relatively easy when the emerging
Extradural Preparation
spinal nerves are respected. In the thoracic region the attached ribs
and the thorax make a lateral approach much more complicated.
Fig. 7 The thin bone layer is mobilized with a sharp hook (insertion). By gently rotat-
ing the hook’s tip, the layers can be lifted and removed with micro-forceps. Once the
bone layer is gone, it is essential not to get confused by fragments of the facet joint
capsule which may appear as damaged nerve stumps. It is advisable to remove
these fragments, as they can interfere with the following steps of surgery
Fig. 8 Once all bone layers and facet joint fragments are removed, the entire
spinal canal and the dura become visible. But the opening is still much too small
to safely show nerve roots or even open the dura. It is necessary to reduce the
lateral edges (dashed lines) with a rongeur. This should be done extremely care-
fully, since damage to the nerve roots should be avoided
Fig. 9 Once the spinal canal has been widely opened, the nerve roots L6 (bold
arrow), L5 (dashed arrow), and L4 (arrowhead) are identified. The dura is incised
with a sharp dura hook. It is advisable not to cut the dura open in the middle but
to create a larger flap on the side where the dura can be retracted with a suture
Although the rongeur’s size might look too big for the resection of
the pedicles, injury to the spinal nerves is rare. Again, annoying
venous bleeding from the bone is easily controlled with bone wax.
Intradural Preparation If intradural surgery is necessary, a sharp dura hook can be used to
open the dura. This should be started cranially, and once a small
Fig. 10 The dura is opened laterally and the flaps are retracted to the muscular
wall with 7-0 dura suture (arrowheads)
distance (e.g., 5 mm) is opened, the edges should be attached to

the paravertebral muscles with a 7-0 suture before continuing
caudally (Fig. 10). This is much easier than opening the whole
dura in one step since the dural edges retract laterally and are hard
to find. A small piece of 2-0 prolene suture held by a dental instru-
ment is used as an atraumatic nerve dissector (Fig. 15).
Identification of nerve roots can either be done by stimulation
or by anatomical landmarks. We used an AD-Instruments setup
and low stimulator settings (e.g., pulse width 100 μs, 1–5 mA and
manual activation) in order to provoke muscle answers in the lower
limbs. While the motoric roots of L4 and L5 are relatively thick
and innervate the legs—stimulation results in movement of the
thigh (L4) and the calf (L5)—L6 is rather thin and innervates the
tail. Anatomically, L6 leaves the spinal canal after the sixth lumbar
vertebra and is thin—L5 and L4 can be identified by counting
cranially. Motoric roots are ventral, and sensory roots dorsal. Small
pieces of 2-0 suture can be placed under the spinal nerve (Fig. 11a)
in order not to lose it. We used this approach for a nerve regenera-
tion study (Fig. 11b) where we performed a microsurgical anasto-
mosis between the L4 and L6 ventral roots. In another project, we
labeled the L4–L6 ventral roots with anterograde neural tracers,
each in a different color. Therefore, we took three small sterile rub-
ber tubes of appr. 3 mm length, opened the top of each tube, care-
fully inserted a nerve root in each tube, closed both sides with
Vaseline® and filled the remaining space with neural tracer (Fig. 12).
This ensured selective tracing of each nerve root with a specific
tracing color.
Fig. 11 (a) The ventral nerve roots L4 und L6 are identified and marked with tiny pieces of 2-0 suture and, in
this study of nerve regeneration, an anastomosis from L4 to L6 is sutured (arrow in b)
Fig. 12 In this picture, the ventral roots of L4–L6 are put in a small rubber tube
each. Both open sides of the tubes are sealed with Vaseline and a neurotracer is
filled in the remaining cavity
3.2.6 Wound Closure After surgery, a small piece of subcutaneous connective tissue is
prepared and sutured to the dura; the cranial and caudal part is
fixed to connective tissue of the facet joints (Fig. 13). Often, after
intraspinal preparation, a CSF-tight dura closure cannot be
achieved without compromising the spinal nerves. In this case the
dura edges were left open and the tight suture of the muscular
fascia prevented dura leaks instead. The paravertebral muscles are
adapted with 5-0 Vicryl suture; tight 5-0 Vicryl subcutaneous
suture closes the skin (Fig. 14). Skin glue (e.g., Dermabond®) seals
the wound.
Fig. 13 Dural closure is difficult to achieve, since the suture can damage the
nerve roots. Therefore, subcutaneous fascia is used to cover the spinal canal. It
is either attached directly to the dura (left side) or to the lateral muscular wall
(right side)
Fig. 14 The skin is attached with 5-0 resorbable, subcutaneous suture as rats
tend to remove any single-knot, cutaneous sutures. If deemed necessary, skin
glue is used
3.2.7 Postoperative Care After surgery, sufficient warming via heat lamp is essential until the
animal wakes up. Anesthesia can be shortened by administering
Antisedan® (atipamezole, apply same volume as of medetomidine).
For 3 days after surgery, the animals receive the oral antibiotic
Borgal® (trimethoprim/sulfadoxine, 15 mg/kg body weight) and
subcutaneous injection of Carprofen (1 mg/kg body weight q24h)
as analgesic.
Fig. 15 A 2-0 or 3-0 suture attached to a dental instrument or a forceps is a safe

way to dissect nerves
4 Notes
There are several risks that the surgeon should be aware of:
● As in humans, it is easy to be misled and operate on the wrong
spinal level. Moving the spinal processes of L6 and S1 against
each other helps identifying the lumbosacral junction.
Orientation is even more critical in the thoracic region.
● When the paravertebral fascia is opened from caudal to cranial,
the scalpel can be misled by the paravertebral tendons and cut
the fascia too lateral. It is much easier to cut from cranial in a
caudal direction.
● Resect subcutaneous tissue before using the drill—otherwise it
might get entangled.
● Sufficient irrigation is essential during the drilling procedure.
Otherwise the rat may die from overheat.
● At the end of laminectomy, a small bone layer should be left
and resected with micro instruments. It is not advisable to try
to open the spinal canal only with the drill as the dura is really
thin and may be injured.
● As soon as neural structures can be touched—especially nerve
roots from L4 and L5, which innervate the legs—the rat should
be positioned with the legs hanging downwards. If the legs
touch the table, their sudden movement may cause the rat to
jump up and the instruments to injure the nerves.
● Bleeding from bones can be hard to control—bone wax is
really helpful.
● During intradural surgery, provide sufficient irrigation to the
nerves—otherwise they may dry out.
● Depending on the type auf anesthesia and the intention of
surgery, muscle relaxants can be used in order to avoid motoric
responses during manipulation of the nerve roots. In some
electrophysiologic settings it is, however, not possible to
apply relaxants as this medication might interfere with the
investigation protocol.
● Subcutaneous injection of 1 ml/h/100 g BW saline is essential
to prevent renal failure.
● Single-knot non-resorbable sutures for wound closure are
critical as rats tend to bite everything off they can. Tight sub-
cutaneous suture and skin glue (e.g., Dermabond®) seem to
be safer.
● Nieto et al. proposed a titanium mesh graft for reconstruction
of the spinal canal in a thoracic laminectomy model [1].
We did not use this mesh as firstly we operated on the lumbar
segment where only peripheral nerves are present without the
risk of myelopathy. Secondly, the lumbar spine shows a higher
mobility compared to the thoracic spine, resulting in an
increased risk of damage to neural structure in case of disloca-
tion of the titanium mesh.
References
1. de Medinaceli L, Wyatt RJ (1993) A method for ical changes of dorsal root ganglion. J Orthop
shortening of the rat spine and its neurologic con- Res 22(1):180–188
sequences. J Neural Transplant Plast 4(1):39–52 5. Hoang TX, Pikov V, Havton LA (2006)
2. Kwon BK, Oxland TR, Tetzlaff W (2002) Functional reinnervation of the rat lower uri-
Animal models used in spinal cord regenera- nary tract after cauda equina injury and repair.
tion research. Spine 27(14):1504–1510 J Neurosci 26(34):8672–8679
3. Ofluoglu EA et al (2007) Implant-related 6. Iwamoto H et al (1995) Production of chronic
infection model in rat spine. Arch Orthop compression of the cauda equina in rats for use
Trauma Surg 127(5):391–396 in studies of lumbar spinal canal stenosis. Spine
4. Kobayashi S, Yoshizawa H, Yamada S (2004) 20(24):2750–2757
Pathology of lumbar nerve root compression: 7. Iwamoto H et al (1997) Lumbar spinal canal
Part 2. Morphological and immunohistochem- stenosis examined electrophysiologically in a
rat model of chronic cauda equina compres- 10. Brookes ZL, Brown NJ, Reilly CS (2000)
sion. Spine 22(22):2636–2640 Intravenous anaesthesia and the rat microcircu-
8. Takenobu Y et al (2001) Model of neuropathic lation: the dorsal microcirculatory chamber. Br
intermittent claudication in the rat: methodol- J Anaesth 85(6):901–903
ogy and application. J Neurosci Methods 11. Nieto JH et al (2005) Titanium mesh implan-
104(2):191–198 tation—a method to stabilize the spine and
9. Xiao CG, Godec CJ (1994) A possible new protect the spinal cord following a multilevel
reflex pathway for micturition after spinal cord laminectomy in the adult rat. J Neurosci
injury. Paraplegia 32(5):300–307 Methods 147(1):1–7
Chapter 10
Stereotaxic Injection into the Rat Spinal Cord

Charla C. Engels and Piotr Walczak
Abstract
Spinal cord is a frequent target for injection of various therapeutic agents including stem cells, growth fac-
tors, small molecules, or genetic constructs. Due to its fragility and specific anatomical localization injec-
tion has to be performed with great deal of care to minimize injury and assure precision of targeting. This
chapter describes two approaches for gaining access to the spinal cord facilitating safe and efficient intra-
spinal injection.
Key words Stereotaxy, Spinal cord, Rat, Neurosurgery
1 Introduction
The estimated global incidence of spinal cord injury (SCI) is 15–40

cases per million. The spinal cord has minimal regenerative capac-
ity, and, with an average age of 38 years at the time of injury onset,
spinal cord injury is rightfully considered one of the most debilitat-
ing conditions, with massive social and economical consequences
for both the individual and society [1–3].
Demyelination, axonal damage, and scar formation contribute
to a loss of motor and sensory function after SCI. Substantial neu-
rological disability with autonomic dysfunction is also associated
with SCI, originating from sudden or sustained trauma or progres-
sive neurodegeneration after injury [1, 2, 4].
The central nervous system (CNS) has an extremely limited
intrinsic regeneration capability [3, 4]; thus, new strategies aimed
at neuroprotection and/or enhancing regenerative potential are
highly desirable. Stereotaxic injection into the spinal cord paren-
chyma is a commonly used technique often used for the delivery of
stem cells or other therapeutic agents [2, 4, 5].
Electronic supplementary material: The online version of this chapter (doi:10.1007/978-1-4939-3730-1_10) contains
supplementary material, which is available to authorized users.
133
134 Charla C. Engels and Piotr Walczak
This chapter details two straightforward approaches for gaining

access to the spinal cord and delineates a technique for precise
stereotaxic injection into the spinal cord parenchyma.
2 Materials
The described protocol is suitable for adult (6–8 weeks old) female
or male rats, weighing between 200 and 300 g.
2.1 Anesthesia 1. Veterinary-grade isoflurane (Aerrane®; Baxter).

2. Induction chamber (Cat# 941444; Vetequip).
3. Mobile animal anesthesia system (Cat# 901807; Vetequip).
4. Rodent anesthesia circuit mask with hose (Cat# 723026;
Harvard Apparatus).
5. Fluovac Waste Gas Control (Cat# 50206; Stoelting).
2.2 Animal Prep 1. Small animal clipper (Cat# 726114; Harvard Apparatus).
2. Nair® Hair Removal Cream.
3. Betadine Surgical Scrub (Cat# 19027132; Fisher Sci.).
4. Infrared heating lamp (optional).
5. Vetericyn Animal Ophthalmic Gel (optional).
2.3 Stereotaxic 1. Surgical microscope (Leica M320 F12).

and Surgical 2. Stereotaxic frame (Leica; Cat# 39463501).
Instruments
3. Spinal cord surgery adaptor (Stoelting Cat# 51695): The
adaptor should be customized by slightly bending (~20°) the
tips, as shown in Fig. 1.
4. Motorized stereotaxic injector (Stoelting; Cat# 53311).
5. Micro drill (Roboz; Cat# RS6300).
6. Microinjection syringe 10 μl (Hamilton Cat# 7635-01).
7. Hamilton needle, 31 G, 1 in. long (Hamilton Cat# 78035).
8. Cotton swabs (Fisher Sci. Cat# 191301518).
9. Adson forceps (FST; Cat# 11027-12).
10. Dumont forceps (FST; Cat# 11251-35).
11. Scalpel handle (FST; Cat# 10004-13).
12. Scalpel blades (FST; Cat# 100020-00).
13. Fine, straight scissors (FST; Cat# 14058-11).
14. Sutures, Silk 4.0 (Ethicon; Cat# 683G).
2.4 Postoperative 1. Ketofen (Cat# 2193; Pfizer).

Period 2. Injectable sodium chloride 0.9 % (Hospira Inc.; Cat# 4094888).
Injection to Spinal Cord 135
Fig. 1 Customized spinal cord surgery adaptor. Arrows point to the position
where the pins are slightly bent downward
3 Methods
3.1 Anesthesia Place the animal in the induction chamber and adjust the isoflu-
rane concentration on the vaporizer to 3.0 %. As soon as the animal
becomes drowsy and the respiratory rate drops to about 60/min,
place the animal on the table for prepping. The table should be
equipped with an anesthesia circuit mask connected to a hose for
continuous anesthesia with 2 % isoflurane.
3.2 Animal Prep 1. Trim the fur over the back of the animal using a clipper.
2. Apply depilation cream, wait for 2–3 min, wipe the cream with
gauze, and then wash with saline. Note: Avoid longer exposure
to depilation cream as it will cause skin damage.
3. Clean the surgical area with 70 % ethanol.
4. Apply Betadine scrub.
5. Move the animal onto the stereotaxic device with the head fac-
ing the operator (Fig. 2a).
3.3 Surgery Selection of the vertebral level is project specific. The exact level
can be conveniently calculated beginning with either palpation of
the final floating ribs and identification of their origin at T13, or by
palpating the most prominent spinal process between the scapulae
corresponding to T2. To prevent hypothermia, the rat should be
placed on a bed of paper towels, thus preventing the animal from
Fig. 2 Gaining access to the spinal cord. (a) The overall surgical setup, including the operating table with a
stereotaxic frame, and an animal with the attached inhalation anesthesia line. (b) Placement of skin incision.
(c) Placement of stereotaxic spinal adaptors. (d) Vertebral segment prepared for laminectomy procedure by
removal of overlying muscles and tendons. (e) Vertebral segment after laminectomy with visible spinal cord.
Arrowhead points to the posterior medial vein. (f) Access to the spinal cord through the intervertebral space.
The spinal cord is visible within the small opening (arrowhead)
being in direct contact with the steel frame of the stereotaxic

device. To maintain constant body temperature during longer sur-
geries, a heat lamp can also be used. To prevent corneal abrasion
during surgery, ophthalmic gel can be applied after anesthesia
induction.
3.4 Laminectomy 1. Using a scalpel, make a 2–3 cm long skin incision in the mid-
line (Fig. 2b).
2. Clear muscles and tendons on the back and sides of two verte-
bral segments using scissors, scalpel, and cotton swabs.
Bilateral, cranial-to-caudal incisions are made with a scalpel in
the paravertebral muscles very near to the spinous processes.
Bluntly separate the paravertebral muscles from the vertebrae
using cotton swabs. Cotton swabs are preferred over forceps
due to the advantage of absorbent properties. Some degree of
bleeding cannot be avoided at this point, but can be reduced
by washing the area with cold saline.
3. Fix spinal adaptors on the stereotaxic device.
4. Using Adson forceps, grasp the sides of vertebral segment just
beneath the transverse processes and lift the animal to the level
of the spinal adaptor pins.
5. Place the vertebra between both pins and tighten such that the
pins hold the animal partly suspended (Fig. 2c).
6. Clear the lamina of the vertebrae thoroughly using cotton
swabs, forceps, and a scalpel to distinctly visualize the bone
edges (Fig. 2d).
7. Use a micro-drill to cut through the bone along both sides of
the spinous process. This procedure is preferably performed
using a surgical microscope, such as a Leica M320 F12 or the
equivalent. When the lamina becomes loose, grasp the spinous
process with the forceps and gently remove the lamina (Fig 3d).
The spinal cord with its dorsal vein (arrowhead) in the midline
should become visible. If the exposed spinal cord area is too
small for adequate manipulation, the laminectomy can be
extended using the micro-drill described above.
3.5 Access Through The initial steps for this procedure are identical to those of the
the Intervertebral laminectomy up to step 6 of Subheading 3.4. The only difference
Space here is that two neighboring vertebrae have to be carefully cleared
from tendons and muscles.
1. Using forceps and scissors, carefully cut and remove the spi-
nous process of the vertebra proximal to the targeted interver-
tebral space.
2. Make an opening in the ligaments between the arches of two
vertebrae using fine forceps, e.g., Dumont 11251-35 (FST).
Fig. 3 Stereotaxic injection procedure. (a) Stereotaxic device with the angle adjusted for the perpendicular
position of the needle with regard to the spinal cord. (b) Vertebral segment after laminectomy with the needle
in place for an intraspinal infusion. Arrowhead points to the needle tip. (c) Intervertebral space with removed
tendons opening access to the spinal cord. Needle tip is placed to the left of the posterior medial vein
(arrowhead)
The spinal cord should be visible a few millimeters below the

bone level (Fig. 2f). If there is no sufficient space between the
vertebrae to clearly see the spinal cord, a spinal adaptor should
be repositioned to achieve extreme kyphosis.
3.6 Stereotaxic 1. Load the Hamilton syringe with the injection fluid. For thick,
Injection viscous solutions, such as cell suspensions, it is preferable to
load from the back of the syringe. Remove the plunger, loosen
the screw holding the needle (one rotation or so is enough),
pull out the needle just enough to break the seal, and use a
100 μl pipette with attached plastic tip to slowly load the solu-
tion into the syringe. Insert the plunger and tighten the screw.
Less viscous solutions can be loaded through aspiration via the
needle. For both methods, it is important to avoid trapping air
in the syringe barrel.
2. Place the Hamilton syringe with the attached needle onto the
motorized injector and tighten well.
3. Set parameters on the motorized injector, including desired
speed and injection volume.
4. Using a surgical microscope, manipulate the stereotaxic device
to place the needle over the spinal segment to be injected.
5. Adjust the angle of the stereotaxic arm, so the needle is per-
pendicular to the surface of the spinal cord (Fig. 3a).
6. Identify the injection site based on the stereotaxic coordinates.
The posterior medial vein is usually a convenient landmark for
identifying the midline; however, in some cases, the vein has a
tortuous course. An alternative landmark is the tip of the adja-
cent spinous process. The lateral coordinate can be adjusted as
required by the application; however, the medial vein should
be avoided, as shown in Fig. 3b for a laminectomy approach or

in Fig. 3c for an intervertebral approach. Puncturing this vessel
causes serious bleeding and may significantly complicate the
procedure.
7. Immediately prior to inserting the needle into the spinal cord,
expel some fluid using the motorized injector, making sure
that the system is working well and eliminating any dead space.
8. The dura mater is a very rigid membrane and cutting through it
with the needle may cause some difficulty. Forcing the needle
through by applying excessive pressure may damage the spinal
cord; thus, it is preferable to drive the needle up and down, gradu-
ally weakening the dura and finally cutting through. Supplementary
Video 1 material demonstrates how to perform this step.
9. Once the dura is penetrated, retract the needle, allowing the
spinal cord to relax, and then slowly bring the needle down
until it touches the surface of the spinal cord. At that point,
reset the dorsal-ventral coordinate and advance the needle fur-
ther into the tissue until the desired depth is reached.
10. Initialize the infusion on the motorized injector. Note that,
due to the small size of the spinal cord, the injection volume
should not exceed 2 μl and the injection speed should be less
than 0.5 μl/min. After completing the infusion, wait for 2 min
to allow for a reduction of pressure and to minimize backflow
of the injected solution.
11. Slowly retract the needle, remove the rat from the stereotaxic frame,
suture the muscle layer, and then suture the skin with an Ethicon
4.0 suture. Apply Betadine solution over the surgical area.
3.7 Postoperative Postoperative analgesia with Ketofen, 5 mg/kg or the equivalent,

Care should be initiated 30 min before the end of inhalation anesthesia
to prevent unnecessary suffering.
Supplementary Video 1 Injection needle is brought into the field of

view of the surgical microscope and moved sideways to avoid injury to
the dorsal vain resulting in significant bleeding. The needle is moved up
and down applying only moderate pressure to the spinal cord. This
gradually weakens and cuts through the dura matter and allows for safe
insertion of the needle into the spinal cord parenchyma.
References
1. Fehlings MG, Vawda R (2011) Cellular treat- tor cell transplants remyelinate and restore
ments for spinal cord injury: the time is right locomotion after spinal cord injury. J Neurosci
for clinical trials. Neurotherapeutics 8(4):704– 25(19):4694–4705
720 3. Bhanot Y, Rao S, Ghosh D, Balaraju S, Radhika
2. Keirstead HS, Nistor G, Bernal G, Totoiu M, CR, Satish Kumar KV (2011) Autologous mes-
Cloutier F, Sharp K et al (2005) Human embry- enchymal stem cells in chronic spinal cord
onic stem cell-derived oligodendrocyte progeni- injury. Br J Neurosurg 25(4):516–522
4. Park SI, Lim JY, Jeong CH, Kim SM, Jun JA, 5. Cummings BJ, Uchida N, Tamaki SJ, Salazar
Jeun SS et al (2012) Human umbilical cord DL, Hooshmand M, Summers R et al (2005)
blood-derived mesenchymal stem cell therapy Human neural stem cells differentiate and pro-
promotes functional recovery of contused rat mote locomotor recovery in spinal cord-injured
spinal cord through enhancement of endoge- mice. Proc Natl Acad Sci U S A 102(39):
nous cell proliferation and oligogenesis. 14069–14074
J Biomed Biotechnol 2012:362473
Chapter 11
Surgical Access to Cisterna Magna Using Concorde-Like

Position for Cell Transplantation in Mice and CNS
Dissection within Intact Dura for Evaluation of Cell
Distribution
Miroslaw Janowski
Abstract
The CSF is increasingly considered as an attractive gateway to the central nervous system (CNS). It is
warranted by the direct delivery of therapeutic agents beyond the blood-brain barrier (BBB) and wide-
spread access to the large areas of the brain and the spinal cord. In small animals access to CSF is not trivial.
The cisterna magna is the largest CSF fluid compartment; thus it was selected as a target. Here, I describe
the surgical procedure for efficient and reproducible access and injection of therapeutic agents such as stem
cells to cisterna magna. Due to hydromechanics, the method is distinct from previously described tech-
niques for CSF withdrawal. Finally, I describe the method for CNS dissection within intact dura for evalu-
ation of cell distribution.
Key words Stereotaxy, Spine, Cisterna magna, Concorde position, Spinal cord, Mouse, Neurosurgery
1 Introduction
The CSF is increasingly considered as an attractive gateway to the

central nervous system (CNS). It is warranted by the direct deliv-
ery beyond the blood-brain barrier (BBB) and widespread access
to the large areas of the brain and the spinal cord. There are three
major routes of access to CSF: intraventricular, suboccipital, and
lumbar. Intraventricular route enables most proximal access with
regard to CSF circulation, but at a cost of invasiveness requiring
break of CNS continuity. The lumbar route is executed by the per-
cutaneous puncture at the level of cauda equina, thus excluding
CNS injury. Being the least invasive it is most frequently route to
CSF employed in humans and large animals. In small animals it is
more complex, and practically not available for mice except of few
people having extremely long training [1]. While the mice are of
specific interest due to abundance of transgenic models, the
141
142 Miroslaw Janowski
detailed description of access to CSF in this species is highly desir-

able. The limited lumbar space is another constraint if larger vol-
umes are expected to be withdrawn or injected. Thus in mice
suboccipital route seems to be optimal, especially for the purpose
of administration of solutions or suspensions to CSF. While the
feasibility of percutaneous approach was indicated [2], due to the
tiny size of cisterna magna in mice and close proximity to vital
structures of brain stem, that approach is considered to be danger-
ous. I have developed surgical approach to cisterna magna in mice,
which additionally enables efficient delivery of solutions and/or
suspensions, with supreme safety due to direct visual control via
operating microscopy. That approach is technically demanding;
however it was reproduced by others [3], and that study supported
the approval of clinical trial for the treatment of multiple sclerosis.
The presented procedure overcomes the reversed leakage of solu-
tions/suspensions, the major limitation of this route which previ-
ously forced researchers to shift from intracisternal to intraventricular
route of cell delivery [4]. Here, I describe a technique of intracis-
ternal cell application in mice, which enabled me to avoid signifi-
cant cell leakage and to achieve a widespread subarachnoid
distribution of transplanted cells [5].
2 Materials
2.1 Animals The procedure was developed using B6SJL mice (Jackson
Laboratories).
2.2 Anesthetics Atropine, ketamine, and xylazine.
2.3 Cells Human bone marrow stromal cells were used for transplantation
[6]. Prior to transplantation, the cells were labeled with 1 μg/ml
Hoechst 33342 (Sigma, St. Louis, MO) and 1 μg/ml 5-(and-6)-
carboxyfluorescein diacetate and succinimidyl ester (CFSE;
Molecular Probes, Eugene, OR) for 30 min at 37 °C, washed twice
with PBS, and further incubated with fresh medium for 24 h in
order to minimize efflux of non-binding dye. Prior to transplanta-
tion the cells were counted and suspended in PBS at a density of
10,000/μl.
2.4 Equipment (a) Stereotaxic apparatus (ASI Instruments, Heidelberg,

Germany).
(b) Micromanipulator arm (ASI Instruments, Heidelberg,
Germany).
(c) Operating microscope (OPMI Pico, Zeiss).
(d) Ultra Micro Pump (WPI).
Concorde for Cisterna Magna 143
2.5 Tools (a) Gauze roller of diameter 10–20 mm, depending on the size of
mouse.
(b) Scalpel.
(c) Adson forceps (FST).
(d) Micro-scissors (FST).
(e) Dumont forceps × 2 (FST).
(f) Syringe RN, 20 μl (Hamilton).
(g) Needle RN, Gauge 30, Point Style 4 (Hamilton).
(h) Staples (3 M, Neuss, Germany).
3 Methods
3.1 Animal Following pre-anesthesia with atropine (0.04 mg/kg s.c.) the ani-
Positioning mals were anesthetized with ketamine (100 mg/kg i.p.) and xyla-
zine (16 mg/kg i.p.). Stereotaxic apparatus was used to fix the
head of the animals. For more convenient placing, the posterior
part of the apparatus was lifted to form a 30° angle with the table
surface. After fixing the head in earbars, we put a gauze roller with
its axis in an anterior-posterior dimension under the mouse. The
size of the roller was adjusted so that the line connecting the most
prominent parts of the cranium and the spine formed a 15° angle
with the horizontal line (Fig. 1). The tooth bar was then used to
press the head down on the nasion, so that the line determining
the facial surface formed a 15° angle with the vertical line (Fig. 1).
The lines together formed a 90° angle. In that position, the cis-
terna magna was nearly at the highest point of the mouse body.
Fig. 1 Photographs of the position of the mouse head in the stereotactic frame according to the “concorde-like
position” after insertion of the transplantation needle
3.2 Surgical After appropriate positioning of animal a midline skin incision was
Procedure made using scalpel and Adson forceps from the superior nuchal
line to the level of C3 vertebrae. Then a strict midline blunt dissec-
tion of suboccipital muscles was performed under the operating
microscope using fine Dumont forceps. As the atlanto-occipital
membrane was exposed, 10 μl of cells (10,000/μl) were taken into
a syringe. The syringe was then placed in the pump, which in turn
was mounted on a micromanipulator arm. Subsequently the nee-
dle attached to the syringe was positioned over the midline of the
atlanto-occipital membrane (where its anterior-posterior dimen-
sion is the longest) to form an angle of 60° with the horizontal
line. The latter line corresponds to the line directly leading towards
the cerebello-medullary fissure, which is visible under the operat-
ing microscope through the membrane. The membrane was
touched by the needle tip midway between the occipital bone and
the posterior arc of atlas. With a quick movement of the dial of the
micromanipulator, the membrane was pierced and the needle was
stopped exactly in half of the needle tip slope. It allowed the excess
outflow of CSF through the needle tip (Fig. 2) giving the space for
the cell suspension to be injected without excessive increase of CSF
pressure. The needle was then moved forward to position the
whole needle tip slope inside the cisterna magna. Under the oper-
ating microscope, it was possible to control the needle ending
inside the cisterna magna to protect the brain stem from injury.
Fig. 2 Placing a drop of saline solution on top of the surgical field in order to avoid
an outflow of the hcell suspension through the needle tract
Next, a large drop of saline (approximately 100 μl) was placed on

top of the surgical field to get a convex meniscus over the wound
(Fig. 3). Cells were injected over a period of 10 min. During the
injection it was indispensable to observe the animal breathing
movements, and as in one case of relenting and subsequent stop-
ping of breathing, the needle had to be partially withdrawn imme-
diately until the animal had recovered. Following injection of the
Fig. 3 Schematic representation of the piercing and injection procedure of the

atlanto-occipital membrane. Scheme 1 represents the situation before penetrat-
ing the atlanto-occipital membrane. The first step (scheme 2) includes the pierc-
ing of the membrane. The needle was stopped exactly in half of the needle tip
slope to allow the outflow of CSF through the needle tip to provide space for the
cell suspension to be injected without excessive increase of CSF pressure.
The needle was then moved forward to position the whole needle tip slope inside
the cisterna magna (scheme 3)
cells the needle was kept in place for ten more minutes before with-
drawal. The wound was closed with staples (3 M, Neuss, Germany).
The herein established transplantation method was subse-
quently used to transplant more than 150 transgenic ALS mice
with various stem cell populations observing no surgery-related
complications applying this technique [6]. Four mice died during
the first night after surgery. This happened only once during the
transplantation period, on two consecutive days. No other animals
died prematurely following surgery; hence we consider the applied
surgical method as being safe.
3.3 CNS Dissection The dissection of the entire CNS within the intact dura mater is
Within Intact Dura pivotal for precise evaluation of transplanted cell distribution.
for Evaluation of Cell Therefore, following standard perfusion with 2 % paraformalde-
Distribution hyde (PFA) the entire mice were postfixed in 2 % PFA overnight.
The CNS was then carefully prepared under the operating micro-
scope to dissect the dura from the bone. Initially the limbs and
internal organs were removed leaving skull and spine intact.
Following muscle detachment, the bone dissection was initiated
under operating microscope in lumbar region.
It begun with cutting of fascia between the posterior and trans-
verse processes of two lumbar vertebral bodies with a forceps under
the microscope. Hence, the posterior process of the upper verte-
bral body was broken with the forceps, exposing the spinal cord
with its whitely lucent dura mater, followed by the removal of both
transverse processes. Subsequently, all processes were removed in
both caudal and cranial direction using this technique, but leaving
the vertebral bodies for stabilizing the spinal cord. After removing
the posterior part of the atlas exposing the atlanto-occipital mem-
brane with the subjacent cisterna magna, the occipital and parietal
bones were removed from the caudal direction after carefully
breaking them into little bits. Next, the skull base was prepared
from caudal paying special attention to the dura, which is tightly
attached to the bones in this region. After exposing the whole skull
base including the cranial nerves, the lateral, frontal, and parietal
bones were removed. Importantly, one had to avoid any level
motion to prevent the bones from boring into the brain paren-
chyma. Finally, the vertebral bodies were removed and the bulbus
olfactorius was prepared by removing the nasal crest, which till
then had served as anchorage point for the finger to lock the CNS
into position. Finally, the remaining processes around the spinal
cord were removed. The so-prepared CNS with intact dura mater
was cryopreserved in 30 % sucrose for 24 h, frozen in isopentane at
−56 °C, and stored at −80 °C.
3.4 The Evaluation Prior to analysis of cell distribution the dissected CNS was cut and
of Cell Delivery thaw-mounted on a cryostat (Leica CM3050 S) at 40 μm thick
to Cisterna Magna slices. The slices were evaluated for the presence of the pre-labeled
cells under an Axiovert 135 inverse fluorescence microscope (Carl

Zeiss, Göttingen, Germany). Estimation of cell numbers within
the subdural space was done by counting Hoechst-labeled cell
nuclei on every third slice throughout the CNS, i.e., ranging from
the bulbus olfactorius back to the lumbar spinal cord. Exact count-
ing was restrained by the often very tight clusters of cells found
within subdural space prohibiting single-cell detection. Where fea-
sible, single-cell nuclei were counted; else estimates based on the
area occupied by the cell clusters compared to the area of single
cells were evaluated.
The hMSCs within the intracranial fluid space were usually
found as clusters of cells of various sizes depending on the com-
partment volume they were found in (Fig. 4). However, the cell
density of every cluster was comparable and accounted for approxi-
mately 5000–10,000 cells per cluster. The highest number of cells
was found in the cerebello-pontine angle and the ambient and
basal cisterns. It ranged from 15,000 to 20,000 cells distributed
over both sides of every mentioned space. Approximately 10,000
cells were found in the spinal cord (90 % in the cervical and 10 % in
the thoracic region), and an average of 5000 cells in the IVth ven-
tricle, prepontine, premedullar, and great cisterns, respectively, as
well as on both sides of the optic chiasm and the olfactory nerve
cisterns (Fig. 4). In both quadrigeminal and third ventricle the
total number of cells did not exceed 1000. No cells were found in
the lateral ventricles and on brain convexity. In total, we found
approximately 80 % of transplanted cells in all compartments
described above. The variation between individual mice was mini-
mal (n = 5), though the indication of an average value and a stan-
dard error would be misleading, because we did only approximate
cell counts due to the high density of the cell clusters prohibiting
single-cell counting.
4 Notes
1. It was important that before the introduction of the needle to

the cisterna magna, the atlanto-occipital membrane should be
strained in order to facilitate its puncture. It could be achieved
by slight increase of the inclination of the animal head under
microscopic control.
2. To avoid cell leakage, which is mostly governed by the gravita-
tional forces, it is critical to position cisterna magna as the
highest point of the mouse body.
3. Due to the position of cisterna magna as the highest part of
body the operative wound was lying in the horizontal plain
enabling us to place and maintain a large drop of saline within
the wound during the injection procedure.
Fig. 4 Representative microphotographs of different sites of the subarachnoidal space of brain (a, olfactory
nerve cistern; b, optic chiasm cistern; c, basal cistern; d, ambient cistern; e, prepontine cistern; f, cerebello-
pontine cistern; g, IV ventricle) and spinal cord (h, ventral surface of cervical spinal cord) showing the distribu-
tion of hMSCs 24 h after transplantation into the cisterna magna using the “concorde-like position” method.
Arrows indicate the transplanted cells stained with Hoechst 33342 (blue) and CFSE (green). Scale bar: 200 m
4. In contrast to other forms of intrathecal or intraparenchymal

cell transplantation, the use of the concorde-like position
causes cell sedimentation from the injection site towards both
the cranial and caudal direction. A significant number of cells
could be retrieved at the skull base and in the fourth ventricle
(which are located against the CSF stream) indicating that the
cell distribution depends not only on gravitation, but also on
the pressure of the administrated fluid. The CSF circulation
seems to be insufficient to carry the large cell clusters during
the first day after transplantation.
5. Although the final cell distribution might depend on the prop-
erties of the transplanted cell type and the lesion of the brain
tissue, the application of the “concorde-like position” for cell
injection into the cisterna magna, as introduced in the present
study, uniformly distributes the transplanted cells within the
subarachnoidal space of the brain and spinal cord. This wide-
spread cell distribution is a prerequisite for potential regenera-
tive treatment strategies in neurodegenerative diseases without
a distinct destruction of large CNS areas, which again promote
cell homing towards the damaged areas. Particularly for chronic
neurodegenerative diseases with a disseminated pattern of cell
loss like ALS or AD, we recommend the “concorde-like posi-
tion” for intrathecal cell transplantation.
References
1. Vulchanova L, Schuster DJ, Belur LR, Riedl Bone marrow stromal cells infused into the
MS, Podetz-Pedersen KM, Kitto KF, Wilcox cerebrospinal fluid promote functional recov-
GL, McIvor RS, Fairbanks CA (2010) ery of the injured rat spinal cord with reduced
Differential adeno-associated virus mediated cavity formation. Exp Neurol 187:266–278
gene transfer to sensory neurons following 5. Janowski M, Kuzma-Kozakiewicz M, Binder
intrathecal delivery by direct lumbar puncture. D, Habisch HJ, Habich A, Lukomska B,
Mol Pain 6:31 Domanska-Janik K, Ludolph AC, Storch A
2. Lee IO, Son JK, Lim ES, Kim YS (2011) (2008) Neurotransplantation in mice: the
Pharmacology of intracisternal or intrathecal concorde-like position ensures minimal cell
glycine, muscimol, and baclofen in strychnine- leakage and widespread distribution of cells
induced thermal hyperalgesia of mice. J Korean transplanted into the cisterna magna. Neurosci
Med Sci 26:1371–1377 Lett 430:169–174
3. Harris VK, Yan QJ, Vyshkina T, Sahabi S, Liu 6. Habisch HJ, Janowski M, Binder D, Kuzma-
X, Sadiq SA (2012) Clinical and pathological Kozakiewicz M, Widmann A, Habich A,
effects of intrathecal injection of mesenchymal Schwalenstocker B, Hermann A, Brenner R,
stem cell-derived neural progenitors in an Lukomska B, Domanska-Janik K, Ludolph AC,
experimental model of multiple sclerosis. Storch A (2007) Intrathecal application of neu-
J Neurol Sci 313:167–177 roectodermally converted stem cells into a
4. Ohta M, Suzuki Y, Noda T, Ejiri Y, Dezawa M, mouse model of ALS: limited intraparenchy-
Kataoka K, Chou H, Ishikawa N, Matsumoto mal migration and survival narrows therapeutic
N, Iwashita Y, Mizuta E, Kuno S, Ide C (2004) effects. J Neural Transm 114:1395–1406
Chapter 12
Animal Models for Experimental Neurosurgery

of Peripheral and Cranial Nerves
Joachim Oertel, Christoph A. Tschan, and Doerther Keiner
Abstract
Common experimental models for investigation of cranial and peripheral nerve function after trauma
include sciatic nerve crush injuries and direct cutting of cochlear or facial nerves. Partial nerve transection,
spinal nerve ligation, and chronic constriction injury are applied in neuropathic pain studies. Although
these models are well established due to their potential to create reliable and reproducible results, an
experimental setup for studying incomplete nerve lesions which resemble the intraoperative surgical condi-
tion was missing for years.
In neurosurgery, manipulation of peripheral or cranial nerves—such as in surgical procedures in the
cerebellar-pontine angle or at the skull base—may lead to severe functional loss despite morphologically
intact nerves. In the past years, different therapeutic agents for regeneration of the functional recovery
have been investigated intensely. The authors’ group has developed animal models to investigate the thera-
peutic potential of various substances in incomplete nerve injuries. In these models, the severity of the
nerve lesion with distinct functional loss and recovery depends on the preset jet pressures.
Key words Waterjet dissection, Cranial nerve, Sciatic nerve, Surgical technique, Animal model, Nerve
regeneration, Traumatic injury
1 Introduction
Today, several animal models for analysis of lesion and regenera-

tion processes after peripheral and—to less extent—cranial nerve
lesions are established. Partial nerve injury is commonly performed
for analysis of neuropathic pain mechanisms or for the assessment
of new drugs [1]. The most frequently employed models include
partial nerve transection and chronic constriction injury [1–3].
These models are limited to the examination of non-traumatic
lesions.
For investigation of traumatic nerve injury, nerve crush [4–8]
or disruptive nerve injuries requiring primary suture or nerve graft-
ing are well established [4, 9–12]. Sciatic nerve crush is the most
commonly studied nerve injury model and has been used to test
151
152 Joachim Oertel et al.
numerous neuroregenerative modalities [5–8]. Today, only few

studies have been published which investigate the functional recov-
ery of cranial nerves after traumatic injury [13, 14].
The functional recovery after the trauma is monitored by
motor and sensory function tests and electrophysiologic evaluation
[12, 15, 16]. For evaluation of functional recovery of the sciatic
nerve, walking track analysis and calculation of the sciatic function
index (SFI; 12) or power production of the hind legs such as exten-
sor postural thrust [16] is performed. For analysis of the morpho-
logical recovery, histological or histomorphometric evaluation and
electron microscopy of the nerves and innervated muscles are com-
monly performed [1, 5].
All animal models provide reliable and reproducible results,
but none of the models evaluates the functional recovery of an
incomplete nerve trauma with a macro-morphologically intact
structure as it is often seen after neurosurgical procedures at the
cranial nerves VII and VIII.
In general, neurosurgical procedures with involvement of the
facial and the vestibulocochlear nerve exhibit a high risk of iatro-
genic traumatic lesion resulting in functional loss of hearing and
facial nerve palsy. In surgical procedures of pathologies located at
the cerebellar-pontine angle such as meningiomas or acoustic neu-
rinomas, the risk of complete functional loss of both cranial nerves
is high even if nerves remain structurally intact. Due to the amelio-
ration of microsurgical techniques and the latest development in
intraoperative monitoring, the proportion of patients with pre-
served facial and vestibulocochlear function has been improved in
the past years [17–21]. But the proportion of patients with preop-
erative good function of the facial and the vestibulocochlear nerve
and significant deterioration after surgical procedure is still at 20 %
(facial nerve palsy; 18) and at 50 %, respectively (hearing loss; 19).
The authors’ group has developed an animal model of trau-
matic nerve injury that allows the investigation of functional loss of
different severity. The application of the waterjet dissection device
with direct contact of the jet and the nerves allows an incomplete
or complete functional loss depending on the preset jet pressure.
Waterjet dissection represents a surgical technique that com-
bines highly precise parenchymal dissection with preservation of
even small vessels without thermal damage to the surrounding tis-
sue. Basically, a water jet is pushed through a small nozzle in various
pressures. Since the early 1980s, waterjet dissection has been inves-
tigated as a new technique in different surgical disciplines having its
beginning in liver [22]. Nowadays it is generally accepted in this
surgical discipline [23–27]. In the following years, waterjet dissec-
tion has been investigated in further surgical disciplines such as
laparoscopic cholecystectomy [28], kidney [29–31], vascular [32],
or ophthalmologic [33, 34] and dermatologic surgery [35].
In 1997, the research group of the authors started to apply a

newly developed waterjet dissection instrument (Helix Hydro Jet,
Erbe Elektromedizin, Tuebingen, Germany) in the field of neuro-
surgery. The instrument has been approved for clinical applications
in Europe and the USA since 1996. First experimental cadaver
studies in the pig and in vivo studies in the rabbit demonstrated
successfully safe and accurate dissection of brain parenchyma with
preservation of blood vessels [36–39]. Thus, the device has been
applied clinically. Having started with meningiomas and metastases
[40, 41], the operative spectrum has constantly increased to vari-
ous intracranial pathologies [42–45]. Today, more than 200 surgi-
cal procedures have been performed including gliomas of all WHO
grades, metastases, meningiomas, vascular tumors, and epilepsy
surgery [45]. In general, the waterjet application enables precise
brain parenchyma dissection in epilepsy surgery as well as tumor-
brain parenchyma dissection under preservation of even small ves-
sels and the pia/arachnoidea mater with pressures below 10 bars
[43, 45]. Tumor debulking in soft and highly vascularized tumors
is possible [44]. The waterjet dissection device can be applied with
high safety; compared with the current literature, a higher inci-
dence of postoperative worsening or an increased risk of deep
infection or tumor spreading is not observed [45].
Since 2005, the possible use of the waterjet dissection
device to protect cranial and peripheral nerves is under evalua-
tion [46, 47].
It was shown that with jet pressures up to 6 bar the structure
and function of rats’ vestibulocochlear nerves remain intact. With
a jet pressure of 8 bar, an incomplete functional lesion was pro-
duced with complete recovery of the auditory brainstem responses
(ABRs) after 12 weeks with complete recovery of ABRs in 60 % of
the animals. Jet pressures of 10 bar resulted in complete functional
and—although the nerve’s continuity was preserved—severe struc-
tural lesion. No recovery was found after 12 weeks [46].
In a second in vivo model, the technique was adapted to trau-
matic lesions of peripheral nerves [47]. In this model, a distinct
and reproducible lesion of different severity was set on sciatic
nerves of rats. The sciatic nerves showed complete functional
recovery after 1 week if waterjet pressures up to 40 bar were used.
Histologically, the nerve’s structure remained intact. After 12
weeks no signs of direct nerve injury and nerve regeneration are
detected. At pressures of 40–80 bar, anatomically and functionally
evidence for nerve injury is found. At 40 and 50 bar, direct nerve
injury and clinical signs of nerve damage are detected, but the
deficits dissolve completely after 1 week. With waterjet pressures
from 60 to 70 bar, clinical and electrophysiologic regeneration
took up to 12 weeks to resolve. Distinct histomorphological char-
acteristics occurred; pathological “mini” fascicles were detected.

Even with a pressure set at 80 bar, the nerve’s continuity was pre-
served, but clinical and electrophysiological regeneration was
incomplete after 12 weeks.
To our knowledge, these are the first models that allow the examina-
tion of a certain therapeutic modality and its potential effect on the
regeneration process in cranial and peripheral nerves after injuries of a
defined force with functional deterioration and structural continuity.
2 Materials
2.1 Animal Model: Nerve type:

Rat, In Vivo 1. Sciatic nerve
2. Vestibulocochlear nerve
Adult male Sprague-Dawley rats (300–400 g for sciatic nerve
dissection, respectively, 300–450 g for vestibulocochlear dissec-
tion) were used as a mammalian model for the in vivo sciatic nerve
experiments. The animals were housed 1–2 per cage with a 12-h
light/dark cycle and had free access to rat chow and water.
All experiments were conducted in accordance with approved
protocols by the Institutional Animal Care and Use Committee
and the German State Committee of Laboratory Animal Research.
2.2 Description For all experiments on the vestibulocochlear nerves, the Helix
of the Waterjet Hydro-Jet (Erbe Elektromedizin, Tübingen, Germany) was used.
Instrument In 2007 it was followed by its successor Erbejet®2 (Erbe
Elektromedizin, Tübingen, Germany) because of additional device
features [48, 49].
For all experiments on the sciatic nerves, the Erbejet®2 by Erbe
Elektromedizin Company (Tuebingen, Germany) was used. The
waterjet is generated via a medium converter with an electronically
controlled mechanical system (double-piston pump) with a pres-
sure ranging from 1 to 80 bar. The medium converter is connected
to a pencil-like handpiece consisting of a narrow nozzle with a
diameter of 120 μm, and a surrounding suction tube. The gener-
ated water jet is a non-rotated thin laminar liquid jet (Fig. 1).
Sterile 0.9 % isotonic saline is emitted as separating medium with a
volume flow of 1–55 ml/min. The suction pressure can be selected
from −100 to −800 mbar with a maximum suction capacity of
25 l/min. The pressure and the suction can be manually adjusted
by preselection. Depending on the surgical procedure, several dif-
ferent settings can be selected. During surgery, the waterjet
application and pressure are adjusted within the preset range by a
foot pedal. The system has been approved by the regulatory
authorities for surgical use in humans in Germany and the USA.
Fig. 1 Handpiece of the Erbejet®2. The jet nozzle diameter of the neurosurgical
standard applicator is 120 μm, which creates a thin laminar liquid jet
2.3 Electro- 1. We use a portable Viking® and a portable Medelec™ Synergy

physiological Device/ (Version 12.2) N-EP—EMG/EP monitoring system with the
Equipment high-frequency filter regulated for 5 kHz and the low-frequency
filter regulated for 3 Hz.
2. For the monitoring of the sciatic nerves, subdermal paired
needle electrodes (Ambu A/S, Ballerup, Denmark) were used
for recording electrodes and for stimulation.
3. For ABR recording (Fig. 2a), subdermal-paired needle elec-
trodes (Ambu A/S, Ballerup, Denmark) were used as record-
ing electrodes (Fig. 2b). For click stimulation, in-ear
headphones were adapted to the small size of the external rat’s
auditory canal (Fig. 2c).
2.4 Surgical 1. Standard surgical microscope (Zeiss, Germany) with photo/

Equipment video documentation.
2.4.1 Sciatic Nerve 2. A no. 15 scalpel blade.
Dissection 3. Standard microsurgical instruments.
4. A wound expander.
5. Vicryl 5-0 sutures (Ethicon, Germany) for muscle fascia.
Fig. 2 (a–c) ABR recording by a portable Medelec™ Synergy (Version 12.2) monitoring system (a). Subcutaneous
needle electrodes are placed over the left and right posterior convexity, vertex, and neck (b). Click stimuli of
80 db are conducted through tubal earphones, inserted into the rat external auditory canal (c)
6. Ethibond 4-0 sutures (Ethicon, Germany) for skin.

7. Chloralhydrate solution.
8. Tramadol (Tramal®).
2.4.2 Vestibulocochlear 1. Standard surgical microscope (Zeiss, Germany) with photo/

Nerve Dissection video documentation.
2. A no. 15 scalpel blade.
3. Standard microsurgical instruments.
4. A wound expander.
5. A diamond drill (1–2 mm).
6. A bipolar forceps (Aesculap, Germany).
7. Tabotamp (Ethicon, Germany).
8. Wound batting (Merocel®, Medtronic, Germany).
9. Vicryl 5-0 sutures (Ethicon, Germany) for muscle fascia.
10. Ethibond 4-0 sutures (Ethicon, Germany) for skin.
11. Ketamine (Ketanest®).
12. Xylazine.
13. Tramadol (Tramal®).
3 Methods
3.1 Waterjet 1. Anesthetize the animals with chloralhydrate solution ip at a

Dissection dose of 36 mg/kg body weight before surgery and perform
on Peripheral Nerves analgesia with tramadol ip at a dose of 50 mg/kg body weight
(the sedation is to be maintained for the duration of each indi-
3.1.1 Surgical Procedure
vidual experiment).
2. For exact positioning of the needle electrodes, both hindlimbs
have to be shaved and disinfected.
3. Perform a posterior-laterally skin incision parallel to the right
femur with the no. 15 scalpel and open the muscle fascia of the
gluteus muscles (Fig. 3a, b).
4. Expose the sciatic nerve carefully at midthigh level with the aid
of a wound expander (Fig. 3c, d).
5. Mobilize the nerve carefully under microscopic view with a
microforceps and microscissors from its surrounding muscle
fascia until it is exposed from the sciatic notch exit to the part,
where the nerve divides into its specific motor branches
(Fig. 3e, f). Care has to be taken to avoid excessive spreading
of the tissue to prevent nerve damage due to tension.
6. Apply the water jet to the right sciatic nerve. Ideally, the rat is
placed on a computer-controlled linear device that moves with
continuous speed and enables to fixate the handpiece of the
Fig. 3 After skin incision parallel to the femur (a), the gluteus muscles are incised (b) and the sciatic nerve is
identified (c). The cut is enlarged (d) and the nerve is dissected carefully from its surrounding tissue under
microscopic view (e, f) leaving the epineurium intact. Finally, the nerve is exposed from the sciatic notch exit
to the part, where the specific motor branches are divided. SC sciatic nerve, P peroneal fascicle, T tibial
fascicle
Fig. 4 Computer-controlled linear device for reliable adjustment of the cutting

distance from the nozzle tip to the nerve surface to obtain accurate and compa-
rable results
waterjet dissector. For this, we use a specially developed

computer-controlled linear device that was employed for accu-
rate and comparable results (Erbe Elektromedizin GmbH,
Tübingen, Germany; Software Servomanager 6.4.1., Parker
Automation). The device allows a standard adjustment (we use
a 2 mm cutting distance from the nozzle’s tip to the nerve
surface and of a cutting speed of 1 cm/s; Fig. 4).
7. The muscle fascia is closed with 5-0 vicryl single stitches fol-
lowed by intracutaneous skin closure preventing wound
gnawing.
8. Postoperatively, oral analgetics (tramadol 50 mg/kg body
weight) are administered po for 1 week.
Fig. 5 (a, b) Perioperative measurement of the distal motor latency and the compound muscle action potential
amplitude for calculation of the motor nerve conduction velocity before (a) and after (b) waterjet lesion
3.1.2 Electro- 1. The distal motor latency and the compound muscle action
physiological Examination potential amplitude are measured and the motor nerve con-
duction velocity is calculated (Fig. 5a, b). All measurements
are taken for the proximal and the distal site of the sciatic nerve
lesion to determine the neurophysiologic decline proximal of
the lesion.
2. Electrophysiological measurements are performed before sur-
gery, after nerve exposure to verify the nerve’s electrophysio-
logical integrity, directly after the wajerjet-induced nerve lesion
and at the end of the surgical procedure after wound closure.
3. If necessary, nerves are moistened with 0.9 % saline solution to
avoid desiccation during the examination.
4. The pre- and postoperative electrophysiologic measurements
are performed on both sciatic nerves.
5. The recording (different) electrode is placed under the skin
above the tibialis anterior muscle and its reference (indiffer-
ence) electrode is placed 1.0–1.5 cm distally above the tendon
of the tibialis anterior muscle. For stimulation, the cathode is
placed at the popliteal fossa for distal stimulation and at the
sciatic notch proximal to the site of the lesion for proximal
stimulation; the anode is placed in the paraspinal muscles. The
ground electrode is placed in the tail.
6. We apply a 20 ms supramaximal stimulus to generate an action
potential, but care is taken to keep the stimulation intensity to
less than 7 mA.
7. We carried out serial follow-up neurophysiologic examinations
at day 1, and at 1 and 12 weeks. For examination, the animals
have to be anesthetized (see Sect. 3.1.1).
3.2 Waterjet 1. Anesthetize the animals with 10 % ketamine ip at a dose of

Dissection 1 ml/kg body weight and 2 % xylazine ip at a dose of 0.5 ml/
on Cranial Nerves kg body weight before surgery (the sedation is to be main-
tained under sedation for the duration of each individual
experiment).
2. The skin of the neck has to be shaved and disinfected.
3. The head of the rats is fixed in anteflexion and prone position
in a stereotactic frame system.
4. In general, the right side (or the best ABR-responding side) is
chosen for surgery.
5. Perform a straight skin incision with a no. 15 scalpel parallel to
the midline, and then incise the neck muscles. The suboccipital
cranium is exposed with the aid of a wound expander.
6. Under microscopic view, perform a lateral suboccipital craniec-
tomy with the aid of a diamond drill of 1–2 mm caliber
(Fig. 6a). The transversal and sigmoid sinus has to be exposed
on the edge of the craniectomy.
7. Open the dura mater and drape it in the direction of the sig-
moid sinus. The cerebellum is retracted upward and to the
middle (Fig. 6b). Use wound batting, if necessary.
8. To get more space, open the cerebellopontine cistern to let out
the cerebrospinal fluid.
9. Expose the vestibulocochlear nerve carefully below the floc-
culus on its course from the brainstem to the internal auditory
canal (Fig. 6c). During this step of nerve exposure changes in
ABR recording are observed in some cases. If this is the case,
the complete recovery of the ABR has to be awaited.
10. Apply the water jet directly to the vestibulocochlear nerves’
surface without the use of the nozzle-integrated suction.
Fig. 6 (a–c) Lateral suboccipital craniectomy by a diamond drill (a). The dura mater is opened and draped in
the direction of the sigmoid sinus (arrow, b). The cerebellum is retracted to the midline. Below the flocculus
the vestibulocochlear nerve is exposed on its course from the brainstem to the internal auditory canal (asterisk, c).
VC vestibulocochlear nerve
11. The dura mater and the skull are closed with autologous fascia
and fibrin gel foam. The muscles and the muscle fascia have to
be tightly sutured by 4-0 ligature. Close the skin with 3-0 liga-
ture single stitches.
12. Observe your animals closely until awake. Oral analgetics (tra-
madol 50 mg/kg body weight or novaminsulfon 20–50 mg/
kg body weight) have to be administered postoperatively for 1
week. Animals presenting with neurological complications
have to be immediately sacrificed.
3.2.2 Electro- 1. The ABRs are recorded preoperatively, intraoperatively before

physiological Examination and directly after lesion and after wound closure.
2. The pre- and postoperative recording has to be performed
bilaterally. Before surgery, select the best responding site.
3. Subcutaneous needle electrodes are placed over the left and
right posterior convexity, vertex, and neck. Click stimuli of
80 db are conducted through tubal earphones that are inserted
into the rat external auditory channel.
4. We carried out serial follow-up examinations bilaterally at day
1, week 1, week 2, and week 6 on anesthetized animals (see
Sect. 3.2.1).
4 Notes
1. For peripheral nerve surgery, it is important to examine ani-

mals of the same weight in the whole experimental trial,
because a direct proportionality between the size of a rat and
the number of sciatic nerve fibers has been shown [50].
2. To obtain comparable results it is of importance to take care of
the same cutting distance and the same cutting speed as it was
received in the authors’ experimental groups by using the
computer-controlled linear device (see Sect. 3.1). There is a
tendency of an increasing lesion with aggravated neurological
deficits in case of a larger distance between the nozzle tip and
the nerve.
3. Heating lamps and/or homeothermic blanket control units
should be used during the surgical procedure and the
neurophysiologic examinations to maintain a body tempera-
ture approximately at 37 °C. Lower temperatures can lead to a
false (positive) pathological distal motor latency or a broad-
ened muscle action potential.
4. During the surgical procedure, iatrogenic excessive spreading
of sciatic nerve surrounding tissue should be avoided to pre-
vent nerve damage due to tension. Prior to the nerve lesion,
the epineurial covering should be left intact to maintain the
nerve integrity.
References
1. Lindenlaub T, Sommer C (2004) Partial sci- peroneal, and posterior tibial nerve lesions in
atic nerve transection. In: Luo ZD (ed) the rat. Plast Reconstr Surg 83:129–136
Methods in molecular medicine, vol 99, Pain 13. Sekiya T, Tanaka M, Shimamura N et al (2001)
research: methods and protocols. Humana, Macrophage invasion into injured cochlear
Totowa, NJ, pp 47–53 nerve and its modification by methylpredniso-
2. Bennet GJ, Xie YK (1988) A peripheral lone. Brain Res 905(1–2):152–160
monoeuropathy in rat that produces disorders 14. Costa HJZR, da Silva CF, Korn GP et al
of pain sensation like those seen in man. Pain (2006) Posttraumatic facial nerve regeneration
33:87–107 in rabbits. Rev Bras Otorrinolaringol 72(6):
3. Ma W, Bisby MA (2000) Calcitonin gene- 786–793
related peptide, substance P and protein gene 15. Martins RS, Siqueira MG, da Silva CF et al
product 9.5 immunoreactive axonal fibers in (2005) Electrophysiologic assessment of
the rat footpad skin following partial sciatic regeneration in rat sciatic nerve repair using
nerve injuries. J Neurocytol 29(4):249–262 suture, fibrin glue or a combination of both
4. Magill CK, Tong AMS, Kawamura D et al techniques. Arq Neuropsiquiatr 63(3-A):
(2007) Reinnervation of the tibialis anterior 601–604
following sciatic nerve crush injury: a confocal 16. Koka R, Hadlock TA (2000) Quantification of
microscopic study in transgenic mice. Exp functional recovery following rat sciatic nerve
Neurol 207(1):64–74 transection. Exp Neurol 168:192–195
5. Pan HC, Chen CJ, Cheng FC et al (2009) 17. Seol HJ, Kim CH, Park CK et al (2006)
Combination of G-CSF administration and Optimal extent of resection in vestibular
human amniotic fluid mesenchymal stem cell schwannoma surgery: relationship to recur-
transplantation promotes peripheral nerve rence and facial nerve preservation. Neurol
regeneration. Neurochem Res 34(3):518–527 Med Chir (Tokyo) 46:176–181
6. Yan JG, Matloub HS, Yan Y et al (2010) The 18. Samii M, Gerganov V, Samii A (2006)
correlation between calcium absorption and Improved preservating of hearing and facial
electrophysiologcal recovery in crushed rat nerve function in vestibular schwannoma sur-
peripheral nerves. Microsurgery 30(2): gery via the retrosigmoid approach in a series
138–145 of 200 patients. J Neurosurg 105(4):527–535
7. Dinh P, Hazal A, Palispis W et al (2009) 19. Samii M, Matthies C, Tatagiba M (1997)
Functional assessment after sciatic nerve injury Management of vestibular schwannomas
in a rat model. Microsurgery 29(8):644–649 (acoustic neuromas): auditory and facial nerve
8. Kalender AM, Dogan A, Bakan V et al (2009) function after resection of 120 vestibular
Effect of Zofenopril on regeneration of sciatic schwannomas in patients with neurofibroma-
nerve crush injury in a rat model. J Brachial tosis 2. Neurosurgery 40(4):696–705
Plex Peripher Nerve Inj 4:6 20. Samii M, Matthies C (1997) Management of
9. Huang JH, Cullen DK, Browne KD et al 1000 vestibular schwannomas (acoustic neuro-
(2009) Long-term survival and integration of mas): the facial nerve-preservation and restitu-
transplanted engineered nervous tissue con- tion of function. Neurosurgery 40(4):
structs promotes peripheral nerve regenera- 684–694
tion. Tissue Eng Part A 15(7):1677–1685 21. Isaacson B, Kileny PR, El-Kashlan H, Gadre
10. Dodla MC, Bellamkonda RV (2008) AK (2003) Intraoperative monitoring and
Differences between the effect of anisotropic facial nerve outcomes after vestibular schwan-
and isotropic laminin and nerve growth factor noma resection. Otol Neurotol 24(5):
presenting scaffolds on nerve regeneration 812–817
across long peripheral nerve gaps. Biomaterials 22. Papachristou DN, Barters R (1982) Resection
29(1):33–46 of the liver with a water jet. Br J Surg
11. Brunelli G, Spano P, Barlati S et al (2005) 69:93–94
Glutamatergic reinnervation through periph- 23. Baer HU, Stain SC, Guastella T et al (1993)
eral nerve graft dictates assembly of glutama- Hepatic resection using a water jet dissector.
tergic synapses at rat skeletal muscle. Proc Natl HPB Surg 6:189–198
Acad Sci U S A 102(24):8752–8757 24. Hata Y, Sasaki F, Takahashi H et al (1994)
12. Bain JR, Mackinnon SE, Hunter DA (1989) Liver resection in children, using a water-jet.
Functional evaluation of complete sciatic, J Pediatr Surg 29:648–650
25. Izumi R, Yabushita K, Shimizu K et al (1993) nozzle types. Acta Neurochir (Wien) 146:
Hepatic resection using a water jet dissector. 713–720
Surg Today 23:31–35 39. Piek J, Wille C, Warzok R et al (1998) Waterjet
26. Rau HG, Schauer R, Pickelmann S et al (2001) dissection of the brain: experimental and first
Dissektionstechniken in der Leberchirurgie. clinical results. Technical note. J Neurosurg
Chirurg 72:105–112 89:861–864
27. Rau HG, Duessel AP, Wurzbacher S (2008) 40. Oertel J, Gaab MR, Piek J (2003) Waterjet
The use of water-jet dissection in open and resection of brain metastases—first clinical
laparoscopic liver resection. HPB (Oxford) results with 10 patients. Eur J Surg Oncol
10:275–280 29:407–414
28. Shekarriz H, Shekarriz B, Kujath P et al (2003) 41. Oertel J, Gaab MR, Warzok R et al (2003)
Hydro-Jet-assisted laparoscopic cholecystec- Waterjet dissection in the brain: review of the
tomy: a prospective randomized clinical study. experimental and clinical data with special ref-
Surgery 133:635–640 erence to meningioma surgery. Neurosurg Rev
29. Hubert J, Mourey E, Suty JM et al (1996) 26:168–174
Water-jet dissection in renal surgery: experi- 42. Piek J, Oertel J, Gaab MR (2002) Waterjet
mental study of a new device in the pig. Urol dissection in neurosurgical procedures: clinical
Res 24:355–359 results in 35 patients. J Neurosurg 96:
30. Penchev RD, Losanoff JE, Kjossev KT (1999) 690–696
Reconstructive renal surgery using a water jet. 43. Oertel J, Gaab MR, Runge U et al (2005)
J Urol 162:772–774 Waterjet dissection versus ultrasonic aspiration
31. Moinzadeh A, Hasan SM et al (2005) Water in epilepsy surgery. Neurosurgery 56:
jet assisted laparoscopic partial nephrectomy 142–146
without hilar clamping in the calf model. 44. Oertel J, Wagner W, Gaab MR et al (2004)
J Urol 174:317–321 Waterjet dissection of gliomas—experience
32. Kobayashi M, Sawada S, Tanigawa N et al with 51 procedures. Minim Invasive Neurosurg
(1995) Water jet angioplasty—an experimental 47:154–159
study. Acta Radiol 36:453–456 45. Keiner D, Gaab MR, Backhaus V, Piek J,
33. Lipshitz I, Bass R, Loewenstein A (1996) Oertel J (2010) Water jet dissection in neuro-
Cutting the cornea with a waterjet keratome. surgery: an update after 208 procedures with
J Refract Surg 12:184–186 special reference to surgical technique and
34. Gordon E, Parolini B, Abelson M (1998) complications. Neurosurgery 67:342–354
Principles and microscopic confirmation of 46. Tschan CA, Gaab MR, Krauss JKK, Oertel
surface quality of two new waterjet-based J (2009) Waterjet dissection of the vestibulo-
microkeratomes. J Refract Surg 14:338–345 cochlear nerve: an experimental study.
35. Siegert R, Danter J, Jurk V et al (1998) Dermal J Neurosurg 110(4):656–661
microvasculature and tissue selective thinning 47. Tschan CA, Keiner D, Müller HD, Schwabe K,
techniques (ultrasound and water-jet) of short- Gaab MR, Sommer C, Oertel J (2010)
time expanded skin in dogs. Eur Arch Waterjet dissection of peripheral nerves: an
Otorhinolaryngol 255:325–330 experimental study of the peripheral nerves of
36. Oertel J, Gaab MR, Knapp A et al (2003) rats. Neurosurgery 67(368):376
Water jet dissection in neurosurgery: experi- 48. Tschan CA, Tschan K, Krauss JK, Oertel
mental results in the porcine cadaveric brain. J (2009) First experimental results with a new
Neurosurgery 52:153–159 waterjet dissector: Erbejet 2. Acta Neurochir
37. Oertel J, Gaab MR, Pillich D-T et al (2004) (Wien) 151(11):1473–1482
Comparison of waterjet dissection and ultra- 49. Tschan CA, Tschan K, Krauss JK, Oertel
sonic aspiration: an in vivo study in the rabbit J (2010) New applicator improves waterjet
brain. J Neurosurg 100:498–504 dissection quality. Br J Neurosurg 24(6):
38. Oertel J, Gaab MR, Schiller T et al (2004) 641–647
Towards waterjet dissection in neurosurgery: 50. Schmalbruch H (1986) Fibre composition of
experimental in-vivo results with two different the rat sciatic nerve. Anat Rec 215(1):71–81
Chapter 13
Surgery of the Brain and Spinal Cord in a Porcine Model

Jan Regelsberger
Abstract
Working time restrictions and economic pressure hinder surgical specialties to implement an adequate and
structured training program. Alternative training forms seem to be requested in which a most realistic
setup is required imitating the daily routine. An in vivo swine model was evaluated for its practical use in
training neurosurgical residents in the past years (Regelsberger et al. Cen Eur Neurosurg 72:192–195,
2010 [2]). Surgical procedures included craniotomy, dura opening, brain surgery with sulcal preparation,
and excision of an artificial tumor as well as laminectomy or other dorsal approaches to the spine with
exposure of the dural sack and nerve roots. Microscopy and bleeding management were an integrated part
of training and were found to be very useful supplements for young neurosurgeons. Our experiences with
these unique in vivo training model are outlined and its advantages and pitfalls described.
Key words Neurosurgical training, Neuroanatomy of the swine, Craniotomy in porcine model, Spine
surgery in pigs, Wet-lab training
1 Introduction
European working time directive (EWTD) may be one of the main

reasons at present why training programs are still judged to be
inadequate and insufficient [3]. Shifting and alternating duties are
in contrast with predictable schedules in the operating rooms and
structured education plans [4]. The high level of medical supply,
pressure of time, and costs may be further strong factors hindering
residents to get skilful surgeons [5]. Therefore adjuvant education
forms seem to be necessary. While hands-on cadaver courses have
spread across the countries we were starting with an in vivo swine
model in cranial and spinal neurosurgery in 2005.
2 Training Model
A swine model was chosen for a neurosurgical training course and

found to be an adequate in vivo object as cranial vault, brain, and
spinal anatomy are easy to mediate allowing young neurosurgeons
165
166 Jan Regelsberger
to focus quickly on their neurosurgical skills. Ideal conditions were

found at the European Surgical Institute (Norderstedt,
Johnson&Johnson, Germany) with a veterinary backup and full
anesthetic equipment where practical experiences in the laboratory
could be exclusively focused on the learning aims [1, 2]. Microscopes
(Möller-Wedel, Wedel, Germany), bipolar coagulation, suction,
and microsurgical instruments (Codman, Johnson&Johnson) were
provided to install an almost realistic copy of the daily situation in
the operating room (OR).
2.1 Neuroanatomy The central nervous system (CNS) of all vertebrates is topologi-
and Brain Surgery cally equivalent and bilaterally, rostrocaudally orientated. This is
in a Porcine Model not different in swine with a developmental architecture of a pros-
encephalon including diencephalon and telencephalon, brain stem
with mesencephalon and rhombencephalon, completing with the
spinal cord caudally. The CNS is covered by the meninges and the
brain is protected by the skull.
Main differences, compared to human anatomy, can be seen in
the skull and cranial base architecture. Looking from above or
anteriorly, a plane forehead and vertex are ending in a high, sloping
crest posteriorly where the neck muscles are assessing (Fig. 1a, b).
The vertex plane is limited laterally by the parietal bones (occipital)
and orbitae (frontal) restricting the craniotomy in the biparietal
width. Midline sagittal suture is identified easily covering the sagit-
tal sinus whereas coronal and lambdoid sutures will not be seen in
a conventional approach. The bone itself will appear soft and less
mineralized, especially in young pigs with a thickness of about
1 cm or more at the lateral edges of the vertex plane.
Fig. 1 (a) Lateral view showing the sloping crest posteriorly and the extent of orbita limiting craniotomy ante-
riorly. The extent of craniotomy is marked black. (b) Anterior view from above with marking of the craniotomy
Neurosurgery in Swine 167
The cerebral hemispheres are divided into lobes and main sulci
comparable to human anatomy. Topographic allocations, cranial
nerves, their origin at the brain stem, and their pathways are very
similar as well (Figs. 2 and 3). The olfactory bulb is somewhat big-
ger but will not be seen following craniotomy. Ventricles contain-
ing CSF are small and do not allow an endoscopic access in an
appropriate way for learning reasons. Compared with humans,
arterial blood supply of the brain in pigs is different as a plexus of
very small vessel discharge into the internal carotid artery. This
circumstance and Willis cross perfusion circumvent cerebral infarc-
tion models. Anterior, middle, and posterior cerebral arteries, lat-
ter one fusing with the basilar artery, are found as well as venous
drainages including sinuses and jugular veins are similar again to
cerebral blood supply in humans.
Following a midline incision starting at the coronal suture and
ending at the sloping crest posteriorly, unilateral and bilateral
approaches to the brain can be performed. Attention has to be
driven to the limited access of about 3.5 cm width and 4 cm length
(anterior-posterior) in bilateral approaches, a thin dura which may
be easily injured during craniotomy and venous bleedings out of
the bone or violated sinus (Fig. 4). Frontal sinuses are large and
will be opened by trepanation. Drills are needed in some cases to
uncover the dura but craniotomy is performed in the ordinary way.
Orbital roof and thickness of bone at the lateral edge may restrict
craniotomy. Bone wax and other hemostyptics should be prepared
in advance as well as bone punches may facilitate a more safe
approach for beginners, especially in bilateral exposures crossing
the sagittal sinus. The approach should be extended by drills and
Fig. 2 Neuroanatomy of the pig with topographic landmarks and anterior-posterior

extent of craniotomy. 1 frontal sinuses, 2 skull, 3 hemisphere, 4 interhemispheric
commissura, 5 chiasm, 6 pituitary, 7 cerebellum, 8 brain stem, 9 medulla, 10
atlas, 11 axis, 12 vertebral column, 13 anterior spinal processes, 14 spinal cord
a Forebrain
MOTOR
Cerebellum
7
6 SOMATIC
SENSORY
1 VISUAL Thalamus
Hypothalamus Midbrain AUDITORY
II 3 Medulla INTO
2 4 ACTION Spinal Cord
5
b 1
1
caudal cranial
view II
view
III
2
V 3
VI
VII
VIII 4 5
6
XII
Fig. 3 (a) Cranial nerves and functional areas of the porcine brain. (b) Cranial nerves in the caudal and cranial
view
Fig. 4 Burr holes and craniotomy with size of about 3.5 × 3.5 cm
bone punches to the lateral, frontal, and occipital limits allowing a

maximum view on the cortical surface later.
Dura is lifted by one suture laterally followed by incision and
opening in a crescent manner leaving the sagittal sinus untouched
(Fig. 5). Arachnoid granulations may bleed and should be pre-
served, otherwise compressed by hemostyptics or transected under
microscopic magnification. Careful microscopic dissection should

be made on bridging veins draining into the sagittal sinus (Fig. 6).
Further preparation may include arachnoid opening, sulcal dissec-
tion, and resection of a gyrus or an artificially inserted tumor
(Fig. 7). Colored fibrin glue (dura sealants, amount of 0.5–1 ml) is
Fig. 5 Dural opening preserving the sagittal sinus in the midline
Fig. 6 Bridging veins are mobilized to get full exposure to the interhemispheric
region
Fig. 7 Sulcal preparation is followed by incision of the white matter. Resection by

suctioning is comparable to the real situation of glioma resection
injected via a needle transcortically into the brain. The cranial base
including cranial nerves and basal arteries is only reached by an
extended bone resection and/or resection of the frontal brain.
Subtemporal, fronto-temporal, or occipital approaches are not
comparable to surgery in humans as bone and soft tissue have to be
prepared extensively in which orbital rims on the one hand and the
entire muscles of the neck have to be dissected.
2.2 Anatomy Spine surgery in swine is limited to the microscopic and minimal
of the Spinal Cord invasive techniques predominantly. Even porcine vertebrae possess
and Surgery on Nerve similar ligamentous structure and facet joint orientation; they are
Roots and Intradural smaller, have anterior processes, and are less mineralized. Screws or
Lesions in a more complex instrumentations will less likely find a sufficient stay
Porcine Model in our experience; therefore human cadaver models may be favored
in these special issues.
The vertebral column in pigs protects the spinal cord by neural
arches composed of lamina with transverse and articular processes,
just like in humans. Therefore porcine spine is an ideal training
model for dorsal or dorsolateral approaches to intraspinal epidural
including exposure of the nerve roots or intradural extra- or intra-
medullary lesions.
Midline incision comprises the extent of three to four spinal
processes in minimum. Paraspinal muscles are of distinctive
strength and have to be removed from the midline to the facet
joints laterally (Fig. 8). Laminectomy is simply performed by bone
punches and rongeurs or by bone saws allowing to mediate the
techniques of laminoplasty and/or laminotomy. Epidural and
intradural lesions are reached if all bleedings from the bone and the
epidural venous plexus are stopped in the conventional manner by
bone wax, drilling without rinsing, bipolar coagulation, hemostyp-
tics, and compression. Comparable to the cranial approaches spinal
dura is thin and may be easily injured (Fig. 9). The dura is opened
Fig. 8 Midline, bilateral exposure to the spinal cord with the possibilities of inter-
laminar approach, transforaminal approach, hemilaminectomy, laminectomy,
and unilateral undercutting procedures
Fig. 9 Interlaminar window with access to the nerve root (here dural opening)
Fig. 10 Laminectomy requires careful bleeding management, especially epidur-

ally. Dural opening allows further preparation on extramedullar lesions following
the nerve roots or the dural sheet as well as intramedullar lesions
under microscopic magnification and fixed by sutures keeping it

open (Fig. 10). Myelotomy is done in the midline following inci-
sion of the arachnoid membrane. While nerve fibers and filum ter-
minale are found in the cauda equina of the lumbar spine, the
myelon impresses more pulpy than in humans.
Transmuscular, tube-assisted approaches to the interlaminar
space and ordinary dorsal approaches to the intervertebral space
may be as well performed. Following these steps disc surgery may
be imitated by flavectomy and/or extended bone removal of the
laminar edges. Once again, bleedings of the epidural venous plexus
have to be carefully managed before nerve roots are followed into
the osseous foramen.
Microscopic dissection and bleeding management of a pulsat-
ing brain and spinal cord with the daily used OR equipment are
particular challenges which can be trained in this setup with calm-
ness and patience. Failure is not life threatening for the patient
(swine). Nevertheless it will end up in an unclear, blood-filled cav-
ity or damaged brain. Suctioning while bleeding, handling the
microscope, changing the instruments, and decision making in the

in vivo model is as complex as in daily practice but their individual
stress factors aggravate the difficulty significantly. Learning is medi-
ated by the successful or disappointing in vivo dissection and
depends on individual experiences.
Team approach can be appointed by the realistic technical
setup in which assistance has to take over the nursing job, has to
anticipate the next surgical step, and is asked to exchange his expe-
riences with the performing surgeon to achieve a most blood-dry
and clear resection cavity. In our point of view the in vivo model
presents as an ideal opportunity for microsurgical training includ-
ing even social skills which are required in a competitive and
achievement-orientated field today.
3 Notes
3.1 Brain Surgery – Midline incision large enough to expose coronal sutures,
in a Porcine Model orbital rims, and sloping crest posteriorly.
– Bilateral parietal burr holes and craniotomy of about 3 by
4 cm, less mineralized bone may require drills, to avoid bleed-
ing from a lacerated sagittal sinus bone punches are more safe
crossing the midline.
– Dura is thin and easy to violate; lift it by one suture before inci-
sion is made; leave the sagittal sinus untouched.
– Ideal training model for learning microneurosurgery, focused
on surgery of the cortex and/or resection of an artificially
inserted brain tumor.
– Endoscopic approaches of the ventricles are limited by the
small and narrow size.
– Vascular procedures exposing the basal arteries require brain
resection and are again limited by the small diameter of arteries
and veins.
– Access for researchers may be the large and safe exposure of
brain.
3.2 Spine Surgery – Predominantly used for microscopic and minimal invasive pro-
in a Porcine Model cedures as instrumentations will less likely find a sufficient stay.
– Midline approach with interlaminar access to nerve roots, lam-
inotomy, laminoplasty, or simple laminectomy to expose the
spinal canal.
– Epidural venous plexus of major concern in bleeding
management.
– Dura is thin and may easily be lacerated.
– Ideal training model for interlaminar approaches to nerve
roots, extra- and intradural, extra- and intramedullary lesions.
References
1. Regelsberger J, Heese O, Horn P, Kirsch M, Eicker 4. Mazotti LA, Vidyarthi AR, Wachter RM,
S, Sabel M, Westphal M (2010) Training micro- Auerbach AD, Katz PP (2009) Impact of duty-
neurosurgery—four years experiences with an hour restriction on resident inpatient teaching.
in vivo model. Cen Eur Neurosurg 72:192–195 J Hosp Med 4:476–480
2. Regelsberger J, Eicker S, Siasios I, Hänggi D, 5. Reulen HJ, Hide RA, Bettag M, Bodosi M,
Kirsch M, Horn P, Winkler P, Signoretti S, Cunha ESM (2009) A report on neurosurgical
Fountas K, Dufour H, Barcia JA, Sakowitz O, workforce in the countries of the EU and associ-
westermaier T, Sabel M, Heese O (2015) In ated states. Task Force “Workforce Planning”
vivo porcinetraining model for cranial neuro- UEMS Section of Neurosurgery. Acta Neurochir
surgery. Neurosurgical review 38(1):157–63 (Wien) 151:715–721
3. Brennum J (2000) European neurosurgical
education—the next generation. Acta Neurochir
(Wien) 142:1081–1087
Chapter 14
Real-Time Convection Delivery of Therapeutics

to the Primate Brain
Dali Yin, Massimo S. Fiandaca, John Forsayeth,
and Krystof S. Bankiewicz
Abstract
Convection-enhanced delivery (CED) has been developed as a drug delivery strategy and represents a
powerful methodology for targeted therapy in the brain. Our group has extensively studied and refined
this approach for distributing various agents, including small molecules, macromolecules, viral particles,
nanoparticles, and liposomal drugs into the brain parenchyma by means of a procedure called real-time
convection-enhanced delivery (RCD). We also defined infusion parameters referred to as “red,” “blue,”
and “green” zones for cannula placements that result in poor, suboptimal, and optimal volumes of distri-
bution, respectively, in the target area of brain of nonhuman primates (NHP). We have defined the scale
differences between NHP brains and those of humans. Furthermore, we applied the ClearPoint® system
to the RCD procedure, which allows RCD to be carried out with a high level of precision, predictability,
and safety. This approach may improve the success rate for clinical trials involving intracerebral drug deliv-
ery by direct infusion. These innovations may have important implications in ensuring effective delivery of
therapeutics into brain targets utilizing NHP stereotactic coordinates translated via stereotactic MRI local-
ization procedures in humans. These delivery innovations should be considered when localized therapeutic
delivery, such as gene transfer or protein administration, is being translated into clinical treatments. In this
chapter, we review recently developed methods that ensure controlled distribution of therapeutic agents in
the brain.
Key words Real-time convection-enhanced delivery, RGB zones, ClearPoint system, Nonhuman
primates
1 Introduction
Convection-enhanced delivery (CED) was an interstitial central

nervous system (CNS) delivery technique [1] that circumvents
the blood–brain barrier in delivering therapeutics into the CNS.
Traditional local delivery of most therapeutic agents into the brain
has relied on diffusion, which depends on a concentration gradi-
ent. The rate of diffusion is inversely proportional to the size of the
therapeutic and is usually slow with respect to tissue clearance.
Thus, diffusion results in a nonhomogeneous distribution of most
175
176 Dali Yin et al.
delivered agents restricted to a few millimeters from the source. In

contrast, CED uses a fluid pressure gradient established at the tip
of an infusion catheter and bulk flow to propagate substances
within the extracellular fluid space [1]. CED allows the extracel-
lularly infused material to further propagate via the perivascular
spaces and the rhythmic contractions of blood vessels acting as an
efficient motive force for spreading of the infusate [2]. As a result,
a higher concentration of drug was distributed more evenly over a
larger area of the targeted structure than what would be seen with
a simple injection. CED has been developed as a drug delivery
strategy and represents a powerful methodology for targeted ther-
apy in the fields of neurodegenerative diseases, such as Parkinson’s
disease [3, 4], and neuro-oncology [5, 6]. Laboratory investiga-
tions with CED cover a broad field of application, such as the
delivery of small molecules [7, 8], macromolecules [1], viral par-
ticles [9], magnetic nanoparticles [10], and liposomes [11].
One of the more recent advances in CED was real-time con-
vective delivery (RCD), which uses MRI to visualize the CED pro-
cess during infusion with the aid of gadolinium-loaded liposomes
(GDL) that co-distribute with the infused therapeutic to the non-
human primate (NHP) target [12, 13]. Our RCD methodology
has evolved through extensive modeling in NHP over the years.
Visualizing infusions in real time allows active feedback on (1) can-
nula placement, (2) physical and anatomic diffusion parameters,
and (3) control of drug delivery for optimizing gene transfer,
thereby reducing the potential for adverse effects. Initially described
by Oldfield and colleagues who employed albumin-linked surro-
gate tracers [14], our current technique of RCD employs interven-
tional MRI (iMRI) to monitor the distribution of therapeutic
agents that are co-infused with gadolinium-based tracers [15]. Our
initial work with GDL [16, 17] has progressed to the co-infusion
of free gadoteridol for predicting the distribution of protein [18]
and AAV2 vectors [15, 17, 19]. A similar strategy was used to co-
infuse therapeutic agent with Gd-diethylenetriamine pentaacetic
acid (Gd-DTPA) in a clinical study in two patients with intrinsic
brainstem lesions at the National Institutes of Health [20]. RCD
allowed us to monitor infusion of liposomal drugs into brain
tumors [21] and viral gene therapy vectors into parenchyma [17].
Visualizing infused drug distribution was necessary to ensure accu-
rate delivery of therapeutic agents into target sites while minimiz-
ing exposure of healthy tissue. Moreover, because infusions could
be visualized, we were able to define quantitative relationships
between infusate volume (Vi) and subsequent volume of distribu-
tion (Vd) for both white and gray matter [12]. This method has
given us the ability to directly monitor the local delivery of thera-
peutic agents and has improved the efficacy of CED in animals.
During RCD, the Vd for a given agent depends on the struc-
tural properties of the tissue being convected, such as hydraulic con-
ductivity, vascular volume fraction, and extracellular fluid fraction
Stereotaxy in Monkeys 177
Fig. 1 Step-design cannula. The length of each infusion cannula was measured to
ensure that the distal tip extended 3 mm beyond the length of the respective guide.
This created a stepped design at the tip of the cannula to maximize fluid distribu-
tion during RCD procedures and minimize reflux along the cannula tract. We refer
to this transition from fused silica tip to a fused silica sheath as the “step”
[22]. It also depends on the technical parameters of the infusion

procedure such as cannula design, cannula placement, infusion vol-
ume, and rate of infusion [23–25], with the overall aim of improving
delivery efficiency while attempting to limit the spread of the thera-
peutic into regions outside the target. Development of the optimal
cannula type for effective CED delivery in the brain has also been
critical. We examined several types of cannulae with respect to size
and design, and concluded that a stepped design (Fig. 1) with a
fused silica tip provided us with the most consistently robust brain
delivery [26–28]. The stepped cannula dramatically reduces reflux
along the infusion device by restricting initial backflow of fluid flow
beyond the step. Furthermore, in our experience, a key component
of successful CED was the site of cannula placement within the tar-
geted area, such as putamen, thalamus, and brain stem. The distance
from cannula step to its entry point in the target region was found
to be critical for optimal distribution of therapeutics. We have devel-
oped the concept of RGB (red, green, and blue zones) zones for
cannula placement during RCD. We defined these zones on the
basis of containment of infusate within the target region. Within
each region so far investigated, we were able to define a subset of
cannula locations associated with complete containment within the
target (green), substantial containment (blue), or poor containment
(red). Infusate escape into nearby ventricles or white matter tracts
was driven by proximity to these structures. We defined three-
dimensional RGB zones in the putamen [29], thalamus, and brain
stem [30] of NHP. To obtain the most effective distribution of the
infused therapeutic within the intended target, it was essential to
understand the optimal site of placement of the step and tip of the
infusion cannula within that target, preferably within the green zone
[29, 30]. Such optimal placement will reduce distribution into sur-
rounding white matter tracts that serve as leakage points [31]. Such
consideration allows more precise delivery of the therapeutic to the
target structure(s) but lessens the risk of inadvertent spread into sur-
rounding brain regions. These factors may also explain some of the
reported failures of CED in both NHP studies and human clinical
178 Dali Yin et al.
trials, which may have been related to suboptimal targeting of the

infusion cannula.
In order to further optimize the RCD technology, the
ClearPoint® system (Surgivision Inc, Irvine, CA) has been adopted
by us to translate targeting from the NHP brain into humans.
ClearPoint was a novel integrated hardware (skull-mounted
SmartFrame device)/software platform for RCD that provides
prospective stereotactic guidance for the cannula placement and
performance of RCD. This platform was based on the concept of
prospective stereotaxy, the alignment of a skull-mounted trajectory
guide within an MRI system [32], already used in clinical studies
to perform brain biopsies [33, 34] and placement of DBS leads
[34–36]. In anticipation of upcoming gene therapy clinical trials,
we adapted this “off-the-shelf” device to RCD of therapeutics via
a customized infusion cannula. The targeting accuracy of this
delivery system and the performance of the infusion cannula were
validated in NHP. The ClearPoint system allows RCD to be per-
formed with a high level of precision, safety, and predictability.
This technique should increase the utility of RCD for expanding
the scope of drug delivery studies. Clinical application of this guid-
ance platform was likely to improve the success of clinical trials
employing intracerebral drug delivery. Although continued refine-
ments in RCD may be expected, our work over the past decade has
resulted in a new paradigm for direct parenchymal delivery that
may find increasing application in the treatment of currently intrac-
table diseases like Alzheimer’s and Parkinson’s disease, brain
tumors, and other movement disorders. Here we describe in detail
our methods for optimal RCD.
2 Materials
2.1 Experimental Normal rhesus macaques and cynomolgus monkeys (aged from 8
Subjects to 18 years; mean age = 11.9 years, weight = 4–9.4 kg) were the
subjects in our study. Experimentation was performed according
to the National Institutes of Health guidelines and to protocols
approved by the Institutional Animal Care and Use Committee at
the University of California San Francisco (San Francisco, CA).
Adult monkeys were individually housed in stainless steel cages.
Each animal room was maintained on a 12-h light/dark cycle and
room temperature ranged between 64 and 84 °F. Prior to assign-
ment to the study, all animals underwent at least a 31-day quaran-
tine period mandated by the Centers for Disease Control and
Prevention (Atlanta, GA).
2.2 Liposome 1. 1,2-Dioleoyl-sn-glycero-3-phosphocholine (DOPC).

Preparation 2. Cholesterol.
3. 1,2-Distear-oyl-sn-glycero-3-[methoxy(polyethylene glycol)-
2000] (PEG-DSG).
4. Chloroform/methanol (90:10, v/v).

5. Ethanol.
6. 0.5 M Gadoteridol (10-(2-hydroxy-propyl)-1,4,7,10-tetraaza-
cyclododecane-1,4,7-triacetic acid).
7. Double-stacked polycarbonate membranes with a pore size of
100 nm.
8. Sephadex G-75 size-exclusion column.
9. HEPES-buffered saline (pH 6.5, 5 mM HEPES, 135 mM
NaCl, pH adjusted with NaOH).
10. Rhodamine.
11. 20 mM Sulforhodamine B in pH 6.5 HEPES-buffered saline.
2.3 Magnetic 1. Periosteal elevator (Fine Scientific Tools, Foster City, CA,
Resonance Imaging USA), rongeur calipers (Fine Scientific Tools, Foster City, CA,
USA), gelfoam (Baxter, Deerfield, IL, USA), dental acrylic,
gauze, syringes (5 and 50 mL), latex gloves, stopwatch timer.
2. Reflux-resistant infusion cannula.
3. Teflon tubing for secondary and loading lines (1.57 mm outer
diameter, 0.76 mm inner diameter; Upchurch Scientific, West
Berlin, NJ, USA).
4. Plastic cannula guide ports.
5. Gadoteridol (ProHance®; Bracco Diagnostics Inc., Monroe
Township, NJ, USA).
6. Skull-mounted aiming device (SmartFrame®, MRI
Interventions Inc., Memphis, TN, USA) and software
(ClearPoint®, MRI Interventions Inc., Memphis, TN, USA).
7. Sterile hardware: Plastic screws, pens, rulers, screwdriver,
dummy catheter (Upchurch Scientific, West Berlin, NJ, USA),
large animal MRI-compatible stereotactic frame (Kopf
Instruments, Tujunga, CA, USA), 3500 Medfusion pump
(Strategic Applications Inc., Lake Villa, IL, USA), Tefzel fer-
rule connectors and Luer-Lock adapters (Upchurch Scientific,
West Berlin, NJ, USA), impaction drill (3.5 mm round drill
bit; Stryker, Portage, MI, USA).
8. 1.5-T MRI scanner (Signa LX; GE Medical Systems, Waukesha,
WI, USA), 5-in. circular surface MRI coil (MR Instruments
Inc., Hopkins, MN, USA).
9. OsiriX® software (v5.5.2; Pixmeo, Bernex, Switzerland).
2.4 Infusion 1. Ketamine and xylazine.

Procedure 2. Isoflurane.
3. MRI-compatible stereotactic frame.
4. MRI-compatible guide cannula.
180 Dali Yin et al.
5. Dental acrylic.
6. Stylet screw.
7. Step-design cannula.
8. Loading line (containing GDL or free gadoteridol).
9. Infusion line with oil and another infusion line with trypan
blue solution.
10. Syringe 1 ml filled with 1 % trypan blue solution.
11. Stereotactic holder.
12. Micro-infusion pump.
2.5 ClearPoint 1. SmartFrame®.

System 2. Infusion cannula.
3. Software system.
3 Methods
3.1 Liposome Separate liposomes were prepared for detection either by MRI or
Preparation by histology. Liposomes containing the MRI contrast agent were
composed of 1,2-dioleoyl-sn-glycero-3-phosphocholine (DOPC)/
cholesterol/1,2-distearoyl-sn-glycero-3-[methoxy(polyethylene
glycol)-2000] (PEG-DSG) with a molar ratio of 3:2:0.3. DOPC
was purchased from Avanti Polar Lipids (Alabaster, AL), PEG-DSG
from NOF Corporation (Tokyo, Japan), and cholesterol from
Calbiochem (San Diego, CA). The lipids were dissolved in chloro-
form/methanol (90:10, v/v), and the solvent was removed by
rotary evaporation, resulting in a thin lipid film. The lipid film was
dissolved in ethanol and heated to 60 °C. A commercial US
Pharmacopeia solution of 0.5 M gadoteridol (10-(2-hydroxy-
propyl)-1,4,7,10-tetraazacyclododecane-1,4,7-triacetic acid)
(Prohance; Bracco Diagnostics, Princeton, NJ) was heated to 60 °C
and injected rapidly into the ethanol/lipid solution. Unilamellar
liposomes were formed by extrusion (Lipex; Northern Lipids,
Vancouver, Canada) by 15 passes through double-stacked polycar-
bonate membranes (Whatman Nucleopore, Clifton, NJ) with a
pore size of 100 nm, resulting in a liposome diameter of 24–124 nm
as determined by quasi-elastic light scattering (N4Plus particle size
analyzer; Beckman Coulter, Fullerton, LA). Unencapsulated gado-
teridol was removed with a Sephadex G-75 (Sigma, St. Louis, MO)
size-exclusion column eluted with HEPES-buffered saline (5 mM
HEPES, pH 6.5, 135 mM NaCl). Liposomes loaded with rhoda-
mine for histological studies were formulated with the same lipid
composition and preparation method as the gadoteridol-containing
liposomes, except that the lipids were hydrated directly with 20 mM
sulforhodamine B (Sigma) in pH 6.5 HEPES-buffered saline by six
successive cycles of rapid freezing and thawing rather than by ethanol
injection. The sulforhodamine B liposomes had a diameter of

90 ± 30 nm (used alone for histological analysis) or 115 ± 40.1 nm
(used for co-infusion with GDL in the MRI-monitoring study). For
the preparation of liposomes containing a DiI-DS fluorescent
probe, 1,1′-dioctadecyl-3,3,3′,3′-tetramethylindocarbocyanine-5,5′
disulfonic acid (DiIC18(3)-DS) (DiI-DS; Molecular Probes,
Eugene, OR) was added to the lipid solution at a concentration of
0.2 mol% of the total lipid. DiI-DS liposomes had a diameter of
110 ± 40 nm. For further detail regarding formulation of liposomes
loaded with chemotherapeutic agents, refer to Krauze et al. [37].
During our studies it became apparent that, even when using the
same method (CED), the volumes of distribution for different com-
pounds were inconsistent. For simple infusates, CED distribution
was significantly increased if the infusate was more hydrophilic or had
weaker tissue affinity [38]. Encapsulation of tissue-affinitive mole-
cules by liposomes significantly increased their tissue distribution.
However, it appeared that liposomal surface properties (cationic ver-
sus neutral liposome, surface charge, and percentage of PEGylation)
affected parenchymal volume of distribution. We found that
PEGylation, which provided steric stabilization and reduced surface
charge, yielded the greatest Vd when compared with volume of infu-
sion (Vi) of all other liposomal formulations [38].
3.2 Quantification The concentration of gadoteridol entrapped in the liposomes was

of Liposome-Entrapped determined from nuclear MR relaxivity measurements. The relation-
Gadoteridol by ship between the change in the intrinsic relaxation rate imposed by
Magnetic Resonance a paramagnetic agent (ΔR), also known as “T1 shortening,” and the
Imaging concentration of the agent is defined by the equation ΔR = r1[agent],
in which r1 = relaxivity of the paramagnetic agent and ΔR = (1/
T1observed − 1/T1intrinsic). The relaxivity of gadoteridol had been
empirically derived previously on the same system, a 2-T Brucker
Omega scanner (Brucker Medical, Karlsruhe, Germany), and had a
value of 4.07 mM−1 s−1. The concentration of the encapsulated
Gadoteridol was then calculated with the following equation:
[Gadoteridol] = [(1/T1wGado) − (1/T1w/oGado)]/4.07.
3.3 Infusion We developed a new stepped design cannula for CED that effec-
Catheter Design tively prevents reflux. Cannula design has been one of the most
neglected features of brain delivery protocols. Reflux was defined
as the phenomenon of the movement of infusate back up the out-
side of the cannula rather than into the tissue. Although earlier
studies showed that smaller cannula diameters permit better deliv-
ery, the crucial problem of reflux was either not assessed or not
measurable. In our early studies, we confirmed that smaller cannula
diameters allowed faster delivery rates but the smallest available
cannulae were associated with increasing reflux when the rate of
infusion exceeded 0.5 μl/min [26], clearly a significant problem
when infusing large volumes. Recently, we have been able to
increase the infusion rate to 5 μl/min without reflux by means of
182 Dali Yin et al.
an innovative stepped cannula [26], which dramatically reduces

reflux along the infusion device by restricting initial backflow of
fluid flow beyond the step. The early metal cannula has been
replaced by one made of silica that also features sharp transitions in
outer diameter that prevent reflux (Fig. 1) [4]. The larger diameter
of the stem of the cannula had an outer and inner diameter of
0.53 mm and 0.45 mm, respectively. The outer and inner diame-
ters of the tip segment were 0.43 mm and 0.32 mm, respectively.
The length of each infusion cannula was measured to ensure that
the distal tip extended 3 mm beyond the length of the respective
guide. This created a stepped design at the tip of the cannula to
maximize fluid distribution during CED procedures and minimize
reflux along the cannula tract. In the text, we refer to this transi-
tion from fused silica tip to a fused silica sheath as the “step,” and
all positioning data are derived from the position of this step
because of its unambiguous visibility on MRI. Pushing flow rates
significantly above 5 μl/min, however, can induce reflux even with
this catheter, as we have shown in canine and primate studies [31].
Robust reflux-free delivery and distribution of liposomes was
achieved with the stepped design cannula in both rats and nonhu-
man primates. This stepped design cannula may allow reflux-free
distribution and shorten the duration of infusion in future clinical
applications of CED in humans.
3.4 Infusion Primates received a baseline MRI before surgery to visualize ana-
Procedure tomical landmarks and to generate stereotactic coordinates of the
proposed target infusion sites for each animal. NHP underwent
neurosurgical procedures to position the MRI-compatible guide
cannula over the target. Each customized guide cannula was cut to
a specified length, stereotactically guided to its target through a
burr hole created in the skull, and secured to the skull by dental
acrylic. The tops of the guide cannula assemblies were capped with
stylet screws for simple access during the infusion procedure.
Animals recovered for at least 2 weeks before initiation of infusion
procedures. Animals were anesthetized with isoflurane (Aerrane;
Ohmeda Pharmaceutical Products Division, Liberty Corner, NJ)
during real-time MRI acquisition. Each animal’s head was placed
in an MRI-compatible stereotactic frame, and a baseline MRI was
performed. Vital signs, such as heart rate and PO2, were monitored
throughout the procedure. Briefly, the infusion system (Fig. 2)
consisted of a reflux-resistant fused silica cannula that was con-
nected to a loading line (containing GDL or free gadoteridol), an
infusion line with oil, and another infusion line with trypan blue
solution. A 1 ml syringe (filled with trypan blue solution) mounted
onto a micro-infusion pump (BeeHive; Bioanalytical System, West
Lafayette, IN) regulated the flow of fluid through the system.
Based on MRI coordinates, the cannula was manually guided to
Loading line
Outer FS (gad, 1m)
sleeve Inner FS
Infusion line
3mm step-design tip tubing (oil/trypan blue, 5m)
Brain
Syringe &
Pump
Fig. 2 Schematic of infusion system. FS fused silica
the targeted region of the brain through the previously placed

guide cannula. After securing placement of the infusion cannula,
the CED was initiated with real-time MRI data simultaneously
acquired (RCD). We used the same infusion parameters for every
NHP infused throughout the study. Infusion rates were as follows:
0.1 μl/min was applied when lowering cannula to targeted area
and increased at 10-min intervals to 0.2, 0.5, 0.8, 1.0, and 2.0 μl/
min. After turning off delivery pumps at the end of each infusion
session, the catheter remained in place for another 5 min to allow
the built-up tissue pressure to abate. Animals received up to three
infusion procedures into the same anatomic location. Each animal
had at least a 4-week interval between each infusion procedure.
3.5 Magnetic NHP were sedated with a mixture of ketamine (Ketaset, 7 mg/kg,
Resonance Image IM) and xylazine (Rompun, 3 mg/kg, IM). After sedation, each
animal was placed in an MRI-compatible stereotactic frame. The
ear-bar and eye-bar measurements were recorded, and an intrave-
nous line was established. MRI data were then obtained, after
which animals were allowed to recover under close observation
until able to right themselves in their home cages. MR images of
brain in NHP were acquired on a 1.5-T Siemens Magnetom
Avanto (Siemens AG, Munich, Germany). Three-dimensional
rapid gradient echo (MP-RAGE) images were obtained with rep-
etition time (TR) = 2110 ms, echo time (TE) = 3.6 ms, flip
angle = 15°, number of excitations (NEX) = 1 (repeated three
times), matrix = 240 × 240, field of view (FOV) = 240 × 240 × 240,
and slice thickness = 1 mm. These parameters resulted in a 1 mm3
voxel volume. The scan time was dependent on the number of
slices needed to cover the extent of infusion and ranges from 9 min
44 s to 11 min 53 s.
MR images were obtained from each RCD and used to mea-
sure distance from cannula step to corpus callosum (CC), internal
capsule (IC) and external capsule (EC) for infusion into the puta-
men, and from the cannula step to the midline, to cannula entry
184 Dali Yin et al.
point in the target region, and to the lateral border of the target
regions for infusion into thalamus or brainstem of NHP. The
measurements were made on an Apple Macintosh G4 computer
with OsiriX® Medical Image Software (v2.5.1). OsiriX software
reads all data specifications from DICOM (digital imaging and
communications in medicine) formatted MR images obtained via
local picture archiving and communication system (PACS). For
each image, the default window and level settings were used
throughout the study; that is, there was no attempt to alter or
manipulate settings from one experiment to another. The distances
from cannula step to each structure mentioned above were manu-
ally defined and then calculated by the software. All the distances
were measured in the same manner on MRI sections. These data
were used to define RGB zones in the putamen, thalamus, and
brain stem.
MR images were also used for volumetric quantification of dis-
tribution of gadoteridol. The Vd of gadoteridol in the brain of
each subject was quantified on an Apple Macintosh G4 computer.
Region of interest (ROI) derived in the target and white matter
tract (WMT) were manually defined, and software then calculated
the area from each MR image and established the volume of the
ROI based on area defined multiplied by slice thickness (PACS
volume). The boundaries of each distribution were defined in the
same manner in the series of MRI sections. The defined ROI
volumes allowed for 3D image reconstruction with BrainLAB
software (BrainLAB, Heimstetten, Germany).
3.6 Coordinates The X, Y, and Z axial values of cannula step location in green zone
for Green Zone were determined with 2D orthogonal MR images generated by
in the Putamen, OsiriX software, where MR images were projected in all three
Thalamus dimensions (axial, coronal, and sagittal). We used midpoint of the
and Brainstem anterior commissure–posterior commissure (AC–PC) line as zero
of Three-Dimensional point (0,0,0) of three-dimensional (3D) brain space. Briefly, AC–
Brain Space in NHP PC line was drawn on mid-sagittal plane of MRI, and the midpoint
of AC–PC line was determined. The horizontal and vertical plane
through the midpoint of AC–PC line was then obtained, and they
could be shown on all the three plans simultaneously. The X, Y,
and Z axial values of cannula step were then obtained by measure-
ments of distance from cannula step to midline on coronal MRI
plane (X value), distance anterior (or posterior) to the midpoint of
AC–PC line of the coronal MRI plane (Y value) and the distance
above (or below) axial plane incorporating the AC–PC line on
MRI (Z value). All the distances were measured (in millimeters) in
the same manner on MRI sections for each case.
The results obtained were used to determine a set of 3D ste-
reotactic coordinates that define an optimal site for infusions into
putamen, thalamus, and brainstem in NHP. Based on the coordi-
nate calculations for the cannula step by MRI, the target for green
zone in the putamen were x = 11.85 ± 0.56 mm lateral (X coordi-

nate), y = 7.36 ± 0.49 mm anterior to the AC–PC midpoint (Y
coordinate), and z = 3.62 ± 0.40 mm superior to the AC–PC axial
plane (Z coordinate). The mean coordinates for placing the step in
the thalamic green zone were x = 6.9 ± 0.7 mm lateral (range 4.1–
10.2), y = 1.2 ± 0.2 mm posterior (range 0.4–1.9 mm), and
z = 3.1 ± 0.4 mm superior (range 1.4–4.7 mm). The mean coordi-
nates for green zone in the brainstem were x = 2.3 ± 0.2 mm lateral
(range, 1.6–3.5 mm), y = 4.0 ± 0.5 posterior (range, 2.6–6.0 mm),
and z = 8.9 ± 1.0 mm inferior (range, 4.8–11.9). We think that can-
nula placement and definition of optimal (green zone) stereotactic
coordinates have important implications in ensuring effective
delivery of therapeutics into the target utilizing routine stereotac-
tic MRI localization procedures.
3.7 Cannula Optimal results in the direct delivery of therapeutics into primate
Placement Guidelines brain depend on reproducible distribution throughout the target
region. In our recent studies, we retrospectively analyzed MRI of
RCD infusions into the putamen, thalamus and brainstem of NHP,
and defined infusion parameters referred to as “red,” “blue,” and
“green” zones (RGB zones) for cannula placements that result in
poor, suboptimal, and optimal volumes of distribution, respec-
tively. The most robust data was achieved in putamen, and the
reason for this is that problematic structures (ventricles, corpus cal-
losum) surround this region. So it was relatively easy to define
RGB zones in this setting (Fig. 3a). In contrast, thalamus and brain
stem, much larger structures in any case, really do not present this
kind of challenge. Accordingly, infusions in thalamus defined G
and B zones but not R (Fig. 3b). In brain stem, we only identified
coordinates that gave excellent containment of infusate (Fig. 3c).
Clearly, each new target region will impose its own anatomical
constraints and optimization of RCD will require empirical deter-
minations to some extent. However, the three regions we have
investigated suggest the following rules of thumb. When infusate
emanates from the tip of stepped cannulae, the infusate forms an
Fig. 3 RGB zones for step placement outlined in the putamen (a), thalamus (b), and brain stem (c) of NHP
186 Dali Yin et al.
a b
MRI-visible skull-mounted
cannula guide base
connections to
hand-held controller
translational
components
infusion
target
Fig. 4 (a) Schematic showing the basic components of the SmartFrame. (b) The fluid stem was aligned to the
target trajectory via both “pitch and roll” axes and an X–Y translational stage
ovoid pattern with the cannula as the vertical axis. The upper
dimension of the ovoid extends upwards about somewhat less than
the length of the step-tip. Thus, a 3 mm step-tip will generate a
little less than 3 mm backflow. In the smaller rat striatum, we
adjusted the cannula tip to 1 mm and placed the step approxi-
mately 1–2 mm from the corpus callosum in order to place the
cannula tip nicely within the striatum while maintaining a clear
separation of the leading edge of the backflow from the entry point
[28]. This rule should be followed for the design of cannulae in
smaller structures. With respect to peri-ventricular zones, we found
in putamen that the cannula should be placed at least 3 mm from
external and internal capsules. In general then, a cannula trajectory
in the monkey that can maintain a distance of 3 mm or more from
sensitive structures seems to be a good place to start. In humans,
of course, these distances are correspondingly enlarged. The size of
the striatum in humans is about fivefold that of the Rhesus monkey
[39], and consideration of such target volume differences is an
important factor in clinical planning.
3.8 ClearPoint The ClearPoint® system consists of the SmartFrame® (Fig. 4a), an
System infusion cannula, and a software system that communicates with
both the MRI console and the operating neurosurgeon in the MRI
suite. The ClearPoint software allows registration of the AC and
PC from an initial MRI scan, selection of a target for cannula tip
placement in AC-PC space, and planning of the cannula trajectory.
Although the entry point was relatively fixed in the NHP due to
use of the adapter plug, in the clinical system the entry point was
modifiable in the pre-craniotomy planning stage as the trajectory
was adjusted. The SmartFrame houses an MRI-visible (gadolinium-
impregnated) fluid stem and integrated fiducials that are detected
by the software. The fluid stem, which also serves as the infusion
cannula guide, was aligned to the target trajectory via both “pitch
and roll” axes and an X–Y translational stage (Fig. 4b). This was
accomplished with an attached hand controller resting at the open-
ing of the MRI bore, according to directions generated by the
software in response to serial T1 MRI sequences, until the fluid
stem alignment matches the chosen target trajectory.
3.9 RCD Infusions were performed in a research magnet shared between

with ClearPoint human and NHP use. Due to institutional regulations prohibiting
System procedures that expose animal blood products in such an area, the
surgical placement and removal of the skull mounted aiming device
(SmartFrame) did not occur in the iMRI suite as would occur in
patients. Two weeks prior to infusion, NHP underwent stereotac-
tic placement of skull-mounted, MRI-compatible, threaded plastic
adapter plugs (12 mm diameter × 14 mm height) (Fig. 5a) for later
attachment of the SmartFrame. After performing bilateral craniec-
tomies, one plug was secured to the skull over each hemisphere
with dental acrylic. After placement of the adapter plugs, animals
recovered for at least 2 weeks before initiation of iMRI infusion
procedures.
3.9.2 Trajectory Planning On the day of infusion, NHP were sedated with ketamine (Ketaset,
and Cannula Insertion 7 mg/kg, intramuscular) and xylazine (Rompun, 3 mg/kg, intra-
muscular), intubated, and placed on inhaled isoflurane (1–3 %).
The plug adapter was prepared sterile and the NHP was placed
supine in an MRI-compatible stereotactic frame. The SmartFrame
Fig. 5 (a) Plastic adapter plug. (b) Sagittal screenshot of target trajectory alignment. The T1 MRI-visible fluid
stem, which holds the infusion cannula, has been aligned by translating the SmartFrame around a fixed pivot
point so that the trajectory meets the target
188 Dali Yin et al.
was attached by screwing the base onto the adapter plug over one
hemisphere. The NHP was moved into the bore and a controller
was attached to the SmartFrame by inserting guide wires into each
of four adjustment knobs. This controller allows the surgeon to
manually “dial in” distance changes to align the cannula to the
desired trajectory in four planes (pitch, roll, anterior-posterior,
medial-lateral) (Fig. 4b) as instructed by the ClearPoint software.
First, a high-resolution anatomical MR scan was acquired for
target identification and surgical planning. Specific details of our
MRI scanning may be found in Fiandaca et al. [40]. The scan was a
9-min 3D Magnetization Prepared Rapid Gradient Echo
(MPRAGE). The MPRAGE images were then transferred to the
ClearPoint system, where the target for cannula tip placement was
selected. Next, rapid scans were obtained that allowed the ClearPoint
software to detect the position and orientation of the SmartFrame
fluid stem. First, a 6-s 2D turbo-spin echo (TSE) was acquired
through the distal fluid stem in an orientation perpendicular to the
desired trajectory. The software used this image to compare the
current SmartFrame trajectory to the target trajectory in order to
calculate an expected error for tip placement and generate instruc-
tions to adjust SmartFrame alignment via the pitch and roll. After
these adjustments were made, the scan was re-acquired to measure
the new expected error and this process was repeated as necessary.
When the expected error fell below 1.0 mm, the pitch and roll
axes on the SmartFrame were locked and a 26-s 2D TSE scan was
acquired along the sagittal and coronal planes of the guide stem for
fine adjustment of the SmartFrame X–Y stage. Seven slices at 1 mm
isotropic resolution were acquired over a 180 × 240 mm FOV with
a TE of 22 ms, a TR of 500 ms, two repetitions, an echo train
length of 7 and a bandwidth of 250 Hz/pixel. The ClearPoint
software used these images to generate instructions for fine adjust-
ment of the trajectory, achieved by dialing in distance changes on
the SmartFrame X–Y stage. This process was repeated until the
software reported an expected error of less than 0.5 mm that typi-
cally required no more than two iterations. The infusion system
included a customized, ceramic, fused silica reflux-resistant can-
nula developed in accord with previously reported principles devel-
oped in our laboratory [16, 26]. For infusions, the cannula was
connected to a loading line containing 1 mM gadoteridol, and
flow was regulated with 1 ml syringe filled with trypan blue,
mounted onto a MRI-compatible infusion pump (Harvard
Bioscience Company). With the aiming device aligned in its final
position, the software reported the expected distance from the tar-
get to the top of the guide stem, and this distance was measured
from the cannula tip and marked on the cannula with a sterile ink
marker. A depth-stop was then secured at the marked location and
the measured insertion distance was verified. The infusion pump
was started at 1 μl/min, and after visualizing fluid flow from the
cannula tip when held at the height of the bore, the cannula was
inserted through the SmartFrame guide stem and into the brain.
When the depth-stop touched the top of the guide stem, it was
secured with a locking screw.
3.9.3 Infusion After cannula insertion, repeated multi-planar Fast Low Angle
and Imaging Shot (FLASH) images were obtained every 5 min throughout the
duration of the infusion. The FLASH images were acquired at an
in-plane resolution of 0.7 × 0.7 × 1 mm with 128 slices over the
180 mm FOV at a TE of 4.49 ms, a TR of 17 ms with two repeti-
tions and a bandwidth of 160 Hz/pixel. The first scan was acquired
with a 4° flip angle to produce a proton-density-weighted image
for visualization of the cannula tip. All subsequent scans were
acquired with a 40° flip angle to increase the T1 weighting and
highlight the signal enhancement from gadoteridol in the infusate.
Upon visualization of gadoteridol infusion at the cannula tip, the
infusion rates were increased from an initial rate of 1 μl/min in a
ramping fashion, 0.5 μl/min every 5 min, to reach a maximum of
3 μl/min. The interface between trypan blue and gadoteridol within
the loading line was also marked at the start and finish of the infusion
in order to verify that the infused volume matched that reported by
the pump. Each NHP first received a small volume infusion in the
thalamic area (16–25 μl) to allow calculation of targeting error, fol-
lowed by a larger volume infusion (187–230 μl) into the ipsilateral
thalamus. In general, the total time under general anesthesia for
NHP that received four sequential infusions was approximately 6 h.
3.9.4 Imaging Data Images obtained during RCD were transferred to the ClearPoint
Analysis system for analysis of targeting error. With the target position hid-
den from view, the location of the cannula tip was manually selected
in the ClearPoint console by identifying the center of the gadoteri-
dol signal in the lower one third of the infusion volume on the first
scan demonstrating convection following cannula insertion
(Fig. 5b). The software then automatically reported the vector dis-
tance between the target site and the actual position of the cannula
tip. The average target error for all infusions was later calculated
and the 95 % confidence interval was determined. Spearman’s rank-
order correlation was used as a non-parametric measure of the sta-
tistical dependence between depth to target and target error.
4 Results of RCD with Application of ClearPoint System
Based on the results of RCD, the ClearPoint system appears to be

highly accurate. Satisfactory cannula placement was achieved on
the first attempt (without the need for repositioning) in all cases
(Fig. 6) (Richardson et al., Stereotactic and Functional Neurosurgery
in Press). The ClearPoint system automatically calculated each tar-
geting error, defined as the three-dimensional distance between the
190 Dali Yin et al.
Fig. 6 Cannula placement and initial infusion in the thalamus with application of
ClearPoint system are shown in panels a for green zone. Panel b shows distribu-
tion of gadoteridol in the thalamus after infusion into green zone. Note that infu-
sion into green zone (b) resulted in tracer distribution in thalamus only
expected cannula tip location and the actual location measured

on post-insertion imaging. The average targeting error for all tar-
gets (n = 11) was 0.8 mm (95 % CI = 0.14 mm). We demonstrated
that this system could place two infusions in close proximity
without producing reflux in the initial cannula tract during the
course of the second infusion. No technical limitations were
encountered in redirecting the cannula for infusing multiple tar-
gets in the same hemisphere. No infusions in any target produced
occlusion, cannula reflux or leakage from adjacent tracts, and no
signs of unexpected tissue damage were observed. In terms of
cannula safety, no MRI-visible hemorrhages occurred during
cannula placement, and no adverse events occurred during
RCD. Standard postoperative care assessments indicated that no
RCD-related side effects were observed over the course of these
experiments (2 months).
The accuracy of ClearPoint system surpasses that of our previ-
ous experience with RCD in NHP where either a guide cannula or
a multiport guide array was placed stereotactically in reference to a
baseline MRI, prior to the iMRI procedure. Infusion data obtained
by these methods were analyzed recently to determine the optimal

zones for cannula placement within the putamen, thalamus, and
brain stem [29, 30] that predict contained distribution within the
target region. Images obtained during RCD in those studies
showed that distribution of gadolinium tracer outside of the target
structure occurred in 64 % of putaminal infusions and in 43 % of
thalamic infusions, clearly demonstrating the need for prospective
stereotaxy. In the current study, no infusions in the thalamus pro-
duced significant extra-thalamic distribution into white matter
tracks or leakage into the CSF space due to poor cannula place-
ment. In fact, the accuracy of this system guarantees insertion of
the infusion cannula at preselected coordinates within a “green
zone” for optimal distribution within target structures. In addition
to the ability to choose the target and trajectory in real time, there
are other advantages to the ClearPoint system that may explain
improved performance over previous experimental studies. In
comparison to the cannula used in our most recent NHP studies
[29], the internal diameter of the current cannula was smaller
(200 mm vs. 324 mm). Additionally, the skull-mounted
SmartFrame provides a rigid housing for the cannula that restricts
axial movement during brain insertion. Therefore, the ClearPoint
system allows RCD to be performed with a high level of precision,
predictability, and safety. This technique should increase the utility
of RCD for expanding the scope of drug delivery studies. Clinical
application of this platform was likely to improve the success rate
for clinical trials employing cerebral drug delivery by direct
infusion.
5 Future Development
The ability of this RCD platform to deliver controlled volumes of

drug to any structure in the NHP brain with highly accurate local-
ization (on the order of 1 mm), and the capacity to monitor the
infusion in real time, expands the utility of the NHP brain to model
human disease and development of novel therapies. The NHP
brain is uniquely suited for neurosurgical investigation of thera-
peutic delivery, due to similarities between human and primate
anatomy and physiology that cannot be closely modeled in other
species [9]. We anticipate that this system will facilitate the creation
of new NHP disease models due to the novel ability for precise
infusion of therapeutic agents within discrete brain regions. Our
recent investigations comparing linear measures in humans versus
NHP will allow us to translate our NHP stereotactic (RGB) target-
ing data to humans [40], thereby facilitating advancement of these
techniques into the clinic. In addition, we expect that ongoing
studies will allow modeling of specific patterns of viral vector distri-
bution and subsequent gene expression in structures to be targeted,
192 Dali Yin et al.
such as mapping infusions in the putamen for Parkinson’s disease.

The evolution of this system also may also include tools that aid the
neurosurgeon in planning, delivering, and anticipating the func-
tional outcome of infusions into multiple brain locations. For
instance, initiatives to incorporate the auto-segmentation of target
structures and surrounding anatomy, as well as auto-segmentation
of infusion volumes in real time, are under way. Eventually, analysis
of retrospective and prospective infusion data in the NHP brain
should allow the development of predictive algorithms that will
ultimately allow the system software to forecast areas of drug dis-
tribution or transgene expression based on a selected location for
cannula placement.
6 Notes
Our strategy for real-time convection delivery (RCD) in the pri-

mate brain is to minimize reflux and leakage of therapeutics infused
while maximizing its distribution in the target. Attention should
be paid to several points during RCD. When the infusion line is set
up, it is necessary to keep all of the lines at the same level as the
cannula and remove air bubble from the lines. When the cannula is
lowered to a targeted area, the pump should be infusing at rate of
1.0 μl/min to prevent occlusion. Poor cannula placement or any
lateral movement during insertion is often a problem during
RCD. It is important to position the cannula in the right area
(green zone) of the target. Real-time MRI and using the ClearPoint
system would improve the accuracy of cannula placement. Infusion
volume (Vi) of therapeutics in the target is also important. A small
Vi would have limited distribution of therapeutics while a Vi that
is too large would result in reflux and leakage. High infusion rate,
such as >5 μl/min, could easily cause reflux and leakage. Therefore,
all these factors need to be considered when RCD is planned.
References
1. Bobo RH et al (1994) Convection-enhanced 5. Kunwar S (2003) Convection enhanced deliv-

delivery of macromolecules in the brain. Proc ery of IL13-PE38QQR for treatment of recur-
Natl Acad Sci U S A 91(6):2076–2080 rent malignant glioma: presentation of interim
2. Hadaczek P et al (2006) “Perivascular pump” findings from ongoing phase 1 studies. Acta
driven by arterial pulsation is a powerful mech- Neurochir Suppl 88:105–111
anism for the distribution of therapeutic mole- 6. Mardor Y et al (2001) Monitoring response to
cules within the brain. Mol Ther 14(1):69–78 convection-enhanced taxol delivery in brain
3. Gill SS et al (2003) Direct brain infusion of tumor patients using diffusion-weighted mag-
glial cell line-derived neurotrophic factor in netic resonance imaging. Cancer Res
Parkinson disease. Nat Med 9(5):589–595 61(13):4971–4973
4. Eberling JL et al (2008) Results from a phase I 7. Carson BS Sr et al (2002) New approach to
safety trial of hAADC gene therapy for tumor therapy for inoperable areas of the brain:
Parkinson disease. Neurology 70(21):1980– chronic intraparenchymal drug delivery.
1983 J Neurooncol 60(2):151–158
8. Lonser RR et al (1999) Convection-enhanced 21. Dickinson PJ et al (2008) Canine model of

selective excitotoxic ablation of the neurons of convection-enhanced delivery of liposomes
the globus pallidus internus for treatment of containing CPT-11 monitored with real-time
parkinsonism in nonhuman primates. magnetic resonance imaging: laboratory inves-
J Neurosurg 91(2):294–302 tigation. J Neurosurg 108(5):989–998
9. Richardson RM, Larson PS, Bankiewicz KS 22. Sykova E (2004) Diffusion properties of the
(2008) Gene and cell delivery to the degener- brain in health and disease. Neurochem Int
ated striatum: status of preclinical efforts in pri- 45(4):453–466
mate models. Neurosurgery 63(4):629–644, 23. Chen MY et al (1999) Variables affecting
discussion 642–644 convection-enhanced delivery to the striatum:
10. Kroll RA et al (1996) Increasing volume of a systematic examination of rate of infusion,
distribution to the brain with interstitial cannula size, infusate concentration, and
infusion: dose, rather than convection, tissue-cannula sealing time. J Neurosurg
might be the most important factor. 90(2):315–320
Neurosurgery 38(4):746–752, discussion 24. Krauze MT et al (2005) Effects of the perivas-
752–754 cular space on convection-enhanced delivery of
11. Krauze MT et al (2006) Real-time imaging and liposomes in primate putamen. Exp Neurol
quantification of brain delivery of liposomes. 196(1):104–111
Pharm Res 23(11):2493–2504 25. Szerlip NJ et al (2007) Real-time imaging of con-
12. Krauze MT et al (2005) Real-time visualization vection-enhanced delivery of viruses and virus-
and characterization of liposomal delivery into sized particles. J Neurosurg 107(3):560–567
the monkey brain by magnetic resonance imag- 26. Krauze MT et al (2005) Reflux-free cannula
ing. Brain Res Brain Res Protoc 16(1-3): for convection-enhanced high-speed delivery
20–26 of therapeutic agents. J Neurosurg 103(5):
13. Saito R et al (2005) Gadolinium-loaded lipo- 923–929
somes allow for real-time magnetic resonance 27. Sanftner LM et al (2005) AAV2-mediated gene
imaging of convection-enhanced delivery in delivery to monkey putamen: evaluation of an
the primate brain. Exp Neurol 196(2):381– infusion device and delivery parameters. Exp
389 Neurol 194(2):476–483
14. Nguyen TT et al (2003) Convective distribu- 28. Yin D, Forsayeth J, Bankiewicz KS (2010)
tion of macromolecules in the primate brain Optimized cannula design and placement for
demonstrated using computerized tomography convection-enhanced delivery in rat striatum.
and magnetic resonance imaging. J Neurosurg J Neurosci Methods 187(1):46–51
98(3):584–590 29. Yin D et al (2011) Optimal region of the puta-
15. Richardson RM et al (2009) Future applica- men for image-guided convection-enhanced
tions: gene therapy. Neurosurg Clin N Am delivery of therapeutics in human and non-human
20(2):205–210 primates. Neuroimage 54(Suppl 1):S196–S203
16. Fiandaca MS et al (2008) Image-guided 30. Yin D et al (2010) Cannula placement for
convection-enhanced delivery platform in the effective convection-enhanced delivery in the
treatment of neurological diseases. Neuro- nonhuman primate thalamus and brainstem:
therapeutics 5(1):123–127 implications for clinical delivery of therapeu-
17. Fiandaca MS et al (2009) Real-time MR imag- tics. J Neurosurg 113(2):240–248
ing of adeno-associated viral vector delivery to 31. Varenika V et al (2008) Detection of infusate
the primate brain. Neuroimage 47(Suppl leakage in the brain using real-time imaging of
2):T27–T35 convection-enhanced delivery. J Neurosurg
18. Gimenez F et al (2011) Image-guided 109(5):874–880
convection-enhanced delivery of GDNF pro- 32. Truwit CL, Liu H (2001) Prospective stereo-
tein into monkey putamen. Neuroimage taxy: a novel method of trajectory alignment
54(Suppl 1):S189–S195 using real-time image guidance. J Magn Reson
19. Su X et al (2010) Real-time MR imaging with Imaging 13(3):452–457
gadoteridol predicts distribution of transgenes 33. Hall WA et al (2001) Brain biopsy sampling by
after convection-enhanced delivery of AAV2 using prospective stereotaxis and a trajectory
vectors. Mol Ther 18(8):1490–1495 guide. J Neurosurg 94(1):67–71
20. Lonser RR et al (2007) Real-time image- 34. Martin AJ et al (2008) Minimally invasive pre-
guided direct convective perfusion of intrinsic cision brain access using prospective stereotaxy
brainstem lesions. Technical note. J Neurosurg and a trajectory guide. J Magn Reson Imaging
107(1):190–197 27(4):737–743
194 Dali Yin et al.
35. Martin AJ et al (2005) Placement of deep in rodent intracranial brain tumor xenografts.
brain stimulator electrodes using real-time Neuro Oncol 9(4):393–403
high-field interventional magnetic resonance 38. Saito R et al (2006) Convection-enhanced
imaging. Magn Reson Med 54(5): delivery of Ls-TPT enables an effective con-
1107–1114 tinuous low-dose chemotherapy against malig-
36. Starr PA et al (2010) Subthalamic nucleus deep nant glioma xenograft model. NeuroOncology
brain stimulator placement using high-field 8(3):205–214
interventional magnetic resonance imaging 39. Yin D et al (2009) Striatal volume differences
and a skull-mounted aiming device: technique between non-human and human primates.
and application accuracy. J Neurosurg 112(3): J Neurosci Methods 176(2):200–205
479–490 40. Fiandaca MS et al (2011) Human/nonhuman
37. Krauze MT et al (2007) Convection-enhanced primate AC-PC ratio—considerations for
delivery of nanoliposomal CPT-11 (irinotecan) translational brain measurements. J Neurosci
and PEGylated liposomal doxorubicin (Doxil) Methods 196(1):124–130
Chapter 15
Focal Cerebral Ischemia by Permanent Middle Cerebral

Artery Occlusion in Sheep: Surgical Technique, Clinical
Imaging, and Histopathological Results
Björn Nitzsche, Henryk Barthel, Donald Lobsien, Johannes Boltze,
Vilia Zeisig, and Antje Y. Dreyer
Abstract
According to the recommendation of international expert committees, large animal stroke models are
demanded for preclinical research. Based on a brief introduction to the ovine cranial anatomy, a sheep
model of permanent middle cerebral artery occlusion (MCAO) will be described in this chapter. The
model was particularly designed to verify several therapeutic strategies during both, acute and long-term
studies, but is also feasible for development of diagnostic procedures. Further, exemplary application of
imaging procedures and imaging data analyses using magnetic resonance imaging (MRI) and positron
emission tomography (PET) are described. The chapter also includes recommendations for appropriate
animal housing and medication.
Key words Large animal model, Sheep, Experimental neurosurgery, Craniotomy, Experimental
stroke, Middle cerebral artery occlusion, MRI, PET
1 Introduction
1.1 The Role Worldwide, ischemic stroke represents a major cause of death and
of Animal Models is the most important reason for permanent disability in adulthood
in Preclinical Stroke [1]. Thrombolysis by recombinant tissue plasminogen activator is
Research currently the only pharmacological approved therapy for this dis-
ease, and the time window for intervention has recently been
extended to 4.5 h [2]. Nevertheless, because of this still narrow
time window and due to rapidly decreasing therapeutic efficacy
within that time [3], the vast majority of stroke patients remain
untreated or only benefits from minor therapeutic effects. Despite
significant research activities and promising results in preclinical
tests, not a single experimental treatment strategy was successfully
translated into clinical routine so far [4]. The underlying reasons
195
196 Björn Nitzsche et al.
are considered multiple and, among others, comprise inappropri-

ate or nonpredictive animal models in preclinical research [5].
The standard species in preclinical stroke research are rats and
mice. Rodent animal models offer numerous advantages such as
easy animal housing, well established methodology including the
availability of genetically modified strains, and excellent tools to
assess functional outcome.
However, international expert committees recommend the use of
large animal models to verify results previously obtained in small ani-
mal studies using neuroprotective [6] or cell-based approaches [7].
Large animals may allow for long-term assessment of safety
and efficacy up to years, whereas the life span of a common labora-
tory rodent is usually restricted to several months post stroke.
Moreover, therapeutic efficacy can be tested in gyrencephalic
brains, and influences related to brain size or bodyweight (e.g.,
after systemic injection of the therapeutic agent) may be assessed
more thoroughly. Large animal models also permit testing in spe-
cific neuroanatomical structures, in particular white matter [6],
and the use of clinical imaging techniques. Moreover, it offers the
possibility to deliver stem cells of several autologous sources. Thus,
large animal models are considered to be of increasing relevance
for preclinical and translational stroke research.
1.2 Large Animal Focal ischemic stroke is predominantly induced by permanent or

Models of Stroke transient occlusion of the middle cerebral artery occlusion
(MCAO). Existing large animal models of focal cerebral ischemia
comprise rabbit [8], canine [9], feline [10], and swine [11] mod-
els. MCAO studies with rabbits have been conducted for a long
time but are restricted by the relatively low weight and small size
of the animals. For anatomical reasons, canine, feline and porcine
stroke models mostly require enucleation to assess the MCA. This
may lead to severe behavioral abnormalities restricting the use of
these models for long-term safety and efficacy assessments.
Nonhuman primate models, which are essential to investigate
acute stroke pathophysiology, are often associated with high mor-
tality rates thus preventing long-term observations [6]. An excep-
tion is the marmoset. However, MCAO studies using this
nonhuman primate [12] did not predict the failure of the neuro-
protectant NYX-059 in clinical trials [13].
An ovine model of permanent distal MCAO (see Note 1) has
been developed avoiding enucleation by using a temporal transcra-
nial approach to the MCA [14]. Ischemic lesion size and subse-
quent functional outcome can be controlled by varying the number
of occluded cortical MCA branches. Further, the sheep model
benefits from the easy availability of clinical imaging modalities.
Widely available clinical scanners, including computer tomography
(CT), magnetic resonance imaging (MRI), and positron emission
tomography (PET) can be used to visualize the progress of the
infarction as well as the effects of potential therapeutic agents.
Permanent MCAO in Sheep 197
A detailed description of the surgical methodology, used mate-

rials, application of sophisticated imaging techniques, and post
mortem pathohistology is given in this chapter. We also briefly
refer to options of stereotaxic or image guided surgery using this
animal model (see Note 2).
2 Materials and Animals
All devices, materials, consumables, and drugs mentioned in the

protocols can principally be replaced by equivalent products from
other suppliers.
2.1 Animal Housing, Some general information regarding animal handling, ovine skull
Care, and Brief anatomy, and cerebral blood supply is needed to ensure adequate
Anatomical and reproducible experimental results. The following paragraphs
Description provide a brief introduction to the mentioned aspects.
of the Ovine Skull
2.1.1 Experimental The herein neurosurgical approach for MCAO induction necessi-
Subjects and Animal tates hornless subjects for easy accessibility of cranial structures.
Housing Merino sheep may be used preferably as many hornless strains can
be found in this widely available breed. Weight (ewe: 75–85 kg;
ram: 120–140 kg) and body size (height at withers: 0.8–0.9 m) of
adult Merino sheep [15] allows relatively easy handling. Species
appropriate housing, feeding (see Table 1) as well as thorough med-
ical inspections and blood screening (see Table 2), medication and
vaccination (see Tables 1 and 3) ensure a significantly reduced risk
of postoperative complications and thereby enhance study quality.
Frequent and early human contact facilitates familiarization and
improves the handling, especially during long-term studies.
2.1.2 Ovine Skull The ovine skullcap is less convex as compared to humans, primates,
Anatomy dogs, and cats. In sheep, the oral rim is comparatively small but the
species has a long pharyngeal cavity with a massive torus linguae
and a long epiglottis, the latter being situated dorsal of the soft
palate. The frontal bone includes widespread air-filled sinuses (that
may be extend up to the origin of the horns, see Fig. 1). Those
sinuses must not be opened due to a possibly fatal sinusitis/
osteomyelitis.
2.1.3 The Ovine Brain The mean weight of an adult sheep brain is 120 g. The mean hori-
zontal circumference including the cerebellum measures about
20 cm, the mean vertical circumference is approximately 12 cm.
When adjusting on BW and age, the cerebral tissue volumes are
51.5 ± 3.9 mL (GM), 35.6 ± 3.2 mL (WM), 29.7 ± 3.3 mL (CSF),
and 87.1 ± 6.0 mL (total brain volume) [16]. Sulci and gyri of the
Table 1
Housing, feeding, and general health care
Food − Water and hay ad libitum

− Add mash daily (avoid sweet corn, use oat or barley) or silage
− Provide mineral food supplements (salt block)
Housing − Flock, separated by sex
− Individual enclosure for 2 days postsurgical with visual contact to flock members
− Space: ≥1.5 m2 per subject
− Ground litter: wooden shaving (sterilized) or straw
− Ambient temperature: 5–28 °C, heating may be required during winter
− No air conditioning is necessary during summer, but shaving of subject before
the summer months
− Outdoor housing for at least 2 h per day (above 5 °C)
− Minimum light intensity 80 lx
− Maximal environmental gas concentration: 3.500 ppm CO2, 30 ppm NH3,
5 ppm H2S
Parasite − Doramectin
prophylaxis − Toltrazuril
Vaccination − Should be completed before entering the experimental facility
− Clostriadia, pneumococcus (e.g., Ovilis Heptavac P, Intervet, Germany): in
postnatal week 3, booster vaccination in week 7, followed by annual boostering
− Footrot (e.g., Footvax® Schering-Plough, Animal Health, Germany): in
postnatal week 3, boostering in week 7, followed by annual boostering
− Ecthyma contagiosum (e.g., Ecthybel ad us. vet., Merial, France): in postnatal
week 7, followed by annual boostering
− Bluetongue disease boostering (e.g., BTVPUR AlSap™ 8, Merial, France): in
postnatal week 4, followed by annual boostering
− Rabies (e.g., Rabisin®, Merial, France): in postnatal week 8, followed by
boostering every third year
Vital parameters − Rumen: normally three fluctuations in 2 min
− Body temperature: 38.0–40.0 °C
− Breathing frequency: 20–100 min−1
− Heart beat rate: 70–110 min−1
− Blood pressure: 80/55/60 mmHg (systolic/diastolic/mean)
ppm parts per million, CO2 carbon dioxide, NH3 ammonia, H2S hydrogen sulfide
neocortex, like in primates, are assembled in an individual configu-

ration in any animal (see Fig. 2; see Note 3).
The pyramidal tracts in primates are of utmost importance for
motor functions. Almost all pyramidal fibers cross to the contralat-
eral hemisphere. In contrast, only about 50 % of the motor fibers
cross at the pyramidal decussation in sheep [17]. Therefore, a uni-
lateral loss of central motor innervation (as seen following stroke)
can partly be compensated. Consequently, even large cortical defi-
cits caused by MCAO in sheep result in hemiparesis, but not hemi-
plegic conditions. Motoric dysfunction can clearly be observed and

quantified, while the animal is still able to move and join a flock.
This dramatically reduces poststroke mortality and complications,
which represents a major advantage in particular for long-term
studies and observations.
2.1.4 Cerebral Blood The blood supply of the ovine circle of Willis (CW) originates from
Supply in Sheep the so called rete mirabile epidurale rostrale. The rete is supplied by
the maxillary artery [18]. This is a major difference to the anatomi-
cal situation found in humans and primates. The rete, being
embedded in a venous sinus, comprises a dense network of inter-
communicating, small arteries. Because of their very small diame-
ter, these arteries do not allow intravascular approaches to the
CW. Only in lambs, the CW is supplied by a carotid artery (origi-
nating from maxillary artery), which obliterates in the first postnatal
months. Thus, a functional intact internal carotid artery does not
exist in adult sheep.
At the level of CW, the blood supply of the ovine brain is com-
parable to the situation in primates and humans. Strong anterior
Table 2
Normal range of relevant blood parameters in sheep
Hematology Blood chemistry
Parameter Unit Value (range) Parameter Unit Value (range)

Leukocytes 109/L 5.0–11.0 Urea mMol/L 2.8–7.1
Erythrocytes 10 /L12
7.0–11.0 Creatinine μMol/L 100–125
Hemoglobin mMol/L 5.6–9.3 Total protein g/L 60–78
HKT L/L 0.27–0.40 ASAT U/L <65
MCV fL 28.0–40.0 ALAT U/L <14
MCH fMol 0.6–0.7 GGT U/L <60
MCHC mMol/L 19.0–23.0 Bilirubin μMol/L <4
Platelets 109/L 280–650 AP U/L <60
Neutrophile granulocytes 10 /L9
0.1–4.8 LDH U/L <600
Lymphocytes 10 /L9
2.2–9.2 CK U/L 13–230
Monocytes 109/L 0–2.0 Na mMol/L 139–152
Eosinophile granulocytes 109/L 0–2.0 Cl mMol/L 95–110
Basophile granulocytes 10 /L9
0–1.0 Fibrinogen g/L 1.0–5.0
ALAT alanine amino transaminase, AP alkaline phosphatase, ASAT aspartate aminotransaminase, CK creatine kinase,
Cl chloride, GGT gamma-glutamyltransferase, HKT hematocrit, LDH lactate dehydrogenase, MCH mean corpuscular
hemoglobin, MCHC mean corpuscular hemoglobin concentration, MCV mean corpuscular volume, Na sodium
Fig. 1 Gross anatomy of the ovine head. Three-dimensional reconstruction was performed using a high resolu-
tion CT data set using OsiriX 3.8.1 [33]. (a) Head of a sheep: The area of surgical access is indicated by the
hemitransparent, elliptic overlay. (b) Topography of the sheep skull: The approximate position of area of surgi-
cal access is indicated by the elliptic overlay. The burr hole is indicated by the hemitransparent circle. The
extensive nasal sinus (1) and the frontal sinus (2) must not be opened during transcranial surgery. The Sella
turcica (3) indicates the level of the middle cerebral artery. The roof of the compact Os parietale (4) is easily
achievable for any kind of cranial surgery in the species. However, the MCA cannot be reached using a high
parietal approach. The atlanto-occipital junction (5) includes a massive axis. Note reconstruction of the endo-
tracheal tube (6) which is placed in the diastema between incisor and premolar teeth
cerebral arteries (ACA) are connected by a communicating branch,

which is present in virtually all subjects. The middle cerebral artery
(MCA) originates from the CW at level of pituitary infundibulum.
In further progress, the MCA runs rostrally to the Lobus pirifor-
mis inside the lateral sulcus (M1 segment). It splits up into two or
three branches (M2 segments). The posterior communicating
arteries of the CW give rise to the posterior cerebral arteries (PCA),
as well as to the rostral cerebellar artery before converging into the
basilar artery [19]. A detailed visualization of the cerebral blood
supply in sheep is given in Fig. 3 (see Note 4).
2.2 Anesthesia, 1. Ketamine, xylazine, diazepam or midazolam, and propofol (see

Arterial, Table 3; see Note 5).
and Venous Access 2. 50 mL 0.9 % sterile sodium chloride solution.
2.2.1 Initial Sedation, 3. Syringes: 3 mL (1×), 5 mL (3×), 10 mL (3×), 20 mL (1×).
Arterial, and Venous 4. 18G needles (10×; Braun Melsungen AG, Germany; see Note 6).
Access
5. 14G venous cannula (1×; length: 80 mm, e.g., Braunüle MT,
Braun Melsungen) with in-stopper (Braun Melsungen).
Fig. 2 Anatomy of the sheep brain and functional organization of the neocortex. (a) Functional areas of the
sheep neocortex are highlighted in a MRI 3D reconstruction [33]. The brain in (b–d) was removed from the
skull after automated perfusion with 20 L 4 % paraformaldehyde, followed by immersion fixation in 4 % para-
formaldehyde for 3 days. (b) Lateral view of the brain: The approximate location of the drill hole is indicated by
the hemitransparent circle overlay. (c) Basal view of a brain. (d) Coronal brain slice at level of optic chiasm.
Legend: (1, 2) somatosensory area I (face, lips, tongue); (3) somatosensory area II (face, forelimb, hindlimb);
(4) auditory area; (5) visual cortex; (6) motor area (eye face, head, forelimb, hindlimb, tongue); (7) lateral sul-
cus; (8) caudal sylvian gyrus; (9) olfactory bulb; (10) lateral rhinal fissure; (11) middle cerebral artery; (12) optic
chiasm; (13) infundibulum; (14) piriform lobe; (ca) corpus callosum; (cc) claustrocortex; (ce) extrem capsule;
(cex) external capsule; (ci) internal capsule; (cr) corona radiata; (nc) caudate nucleus; (p) putamen; (pa) globus
pallidus; (sn) septal nuclei
Fig. 3 Cerebral blood supply in sheep. (a) MRI time-of-flight (TOF) visualization of the cerebral blood supply at
the skull base. (b) Dorsal view into the cranial cavity (3D reconstruction of a CT-angiography). The major arte-
rial vessels were highlighted digitally. (c) Corrosion cast model made by intra-arterial delivery of Mallocryl M®,
basal view. Legend: (1) anterior cerebral artery with communicating branches; (2) occluded middle cerebral
artery (note missing ipsilateral capillaries); (3) caudal communicating branches of the Circle of Willis; (4) rete
mirabile epidurale rostrale; (5) basal artery
6. 18G LEADER-Catheter set (1×; diameter: 1.2 mm, length:

18 cm, Vygon GmbH & Co.KG, Germany).
7. Tube guide rod (aluminum, semiflexible, diameter: 2–3 mm,
length: 60 cm).
8. Laryngoscope (Heine AG, Germany) with Miller blade, size 4
(Heine Classic F.O., Germany).
9. Endotracheal tube (size: 10 or 11 mm, length: 140 mm;
Pharmazeutische Handelsgesellschaft, Germany) and bandage
(e.g., Idealast® C, Paul Hartmann AG).
10. Pulse oximeter (PM-60, Mindray Ltd., China) with tongue
probe (Eickemeyer KG, Germany).
11. Stethoscope (Littmann®, 3M GmbH, Germany).

12.
Temperature probe/clinical thermometer (Henry-Schein
GmbH, Germany).
13. Animal balance (e.g., Kern EOS300 K200XL, Wägetechnik
Koch GmbH, Germany).
14. Small (size: 1.6 mm e.g., Golden A fine, Averde GmbH,
Germany) and large (size: 2.8 mm, e.g., Golden A medium)
electric clippers (e.g., Oster Power UltraPro, Averde GmbH,
Germany).
2.2.2 Inhalative 1. 2 L sterile Ringer-lactate (see Table 3).

Anesthesia During Surgery 2. Medical gases: oxygen and isoflurane (see Table 3; see Note 7).
and Imaging Procedures
3. Eye ointment (e.g., Corneagel®, Bausch & Lomb GmbH,
Germany).
4. Respirator (e.g., Draeger Primus®, Draeger AG, Germany)
with oxygen, isoflurane vapors absorber, capnography sensor,
and breathing filter with heat/moisture exchanger
(ClearGuard3, Intersurgical GmbH, Germany). A MR-
compatible machine is required for ventilation anesthesia dur-
ing MRI (see Note 8).
5. Electrocardiogram (ECG) and vital signs monitor (e.g., Infinity
Gamma XXL, Draeger AG) with crocodile clips (Eickemeyer
KG), noninvasive blood pressure (NIBP) cuff (size: pediatric)
and rectal temperature probe (Draeger AG).
6. Automated infusion system (Infusomat FM®, Braun Melsungen
AG) and tubing.
7. Heidelberger extension and 3-way-stopcock (Discofix®, Braun
Melsungen AG) (Table 3).
Table 3
Medication schemes
Purpose Recommended drug Supplier Route Dose Time of application Duration

Parasite prophylaxis Doramectin (e.g., Pfizer i.m. 0.2 mg kg−1 After animal arrival at Single injection
Dectomax®) experimental facility
Toltrazuril (e.g., Baycox Bayer per os 20 mg kg−1 Single application
5 %®)
Antibiosis Enrofloxacin (e.g., Bayer i.m. 5 mg kg−1 7× daily post MCAO Single injection
Baytril®)
Antiphlogistic and Flunixin meglumin CP Pharma i.m. 2.2 mg kg−1 5× daily post MCAO Single injection
analgesia −1
Buprenorphine (e.g., RB Pharmaceuticals i.m. 0.01 mg kg Every 8 h for 2 days Single injection
Temgesic®) postsurgical
Induction of 2 % xylazine hydrochloride Ceva Sante Animal i.v. bolus 0.1 mg kg−1 Prior to anesthesia Single injection
anesthesia (e.g., Xylazin®)
(surgery and
Ketamine hydrochloride Medistar i.v. bolus 4 mg kg−1 Prior to anesthesia Single injection
imaging)
(e.g., Ketamin®)
Diazepam (e.g., Faustan®) Temmler Pharma i.v. bolus 0.2 mg kg−1 Prior to surgery Single injection
or
Midazolam (e.g., Braun Melsungen i.v. bolus 0.2 mg kg−1 Prior to imaging Single injection
Midazolam®)
Inhalation Isoflurane CP Pharma Ventilation 1.5–2.0 % During surgery Up to 5 h
anesthesia
Oxygen Linde Medical Gases 20– 40 %
(continued)
Permanent MCAO in Sheep
203
Table 3
204
(continued)
Purpose Recommended drug Supplier Route Dose Time of application Duration

Infusion anesthesia Midazolam Braun Melsungen i.v. infusion 0.1 mg kg−1 h−1 During imaging Up to 5 h
−1 −1
Ketamine hydrochloride Medistar i.v. infusion 2 mg kg h
(e.g., Ketamin®)
2 % propofol (e.g., Braun Melsungen i.v. infusion 6 mg kg−1 h−1
Björn Nitzsche et al.
Propofol Lipuro®)
Hydration Physiological saline Braun Melsungen i.v. infusion 3 mL kg−1 h−1 During imaging Up to 4 h
solution
Ringer-Lactate Braun Melsungen i.v. infusion 3 mL kg−1 h−1 During surgery
Ruminal stimulation Propionic acid Raiffeisen per os 12.5 g 5× daily post MCAO Twice a day
Butafosfan (e.g., Catosal Bayer Healthcare i.m. 0.5 mL kg−1 Single injection
10 %®)
Menbuton (e.g., Genabil®) Boehringer i.m. 5 mL per animal Single injection
Ingelheim
Amynin CP Pharma i.v. 5 mL kg−1 Single injection
−1
Sacrifice Pentobarbital (e.g., Veterinaria i.v. Bolus 80 mg kg Prior to decapitation; Single injection
Eutha77®) only during
anesthesia!
i.m. intramuscular, i.v. intravenous, kg−1 (× h−1): per kilogram body weight (and hour)
2.3 Surgery 1. 2 L 0.9 % sterile sodium chloride solution (Braun Melsungen
and Postsurgical Care AG).
2.3.1 Surgical Approach 2. Iodine-containing solution for skin disinfection.
for MCAO 3. Folio drape (1 × 1 m, 2×), adhesive folio drape (0.3 × 0.3 m,
1×) and fenestrated folio drape (0.9 × 0.9 m, 1×; all Heiland
VET GmbH, Germany).
4. Adhesive tape (1×, Eickemeyer KG).
5. Eye swab (minimum: 10×) and surgical pads (minimum: 30×,
all Eickemeyer KG).
6. Bone wax (1 package, Braun Melsungen AG) and neurosorb®
patties (2 packages, Vostra GmbH, Germany).
7. 0-0, 2-0, and 6-0 resorbable filaments (2× each, Ethicon Ltd.,
Germany).
8. Electrosurgery and cauterization device (ME 411, KLS Martin,
Germany) with straight and bayonet-shaped neurosurgical
bipolar forceps (1× each, Aesculap AG, Germany).
9. Electric power system and surgical motor (e.g., Microspeed®
uni mini inclusive straight handpiece HiLan XS size II, scil ani-
mal care company GmbH, Germany) with 4 mm Rosen burr
and 6 mm Barrel burr (size II, scil animal care company
GmbH).
10. Standard surgical instruments including Williger raspatories
(size: 2 × 2 mm (1× sharp, 1× blunt) and 1 × 4 mm), atraumatic
retractor (1×), curved Wullstein retractor (1×), dura hook
(Fisch dura retractor, length: 185 mm, 1×, e.g., catalog-no:
FD376R, Braun Melsungen AG), surgical and anatomical for-
ceps (3× each), atraumatic Adson-Brown forceps (1×),
bayonet-shaped neurosurgical forceps (length: 240 mm, 1×),
Roberts artery forceps (1×), curved and straight Metzenbaum
scissor (1× each), straight and angled spring type micro scissors
(length: 120 mm, 1× each), ligature scissor (1×), scalpel (size
22, 1×), Kerrison rongeurs (size: 2 and 4 mm, 1× each),
Backhaus towel clamps (12–20×), standard needle holder (e.g.,
Mathieu Durogrip, 1×).
11. Head light system with magnifiers (e.g., 3s LED Headlight
PR, Heine Optotechnik GmbH & Co.KG, Germany).
2.3.2 Postsurgical Care 1. Drugs: enrofloxacin, flunixine-meglumine, buprenorphine

(see Table 3).
2. Aseptic wound spray (Betadona®, Mundipharma GmbH,
Germany).
3. Wound cover spray (Alu spray silver®, Eskadron GmbH,
Germany).
4. 5 mL syringes (2×) and 18G needles (10×, Braun Melsungen
AG).
5. Adhesive tape (1×, Eickemeyer KG, Germany).

6. Elastic bandage (2×), cotton (1×) and tape (1×, Eickemeyer
KG).
2.4 Imaging 1. Adhesive tape (Eickemeyer KG).

Procedures 2. Folio drape (2× minimum, 1.5 × 1.5 m, Heiland GmbH) for
2.4.1 General Materials covering.
for Imaging 3. Neck crest (height: 0.15–0.25 m).
2.4.2 MR Imaging 1. 1.5 T Scanner (e.g., Gyroscan Intera, Philips, Netherlands)

with a flexible double loop RF-coil (Sense Flex M, Philips) or
2. 3 T Scanner (e.g., Magnetom Trio, Siemens, Erlangen,
Germany) with a four channel flex coil.
2.4.3 PET Imaging 1. High-resolution clinical PET scanner (e.g., ECAT EXACT
HR+; Siemens/CTI, USA).
2. NeuroShield® (Scanwell Systems, Canada).
3. [15O]H2O; synthesized by a catalyst-mediated reaction between
[15O]O2 and H2 (from PETtrace cyclotron, GE Healthcare,
USA), followed by dialysis exchange in an automated system
(Veenstra, The Netherlands), that performs tracer injection
subsequently.
4. [18F]Fluordesoxyglucose (FDG), synthesized by a standard
nucleophilic substitution with alkaline hydrolysis.
5. Blood sampler (e.g., ALLOGG AB, Allogg Mariefred,
Sweden).
6. MRI data set with individual anatomical information for coreg-
istration with functional PET data; preferentially use T1 or T2
T2 (turbo spin echo (TSE), fluid attenuation inversion recov-
ery (FLAIR)) sequences.
2.4.4 Imaging Data 1. Vendor-specific MRI postprocessing software for magnetic

Interpretation and Analysis resonance angiography (MRA, e.g., Syngo, Siemens) and for
diffusion tensor imaging (DTI; e.g., DTI studio, Center for
Imaging Science, Johns Hopkins University, Baltimore, USA).
2. ImageJ image postprocessing software (National Institute of
Health, USA).
3. Vendor-specific PET postprocessing software: PMOD soft-
ware (version 3.0, PMOD Technologies, Ltd., Zürich,
Switzerland) for combined interpretation of image data sets
derived from different modalities (PET, MRI, and others).
Further, the PMOD software can be used for kinetic modeling
like cerebral blood flow (CBF) quantification.
2.5 Sacrifice 1. Pentobarbital (see Table 3).

and Post Mortem 2. 3 L Phosphor-buffered sodium (PBS) solution (pH 7.4, store
Analyses at 4 °C).
3. 20 L buffered paraformaldehyde (PFA, pH 7.4, store at 4 °C).
4. 5 L 30 % sucrose in PBS (pH 7.4).
5. 12G steel cannula (2×, Carl-Roth AG, Germany), 100 or
500 mL syringes (2×, Braun Melsungen AG), large surgical
(2×) and anatomical (2×) forceps, stout thread (4×, length:
0.25 m), knife with 180 mm blade (2×), oscillating saw (1×,
e.g., HEBUmedical GmbH, Germany).
6. Roller pump (pericyclic pump) with two rollers (e.g., Cyclo
II®, Carl-Roth GmbH).
7. 20 L canister and adapter for tubing (Carl-Roth GmbH).
8. Tubing (2 × 3 m).
9. Embedding system (e.g., HyperCenter®, Shandon, Germany).
10. Antibodies and histological staining reagents according to
desired analyses. For more details and suggestions, please refer
to [14].
3 Methods
3.1 Animal Housing 1. Prior to the trail, subject all animals to a familiarization period
of at least 7 days (14 days for long-term studies) in the experi-
mental facility.
2.
Perform parasite prophylaxis immediately upon arrival
(see Table 3).
3. Collect blood samples for routine hematological and general
health screening prior to any trial.
4. Exclude subjects with nonphysiological values in routine
hematological screening (see Tables 1 and 2), obvious preexist-
ing neurofunctional deficits, or other illnesses including parasite
infestation.
5. Prior to MCAO or any anesthesia, deprive subjects of food for
at least 18 h (optimum: 24 h) by using common multiperfo-
rated calf muzzle (diameter 20 cm).
6. Allow ad libitum water access (drinking is possible with the
multiperforated muzzle).
7. For venous access via the jugular veins, shave the lateral neck
around the jugular sulcus on each side using electric clippers,
followed by disinfection with 70 % alcohol. To prepare arterial
access, shave the area of the tarsus.
8. Check the weight of the animal prior to sedation for adequate
drug dosages (see Table 3).
3.2 Anesthesia, 1. Prepare syringes for anesthesia: 1× ketamine and xylazine mix
Arterial, (can be used in one syringe, so called “Hellabrunner mix”), 1×
and Venous Access diazepam or midazolam, 3× propofol according to individual
dosages (see Table 3), and 20 mL of 0.9 % sterile sodium chlo-
3.2.1 Initial Sedation,
ride solution.
Arterial, and Venous
Access 2. Repeat disinfection of both jugular sulci with 70 % alcohol.
3. Sedate the subject via slow intravenous injection of ketamine
and xylazine directly into the right jugular vein. Wait until ani-
mal loses consciousness and catch it when it falls down.
4. Place the animal in a lateral position on the right side. Always
maintain this position for transportation and during the surgi-
cal approach to avoid torsion of stomach. Place animal in prone
position through imaging.
5. Cannulate the left jugular vein with the venous catheter, fix the
catheter with skin suture and place the in-stopper (see Note 7).
6. Slowly inject diazepam (for surgery) or midazolam (for imag-
ing) via the venous catheter.
7. Intubate the animal, using the guide rod and laryngoscope.
The guide rod is useful to lift the large soft palate which often
interferes direct tube insertion. In case the swallowing reflex
still persists, administer a propofol bolus intravenously. Finally,
inflate the blocker balloon of the endotracheal tube, and fix the
tube using elastic bandage.
8. Place a naso-oesophageal reflux collector (diameter: 1 cm)
below the ventral conches. Guide along the tracheotubes
through the diastema between incisor and premolar tooth.
9. Continuously monitor breathing frequency using the stetho-
scope. Control for heart rate and oxygen saturation using the
pulse oximeter and the tongue clip. Make sure the tongue clip
is in an adequate position.
10. Check rectal temperature using the thermometer every 10 min.
11. For arterial access, disinfect the tarsus area and (see step 2 of
this section) and place the LEADER-catheter in the tarsal
artery using the Seldinger technique. The arterial line is used
for invasive blood pressure measurements (IBP) during sur-
gery and for blood sampling during PET. Carefully fix the
catheter with skin suture and tape.
12. During any transportation of the animal (to the imaging facil-
ity or to the operation room), maintain continuous monitor-
ing, and protect venous and arterial lines. You can use a
standard stretcher for transportation.
3.2.2 Inhalative 1. Recommended concentrations for isoflurane and oxygen are

Anesthesia During Surgery given in Table 3. These recommendations may be ignored and
and Imaging Procedures be replaced by alternative concentrations in case the anesthetist
will find this appropriate. During surgery, in particular trepa-

nation, do not use less than 1.5 % of isoflurane.
2. Set breathing volume to 15 mL kg−1 bodyweight and respira-
tion frequency to 12–18 min−1. Use oxygen concentration
between 20 and 40 % (see Note 7).
3. Set inspiratory pressure to 18 cm H2O.
4. Infuse the Ringer-lactate intravenously with 2–10 mL h−1 kg−1
bodyweight.
5. Use eye ointment to protect eyes from drying.
6. Monitor the peripheral pulse and oxygen saturation during
MCAO and imaging. Further monitor capnography, ECG,
NIBP, IBP, rectal temperature (see Table 1) during MCAO
surgery using the mentioned monitoring systems.
7. Alternatively, maintain anesthesia by infusion method if appli-
cable (see Note 8).
3.3 Surgery 1. Place the animal in a right lateral position.

and Postsurgical Care 2. Ensure deep unconsciousness by testing for the blink reflex
3.3.1 Surgical Approach and ensure anesthesia by repeated moderate pain stimuli (e.g.,
for MCAO use a small needle to stab between the claws of a forelimb).
3. Apply the eye ointment in both eyes. Then attach the ear
straight behind the neck using adhesive tape. Fix the head
thoroughly using adhesive tape.
4. Shave the area between ear and eye and use iodine-containing
solution for thorough disinfection. Wait for at least 3 min
before the next step.
5. Cover the head with the folio drape surgical covers, only expos-
ing the area between ear and eye. Also cover the body to avoid
secondary contamination.
6. Remove an elliptic skin lobe at the temporal region between
lateral eye rim and ear (see Note 9).
7. Carefully expose the superficial temporal artery and the corre-
sponding vein. First occlude the artery electrosurgically using
the bipolar forceps, followed by the vein (see Note 10).
8. Dissect the retroorbital fat directly behind the orbital rim to
fully expose the temporal muscle.
9. Incise the temporal muscle at the temporal line. Importantly,
make sure to leave a small rim of connective tissue at the bone
to readapt the muscle after MCAO. Then carefully elevate the
temporal muscle using a Wullstein retractor (see Fig. 4; see
Note 11). The coronoid process of the mandibula is covered
by the tissue of the temporal muscle (see Note 12), and must
not be exposed. Thus proceed with step 10.
Fig. 4 Detailed scheme of surgical access to the ovine cranium and illustration of relevant steps during MCAO
surgery. (a) Three-dimensional reconstruction of the skull from a CT data set: (1) area for trepanation, (2) the
coronoid process, and (3) bone suture between temporal and parietal skull plate. (b) The approximate position
of superficial temporal artery is indicated by (4). The artery and its accompanying vein are located in the surgi-
cal field and need to be cauterized after elliptic skin resection (5). (c) The coronoid process of the mandibula
(2), being situated within the temporal muscle (6), needs to be elevated with the muscle after cutting the
temporal muscle at the temporal line (7; black line). (d) Detailed visualization of the craniotomy: Due to the
limited space directly behind the orbital rim (8), the coronoid process of the jaw (2) needs to be lifted with the
temporal muscle (lifting direction is indicated by white arrow). The area of trepanation (9, indicated by hemi-
transparent overlay) is limited in the depth by the suture between temporal and parietal skull plates (3). (e)
After extending the trepanation and opening of the dura mater, the MCA can be found in the cranioventral field
of trepanation (10). Place a brain cotton pad (11) on the brain surface to protect it during further
manipulation
10. The temporal muscle is fixed to the bone by connective tissue.

Remove this connection using a 2 mm Williger raspatory. Then
further lift the temporal muscle. Repeat these steps until the
surgical field is wide enough for further proceeding. This is the
case when the bone suture between temporal and parietal skull
plates is clearly visible.
11. Further expose the skull bone surface behind the orbital rim
using the Williger retractor. The area of craniotomy is placed
directly behind the orbital rim, ventral to the temporal line and
dorsal to the bone suture between temporal and parietal skull
plates (see Fig. 4). Ensure elevation of the major part of the tem-
poral muscle. This requires removal of connective tissue fixing
the muscle also in the caudal part (see step 10 of this section).
12. Perform craniotomy with a 6.0 mm Barrel burr at 10,000 rpm
in the specified region (see Figs. 1 and 4). Hold the burr at an
angle of 30°–45° to the bone surface. Use the headlight and
magnification glasses for better control and visualization.
Avoid any opening the retroorbital space (see Note 13).
13. Extend the drill hole to all directions using Kerrsion rongeurs. If
necessary, remove dura connections to the inner side of the skull
bone using the 2 mm Williger raspatory. Make sure not to open
the dura at this stage! In particular, extend frontal part of the
transcranial approach. Remove all bone fragments (see Note 14).
14. Lift the dura with the Adson forceps. Carefully open the dura
mater with a dorsoventral incision using the microscissors.
Lifting is important to avoid accidental damage of underlying
cortical vessels (see Note 15). Carefully widen the dura hole
using the scissors. Collect the cerebrospinal fluid using eye
swab or surgical pads.
15. Place the neurosorb® patties on the brain surface. Use the back
of the blunt 2 mm Williger raspatory in your left hand to slowly
apply very gentle pressure to the brain surface (always covered by
neurosorb® patties). This may help to visualize the MCAO. DO
NOT apply intense pressure and DO NOT move rapidly. Keep
the Williger raspatory in place manually until step 18.
16. Collect cerebrospinal fluid using eye swab or surgical pads
immediately before electrosurgical occluding the MCA. Use
the nonadherent bayonet-shaped neurosurgical bipolar forceps
and apply <50 W. See Note 16 for MCAO.
17. You may occlude the proximal MCA branch for a large territo-
rial stroke or distal MCA branches for smaller stroke lesions.
For details, please refer to [14].
18. Cover the brain by repositioning of the dura mater.
19. It is possible to suture the dura (use 6-0 surgical threats) and/
or close the bone defect with bone cement. However, this will
result in dangerous intra cranial pressure (ICP) peaks due to
the concomitant brain edema after larger strokes in the sub-
acute phase following MCAO. Thus, just reposition the dura
in case of a large territorial stroke. Step 20 is sufficient to pre-
vent any damage to the area.
20. Relocate the temporal muscle, now covering the drill hole, and
fix it to connective tissue that was left at the muscle insertion at
the temporal line. Use 2-0 resorbable filament. A Kirschner
suture with 2-0 filament should be performed for readapting
of the subcutis, followed by a Reverdin suture to close the skin
wound.
3.3.2 Postsurgical Care 1. Perform antibiotic and analgesic treatment by intramuscular injec-
tion of enrofloxacin and flunixin-meglumine before surgery and
for at least 5 days (optimum: 7 days) postsurgical (see Table 3).
2. Intramuscular injections of buprenorphine (see Table 3) are
performed for additional analgesic treatment for 48 h follow-
ing MCAO. Repeat injection every 8 h (see Note 17).
3. Perform wound treatment with disinfection wound spray fol-
lowed by covering with silver or aluminum wound spray.
4. Protect venous lines by covering with cotton, followed by a
bandage. The bandage must then be covered with isolator tape
to avoid damage induced by the subject or flock mates.
3.4 Imaging 1. Wrap the animal in folio drape to protect it from cooling down.
Procedures The drape also prevents soiling of the scanner.
3.4.1 General Procedure 2. Place the subject in a prone position upon the scanner table
with the head resting on the neck crest. The nose faces into the
scanner bore (see Fig. 5).
3. Fix the head and the body of the animal to the table using
adhesive tape to avoid any movement artifacts. Be sure not to
hinder breathing movements.
3.4.2 MR Imaging MRI is used to monitor the impact of MCAO, to visualize the
lesion and to control its development by repeated measurement
during a longer observation period. All means of modern MR
imaging [20, 21] can be applied to the sheep model (see Table 4).
While a 1.5 T is sufficient to monitor the lesion with basic param-
eters like T2 TSE, DWI, perfusion weighted imaging (PWI), or
time of flight (TOF)-MRA, application of 3.0 T MRI provides fur-
ther options including diffusion tensor imaging (DTI) for fiber
track reconstruction [22].
Fig. 5 Animal preparation and positioning for MRI session. (a) Animal on table of a clinical scanner in prone
position with head facing into the scanner: The animal is not yet covered with folio drape. (b) Three-dimensional
CT dataset reconstruction of animal before imaging: (1) Intravenous perfusion as described in Note 8. The use
of a (2) neck-rest fixates the head. To prevent movement artifacts during imaging, the head should further be
fixed by friction tape. Intubation (3) is recommended during the scanning procedure. (c) Illustration of
SenseFlexM coil placement during MRI procedure: (4) Scheme of head coil SenseFlexM placement in a CT-MRI
reconstruction corresponding to the position of the brain (5)
Table 4
Recommended imaging protocol for 3 T MRI Scanner
Sequence Parameters
T2 TSE Voxel size: 0.5 × 0.4 × 2.0 mm; slices: 50; TR: 6000; TE: 105; acquisition time: 8:26 min
T2* Voxel size: 0.8 × 0.7 × 3.0 mm; slices: 35; TR: 700; TE: 20; acquisition time: 5:24 min
TOF MRA Voxel size: 0.5 × 0.3 × 0.5 mm; slices: 40; TR: 24; TE: 4.43; acquisition time: 13:50 min
DTI Voxel size: 1.9 × 1.9 × 1.9 mm; slices: 70; TR: 10600; TE: 100; acquisition time:
12:02 min; diffusion directions: 64
PWI Voxel size: 1.6 × 1.6 × 5 mm; slices: 13; TR: 1450; TE: 45; measurements: 50;
acquisition time: 1:18 min
T1 Voxel size: 0.6 × 0.6 × 1.0 mm; slices: 160; TR: 1900; TE: 2.83; acquisition time:
MPRAGE 7:58 min
Planning the sequences includes three short T2 HASTE sequences, performed in addition to the scout imaging
TSE turbo spin echo, TOF MRA time of flight magnetic resonance angiography, DTI diffusion tensor imaging, PWI perfu-
sion weighted imaging, MPRAGE magnetization prepared rapid gradient echo, TR time of repetition, TE time of echo
Preparation for MRI 1. Fix a MRI-positive marker laterally at the head to ensure ade-
Acquisition quate spatial orientation after imaging (see Note 18).
2. Put the coil around the head and fix it using adhesive tape
(see Fig. 5 for positioning).
3. Focus the scanner to the sheep brain.
Imaging Process 1. Imaging starts with a scout image plus three additional T2
Half-Fourier Acquisition Single-shot Turbo Spin Echo
(HASTE) sequences in three different planes, which facilitate
exact planning of the scans. Always plan scans in orthogonal
direction (corresponds to the axis of the brain and brain stem).
2. For basic monitoring, perform scanning sequences as given in
Table 4 (see Note 19).
3. Proceed with MRA to identify the occlusion of the MCA
(postsurgical control, compare to Fig. 3).
3.4.3 PET Imaging 1. Measure tissue attenuation using three rotating 68Ge rod
sources previous to tracer injection (transmission scan, scan
General Remarks
time: 10 min).
2. Place a NeuroShield® in the neck region of the sheep for brain
imaging to minimize scatter radiation.
3. Extend the venous line to the animal with additional tubing.
Pay attention for the tubing to be placed outside of the field of
view during tracer administration.
4. Use [15O]H2O for CBF measurement. Inject a dose of
~1000 MBq per sheep and scan. The injection system (see
Sect. 2.4.3) realizes tracer administration automatically (over
~30 s).
Table 5
Exemplary blood sampling protocols for PET analysis
[15O]H2O
Sample no. 1 2 3 4 5 6
Time p.i. (min) 0:00–2:00a 2:50 3:00 3:50 4:00 5:00
[18F]FDG
Sample no. 1 2 3 4 5 6
Time p.i. (min) 0:00–2:45 a
3:00 4:00 5:00 7:00 10:00
Sample no. 7 8 9 10 11 12
Time p.i. (min) 15:00 20:00 30:00 40:00 50:00 60:00
no. number, p.i. post injection

a
Time period of continuous blood sampling
5. Use [18F]FDG to measure cerebral metabolic rate for glucose

(CMRGlu). Inject a dose of ~370 MBq per sheep and scan.
After i.v.-injection (over approximately 90 s), the perfusion
line is washed with 0.9 % sterile sodium chloride solution.
6. Link every PET scan with arterial blood sampling (see Table 5),
which is needed for further kinetic modeling, especially if abso-
lute CBF quantification is desired. If conducted manually, arte-
rial blood samples should be taken in meaningful intervals with
regard to individual tracer metabolism (see Note 20).
Basic Information on PET 1. Perform dynamic emission scans according to the following
Data Processing parameters: axial field of view: 155 mm; number of parallel
transverse slices: 63; slice thickness: 2.4 mm; image resolution:
7.1 mm (transverse), 6.7 mm (axial); matrix: 128 × 128; acqui-
sition mode: 3D; acquisition time: 60:00 min for [18F]FDG
and 5:00 min for [15O]H2O.
2. PET data obtained must initially undergo standard correction
for radioactive decay, death time, scatter, and attenuation.
Images are finally reconstructed by means of iterative Ordered
Subset Expectation Maximisation (OSEM) algorithm (for
instance 10 iterations, 16 subsets) (see Note 21).
3.4.4 Imaging Data Relevant examples of frequently used imaging data analysis are
Interpretation and Analysis given in this section.
MR-Based Analyses A territorial infarct results in cerebral edema. Therefore, calcula-

of Hemispherical Atrophy tion of the lesion should always be performed corresponding to
the remaining brain tissue [23]. Numerous software-based meth-
ods have been developed for automated or semiautomated calcula-
tion. An approach using the open source software ImageJ is feasible
for many researchers, and can be used even in case professional

(and expansive) MR image processing and analyzing software is
not available (for alternatives see Note 22). The work flow is as
follows:
1. Open all images of a certain MRI sequence (e.g., T2 TSE)
using stack image.
2. Normalize gray-scale values by automatic histogram function.
3. Use threshold function to separate different regions step-by-
step. (1) Left and right hemispheres including ventricles and
infarction, (2) the infarct area, and (3) left and right ventricle.
Use always a native stack for each structure (e.g., left ventri-
cle). Save all resulting stacks separately in binary file-mode and
name them for a randomized code (see Note 23).
4. Use the “wand-tracing-tools” to select the separated structure
in the binary file and transfer it to the region of interest (ROI)-
manager (see Note 24).
5. Measure ROI size and calculate the volume of a structure by
the formula:
Ai −1 + Ai + Ai −1 × Ai n
Vpart ,i = × di −1,i and Vtotal = ∑Vpart ,i
3 i =1
(n—number of partial volume; Vpart,i—partial volume; Vtotal—total
volume; Ai—Area of specific ROI measurement; di−1,i—dis-
tance between two slices)
6. Calculate hemispherical atrophy using the following formula:
Hemispherical atrophy = (Vleft hemisphere − Vinfarction − Vleft )/(Vright
ventricle
hemisphere − Vright ventricle)
(V—Volume)
Postprocessing of DTI A fast and practicable way for Fractional Anisotropy (FA) and
Sequences Apparent Diffusion Coefficient (ADC) analysis with DTI studio
(Version 3.0.2) is as follows (see Fig. 6):
1. Open DTI Studio and select “File” > “DTI Mapping.”
2. Specify your data according to your scanner (i.e., Siemens
Mosaic) > “Continue.”
3. Check slice orientation, slice sequencing, slices to be processed
(we use “all slices”).
4. Specify the b-values according to your data (important for cor-
rect ADC calculation).
5. Select the folder with your DICOM data by choosing “Add a
Fold.”
6. Click “Get gradient from DICOM file header” followed by
“OK.”
Fig. 6 MRI of the ovine brain following stroke. Images (a–c) were obtained 24 h after MCAO, whereas (d and
e) represent scans conducted 6 weeks following experimental stroke. All images were obtained at 3 T MR. (a)
Fiber tracking after acute MCAO in sheep: Fiber reconstruction based on Diffusion Tensor Imaging (DTI) fused
with an anatomical 3D T1 and Diffusion Weighted Imaging (DWI). High signal intensities in DTI show a higher
anisotropy of diffusion, indicating a higher density and homogeneity of fiber tracts. DTI signal loss in the region
of a high DWI signal occurs due to post ischemic tissue destruction. (b) Apparent diffusion coefficient (ADC)
map of DWI in acute stage of the ovine infarction: The typical dark signal indicating impaired diffusion can be
seen at the site of the infarction. (c) DWI of acute MCAO: The circumscribed, increased signal intensity at the
altered hemisphere indicates reduced diffusion. (d) ADC map of a DTI sequence in chronic stage of stroke:
Higher signal intensity indicates a higher diffusivity. The tissue defect in the impaired hemisphere is filled with
cerebrospinal fluid. Thus, the signal is equal to the signal within the ventricles. (e) DWI of a chronic infarct:
Compare the signal characteristics with (c) to note the differences of the imaging signs at both stages
7. In the next window click on the “DTiMap”-tab and generate

maps of FA (Tensor, Color Map, etc.) and ADC (ADC-Map)
using the defaults.
8. Change to the tab “Image” and choose the map to be
analyzed.
9. Change to the tab “ROI” and draw the ROI in the regions to
be analyzed.
Perform tractography (see Fig. 6) using Siemens Syngo MR B15
(Vendor Specific) as follows:
1. Load DTI data and 3D MPRAGE into “Neuro 3D”
application.
2. Click “fusion” followed by double click on “3D.”
3. Draw a ROI into the region where fiber tracking starts by
simultaneously pressing “Control + left mouse button.”
4. Select the ROI and press the right mouse button. Choose
“Start Tractography.”
PET-Based Analysis [15O]H2O PET is the gold standard method for determining CBF. In
of Brain Perfusion case of ischemic stroke, follow-up examinations allow for the deter-
mination of different infarct stages (see Fig. 7). It is also of use to
investigate acute infarct evolution. Because of its short half-life time
(approximately 2 min for 15O), PET imaging with [15O]H2O can be
performed in a serial scanning mode. Note that follow- up PET
should not be started earlier than 20 min (10 half-life times of the
tracer) after the previous tracer application. Data processing is per-
formed as follows:
1. For a semiquanitative approach, sum up image frames with
early cerebral tracer uptake up to a total time of 1 min (start
with frame of first observable tracer activity in brain tissue).
The interpretation of these data should be preferentially based
on standardized uptake values (SUVs). Normalize images for
injected activity (ID) and bodyweight as follows: SUV = ROI
activity [Bq mL−1] × bodyweight [g] × injected dose (ID)
[Bq]−1.
2. To obtain absolute CBF values (units: mL min−1 100 g−1), use
dynamic PET data in conjunction with individually derived arte-
rial input function (see Note 25) and quantify CBF by applying
the method described by Alpert et al. [24] (see Note 26).
3. As most therapeutically interventions aim to preserve and/or
rescue the ischemic penumbra, the visualization of that “tissue
at risk” is of special interest. CBF quantification offers the
opportunity to separate stroke-related tissue regions by apply-
ing established CBF thresholds [25] <8 mL 100 g−1 min−1 for
infarction core, 8–22 mL 100 g−1 min−1 for the ischemic pen-
umbra and >22 mL 100 g−1 min−1 for normal brain tissue (see
Fig. 7, see Note 27).
PET-Based Assessment Especially for long-term studies of brain vitality, indicated by glu-
of Cerebral Glucose cose metabolization, it is useful to perform [18F]FDG-PET imag-
Consumption ing. Data processing is performed as follows:
1. Create a summed image from the late frames of the PET scan
(frames should cover the last 30 min of acquisition time),
which can be used for semiquantitative analysis approaches.
2. Perform the SUV quantification for subsequent qualitative and
quantitative assessment of glucose consumption: This method
is easy to implement in ovine FDG-PET image analysis with-
out the need of arterial blood sampling.
3. For that purpose, convert the prepared single-frame image
(the sum-image from late frames) to SUV data by normalizing
data for bodyweight and injected activity (see semiquantitative
[15O]H2O-PET analysis in Sect. “PET-Based Analysis of Brain
Perfusion”).
Fig. 7 [15O]H2O PET of the ovine brain after MCAO. (a) Parametric CBF map, derived from [15O]H2O PET 2 h after
MCAO. Left hemispheric infarct resulted in different perfusion-based stroke compartments: infarct core (white
line, CBF <8 mL min−1 100 g−1), ischemic penumbra (8–22 mL min−1 100 g−1), and normal brain tissue
(>22 mL min−1 100 g−1). Outer delimination of brain tissue was done by means of individually superimposed
MRI images (not shown). (b–g) show results from a long-term ovine stroke study including follow-up [15O]
H2O-PET imaging. Representative coronal brain slices of [15O]H2O PET at day 1 (b), 14 (c), and 42 (d) correlated
well with findings obtained with MR imaging using T2 TSE at day 1 (e) and 14 (f), and FLAIR on day 42 (g),
respectively. The white arrows mark the area of perfusion deficit in (b–d) and corresponding MR findings in
(e–g). The value bar of PET images represent 0 (bottom) to 120 (top) mL min−1 100 g−1
4. Outline the stroke-affected volume of interest (VOI) with

decreased glucose consumption and low SUV units with
PMOD software, using automated threshold function. Take
the unaffected hemisphere as a reference region or mirror the
selected VOI to the contralateral side. With a calculation of
SUV ratios (SUVR = SUVstroke VOI/SUVreference VOI), regional def-
icits of CMRGlu can be quantified (see Note 28).
3.5 Sacrifice 1. Sacrifice subjects in deep anesthesia (see Sect. 3.2) by intrave-

and Post Mortem nous injection of pentobarbital (see Table 3). Control the
Analyses absence of cardiac action, breathing, and reflexes over a period
of at least 2 min.
2. Turn the animal on the back and decapitate at the atlanto-
occipital junction using a sharp and robust knife. Dispose the
trunk (or use it for any kind of other investigation) and pro-
ceed with head preparation.
3. Carefully expose both external carotid arteries, place steel can-
nulas in both carotids and fix them with stout thread.
4. Perfuse the head manually via cannulas with 3 L PBS using
100 or 500 mL syringes (see Note 29). In the case vitality
staining is requested see Note 30.
5. Connect the roller pump system to the steel cannulas and per-
fuse the head with 20 L 4 % PFA (rate: 15 mL min−1).
6. Remove the skin and muscles from the cranial roof and the
neck using a sharp knife.
7. Remove the upper part of the skull with an oscillating saw.
Avoid damaging the dura mater and the brain.
8. Incise the dura mater and carefully expose the brain.
9. Perform immersion fixation of the entire head with 4 % PFA
for at least 48 h.
10. Carefully remove the brain from the cranial cavity and perform
further immersion fixation in 4 % PFA for 3 additional days.
11. Cut the brain consecutively into coronal slices of 4 mm thick-
ness. Photograph and number the specimens together with a
scale. Measure ventricle, tissue, and infarct of all photographed
specimens (see Note 31).
12. Compartmentalize the specimens for further embedding pro-
cedure (see Fig. 8). Paraffin embedding after 18 h dehydration
of the specimens is recommended for conventional histology,
as well as for immunohistochemistry using antibodies for par-
affin embedded material (e.g., rabbit-anti-GFAP, code: Z0344,
Dako Cytomation AG, Germany, or mouse-anti-NSE, code:
M 0873, Dako Cytomation AG) (see Note 32). Specimens are
stored in 30 % sucrose solutions until further processing.
Cryoprotection is recommended when labeling with
fluorescence-based techniques will be performed.
Fig. 8 Gross brain pathology and examples for histological evaluations. Pathologic-anatomical (a–c) and his-
tological findings (d–f) following MCAO-caused infarction (white arrows) in sheep: (a) Ischemic brain tissue 6 h
after MCAO: The infarct area in the coronal brain slice (at the level of the optic chiasm) is indicated by absent
TTC staining. (b) A sharp demarcation of the infarct area can be seen macroscopically 6 weeks after
MCAO. Atrophy in the impaired hemisphere is clearly associated with an enlargement of the lateral ventricle.
For further analysis, a compartmentalization of specimen (gray lines) is recommended. (c) Macroscopic find-
ings were confirmed 26 weeks after MCAO. Histological findings in the area next to the infarct (box insert in b)
can be described as follows: (d) 6 week after MCAO, Nissl staining reveals alterations of neurons (white arrow
head) and neuropil. (e) Labeling with an antibody against neuron specific enolase (monoclonal, mouse-anti-
NSE) shows axonal alterations (white circle) in the white matter next to the impaired area (white arrows in d–f).
(f) Astrogliosis (white edged arrow head) is indicated by increased glial fibrillary acid protein (polyclonal, rab-
bit-anti-GFAP) in the cortex next to the infarct
4 Notes
4.1 Introduction 1. Recently, a transient MCAO model using sheep was developed
by an Australian group [26]. This model uses an aneurysm clip
to induce a 2-h MCA blockage resulting in lesion extension
and morphology being comparable to those seen in the
described approach. The model was also reported to be highly
reproducible and is also expected to be associated with a low
incidence of intraoperative complications.
2. In addition to conventional surgical techniques, stereotaxic
interventions become popular. The sheep model can also be
adapted to hemorrhage induction by application of autologous
blood using the Brainsight™ stereonavigation and stereotaxic
system (Rogue Research Inc., Canada). The system can further
be used to administer therapeutic compounds or cells stereo-
taxically [27].
4.2 Topography 3. Functional organization of the neocortex is described else-

where [28–30] (Please also see Fig. 2).
4.3 Anesthesia 4. Considering the venous drainage system an interspecies com-

parison between sheep, dogs, and rats suggests that the cere-
bral venous angioarchitecture in large animals is better
comparable with the human anatomy although substantial dif-
ferences remain [31].
5. Detomidine (0.02–0.06 mg kg−1 i.v., Domosedan®, Orion
Corporation, Finland) can be used instead of xylazine. Note the
high receptor affinity of alpha-2-agonists for small ruminants.
6. Choose a 16G butterfly cannula (Braun Melsungen AG) to
access the V. saphena lateralis distal to the tarsus in case an
additional venous line is needed, especially during surgery.
7. Never apply carrier gas (such as nitrous oxide) for any reason
as the gas will accumulate in the rumen.
8. Alternatively, intravenous infusion during MR imaging can be
performed in case a MR-compatible respirator with isoflurane
vapor is not available. For this, fill one 50 mL syringe (Original-
Perfusor® syringe, Braun Melsungen AG) with propofol.
Prepare another one with 3 mL 10 % ketamine, 3 mL mid-
azolam ad 30 mL 0.9 % sterile sodium (see Table 3). Place the
syringes in the perfusion pumps (e.g., Perfusomat® compact S,
Braun Melsungen AG) and connect the demanded length of
perfusion lines (Braun Melsungen AG) with the 3-way-stop-
cock (Discofix®, Braun Melsungen AG) upon the vein cannula.
Apply the propofol at 0.2 mL h−1 kg−1 BW and the ketamine-
midazolam mix at 0.26 mL h−1 kg−1 BW. If necessary, first
increase infusion speed of the ketamine-midazolam mix step-
wise, then that of the propofol. Ensure continuous monitoring
of adequate breathing!
4.4 Surgery 9. The elliptic wound helps to avoid post surgical complications.
Readapting the skin will result in a tight suture that prevents
seroma formation.
10. The tissue which covers the region of the transcranial access is
supplied by the Nervus auriculopalpebralis (originates from
the facial nerve). The nerve can be cut.
11. Dissect carefully! The artery and veins supplying the temporal
muscle should not be damaged during the procedure.
12. Dissect carefully! The coronoid process of the jaw (see Fig. 4)
must not be exposed from the covering muscle.
13. Avoid incision of the retroorbital space, as a massive bleeding
can result from the venous sinuses situated in the retroorbital
space.
14. The thickness of the dorsocaudal part of the Os parietale is only

2–3 mm while the ventrofrontal part is about 5–10 mm (Fig. 4).
First, carefully ablate the bone stepwise in the area of craniot-
omy by circling the burr without applying much pressure. Use
a 4 mm Rosen burr to widen the area. Finally, only a very thin
bone lamina is remaining. Now open the cavity in the middle of
the ablation by using forceps or by very gentle drilling.
In the case bone bleeding occurs: Stop the bleeding using bone wax
before opening the dura.
15. Due to the relative small space for the surgical access, Adson
forceps are sometimes not feasible for dura lifting (not enough
space to pick the dura). In such cases, the dura can be lifted
alternatively by a dura hook. The ovine dura has multiple “lay-
ers,” so the hook can be placed safely in an upper layer for lift-
ing the entire dura without uncontrolled incision and/or
alterations of the underlying brain tissue and vessels (use sharp
and angled Fisch dura retractor, length 185 mm, catalog-no:
FD376R, Braun Melsungen AG for that purpose, see http://
www.surgical-instruments.info/en/products.html). Place the
hook at the dura in an angle of <10° in relation to the dural
surface. After penetrating of an upper dura layer with the
retractor tip, lift the hook about 5 mm and do not move it
anymore! Cut the dura in the described manner. Make sure to
lift the flap all the time while cutting the dura.
16. In a few cases of own experiments, a noticeable variation of
MCA architecture (doubling at the origin of the CW, preferen-
tially in female subjects) necessitated occlusion of both MCA
branches to induce a stroke.
17. After occlusion of the MCA main branch, a larger stroke and
more severe brain edema may occur in the first days following
stroke. An open burr hole reduces the intracranial pressure,
therefore, lethal herniation is a very rare complication. In the
case of a reduced state of activity and depressed motor func-
tions (usually to be seen between day 2 and 5 post MCAO),
provide stimulation of ruminal/gastrical digestion with propri-
onic acid (oral), butafosfan (intramuscular), menbuton (intra-
muscular), and Amynin® (intravenous infusion) (see Table 3).
4.5 MR Imaging 18. The use of small capsules of glycerol nitrate is recommended
for this purpose.
19. Due to the size of the sheep brain, different sequences should
be used to describe the infarct status or other brain alterations
(e.g., DWI, PWI, T2 TSE) [32].
4.6 PET Imaging 20. Alternatively, an automated sampling device can be used.
Latter equipment (blood sampler) is highly commended for
short acquisition times, as in the case of [15O]H2O-PET (scan

time: 5 min).
In contrast to small animal models, the ovine blood volume
is similar to that of humans (approximately 60–70 mL kg−1
bodyweight), allowing blood sampling (see Table 5) without
affecting circulation.
21. Other methods, like filtered backprojection algorithms result
in lower spatial resolution and should therefore be replaced by
iterative reconstruction algorithms.
4.7 Imaging Data 22. Alternatively, freeware (OsiriX, Pixmeo, Geneva, Switzerland)
Interpretation can be used on another computer. The scans have to be
and Analysis exported from the scanner either via burning them to a CD-
ROM or sending them to a picture archiving system, depend-
ing on local configurations [33].
23. Using the OpenSource software ImageJ offers several plugins
for MRI analyses (e.g., Quickvol 2) [34]. Alternatively, OsiriX
can be used.
24. ImageJ names the created ROI automatically. The ROI name
identifies region of interest in a specific slice. Manually changes
of the ROI name disturb the defined allocation!
25. Various methods for CBF quantification not requiring arterial
blood sampling have been published. To avoid the need for
arterial input function data, most alternative methods deter-
mine several parameters for kinetic modeling, such as the parti-
tion coefficient (Vd) [35]. However, due to the existence of
heterogenic infarct compartments (Vd differs relevantly in
infarct core, penumbra and tissue of benign oligemia), these
alternative approaches may not give reliable results in acute
stroke imaging.
26. Use the PMOD tool “PKIN” for that purpose. The software
corrects individual arterial input functions for delay and disper-
sion and creates parametric image data for CBF (voxel val-
ues = mL 100 g−1 min−1) and distribution volume (Vd; [mL g−1]).
These three-dimensional parametric images can be processed
further (VOI based, voxel based, or other approaches).
27. Use automated threshold function of PMOD for an operator-
independent VOI determination (see Fig. 7).
28. As FDG is intensively taken up into the brain, local deficits
show a high target-to-background image contrast. Considering
the fact of increased glucose consumption is a typical indicator
of inflammatory processes as well as the knowledge about post
stroke luxury perfusion and inflammation, it is recommended
to interpret FDG-PET images from acute infarct stages with
special care. For long-term stroke studies, FDG-PET offers
reliable data on neuronal integrity, i.e., brain tissue viability.
4.8 Post Mortem 29. Perfusion should be done simultaneously via both cannulas.
Specimen Processing 30. Immediate vitality staining using triphenyltetrazolium chloride
(TTC) of acute cerebral infarction can be performed alterna-
tively by incubation of 4 mm brain specimens in 1 % TTC
(diluted in PBS) at 37 °C for 1.5 h (see Fig. 8). This requires
removal of the brain immediately after perfusion. Do not apply
an additional fixation period. Remove the brain very carefully!
31. Gross pathology: Acute infarctions (up to 10 h following
MCAO) can hardly be discriminated from the surrounding tis-
sue without any staining procedures. Therefore, TTC staining
can be used to visualize ischemic altered gray and white matter
(see Fig. 8).
32. Histological findings 6 weeks after MCAO are comparable to
those of human [36] and nonhuman primate species [37].
Common histological staining procedures as well as immuno-
histochemistry and fluorescent techniques using a portfolio of
antibodies against neurons (e.g., NSE), astrocytes (e.g.,
GFAP), microglia (e.g., CD11b, IBA1), vessels (e.g., Glut-1),
and collagen can be performed (see Fig. 8).
Acknowledgments
The authors want to thank Dr. Karl-Titus Hoffmann, professor of

neuroradiology, and Dr. Osama Sabri, professor of nuclear medi-
cine, for the allowance to use the scanners in their departments at
Leipzig University. The authors are further grateful to Dr. Uwe
Gille, Dr. Johannes Seeger, and Dr. Heinz-Adolf Schoon, profes-
sors at the Faculty for Veterinary Medicine at Leipzig University.
References
1. O’Donnell MJ, Xavier D, Liu L et al (2010) 6. STAIR-group (1999) Recommendations for
Risk factors for ischaemic and intracerebral standards regarding preclinical neuroprotective
haemorrhagic stroke in 22 countries (the and restorative drug development. Stroke
INTERSTROKE study): a case-control study. 30:2752–2758
Lancet 376:112–123 7. Savitz SI, Chopp M, Deans R et al (2011)
2. Hacke W, Kaste M, Bluhmki E et al (2008) Stem cell therapy as an emerging paradigm for
Thrombolysis with alteplase 3 to 4.5 hours stroke (STEPS) II. Stroke 42:825–829
after acute ischemic stroke. N Engl J Med 8. Amiridze N, Gullapalli R, Hoffman G et al
359:1317–1329 (2009) Experimental model of brainstem
3. Wahlgren N, Ahmed N, Davalos A et al (2008) stroke in rabbits via endovascular occlusion of
Thrombolysis with alteplase 3–4.5 h after acute the basilar artery. J Stroke Cerebrovasc Dis
ischaemic stroke (SITS-ISTR): an observa- 18:281–287
tional study. Lancet 372:1303–1309 9. Kang BT, Lee JH, Jung DI et al (2007) Canine
4. Hachinski V, Donnan GA, Gorelick PB et al model of ischemic stroke with permanent mid-
(2010) Stroke: working toward a prioritized dle cerebral artery occlusion: clinical and histo-
world agenda. Cerebrovasc Dis 30:127–147 pathological findings. J Vet Sci 8:369–376
5. Traystman RJ (2003) Animal models of focal 10. Garcia JH, Kalimo H, Kamijyo Y et al (1977)
and global cerebral ischemia. ILAR J 44: Cellular events during partial cerebral ischemia.
85–95 I. Electron microscopy of feline cerebral cortex
after middle-cerebral-artery occlusion. 25. Baron JC (2001) Perfusion thresholds in

Virchows Arch B Cell Pathol 25:191–206 human cerebral ischemia: historical perspective
11. Imai H, Konno K, Nakamura M et al (2006) A and therapeutic implications. Cerebrovasc Dis
new model of focal cerebral ischemia in the 1:2–8
miniature pig. J Neurosurg 104:123–132 26. Dreyer A, Stroh A, Nitzsche B et al (2012)
12. Marshall JW, Ridley RM (2003) Assessment of Frameless stereotaxy in sheep—neurosurgical
cognitive and motor deficits in a marmoset and imaging techniques for translational stroke
model of stroke. ILAR J 44:153–160 research. INTECH Open Access Publisher,
13. Shuaib A, Lees KR, Lyden P et al (2007) NXY- London
059 for the treatment of acute ischemic stroke. 27. Wells AJ, Vink R, Blumbergs PC, Brophy BP,
N Engl J Med 357:562–571 Helps SC, Knox SJ, Turner RJ (2012) A surgi-
14. Boltze J, Forschler A, Nitzsche B et al (2008) cal model of permanent and transient middle
Permanent middle cerebral artery occlusion in cerebral artery stroke in the sheep. PLoS One
sheep: a novel large animal model of focal cere- 7(7):e42157
bral ischemia. J Cereb Blood Flow Metab 28. Cooley KR, Vanderwolf CH (2004) The sheep
28:1951–1964 brain—a photographic series. A.J. Kirby Co,
15. Behrens H, Ganter M, Hiepe T (2009) London, ON
[Textbook of ovine diseases]. Parey, Stuttgart, 29. Mitchell JF (1958) The characteristics of some
Germany, Chapter in German points of cardiovascular and respiratory repre-
16. Nitzsche B, Frey S, Collins L et al (2015) A sentation in the cerebral cortex of sheep.
stereotaxic, population-averaged T1w ovine J Physiol 144:17–8P
brain atlas including cerebral morphology and 30. Gierthmuehlen M, Wang X, Gkogkidis A et al
tissue volumes. Front Neuroanat 9:69. (2014) Mapping of sheep sensory cortex with a
doi:10.3389/fnana.2015.00069 novel micro-electrocorticography grid. J Comp
17. Salomon FV, Geyer H, Gille U (2008) Neurol 522(16):3590–3608
[Textbook of anatomy for veterinary medi- 31. Hoffmann A, Stoffel MH, Nitzsche B et al
cine]. Enke Verlag, Leipzig, Germany, Chapter (2014) The ovine cerebral venous system:
in German comparative anatomy, visualization, and impli-
18. Forschler A, Boltze J, Waldmin D et al (2007) cations for translational research. PLoS One
[MRI of experimental focal cerebral ischemia 9(4):e92990
in sheep]. Rofo 179:516–524, Article in 32. Latchaw RE (2004) Cerebral perfusion imag-
German ing in acute stroke. J Vasc Interv Radiol
19. Ashwini CA, Shubha A, Jayanthi KS (2008) 15:29–46
Comparative anatomy of the circle of Willis in 33. Rosset A, Spadola L, Ratib O (2004) OsiriX:
man, cow, sheep, goat, and pig. Neuroanatomy an open-source software for navigating in mul-
2008:54–65 tidimensional DICOM images. J Digit Imaging
20. Muir KW, Santosh C (2005) Imaging of acute 17:205–216
stroke and transient ischaemic attack. J Neurol 34. Schmidt KF, Ziu M, Schmidt NO et al (2004)
Neurosurg Psychiatry 76:19–28 Volume reconstruction techniques improve the
21. Wechsler LR (2011) Imaging evaluation of correlation between histological and in vivo
acute ischemic stroke. Stroke 42:12–15 tumor volume measurements in mouse models
of human gliomas. J Neurooncol 68:207–215
22. Nucifora PG, Verma R, Lee SK et al (2007)
Diffusion-tensor MR imaging and tractogra- 35. Watabe H, Itoh M, Cunningham V et al (1996)
phy: exploring brain microstructure and con- Noninvasive quantification of rCBF using posi-
nectivity. Radiology 245:367–384 tron emission tomography. J Cereb Blood
Flow Metab 16:311–319
23. Swanson RA, Morton MT, Tsao-Wu G et al
(1990) A semiautomated method for measur- 36. Mena H, Cadavid D, Rushing EJ (2004)
ing brain infarct volume. J Cereb Blood Flow Human cerebral infarct: a proposed histopath-
Metab 10:290–293 ologic classification based on 137 cases. Acta
Neuropathol 108:524–530
24. Alpert NM, Eriksson L, Chang JY et al (1984)
Strategy for the measurement of regional cere- 37. Garcia JH, Kamijyo Y (1974) Cerebral infarc-
bral blood flow using short-lived tracers and tion. Evolution of histopathological changes
emission tomography. J Cereb Blood Flow after occlusion of a middle cerebral artery in pri-
Metab 4:28–34 mates. J Neuropathol Exp Neurol 33:408–421
Chapter 16
A Nonhuman Primate Model of Delayed Cerebral

Vasospasm After Aneurismal Subarachnoid Hemorrhage
Ryszard M. Pluta, John Bacher, Boris Skopets, and Victoria Hoffmann
Abstract
Animal modeling of human disease has a long history but remains controversial especially when using a
sensitive species. Despite these controversies, in a stroke research, a nonprimate model has been recog-
nized as a most successful and useful for developing new treatments. Among stroke models, a nonhuman
primate model of delayed cerebral vasospasm after subarachnoid blood clot placement has a very unique
position as it revealed important pathomechanisms and led to development of several crucial clinical trials.
In 1989, we adopted this model to study pathophysiology and develop a treatment against delayed cere-
bral vasospasm after intracranial aneurysm rupture (aSAH). In this chapter, we presented detailed descrip-
tions of the animal treatment according to the National Institutes of Health guidance, techniques of
anesthesia, cerebral arteriography, suboccipital puncture for cerebrospinal fluid collection, a surgical clot
placement along the middle cerebral artery as well as postoperative care, euthanasia, and autopsy. Moreover,
to accommodate the recent clinical findings, strongly suggesting a limited role of delayed cerebral vaso-
spasm on the outcome of aSAH, we proposed a modification of the model, which addressed some mecha-
nisms of ultra and early damage to the brain evoked by an intracranial aneurysm rupture.
Key words Animal modeling, Human disease, Stroke research, Nonprimate model
1 Introduction
1.1 Aneurismal While securely located in the skull, brain is additionally protected
Subarachnoid by three layers of connective tissue of different thicknesses known
Hemorrhage (aSAH) as a dura matter, arachnoid, and pia; the latter being the closest to
and Delayed Cerebral the brain surface. Under normal conditions, there is no space
Vasospasm: between the bone and dura matter and only a hairline thick space
Challenges between the dura and arachnoid membrane. The space between
and Solutions the arachnoid and pia known as subarachnoid space consists of sev-
eral cisterns, which in man contains about 140 ml of cerebrospinal
spinal fluid (CSF) produced by the intraventricular choroid plexus
at about 0.5 ml/min. Detailed surgical anatomy of these cisterns
has been described in a seminal paper by Professor Gazhi Yasargil
[1]. The CSF plays several physiological and pathophysiological
227
228 Ryszard M. Pluta et al.
roles by providing mechanical support for the brain, an exchange

pathway for brain metabolites, as well as providing a space for cere-
bral arteries and veins running on the surface of brain. Because of
its presence, pressure, and pulsatile flow, the CSF influences cere-
bral blood flow, facilitates metabolite exchange between blood and
brain, and when obtained via lumbar puncture or ventricular drain-
age it serves as useful diagnostic tool of diseases affecting the brain.
Furthermore, it provides a route for drug administration but can
also facilitate spreading of infection and some tumors [2–23].
Arteries supplying blood to brain enter the skull and pierce the
dura matter to access the subarchnoid space and the cisterns; ver-
tebral arteries enter the cisterna magna and internal carotids enter
the basal cisterns. In those cisterns, both vertebral arteries give off
extracranial, intracranial, and spinal branches before coalescing
into the basal artery. Internal carotids after sending the posterior
communicating artery along the edge of tentorium to the poste-
rior cerebral artery bifurcate and become the anterior and middle
cerebral arteries. All these cerebral vessels run in the subarachnoid
space and cisterns on the base of brain. In some people, an intra-
cranial aneurysm develops at the bifurcation of these conductive
vessels. Most often, the first symptom of such an aneurysm is the
worse ever experienced headache when the aneurysm ruptures in
about 2–10 of 100,000 people [24, 25]. Such a rupture produces
aneurismal subarachnoid hemorrhage (aSAH). At the beginning of
aSAH, violently flowing arterial blood, which is highly oxygenated
and under high pressure, rapidly fills the subarachnoid cisterns.
Rapidly clotting blood displaces cerebrospinal fluid and increases
intracranial pressure until it almost equals the systemic blood pres-
sure, which effectively stops the blood outflow from the aneurysm
but at the same time ceases blood flow in the artery harboring the
aneurysm and adjacent or little bit more distant vessels evoking so-
called stop flow phenomenon [26–28]. In about 50 % of cases
platelet thrombus plugs the opening in the aneurysm allowing for
clot formation and the patient survives the initial aSAH [25].
These lucky people when admitted to emergency room are diag-
nosed by CT, CT arteriography, or digital cerebral arteriography as
having intracranial aneurismal bleeding and their aneurysms can be
successfully treated with endovascular coiling or surgical clipping
[29–36]. Unfortunately, even those patients are not out of the
woods yet because the presence of blood clot and a release of vaso-
genic metabolites acting via mechanical and chemical pathways
evoke constriction of the cerebral arteries, which causes a delayed
cerebral vasospasm [37–46]. For a well over a half of century, this
enigmatic phenomenon has been widely accepted as a central
culprit responsible for a poor outcome after aSAH due to delayed
ischemic neurological deficits. But this view on the role of delayed
cerebral vasospasm has been recently changing as new venues of
aSAH-related research are pursued [40, 47, 48]. This renewed
A Nonhuman Primate Model of Delayed Cerebral Vasospasm After Aneurismal… 229
interest in mechanisms contributing to development of delayed

cerebral vasospasm, separately from ischemic neurological deficits,
has increased the interest in aSAH models.
Many experimental models of aSAH have been developed dur-
ing the second part of the last century in pursuit of elucidating
pathomechanism(s) of vasospasm. The review of these models is
beyond the scope of this chapter but interested readers may find
excellent summaries, descriptions, and analyses of aSAH models in
several recent reviews [43, 49] and original papers. Among these
models, a nonhuman primate model with a direct surgical place-
ment of blood clot around the middle cerebral artery, as developed
by Espinoza in Bryce Weir laboratory, has proven to be the most
consistent in mimicking development of vasospasm after aSAH
[43, 49–52]. This model led to current development of two of the
most promising clinical trials focused on prevention of vasospasm:
CONSCIOUS-1/3 (Clazosentan to Overcome Neurological
iSChemia and Infarction OccUring after Subarachoid hemorrhage
Study) [29, 53] and sodium nitrite for prevention of vasospasm
currently at Phase IIB (www.clinicaltrials.gov). The Clazosentan
study showing a clinical efficacy confirmed the earlier—but brushed
away—observations [40, 47, 48, 54] that vasospasm, despite being
a tremendous challenge and a possible contributor to a poor out-
come, may not be as crucial as it has been thought for years. New
culprits, including early ischemia and vasospasm, inflammation,
gene, and protein changes as well as combination of cortical spread-
ing depressions and cortical spreading ischemias [47, 48, 55–57]
have been proposed to be responsible for a poor outcome after
successful elimination of an aneurysm surgically or using endovas-
cular techniques. This new perception of aSAH and its conse-
quences has opened new perspectives for research and new venues
for aSAH modeling.
2 Nonhuman Primate Subarachnoid Hemorrhage Model
2.1 Animal Selection In 1984, F. Espinoza in the Bryce Weir’s laboratory in Alberta,
and Preparation Canada published his seminal description of a surgical subarach-
noid hemorrhage model [50] that quickly become recognized as a
best fitting standard experimental model to study a delayed cere-
bral vasospasm after aSAH [49]. This model incorporated com-
bined advantages of (1) providing superior control of SAH
conditions, (2) allowing for repeated cerebral arteriographies, (3)
producing reliable and consistent the middle cerebral artery spasm
in above 90 % of animals without (4) evoking neurological deficits.
Despite it costs, it become widely accepted as the model of choice
to study vasospasm pathophysiology and has been used to validate
new proposed treatments in many preclinical studies. As men-
tioned above, positive results of studies using a nonhuman primate
model of aSAH led to development of promising current clinical

trials with Clazosentan [53], nitroprusside [58], and sodium nitrite
[59]. Nevertheless, as mentioned above, the results of
CONSCIOUS-1/3 studies dramatically exposed the weakness of
this model that for a long time was treated as its strength.
Clazosentan successfully, as predicted from primate studies, pre-
vented development of vasospasm but, unfortunately, it did not
improve the overall outcome in patients after aSAH [60]. This
result that was further investigated and analyzed in the
CONSCIOUS-2 and -3 studies [61] opened the discussion about
the value of disease modeling especially using sensitive species [47,
48]. Nevertheless, despite some still existing controversy, it has
become clear that the nonhuman primate model successfully deliv-
ered what it was design for. The investigators were able to establish
pathomechanisms of delayed cerebral vasospasm and subsequently
addressed them successfully [60]. The problem was, and remains
valid, that some early warning reports that vasospasm might not be
the only, or even the most important, contributor to poor out-
come after technically perfect treatment of a ruptured intracranial
aneurysm, were overlooked [62]. So now, the research is refocus-
ing on explaining this unexpected phenomenon and the nonhu-
man primate model of aSAH should continue to play a leading role
in testing hypotheses and treatments.
2.2 Animals In the Espinoza’s pioneered model, the nonhuman primates,

Cynomologus monkeys (Macaca fasccularis) were most often used
because they are smaller, easier to handle, and in the 1980s were
cheaper than rhesus and other nonhuman primates. These mon-
keys vascular neuroanatomy corresponds to that of human except
that they have, a characteristic for primates, the single pericallosal
artery. For our experiments, initially we used wild caught animals
but later when breeding colonies were established in the United
States we used purpose bred animals. We have also used, if ade-
quate, less expensive recycled monkeys obtained from different
sources.
2.3 Quarantine, Adult monkeys (6–17 years old) of both sexes were used and their
Husbandry, weights ranged from 2.5 to 7.5 kg. Animals come from four
and Treatments of SAH domestic breeding/quarantine facilities. Before arrival to the
Monkeys research facility on main NIH campus in Bethesda, MD, animals
underwent at least 13-weeks quarantine. During quarantine, the
animals were examined at least twice by a facility veterinarian,
TB-tested five times at 2 week intervals, received two courses of a
broad spectrum parasiticide, and their blood was tested for hema-
tology and chemistry as well as by serology and/or PCR for a com-
prehensive viral panel (i.e., serology for measles, Herpes B, SRV-1,
2, 3, and 5, SIV, STLV-1; and PCR for SRV-1, 2, and 5). Animals
that did not have protective measles antibodies were vaccinated.
Additional tests and treatment were conducted if necessary depend-

ing on animal health problems at discretion of facility veterinarian.
Within 2 weeks of arrival to the quarantine facility each animal had
a behavior examination, which became a part of the individual ani-
mal record. During quarantine, caretakers wore a respirator, shoe
covers, Tyvek coveralls, eye protection, cap or bonnet, and latex
gloves when entering the quarantine rooms or working with mon-
keys. Protective clothing was immediately disposed of upon leav-
ing the room.
After completion of quarantine and upon arrival to the research
facility on main NIH campus, animals received an entrance physi-
cal examination by a facility veterinarian and started on a three
times per year TB-testing program.
During and after quarantine, animals were singly housed in
appropriate size cages and their health checks were performed at
least twice daily. Any problems were brought to the immediate
attention of the supervisor and the veterinarian. Animals were fed
8788 Harlan/Teklad primate diet twice daily in quantities devel-
oped by a NIH Laboratory Animal Nutritionist according to ani-
mal body weights. For example, animals weighing 3–6 kg received
5–8 biscuits and animal weighing 6–10 kg received 8–11 biscuits
of 8788 diet. Animals had constant access to water, which was sup-
plied through automatic watering system and were given daily fruit
supplements. Cages including pans were rinsed at least once daily.
Cages were sanitized at least every 14 days by passing through the
cage wash. Animal room floors were mopped at least once daily
with appropriate Tuberculocidal disinfectant/detergent.
As a part of environmental enrichment program, each animal
was offered at least two toys: one inside and one outside the cage.
The toys were rotated every 2 weeks at the times of cage changes.
Additionally, approximately once weekly, animals were shown car-
toons or nature movies on a TV/DVD screen mounted on a cart.
2.4 Preparation Cynomologus monkeys weighing from 2.5 to 7.5 kg were used as
and Anesthesia an animal model to study the delayed cerebral vasospasm after
aSAH. Monkeys were held off food overnight before surgery. The
morning of surgery the animals were transported to the surgical
facility. The transport cage was placed inside a Plas-Labs Intensive
Care System (Plas-Labs, Inc. Lansing, MI 48906) and the oxygen
was turned on and set at 8 l/min and the temperature set at 85 °F
to stabilize the animals temperature prior to surgery. Next, the
monkey was given Ketamine (10 mg/kg IM) and atropine
(0.04 mg/kg IM) as a preanesthetic. Once sedated, the animal was
moved to the preparation room and the groin(s) and head areas
were clipped in preparation for surgery. During preparation, the
animal’s body temperature was maintained by a heat lamp sus-
pended over the preparation table as well as an electric heating pad
placed on the tabletop. A 22 gauge, 1-in. angiocatheter was
inserted into the cephalic or saphenous vein for hydration with

normal saline (10–15 ml/kg/h) and drug delivery. Ketofen
(2 mg/kg IM) and Cefazolin (50 mg/kg IV) were given. Propofol
(10 mg/ml) to effect was given at 0.2 ml increments until proper
level of anesthesia was reached for intubation. Depending on the
size of the monkey a 4 or 5 mm endotracheal tube was inserted to
maintain the airway. Animals were then transported to the surgery
room and placed in supine for cerebral arteriogram or left lateral
recumbence for the subarachnoid hemorrhage clot placement.
For cerebral angiograms, the head was placed in a homemade
head holder made of foam, which had a radiopaque ruler on top so
accurate measurements of the middle cerebral artery could be
made (Fig. 1). The cerebral arteriograms were done using an OEC
mobile digital imaging system (Series 9600) set on the subtraction
mode at eight frames per second (Fig. 2). Arteriographic runs were
played back and one or two images were selected for measurement
of the middle cerebral artery.
The surgical sites were scrubbed with Alcare (Steris
Corporation, St. Louis, MO 63110) or chlorhexidine surgical
scrub and draped for aseptic surgery. During surgery, anesthesia
was maintained with a Narkomed 2B anesthesia machine (Fig. 3)
using Isoflurane (0.5–2 %) and oxygen at a total flow rate of 2 l/
min. A 40-in. Universal F anesthetic hose, which connected the
Fig. 1 Head positioner with radiographic ruler used for cerebral angiograms
Fig. 2 General Electric OEC (9600) C-arm used for doing cerebral angiograms
anesthesia machine to the animal, was used to conserve heat loss

and maintain proper humidity of gasses delivered to the monkey.
The use of these hoses prevents the need for the addition of a heat
and moisture exchanger and prevents the problems associated with
the added dead space and increased flow resistance. Body tempera-
ture was maintained during surgery by use of a piped-in warm
water heating blanket (Hemotherm by Cincinnati Sub-Zero
Cincinnati, OH 45241) set at 44 °C, Hotline Fluid Warmer to
maintain IV fluids at 37–42 °C (Smith Medical, Rockland, MA
02370), and portable heat lamps as needed. During surgery, ECG,
SpO2/Pleth, end tidal carbon dioxide (ETCO2), respiratory rate,
minute volume, airway peak pressure, tidal volume, I:E ratio,
oxygen percentage, temperature, and direct or indirect blood pres-
sures as appropriate were monitored.
Fig. 3 Narkomed 2B anesthesia machine and Hewlet Packard monitors used for
surgical procedures
3 Cerebral Arteriography
Each animal after quarantine but before being accepted for a pro-
tocol had a baseline arteriography and, if necessary, a baseline tran-
scranial Doppler study.
3.1 Equipment Ketamine-Xylazine-atropine induction, clippers, alcohol wipes, 22

gauge IV angiocath, saline, propofol, endotracheal tube size
4–5 mm, clippers, antibiotic, analgesic, rebreathing anesthesia
machine with isoflurane anesthetic, scrubbing solution, cut down
tray, sterile drapes, #10 or #15 scalpel blade, iris scissors, fine dis-
secting scissors, vascular forceps, bipolar coagulation, vascular
retractor, straight and L-shaped vascular forceps, 3, 5, and 12 cc

syringes, 18–25 gauge needles, 3 or 5 French catheter preshaped
to the “lazy S” with guide wires, basin, local anesthetics, gealfoam,
papaverine, heparin, half-by half inch cottonoids (patties), 5 ml of
iodine contrast, surgical loupes, head holder with radio-opaque
ruler, digital arteriography C-arm, 3-0 silk ties, 4-0 Vicryl sutures
for skin closure, bandage material for pressure dressing, and
Buprenorphine.
4 Technique of Cerebral Arteriography
Animal is placed in a supine position on the operating room table

with the lower extremities slightly bent, rotated externally, and
secured with small sand bags and surgical ties. Usual skin prepara-
tion is performed and surgical field is covered with four small ster-
ile drapes and then with a larger pediatric laparotomy drape. The
French 3–5 angiographic catheter is flushed with heparinized saline
and put with the guide wire in a sterile basin with saline. Sterile
saline (10 ml) with 100 units of heparin is prepared in a small cup
for periodic flushing of the catheter. Five milliliter of contrast
(iopamidol injection 61 %, Bracco Diagnostics) is drawn up into a
6 ml syringe. A 2-in. long skin incision is performed with a #10
blade about half an inch above the knee in the crease between a
vastus medialis and an abductor muscle. Bupivicaine (1–2 cc of
0.25 %) is injected subcutaneously around the incision site for addi-
tional analgesia. Dissection is continued between the muscles to
expose the neurovascular bundle and the femoral artery is isolated
with small dissecting scissors. Two 3-O silk ties are placed under
the artery about 1 in. (2 cm) apart. If the artery gets into spasm, a
half-by-half inch patty soaked in papaverine is placed on the vessel
for 1–2 min. Then, a small V-shaped incision is made in the artery;
blood flow is controlled by tension on both ties and the intra-
arterial catheter with the wire exposed by 1–2 mm is advanced into
the artery. To avoid accidental perforation and use a curvature on
the tip of the catheter, the wire is moved back about 1 in. as soon
as the catheter is safely in the artery. After advancing the catheter
to the level of the heart under radiological visualization, the tip of
catheter is rotated up and catheter advances to the right brachioce-
phalic trunk and then to the common carotid artery. At this point,
the head of the monkey is rotated to the right and the tip of the
catheter is guided to the internal carotid artery (Fig. 4). The head
is moved back to the straight position and placed in the head
holder. 0.2 ml of contrast in injected to confirm correct position of
the catheter tip in the internal carotid artery. The catheter is flushed
with 1 ml of heparinized saline followed by 1 ml of contrast agent,
which is injected during the eight frames per second subtraction
run for arteriography (Fig. 5a, b). After obtaining the arteriogram,
Fig. 4 The angiocatheter is advanced to he common carotid and the head of

monkey is rotated to the right to facilitate the navigation of catheter into the right
internal carotid artery
Fig. 5 (a) Right internal carotid arteriogram. Sometimes all four cerebral arteries of circle of Willis can be
visualized with 1 ml of contrast as on this baseline arteriograms. (b) This animal developed moderate
vasospasm of the right middle cerebral artery (arrows) on the 7th day after a clot placement
the catheter is quickly removed and the opening in the artery is

covered with a tiny piece of gelfoam and kept under pressure
through a small half-by-half inch cottonoid patty for about 5 min.
After the patty is removed, the skin incision is closed with an intra-
dermal a 4-0 vicryl suture. The wound is covered with antibiotic
cream and a pressure dressing is applied. Gas anesthesia is
discontinued.
The monkey is allowed to gain a gag reflex and then the endo-
tracheal tube, rectal temperature probe, ECG leads, SpO2, and IV
line are removed and the monkey is returned to the transport cage
that remains in the Plas-Labs Intensive Care Unit. The cage tem-
perature is maintained at 85 °F unless the monkey is hypothermic.
Oxygen is delivered to the cage at the rate of 8 l/min. Buprenorphine
(0.01 mg/kg IM) is given postoperatively for additional analgesia.
The animal is monitored until recovery is complete and then
returns to its home cage.
Usually, arteriography is repeated on the 7th day after a clot
placement (SAH) and subsequently, if necessary on days 14, 21,
28. For subsequent arteriograms, the same side femoral artery is
exposed but 1-in. higher than the initial incision. Usually, after the
incision of the artery with iris scissors and removal of the white
clot, the patency of the vessel is restored allowing for catheter
insertion and navigation. If this is too difficult and survival of the
animal is anticipated then the left femoral artery is exposed and
catheterized.
If this is a terminal arteriogram, the dissection of the right
femoral artery could be continued higher along the femoral artery
to the iliac arteries or if this is impossible and the left femoral artery
is also thrombosed, a direct access to the common carotid artery
can be used. We have also used the brachial artery in a couple of
instances to gain access to the internal carotid.
For some experiments when repeated arteriographies are nec-
essary, contrast reinjections are performed within minutes or hours
after the baseline. Between injections the catheter has to be moved
back to the common carotid to avoid thrombosis and while not
used it is slowly infused with heparinized saline. Each subsequent
contrast injection is with no more than 0.75 ml of contrast and the
animals are properly hydrated with continuous monitoring of
blood pressure and heart rate.
5 Arteriographic Measurements: Vasospasm Assessment
All cerebral arteriograms were performed as described above. The

two dimensional area of the proximal 14 mm of the right middle
cerebral artery on the anterioposterior projection was recorded for
all arteriograms using an image analysis system (NIH Image, ver-
sion 1.62 or NIH Image J; National Institutes of Health, Bethesda,
MD). First, the ruler on the arteriogram was measured three times
and the mean value of measurements was assessed and calibrated.
Then, the length of the right middle cerebral artery from the inter-
nal carotid bifurcation to the most lateral exposed M2 part was
measured on the baseline arteriogram. Next, the area of the exposed
to clot middle cerebral artery was outlined and measured. Then, all
the steps were repeated with the post SAH arteriogram making
sure that the starting point of the length measurement was exactly
the same as on the baseline arteriogram. The degree of vasospasm
for each animal at the time of assessment of the vascular responses
was determined by comparing to the initial preoperative baseline
arteriogram. The presence of significant vasospasm was defined as a
25 % or greater reduction in the proximal 14 mm of the right mid-
dle cerebral artery area as measured on the anterioposterior projec-
tion of cerebral arteriogram. Arteriographic results are reported as
the average of measurements performed by three independent,
blinded observers before the experimental code is broken.
6 Cerebrospinal Fluid Collection via Suboccipital Puncture
6.1 Equipment General anesthesia or heavy sedation, clippers, scrubbing solutions,

alcohol, drape, 1 1/2 in. long 22 gauge short bevel needle or 2 in.
25 gauge needle, 3 ml syringe, 5 ml red top blood collection tube.
6.2 Technique Animal under general anesthesia (with or without intubation) is

shaved and prepped in usual way and then positioned on the right
side with the head bent to the chest. A 1 1/2 in. 22 gauge needle
is attached to the 3 ml syringe. The needle is introduced in the
neck muscles a short distance below the palpable edge of the occip-
ital protuberance, in the midline and slowly advanced toward the
cistern magna while a slight negative pressure is kept in the syringe.
If resistance is encountered, the needle is turned more caudally and
advanced 2–3 mm until the CSF flaws in the syringe. One to 3 ml
of CSF is collected. The needle is then removed and skin covered
by antibiotic cream. If blood enters the needle, the fluid is imme-
diately centrifuged and only supernatant is removed and used for
further investigation.
7 A Clot Placement, Subarachnoid Hemorrhage
7.1 Anesthesia During the subarachnoid hemorrhage procedure, the animal is

placed on the pediatric ventilator with the tidal volume adjusted to
approximately 15 ml/kg and respiratory rate set to 10–12 breaths/
min. Once the peak pressure has stabilized, the bellows on the
ventilator is adjusted to reach a peak pressure of 13 cm H2O and
later adjusted to maintain an ETCO2 of about 40. Prior to expos-
ing the brain the ETCO2 is lowered to about 22 mmHg by increas-

ing the respirations to 17–22 and/or increasing the tidal volume
by raising the bellows to decrease blood flow and to limit cranial
swelling. This level of ETCO2 is maintained until closure when the
ETCO2 is allowed to go back to normal. When necessary dexa-
methason (1–2 mg/kg IV or IM) or furosemide (1–2 mg/kg IV)
is used to decrease cranial swelling during surgery or during the
recovery phase.
7.2 Equipment Ketamine-Xylazine-atropine induction, clippers, alcohol wipes, 22

gauge IV angiocath, NaCl fluids for IV administration, propofol,
endotracheal tube size 4–5 mm, antibiotic, analgesics, antibiotic,
rebreathing anesthesia machine with isoflurane anesthetic, tape to
secure head on foam cushion, scrubbing solution, general instru-
ment tray, micro instruments, microforceps, microscissors curved
and straight, arachnoid knife and #15 scalpel blade, sterile drapes,
rubber bands, monopolar and bipolar coagulations, suction tub-
ing, cottonoids, gelfoam, Hall drill with cutting and diamond
burrs (1–3 mm), 1, 3, and 6 cc syringes, 25 gauge angiocath for
CSF collection, local anesthetics, gelfoam, 1/4 × 1/4 in. cotto-
noids (patties), surgical loupes and operating microscope, bone
wax, sutures for closure (monofilament 7/8-0 for dural closing;
vicryl 3-0 for muscle closing, 4-0 for intradermal skin closure),
topical antibiotic ointment for the incision site, and Buprenorphine.
8 SAH: Surgical Technique
Intubated monkey is placed on the heating pad on the operating

room table in a left lateral oblique position. Extremities are secured
to the sides of the operating table exposing the left groin area. The
head is positioned in a homemade foam head holder with several
layers of 4 × 4 gauze put in the cut out area of the foam to tip of the
head to a 10–15° angle. The position of the head is secured with
masking tape running through the base of the nostrils (Note 1).
The head and both groins, which are shaved immediately follow-
ing induction of anesthesia, are prepped for aseptic surgery.
Bupivicaine (11/2 to 2 cc of 0.25 %) is injected subcutaneously
around incision site for additional analgesia. After placement of
sterile drapes, a C-shaped skin incision is made in the right supra-
orbital area from the zygomatic arc to close to the midline just
above the orbital ridge. The skin flap is turned up over the eye and
kept in position by 3-0 vicryl suture. Next, a shallow T-shaped inci-
sion with a Bovie electocautery is made through the fascia from the
zygomatic arch to 1–1 1/2 in. above the “key-hole” area at the
base of the orbital rim and then the muscle incision is carried out
from the orbital rim perpendicularly to the first one above an area
of the Sylvian fissure. The frontal part of the temporalis muscles is
separated from the bone and after covering with wet gauze secured
with the 2-0 vicryl suture. Then, a deeper incision through the
superficial temporalis muscle toward the zygomatic arc is per-
formed and superficial muscle flap is separated and secured with
the suture/rubber band to the drapes. The incision is carried down
to the temporal bone and after resection of the part of the deep
temporalis muscle and coagulation of the deep branch of temporal
artery the rest of the muscle is secured with the superficial muscle
(Note 2). A 2 in. long and 1 in. wide bone flap with its base at the
bottom on the orbital ridge is marked with the cutting drill bit and
then drilled to the internal bone plate under constant irrigation
(Note 3). Any bone bleeding is topped with bone wax. A diamond
drill is used to remove the internal plate. At this moment, 1–2 mg
of furosemide is administrated intravenously (Note 4) and hyper-
ventilation is started to lower and then maintain the ETCO2 at
22 mmHg. With Penfield number 2 or 4, depending on the thick-
ness of the bone, the bone flap is carefully elevated and removed.
The sphenoid wing bone is drilled out under magnification of the
operating microscope (Fig. 6). Next, under microscope magnifica-
tion, the dura is incised about ½ in. above the orbital ridge with a
#15 blade avoiding the opening of the arachnoid. The incision is
carried up to the orbital ridge above the frontal lobe and then
down above the Sylvian fissure toward tip of the temporal lobe
under the zygomatic arch (Fig. 7). The orbital edge of the dura is
turned up and if necessary kept against the bone with wet cotto-
noid. After identification of the upper, frontal lip of the Sylvian
fissure, the arachnoid knife (a 27-gauge slightly bend needle) is
used to open the arachnoid, which then is further opened sharply
with microscissors or divided with microforceps. The CSF is gently
collected using 1-in. long 25-gauge IV catheter on a 3 ml syringe.
The frontal lobe is gently elevated using a Penfield retractor #4 and
single micro-patties are individually placed for retraction of the
brain partially achieved by furosemide and CSF removal. The same
maneuver is repeated with the temporal lobe; usually four to five
micro-patties are placed under each lobe. Occasionally, small per-
forating veins crossing from the temporal lobe need to be coagu-
lated with bipolar coagulation and divided. Then the arachnoid
over the Sylvian fissure is removed exposing the middle cerebral
artery (Fig. 8), its branches, and the internal carotid artery down
below the anterior clinoid. Attention is then shifted to prepare the
blood clot for placement around the middle cerebral artery. A skin
incision is made to expose the femoral artery in the left groin simi-
lar to what was done for the cerebral arteriography. After exposure
of femoral artery, a 25-gauge IV angiocath is introduced into the
artery and 5 ml of blood is removed and set aside to clot. The
artery and the wound are secured as described for arteriography. At
this moment, hyperventilation is stopped and ETCO2 gradually
increases back to normal of 38–40 mmHg. Under magnification,
the micro-patties are removed, CSF is collected, and the first small
Fig. 6 The operating microscope is in position for dissection of the Sylvian fissure
to expose the right middle cerebral artery
pieces of a clot are “locked” in the front and back of the internal
carotid artery below the bifurcation (Note 5). Then, bigger pieces
of the clot are inserted under the frontal and temporal lobes to
keep the Sylvian fissure open, and the last piece of clot is placed on
the middle cerebral artery and if a significant subdural space per-
sists it is filled up with the rest of clot (Fig. 9). Usually a clot from
5 ml of blood is enough to cover adequately the middle cerebral
artery and it has been enough to produce vasospasm in 95 % of
animals. The next step is closing the dura with 7-0 or 8-0 mono-
filament continuous suture (Note 6). After removal of the micro-
scope, the temporalis muscle is approximated with 3-0 vicryl suture
and the skin is sutured (Note 7) with 4-0 intradermal sutures
(Note 8). The wound is covered with a topical antibiotic cream,
anesthesia is turned off, and the monkey extubated after regaining
a gag reflex. An initial neurological assessment is done before the
animal is sent back to the heated cage.
Fig. 7 The base of skull after removal. Arrows point to the dural suture line. Note
hemosiderin deposits in the dura of anterior and temporal fossae
Fig. 8 The base of the brain, arrows indicate the area of removed arachnoid over
the right middle cerebral artery
Fig. 9 The fresh clot covering the right middle cerebral artery
9 Modification of Model to Examine aSAH-Evoked Delayed Ischemic Deficits
As mentioned above, the results of the recent study and earlier

observations suggest that vasospasm may not be as crucial in
development of delayed ischemic deficits as has been thought for
a long time. To provide an insight in the aSAH-related mecha-
nisms responsible for delayed ischemic deficits and/or delayed
cerebral vasospasm, we propose a slight modification of the
model, in which after opening the dura as described before with
the #15 blade, the culvinear incision of dura is performed using
a microscissors 1 in. above the frontal bone rim extending from
the frontal lobe to the edge of the temporal pole. The crucial
part of this surgery is that the incision has to be carried out
avoiding the opening the arachnoid. Then, a small nick in the
arachnoid is performed with the arachnoid knife just above the
edge of the Sylvian fissure. The tip of 25-gauge intravenous
catheter attached to 6 cc syringe with normal saline is introduced
through the arachnoid opening and 1–2 ml of warm saline is
slowly infused under lower magnification toward the frontal
lobe. This maneuver usually produces an elevation of arachnoid
over the posterior aspect of the inferior and/or middle frontal

gyrus. After another small nick of the elevated arachnoid, using
an arachnoid knife or microscissors and microforceps in an atrau-
matic way, at least a 1 × 1 cm area of arachnoid over the frontal
gyri is removed without any bleeding. If any bleeding starts, one
or two micro-patties are inserted under the arachnoid and left
for couple minutes. Then the removal of arachnoid is continued
in this area or if more convenient in another close-by area. After
removal of arachnoid over a small aspect of the lateral frontal
lobe, the attention is directed toward opening the Sylvian fissure
and exposure of the middle cerebral artery. At first, 1 ml of saline
is injected through the previous opening in arachnoid above the
Sylvian fissure which allows for separation of arachnoid from the
pia usually more on the side of the frontal than temporal lobe.
Then, using a microscissors/microforceps and/or arachnoid
knife the arachnoid is opened along the distal first and then
proximal MCA, the ICA bifurcation, and the ICA below the ten-
torial edge; if easily achieved, the Lilliequist membrane is also
sharply opened. The arachnoid is removed from above the ves-
sels. Each time, an additional part of arachnoid, if safely accessi-
ble, is removed exposing additionally about 0.5–1 cm of a
posterior frontal and sometimes temporal lobe parallel to the
Sylvian fissure. After exposure of the internal carotid artery and
its bifurcation, further dissection and removal of arachnoid is
focused on the orbital part of frontal lobe and under the highest
magnification the arachnoid over the anterior cerebral artery and
pericallosal artery is removed exposing additional part of orbital
gyrus and olfactory nerve. The area of the brain depleted of
arachnoid is described carefully in the operative notes to assure
an easy recognition of the relevant areas during autopsy and his-
topathological studies. This is especially important for the con-
trol group because in the SAH group at the time of autopsy, the
denuded area, which has been exposed to blood has a signifi-
cantly more yellowish tint from hemosiderin than the area that
remained covered by arachnoid.
Upon completion of surgical procedure the monkey is allowed
to regain a gag reflex and then the endotracheal tube, rectal tem-
perature probe, ECG leads, SpO2, and IV line is removed and ani-
mal returned to the transport cage that is still in the Plas-Labs
Intensive Care Unit. The cage temperature is maintained at 85 °F
unless the monkey is hypothermic at which time the cage tempera-
ture is adjusted between 85 and 100 °F (Note 9). Oxygen is deliv-
ered to the cage at the rate of 8 l/min. Buprenorphine (0.01 mg/
kg IM) is given postoperatively for additional analgesia. The animals
are monitored until recovery is complete and then returned to
their home cage (Note 10).
10 Posttreatment Handling
Upon return from the surgery the animals are started on a treat-
ment course consisting of an analgesic Buprenorphine at 0.03 mg/
kg twice a day for total of 3 days and antibiotic Cefazolin at 25 mg/
kg IM twice a day for 7 days. Animals are closely monitored for
alertness, activity level, integrity, swelling or infection of surgical
incisions, hydration, appetite, urination, defecation, and presence
of neurological deficits. If animal appears in pain, Ketoprofen is
administered at 2 mg/kg IM as needed. In cases of possible infec-
tion of incision site, Gentamicin is added to the treatment plan at
4 mg/kg IM twice a day for 7 days. If animal appears dehydrated,
100–200 ml of Lactated Ringers Solution is given subcutaneously.
If animal appears hypoactive, it is placed in the temperature and
relative humidity controlled oxygen cage with oxygen flow up to
10 l/min. Additional diagnostic workup (e.g. CBC, blood chemis-
try, X-rays) is ordered as needed.
11 Experiment Termination and Autopsy
The procedure for performing necropsies on macaques and other

nonhuman primates is similar to that for other mammals. However,
additional safety considerations are necessary. Monkeys can harbor
many zoonotic agents such has various Mycobacterium species.
The most important pathogen is Cercopithicine herpesvirus 1
(Herpes B). This virus can cause oral and genital ulceration in
macaque species or be carried asymptomatically. Animals can har-
bor the virus and remain serum negative, so, there is no way to be
sure that an animal in not infected. Herpes B is highly fatal in
humans with 70–80 % fatality rate reported in untreated human
cases. Fatalities are much lower in cases treated immediately with
antiviral drugs.
Because of the zoonotic risk associated with primate necropsies
appropriate personal protection (PPE) and biosafety equipment is
necessary. PPE includes a full-face shield, surgical mask, a labora-
tory coat made of nonabsorbent material, surgical scrubs, shoe
covers, and disposable gloves. Steel gloves are also available for the
nondominant hand. For removing tissue from the central nervous
system (CNS), a Biosafety cabinet is recommended. Necropsies
can be also performed on down draft tables; so, that pathogens are
less likely to be spread by air currents.
Necropsies are performed with the cadaver in supine position.
First an external exam is performed noting general body condition,
hydration status, surface lesions, eyes, and oral cavity. A ventral
midline incision is made in the skin from the mandibular symphysis
to the pubic bone and the skin is reflected from underlying body
wall. A ventral midline incision is made through the abdominal

muscles. An incision is made through the diaphragm following the
contour of the rib cage. Bone cutting shears are used to remove the
rib cage. After the thoracic and abdominal cavities are opened, a
systematic examination is performed including lymph nodes, tho-
racic viscera, and abdominal viscera. Any abnormal tissues are
noted. Tissue samples then can be collected as needed for study
protocol as well for diagnostic purposes in the case of unexpected
lesions. Tissues of interest can be also sliced at 1 cm intervals for
the presence of deep parenchymal lesions.
To remove the brain, the head is severed from neck. The skin
and muscles overlying the cranial vault are removed. A vibrating
(or Stryker) saw is used to cut through the skull starting with a
transverse cut through the frontal bone, extending ventrally, and
caudally through the temporal bone and then the occipital bone to
the foramen magnum. The top of the skull in pried off and the
brain gently dissected away from the cranial floor. The remaining
skull can be hemisectioned if necessary to examine the nasal cavity
and remove the pterygopalatine ganglia. Because the brain is soft
and easily macerated, sectioning of the brain is recommended until
after tissue fixation. After fixation, the brain can be sectioned trans-
versely beginning at the frontal cortex and slices from the various
levels examined for abnormalities.
12 Notes
1. As with a surgical approach to the intracranial aneurysm, the

head position is crucial to facilitate the approach and access to
the basal cisterns with no or only minimal brain retraction. But
the problem is we do not have a Mayfield frame to facilitate
and stabilize the head. So, make sure that the monkey head is
properly positioned and secured with the tape.
2. The incision of temporalis fascia and removal of the deep tem-
poralis muscle should be performed very carefully for a water-
tight closing later on.
3. When you start drilling the bone, start hyperventilation to give
the brain time for relaxation. While removing the bone flap,
make sure that you carefully secure all the bleeds from the
muscle and the bone. It is crucial to avoid any trauma to the
brain.
4. Before the dura incision, give the dose of furosemide. While
opening the dura, it is crucial to keep the arachnoid intact and
the surgical field bloodless. Any bleeding from the muscle,
bone, or dura should be stopped before dura opening.
Coagulation of the middle meningeal artery should be per-
formed directly before its cutting; avoid evoking the shrinkage
of dural edges. This part is crucial for a water-tight closure of

dura.
5. While placing the clot around the middle cerebral artery,
remember about putting the tiny pieces of clot around the
internal carotid artery below the optic nerve and below the
posterior communicating artery. This maneuver will protect
the rest of clot from being “washed” away from the surface of
cerebral arteries. As the second step, lay a significant amount of
clot under the frontal and temporal lobes to keep the Sylvian
fissure open, and then cover the middle cerebral artery with
the rest of clot.
6. The dural closure had to be water tight. If you need, use the
fascia/muscle graft or tissue glue to secure it.
7. The anatomical closure of the muscle and fascia is important
for several reasons; it accelerates wound healing and animal
recovery, it allows animal for faster restarting drinking and eat-
ing and it protects against collection of fluid under the skin
alleviating the risk of infection.
8. ALWAYS use the intradermal sutures on the skin.
9. During preparation, surgery, and after surgery remember that
monkey has only a thin layer of fat; it may become hypother-
mic very easily; so, monitor closely the body and environment
temperature to avoid it.
10. During the first 24 h, do not leave any easy to choke on things
or food in the cage. Make sure that animal is regularly checked
and given a proper analgesia. If the animal does not drink/eat,
try a softer food (banana) and orange juice before starting the
subcutaneous hydration and tube feeding.
Acknowledgment
This research was supported in part by the Intramural Research

Program of the NIH, NINDS.
References
1. Yasargil M, Kasdaglis K, Jain K, Weber H hemorrhagic cerebro-spinal fluid on cat pial
(1976) Anatomical observations of the sub- arterioles in situ. J Neurosurg 54:351–356
arachnoid cisterns of the brain during surgery. 4. Clark R, Carcillo J, Kochanek P, Obrist W,
J Neurosurg 44:298–302 Jackson E, Mi Z, Wisniewski S, Bell M, Marion
2. Bergstrand G, Larsson S, Bergstrom M, D (1997) Cerebrospinal fluid adenosine con-
Eriksson L, Edner G (1983) Cerebrospinal centration and uncoupling of cerebral blood
fluid circulation: evaluation by single-photon flow and oxidative metabolism after severe
and positron emission tomography. AJNR Am head injury in humans. Neurosurgery
J Neuroradiol 4:557–559 41:1284–1293
3. Brandt L, Ljunggren B, Anderson K et al 5. Fuijmori A, Yanagisawa M, Saito A, Goto K,
(1981) Vasoconstrictive effects of human post- Masaki T, Mima T, Takakura K, Shigeno T
(1990) Endothelin in plasma and cerebrospinal 17. Sato S, Ishihara N, Yuniki K (1990) The
fluid of patients with subarachnoid haemor- changes in endothelin concentration in CSF
rhage. Lancet 336:633 following subarachnoid hemorrhage. In: Sano
6. Hardemark HG, Almqvist O, Johansson T, K et al (eds) Cerebral vasospasm. University of
Pahlman S, Persson L (1989) S-100 protein in Tokyo Press, Tokyo, pp 269–271
cerebrospinal fluid after aneurysmal subarach- 18. Shore PM, Berger RP, Varma S, Janesko KL,
noid haemorrhage: relation to functional out- Wisniewski SR, Clark RS, Adelson PD, Thomas
come, late CT and SPECT changes, and signs NJ, Lai YC, Bayir H, Kochanek PM (2007)
of higher cortical dysfunction. Acta Neurochir Cerebrospinal fluid biomarkers versus glasgow
(Wien) 99:135–144 coma scale and glasgow outcome scale in pediat-
7. Jung CS, Oldfield EH, Harvey-White J, Espey ric traumatic brain injury: the role of young age
MG, Zimmermann M, Seifert V, Pluta RM and inflicted injury. J Neurotrauma 24:75–86
(2007) Association of an endogenous inhibitor 19. Suzuki H, Muramatsu M, Tanaka K, Fujiwara
of nitric oxide synthase with cerebral vasospasm H, Kojima T, Taki W (2006) Cerebrospinal
in patients with aneurysmal subarachnoid hem- fluid ferritn in chronic hydrocephalus after
orrhage. J Neurosurg 107:945–950 aneurysmal subarachnoid hemorrhage.
8. Kim P, Yaksh T, Romero S, Sundt T (1987) J Neurol 253:1170–1176
Production of uric acid in cerebrospinal fluid 20. Suzuki Y, Osuka K, Noda A, Tanazawa T,
after subarachnoid hemorrhage in dog: investi- Takayasu M, Shibuya M, Yoshida J (1997)
gation of possible role of xanthine oxidase in Nitric oxide metabolites in the cisternal cere-
chronic vasospasm. Neurosurgery 21:39–44 bral spinal fluid in patients with subarachnoid
9. Kırış T, Erden T, Sahinbas M, Omay B, Esen F hemorrhage. Neurosurgery 41:807–812
(2005) CSF drainage for prevention and rever- 21. Voldby B, Petersen OF, Buhl M, Jakobsen P,
sal of cerebral vasospasm after surgical treat- Ostergaard R (1984) Reversal of cerebral arte-
ment of intracranial aneurysms. In: Macdonald rial spasm by intrathecal administration of a
RL (ed) Cerebral vasospasm. Thieme, calcium antagonist (nimodipine). Acta
New York, pp 255–258 Neurochir (Wien) 70:243–254
10. Latour LL, Warach S (2002) Cerebral spinal 22. Woszczyk A, Deinsberger W, Boker A-K
fluid contamination of the measurement of the (2003) Nitric oxide metabolities in cisternal
apparent diffusion coefficient of water in acute CSF correlate with cerebral vasospasm with a
stroke. Magn Reson Med 48:478–486 subarachnoid hemorrhage. Acta Neurochir
11. Moriyama E, Matsumoto Y, Meguro T, 145:257–264
Kawada S, Mandai S, Gohda Y, Sakurai M 23. Xu L, Wang B, Kaur K, Kho MF, Cooke JP,
(1995) Combined cisternal drainage and intra- Giffard RG (2007) NOx and ADMA changes
thecal urokinase injection therapy for preven- with focal ischemia, amelioration with the
tion of vasospasm in patients with aneurysmal chaperonin groel. Neurosci Lett 418:201–204
subarachnoid hemorrhage. Neurol Med Chir 24. Kassell NF, Torner JC (1984) The interna-
(Tokyo) 35:732–736 tional cooperative study in timing of aneurysm
12. Naff N, Williams M, Rigamonti D, Keyl P, surgery—an update. Stroke 15:566–570
Hanley D (2001) Blood clot resolution in 25. Stapf C, Mohr J (2004) Aneurysms and sub-
human cerebrospinal fluid: evidence for first- arachnoid hemorrhage—epidemiology. In: Le
order kinetics. Neurosurgery 49:614–621 Roux PD, Winn HW, Newell DW (eds)
13. Penn RD (2003) Intrathecal medication deliv- Management of cerebral aneurysms. Elsevier
ery. Neurosurg Clin N Am 14:381–387 Inc, Philadelphia, PA, pp 183–187
14. Pluta R, Boock R, Afshar J, Clouse K, Bacic M, 26. Asano T, Sano K (1977) Pathogenic role of no
Ehrenreich H, Oldfield E (1997) The source reflow phenomenon in experimental subarach-
and cause of endothelin 1 release to CSF after noid hemorrhage in dogs. J Neurosurg
SAH. J Neurosurg 87:287–293 46:454–466
15. Pluta RM (2006) Dysfunction of nitric oxide 27. Nornes H (1973) The role of intracranial pres-
synthases as a cause and therapeutic target in sure in the arrest of hemorrhage in patients
delayed cerebral vasospasm after SAH. Neurol with ruptured intracranial aneurysms.
Res 28:730–737 J Neurosurg 39:226–234
16. Proescholdt MG, Hutto B, Brady L, 28. Trojanowski T (1984) Early effects of experi-
Herkenham M (2000) Studies of cerebrospinal mental arterial subarachnoid haemorrhage on
fluid flow and penetration into brain following the cerebral circulation. Part II: Regional cere-
lateral ventricle and cisterna magna injections bral blood flow and cerebral microcirculation
of the tracer [14c] inulin. Neuroscience after experimental subarachnoid haemorrhage.
95:577–592 Acta Neurochir (Wien) 72:241–259
29. Macdonald RL, Higashida RT, Keller E, Mayer 41. Jahromi B, Macdonald R (2004) Vasospasm:
SA, Molyneux A, Raabe A, Vajkoczy P, Wanke diagnosis and medical treatment. In: Leroux P,
I, Frey A, Marr A, Roux S, Kassell NF (2010) Winn W, Newell D (eds) Management of cere-
Preventing vasospasm improves outcome after bral aneurysms. Elsevier Inc., Philadelphia, PA
aneurysmal subarachnoid hemorrhage: ratio- 42. Kassell N, Sasaki T, Colohan A, Nazar G
nale and design of CONSCIOUS-2 and (1985) Cerebral vasospasm following aneurys-
CONSCIOUS-3 trials. Neurocrit Care mal subarachnoid hemorrhage. Stroke
13:416–424 16:562–572
30. Al-Tamimi YZ, Ahmad M, May SE, Bholah 43. Macdonald R, Weir B (2001) Cerebral vaso-
MH, Callear J, Goddard A, Quinn AC, Ross spasm. Academic, San Diego, pp 449–450
SA (2010) A comparison of the outcome of 44. Mayberg M (1998) Cerebral vasospasm.
aneurysmal subarachnoid haemorrhage before Neurosurg Clin N Am 9:615–627
and after the introduction of an endovascular
service. J Clin Neurosci 17:1391–1394 45. Ohta T (2004) Cerebral vasospasm revisited:
SAH syndrome. In: Macdonald RL (ed)
31. Alnaami I, Saqqur M, Chow M (2009) A novel Cerebral vasospasm: advances in research and
treatment of distal cerebral vasospasm. A case treatment; 8th international symposium of
report. Interv Neuroradiol 15:417–420 cerebral vasospasm (Chicago, IL, 2003).
32. Ryttlefors M, Enblad P, Kerr RS, Molyneux AJ Thieme, New York, pp 106–111
(2008) International subarachnoid aneurysm 46. Weir B, Grace M, Hansen J et al (1978) Time
trial of neurosurgical clipping versus endovas- course of vasospasm in man. J Neurosurg
cular coiling: subgroup analysis of 278 elderly 48:173–178
patients. Stroke 39:2720–2726
47. Macdonald RL, Pluta RM, Zhang JH (2007)
33. Kaku Y, Watarai H, Kokuzawa J, Tanaka T, Cerebral vasospasm after subarachnoid hemor-
Andoh T (2007) Cerebral aneurysms: conven- rhage: the emerging revolution. Nat Clin Pract
tional microsurgical technique and endovascu- Neurol 3:256–263
lar method. Surg Technol Int 16:228–235
48. Pluta RM, Hansen-Schwartz J, Dreier J,
34. Lanzino G, Fraser K, Kanaan Y, Wagenbach A Vajkoczy P, Macdonald RL, Nishizawa S,
(2006) Treatment of ruptured intracranial Kasuya H, Wellman G, Keller E, Zauner A,
aneurysms since the international subarach- Dorsch N, Clark J, Ono S, Kiris T, Leroux P,
noid aneurysm trial: practice utilizing clip liga- Zhang JH (2009) Cerebral vasospasm follow-
tion and coil embolization as individual or ing subarachnoid hemorrhage: time for a new
complementary therapies. J Neurosurg world of thought. Neurol Res 31:151–158
104:344–349
49. Megyesi JF, Vollrath B, Cook DA, Findlay JM
35. Wijdicks EF, Kallmes DF, Manno EM, (2000) In vivo animal models of cerebral vaso-
Fulgham JR, Piepgras DG (2005) Subarachnoid spasm: a review. Neurosurgery 46:448–460,
hemorrhage: neurointensive care and aneu- discussion 460–441
rysm repair. Mayo Clin Proc 80:550–559
50. Espinosa F, Weir B, Overton T, Castor W,
36. Flett LM, Chandler CS, Giddings D, Gholkar Grace M, Boisvert O (1984) A randomized
A (2005) Aneurysmal subarachnoid hemor- placebo-controlled double-blind trial of
rhage: management strategies and clinical out- nimodipine after SAH in monkeys. Part I:
comes in a regional neuroscience center. AJNR Clinical and radiological findings. J Neurosurg
Am J Neuroradiol 26:367–372 60:1167–1175
37. Asano T, Matsui T (1993) Cerebral vasospasm: 51. Macdonald RL, Weir BKA, Runzer TD, Grace
a disorder of the local mechanotransduction MGA, Findlay JM, Saito K, Cook DA, Mielke
system in cerebral arteries. Crit Rev Naurosurg BW, Kanamaru K (1991) Etiology of cerebral
3:284–294 vasospasm in primates. J Neurosurg
38. Dorsch N (1994) A review of cerebral vaso- 75:415–424
spasm in aneurysmal subarachnoid hemor- 52. Pluta R (2005) Delayed cerebral vasospasm
rhage. III. Mechanisms of action of calcium and nitric oxide: review, new hypothesis, and
antagonists. J Clin Neurosci 1:151–160 proposed treatment. Pharmacol Ther
39. Fox J (1983) Vasospasm. II. Clinical consider- 105:23–56
ation. In: Fox J (ed) Intracranial aneurysms. 53. Vajkoczy P, Meyer B, Weidauer S, Raabe A,
Springer, New York, pp 250–271 Thome C, Ringel F, Breu V, Schmiedek P
40. Hansen-Schwartz J, Vajkoczy P, Macdonald (2005) Clazosentan (axv-034343), a selective
RL, Pluta RM, Zhang JH (2007) Cerebral endothelin a receptor antagonist, in the pre-
vasospasm: looking beyond vasoconstriction. vention of cerebral vasospasm following severe
Trends Pharmacol Sci 28:252–256 aneurysmal subarachnoid hemorrhage: results
of a randomized, double-blind, placebo- M, Thomas JE (2000) Intrathecal sodium

controlled, multicenter phase IIa study. nitroprusside improves cerebral blood flow and
J Neurosurg 103:9–17 oxygenation in refractory cerebral vasospasm
54. Treggiari M, Walder B, Suter P et al (2001) and ischemia in humans. Stroke
Systematic review of the prevention of delayed 31:1195–1197
ischemic neurological deficits with hyperten- 59. Pluta RM, Dejam A, Grimes G, Gladwin MT,
sion, hypervolemia, and hemodilution therapy Oldfield EH (2005) Nitrite infusions prevent
following subarachnoid hemorrhage. cerebral artery vasospasm in a primate model of
J Neurosurg 98:978–984 subarachnoid aneurismal hemorrhage. JAMA
55. Dreier JP, Woitzik J, Fabricius M, Bhatia R, 293:1477–1484
Major S, Drenckhahn C, Lehmann TN, 60. Macdonald RL (2008) Clazosentan: an endo-
Sarrafzadeh A, Willumsen L, Hartings JA, thelin receptor antagonist for treatment of
Sakowitz OW, Seemann JH, Thieme A, vasospasm after subarachnoid hemorrhage.
Lauritzen M, Strong AJ (2006) Delayed isch- Expert Opin Investig Drugs 17:1761–1767
aemic neurological deficits after subarachnoid 61. Macdonald RL, Kassell NF, Mayer S,
haemorrhage are associated with clusters of Ruefenacht D, Schmiedek P, Weidauer S, Frey
spreading depolarizations. Brain A, Roux S, Pasqualin A (2008) Clazosentan to
129:3224–3237 overcome neurological ischemia and infarction
56. Leng LZ, Fink ME, Iadecola C (2010) occurring after subarachnoid hemorrhage
Spreading depolarization: a possible new cul- (CONSCIOUS-1): randomized, double-blind,
prit in the delayed cerebral ischemia of sub- placebo-controlled phase 2 dose-finding trial.
arachnoid hemorrhage. Arch Neurol Stroke 39:3015–3021
68:31–36 62. Schatlo B, Dreier JP, Glasker S, Fathi AR,
57. Sehba F, Bederson J (2006) Mechanisms of Moncrief T, Oldfield EH, Vortmeyer AO,
acute brain injury after subarachnoid hemor- Pluta RM (2010) Report of selective cortical
rhage. Neurol Res 28:381–398 infarcts in the primate clot model of vasospasm
58. Vajkoczy P, Hubner U, Horn P, Bauhuf C, after subarachnoid hemorrhage. Neurosurgery
Thome C, Schilling L, Schmiedek P, Quintel 67:721–728, discussion 728–729
INDEX
A 165–168, 170–172, 175–178, 181, 183–185, 189, 191,

192, 196–199, 201, 210–214, 216–220, 222–224, 227,
Abercrombie equation ...................................................... 116 239, 240, 242, 244, 246
Adson forceps ...................... 40, 134, 137, 143, 144, 211, 222 Brain stem ................... 78, 160, 167, 176, 177, 184–185, 191
Amyotrophic lateral sclerosis (ALS)......................... 146, 149 Bregma ...............................7, 9, 10, 32, 36, 43, 44, 48, 50, 79
Anesthesia ..................................5–9, 26, 40–43, 48, 50, 56, 67, Bulbus olfactorius ..................................................... 146, 147
128, 134–137, 139, 189, 200–204, 207–209, 219, 221, Bupivacaine ........................................................................ 58
231–234, 237–239 Buprenorphine................................ 40, 50, 67, 116, 203, 205,
Aneurismal subarachnoid hemorrhage ............ 228–232, 234, 212, 235, 237, 239, 244, 245
237–240, 243–247 Burr hole.....................44, 45, 60, 64, 168, 172, 182, 200, 222
Anterior cerebral artery (ACA) ........................ 199–201, 244
Anterior clinoid ................................................................ 240 C
Arachnoid granulations .................................................... 168
Cadaveric ................................................................ 91, 97, 98
Arteriotomy
Cannula ................................ 43, 46–48, 50, 51, 87, 176–190,
Ascorbic acid ................................................................ 33, 50
192, 200, 207, 221
Astrocytomas ...................................................................... 87
Catheter............................................ 108, 110, 176, 181–183,
Astrogliosis ............................................................... 115, 220
208, 235–237, 240, 243
Atlanto-occipital junction......................................... 200, 219
Cauda equina ............................................................ 141, 171
Atlanto-occipital membrane ..................................... 144–147
Caudate ............................................... 14, 16, 26, 70, 98, 201
Auditory brainstem responses (ABRs) .................... 153, 155,
Caudate–putamen unit (CPu) .................... 21, 32–37, 55, 63
156, 160, 161
Cavalieri method .............................................................. 116
Auditory stimulation .................................................... 74, 82
Cavitation ................................................................. 114–116
B Cell distribution ................................141–144, 146, 147, 149
Cell suspension .......................... 40, 45, 47, 48, 138, 144, 145
Baboon ..................................................14, 16, 19–21, 24–26 Central nervous system (CNS) .........................14, 16, 18, 28,
Basal ganglia ....................................................................... 37 133, 141–144, 146, 147, 149, 166, 175, 245
Basilar artery............................................................. 167, 200 Cerebellar-pontine angle (CPA)....................................... 152
Basilar artery....................................................................... 14 Cerebello-medullary fissure .............................................. 144
Biological responses ............................................................ 13 Cerebello-pontine cistern ......................................... 148, 160
Bipolar coagulation ........................................... 166, 234, 240 Cerebral blood flow (CBF)................ 213, 217, 218, 223, 228
Bipolar coagulations ......................................................... 239 Cerebral blood vessels ......................................................... 64
Blood brain barrier (BBB) .................................. 15, 141, 175 Cerebral hemispheres ....................................................... 167
Blood clot ......................................................... 228, 229, 240 Cerebro-spinal fluid (CSF) ..............................58, 61, 70, 86,
Blood flow ............................... 5, 62, 65, 66, 70, 71, 235, 239 116, 141, 142, 144, 145, 149, 160, 167, 191, 197, 211,
Blood gases ..................................................................... 6, 41 216, 227, 228, 238
Blood pressure ....................................... 6, 198, 228, 233, 237 Cerebrovascular disease
Blood-brain barrier (BBB) ............................................... 141 Cervical spinal column ..................................................... 113
Bone ................................. 4–10, 22, 28, 40, 43, 60, 78–80, 83, Chloral hydrate ..................................................................... 5
109, 111, 113–115, 121, 122, 124, 125, 130, 137, 138, Circle of Willis (CW) ...................................... 199, 201, 236
142, 144, 146, 166, 167, 170–172, 197, 205, 209–211, Cisterna magna......................................... 141–147, 149, 228
222, 227, 239, 243, 245, 246 Clazosentan .............................................................. 229, 230
Brain ....................................... 1–11, 14, 15, 17, 18, 23, 25, 27, ClearPoint ................................................ 178, 180, 186–192
28, 31, 33, 35, 37, 38, 43, 46, 47, 49–51, 61, 62, 69, 70, Clip............................................. 40, 48, 65, 66, 71, 111–117,
78, 79, 83, 85, 86, 99, 141, 144, 146–149, 153, 202, 208, 220
DOI 10.1007/978-1-4939-3730-1, © Springer Science+Business Media New York 2016
251
EXPERIMENTAL NEUROSURGERY IN ANIMAL MODELS
252 Index
Closed head injury (CHI) ................................................ 4, 5 Electrocoagulate ................................................................. 70

Cobalt-60 ........................................................................... 14 Electrocoagulation ........................... 56, 57, 59, 61–64, 66, 70
Collagen ..................................................................... 81, 224 Electrode ........................................ 31, 38, 74, 78, 79, 81–84,
Collagenase ........................................................................ 81 155–157, 159, 161
Collimator ............................................. 14, 16, 19, 20, 23, 27 Electrophysiology .....................................74, 76–77, 84, 116,
Colored fibrin glue ........................................................... 169 152, 153, 155, 159, 161
Common carotid artery (CCA) ................................ 235, 237 Endothelin-1 .................................................... 56, 57, 64–65
Computer tomography (CT) ........................86, 87, 196, 200, Endovascular .................................................................... 229
210, 212, 228 Endovascular coiling......................................................... 228
Concorde .................................................. 141–144, 146–149 Enucleation ...................................................................... 196
Connexin-43 ...................................................................... 17 Epidural venous plexus ............................................. 170–172
Controlled cortical impact (CCI) ............................... 2–8, 10 Epilepsy ................................................................ 17, 19, 153
Contusion-compression.................................... 111, 114–117 Epineurium ...................................................................... 157
Convection-enhanced delivery (CED) .................... 175–177, Excitotoxicity................................................................ 37, 41
181, 183 External carotid artery (ECA) .......................................... 219
Convex meniscus .............................................................. 145
Coordinate................................ 23, 26, 27, 31, 32, 36, 43–45, F
47, 48, 50, 51, 82, 86, 138, 139, 182, 184–185, 191 Facet joint ..........................................113, 123–125, 127, 170
Cortex......................................... 3, 10, 16, 33, 55, 63, 75, 82, Faraday cage ................................................................. 82, 84
172, 201, 220, 246 Fiducial markers ..................................................... 22, 23, 27
Corticosteroids ................................................................... 15 Filum terminale ................................................................ 171
Cranial nerves ............................ 146, 151–161, 167, 168, 170 Fisch dura retractor .................................................. 205, 222
Craniectomy ............................. 43–45, 47, 55, 56, 58–62, 64, Flavectomy ....................................................................... 171
66–68, 70, 187 Fluid percussion injury (FPI) ......................................... 3–11
Craniotomy ............................ 3, 5–10, 22, 27, 43, 60, 63, 79, Fluordesoxyglucose (FDG) .............................. 206, 214, 223
83, 98–100, 166, 167, 172, 210, 222 Fluorescence ............................................................. 147, 219
Cranium ..................................................... 31, 143, 160, 210 Foramen magnum ............................................................ 246
Cunningham adaptor ....................................... 40, 41, 48, 49 Foramen ovale .............................................................. 60, 63
Cytokine ....................................................................... 15, 18 Forceps ................................... 7, 9, 40, 43, 44, 59–62, 70, 77,
79–81, 84, 97, 99, 108–110, 114, 117, 121–123, 129,
D
137, 146, 156, 205, 207, 209, 211, 222, 234
Delayed ischemic neurological deficits (DIND)............... 228 Friedman-Pearson rongeurs.............................................. 114
Demyelination ...........................................19, 21, 24, 27, 133 Frontal lobe ............................... 14, 16, 22, 27, 240, 243, 244
Dental cement ............................................................ 7, 9, 83 Frontal sinus ............................................................. 167, 200
Diffuse axonal injury (DAI) ............................................. 4, 6 Frontotemporal skull .......................................................... 60
Diffusion tensor imaging (DTI) ...................... 206, 212, 213, Furosemide ....................................................... 239, 240, 246
215–217
Diffusion weighted imaging (DWI) ............... 68, 212, 216, 222 G
Dogs ..................................................................... 1, 197, 221 GABA ................................................................................ 37
Dopamine ............................................................... 32, 33, 38 Gadoteridol ............................... 176, 179–182, 184, 188–190
Dormicum ............................................................................ 6 Gamma knife.................................................... 14–17, 19–24
Drill ...................................... 7, 9, 22, 27, 40, 42–44, 60, 64, 74, Ganciclovir ......................................................................... 16
78–80, 98, 121, 123, 130, 134, 156, 160, 167, 170, 172, Ganglia ............................................................................... 32
201, 211, 222, 239, 240, 246 Glass micropipette .............................................................. 47
Dull spinal hook ............................................................... 111 Glioma ....................................................17, 18, 99, 153, 169
Dumont forceps.........................................7, 8, 134, 143, 144 Gliosis ................................................................................ 15
Dura mater ................................. 2, 3, 7, 9–11, 22, 27, 79, 80, Graft .......................................... 38, 45, 47–49, 130, 151, 247
139, 146, 160, 210, 211, 219 Gyrus ................................................................ 169, 201, 244
Dural sack ........................................................................ 125
H
E
Halothane ............................................................................. 5
Edema ....................................... 14, 15, 18, 86, 211, 214, 222 Hamilton ........................... 40, 43, 45–47, 110, 134, 138, 143
EEG ................................................................................... 17 Haptic......................................................91, 92, 95, 102, 103
Electrocautery............................................................... 85, 97 Hcell suspension ............................................................... 144
Index
253
Head ................................3–5, 7, 8, 10, 22, 24–27, 32, 41–43, L
48, 58, 59, 70, 74, 75, 78, 82, 87, 98, 100, 136, 143,
160, 200, 201, 205, 209, 212, 213, 219, 220, 231, 232, Lambda .......................................................43, 48, 75, 78, 79
235, 236, 238, 239, 246 Lambdoidal suture.............................................................. 77
Hematoma Laminectomized ....................................................... 108, 113
Hemiparesis ...................................................................... 198 Laminectomy............................................108, 112–114, 119,
Herniation ........................................................................ 222 122–125, 130, 136–139, 170–172
High speed drill .................................................................. 44 Laminoplasty .................................................................... 170
Hippocampus ..................................................... 2, 17, 19–21 Laminotomy ............................................................. 170, 172
Histology .............................. 14, 16, 19, 20, 24, 26, 152, 153, LASER Doppler
180, 207, 219, 220, 224 Lateral suboccipital craniectomy ...................................... 160
Hooke’s law ...................................................................... 115 Leakage ........................................................... 116, 142, 147,
Human–machine interface ....................87, 94–100, 102, 103 177, 190–192
Huntington’s disease (HD)......................... 32, 36–38, 43, 45 Leksell Gamma Plan®................................................... 23, 24
Hypothalamus .................................................................... 26 Lenticulo striate artery (LSA) ................................ 61, 65, 70
Hypothermia .......................................................... 6, 41, 136 Lesion........................................... 5, 14, 16, 19–21, 27, 32–36,
38–40, 43, 45, 49–51, 66, 70, 85, 86, 89, 107–112, 114,
I 115, 117, 149, 151–153, 159, 161, 170–172, 176, 211,
212, 214, 220, 245
Image-guided ....................................................... 91, 92, 103
Ligature ........................................................ 65, 71, 161, 205
ImageJ .............................................................. 206, 214, 223
Lilliequist membrane........................................................ 244
Imaging ...................................... 6, 14, 16, 26, 28, 31, 85–87,
Linear accelerator (LINAC) ............................. 14–16, 19, 20
90–92, 95, 96, 99, 102, 103, 179, 181, 184, 189, 190,
Liposome .......................................................... 176, 178–182
196, 197, 202–204, 206, 208–209, 212–219, 221–223
Lobus piriformis ............................................................... 200
Inferior cerebral vein (ICV) ............................. 61, 63, 64, 66
Locomotor recovery.......................................................... 107
Inferior colliculus (IC) .................................74, 75, 78, 80, 83
Lumbar ............................................................................. 171
Informatic surgery .............................................................. 86
Lumbar puncture .............................................................. 228
Injury .......................................... 1–11, 14, 15, 28, 44, 68, 97,
Lumbar spine ................................................... 120–129, 131
125, 133, 141, 144, 151–154
Lumbosacral junction ............................................... 123, 130
Internal capsule (IC) ........................................... 14, 16, 183,
186, 201 M
Internal cerebral artery (ICA)........................................... 244
Intervertebral space .......................................... 136–138, 171 Macaques.................................................................. 178, 245
Intracerebral ................................................... 14, 32, 40, 178 Magnetic resonance imaging (MRI) .................... 15, 16, 19–21,
Intracerebral cell ................................................................. 38 25, 68, 176, 178–180, 182–188, 190, 192, 196, 201, 202,
Intracranial pressure (ICP) ........................... 5, 211, 222, 228 206, 212, 213, 215, 216, 218, 223
Intraventricular ................................................. 141, 142, 227 1-Methyl-4-phenyl-1,2,5,6-tetrahydropyridine
Irradiation......................................................... 14–17, 19–21 (MPTP).................................................................. 32
Irrigation-suction device................................................... 123 Medial forebrain bundle (MFB)....................... 32–36, 43, 50
Ischemia ...................................................55, 62–65, 68, 115, Microaneurysm .................................................................. 71
195–224, 229 Microaneurysm clips .......................................................... 66
Ischemic lesion ..................................................... 55, 70, 196 Microdialysis .................................................................. 6, 38
Isodose line ................................................................... 23, 27 Microforceps........................................ 64, 157, 239, 240, 244
Isoflurane ............................ 5, 40–42, 56, 107, 111, 116, 134, Micromanipulator ........................................ 74, 82, 142, 144
135, 179, 182, 187, 202, 203, 208, 221, 232, 234, 239 Microneurosurgery ..................................................... 91, 172
Microtransplantation .................................................... 47–49
J Middle cerebral artery (MCA) ........................... 55–57, 59, 60,
62, 64–70, 196, 200, 201, 210, 211, 213, 220, 222, 229,
Jet nozzle .......................................................................... 155
232, 236–238, 240–244
Jugular vein ....................................................... 167, 207, 208
Middle cerebral artery occlusion (MCAO) ................. 55, 56,
K 59, 60, 62, 64, 66, 68, 70, 196, 197, 199, 202–224
Middle meningeal artery .................................................. 246
Kainic acid .......................................................................... 17 Midline incision ...........................................43, 97, 121, 167,
Ketamine .................................. 27, 40, 41, 73, 119, 142, 143, 170, 172, 245
179, 183, 187, 200, 203, 208, 221, 231, 234, 239 Minimally invasive ............................................................. 13
Kyphosis ................................................... 116, 119, 120, 138 Mini-osmotic pump ................................................. 107, 110
254 Index
Monkey ............................................ 19, 21, 24–26, 178, 186, Penfield retractor .............................................................. 240
230–231, 233, 235–237, 239, 244–247 Pentobarbital ..................................... 5, 15, 83, 204, 207, 219
Morris water maze.............................................................. 19 Pericallosal artery...................................................... 230, 244
Motorized injector............................................ 134, 138, 139 Peripheral nerves .............................................. 131, 151–161
Mouse............................................ 22, 27, 111, 143, 147, 216 Pia/arachnoidea mater ...................................................... 153
MR Operating Theatre ...................................................... 96 Picospritzer........................................................................... 4
Myelin .............................................................................. 111 Poly ornithine coated monofilament
Myelotomy ....................................................................... 171 Pons .............................................................................. 14, 16
Positron emission tomography (PET) ............. 196, 206, 208,
N 213–214, 217–219, 222–223
Nanoparticles.............................................................. 98, 176 Posttraumatic ischemia ..................................................... 115
Nasion .............................................................................. 143 Preclinical research ........................................................... 196
Necrosis ............................................................ 14, 15, 19, 22 Primate ....................... 22, 176–192, 197–199, 224, 228–232,
Needle tip ................................................... 51, 138, 144, 145 234, 237–240, 243–247
Neocortex ........................................................... 33, 201, 221 Propofol ................................................ 24, 25, 200, 208, 221,
Neonatal ................................................32, 36, 38–41, 48–50 232, 234, 239
Neuroanatomy .......................................... 166–170, 196, 230 Pterygopalatine artery (PPA)............................................ 246
NeuroArm ............................................................ 87, 91–102 Pyramidal tracts ................................................................ 198
Neurodegeneration ........................................................... 133
Q
Neurodegenerative disorder .......................................... 32, 36
Neuroinflammation .......................................................... 115 Quinolinic acid (QA) ...................... 37, 39–41, 43, 45, 46, 51
Neuronal regeneration ...................................................... 119
Neuropathic pain .............................................................. 151 R
Neuroprotective ........................................................ 5, 6, 196 Rabbit ................................................................. 14, 153, 196
Neurosurgeon ........................................ 85, 98, 165, 186, 192 Radiation ................................. 13–17, 23, 24, 27, 38, 84, 213
Neurosurgery .................................. 13, 28, 85–87, 91, 92, 94, Radiosensitivity ............................................................ 13, 15
96–100, 102, 103, 151–161, 165, 189 Radiosurgery................................................................. 23–25
Neurotransplantation........................................ 31, 32, 38, 40 Rat ......................................1–3, 5, 7, 10, 14, 16–18, 20–24, 26,
Nitrous oxide .................................................. 5, 56, 112, 221 32–33, 36–39, 41, 48, 50, 56, 66–70, 73–76, 78–80, 82,
NMDA receptor ........................................................... 37, 40 83, 88, 107–112, 117, 119–121, 129, 130, 133–139,
Non-human primate (NHP) ................... 176–178, 182–187, 153, 154, 160, 161, 182, 186, 196, 221
189–192, 196, 224, 228–232, 234, 237–240, 243–247 Recombinant tissue plasminogen activator ....................... 195
Nucleus accumbens ...................................................... 33, 35 Reflux ................................................ 177, 181, 190, 192, 208
Regional cerebral blood flow (rCBF) ............................... 225
O
Reperfusion .................................................................. 65, 66
Occipital protuberance ............................................. 113, 238 Rete mirabile .................................................................... 201
6-OHDA ........................................ 21, 32–36, 40, 43, 45, 50 Retroorbital space ..................................................... 211, 221
Olfactory bulb .......................................................... 167, 201 Rhesus monkey........................................................... 19, 186
Olfactory tract ............................................ 61, 62, 64–67, 70 Robotics.............................................86–91, 96, 97, 100–103
Operating microscope .......................... 39, 40, 43, 58, 65, 71, Robots ......................85, 86, 91, 92, 94, 96–98, 100, 102, 103
85, 97, 99, 119, 142, 144, 146, 239–241 Rodents ................................ 1, 4, 5, 8, 31–33, 35–38, 41, 49,
Operating microscopy ...................................................... 142 50, 55, 97, 102, 134, 196
Optic nerve ....................................................................... 247
Orbitae ............................................................................. 166 S
Orbital ridge ............................................................. 239, 240 Sagittal sinus............................................... 78, 166–169, 172
Ordered Subset Expectation Maximisation Sagittal suture ....................................................... 32, 77, 166
(OSEM) ............................................................... 214 Saline .....................................7, 33, 46, 48, 50, 58, 60, 67, 70,
OsiriX ............................................................... 184, 200, 223 116, 130, 135, 137, 144, 145, 147, 154, 159, 179, 180,
204, 232, 234, 235, 237, 243, 244
P
Scalpel .................................7, 40, 43, 77, 107, 113, 121, 122,
Papaverine ........................................................................ 235 130, 134, 137, 143, 144, 155–157, 205, 234, 239
Paravertebral muscles ............................... 121, 126, 128, 137 Sciatic function index (SFI) .............................................. 152
Parkinson’s disease (PD) .............................32–33, 35, 36, 45, Sciatic nerve ......................................151, 153–157, 159, 161
176, 178, 192 sheep.................................. 196–201, 212–214, 216, 220–222
Index
255
Sheep ........................................ 196, 197, 199, 202, 205–209, T
211–215, 217, 219, 221–224
Short-hairpin RNA ............................................................ 38 Telencephalon............................................................. 78, 166
Sigmoid sinus ................................................................... 160 Temporal lobe...................................... 20, 240, 241, 244, 247
Skull ..................................3–9, 22–24, 32, 41, 43, 44, 48, 56, Temporal muscle ...............................6, 59–62, 209–211, 221
60, 69, 74–84, 146, 161, 166, 167, 182, 187, 197–201, Thalamus ..................... 2, 14, 16, 26, 177, 184–185, 189–191
210, 211, 219, 227, 242, 246 Thrombolysis.................................................................... 195
Skull base..................................... 28, 59–60, 63, 70, 146, 149 Thrombus ......................................................................... 228
SmartFrame .......................................178, 180, 186–188, 191 Thymidine kinase (TK) ...................................................... 16
Sphenoid wing bone ......................................................... 240 T-maze ............................................................................... 38
Spinal cord.......................... 14, 108, 110, 114–115, 121–129, Trajectory ...................43, 45, 47, 98, 101, 178, 186–189, 191
134–139, 141, 146–149, 165–167, 170–172 Transplantation........................ 42, 45–49, 141–144, 146–149
Spinal cord impactor ........................................................ 109 Transplantation coordinates ............................................... 49
Spinal cord injury (SCI) .................................. 107, 108, 110, Transverse processes ......................................... 137, 146, 170
111, 114–117, 133 Transverse sinus .......................................................... 48, 160
Spinal process ................................... 113, 114, 116, 121–122, Trauma ...............................2, 3, 5–7, 9, 10, 47, 133, 152, 246
130, 136, 167, 170 Traumatic brain injury (TBI) ......................................... 1–11
Spine ......................................... 119, 123, 143, 146, 170–172 Trephine ....................................................................... 5, 7, 9
Spinous process ......................... 107, 108, 121–123, 137, 138 Trigeminal nerve .................................................... 19, 20, 26
Spiny neurons ............................................................... 37–39 Trigeminal neuralgia..................................................... 17, 19
Stepped cannula ............................................... 177, 181–185 2,3,5-Triphenyltetrazolium chloride (TTC) ................. 68, 69
StereoInvestigator ............................................................. 116 Trypan blue ................................... 45, 49, 180, 182, 188, 189
Stereologic Tumor ............................................. 15, 17, 18, 23, 27, 28, 38,
Stereotactic apparatus ............................ 31, 32, 39–43, 45, 48 169, 172, 176, 178, 228
Stereotactic frame ................................... 2, 22, 23, 27, 40–43, Tumor necrosis factor alpha (TNF-α) .......................... 15, 17
46, 143, 160, 179, 182, 183, 187 Tyrosine hydroxylase (TH) ..................................... 34, 36, 39
Stereotactic guidance .................................................. 13, 178
V
Stereotactic surgery ................... 31, 32, 34–41, 43–45, 47–52
Stereotaxis .......................................................................... 31 Vascular response ........................................................ 15, 238
Stop flow phenomenon .................................................... 228 Vasospasm ......................... 228–232, 234, 236–240, 243–247
Striatum.......................................... 22, 27, 32–35, 38, 39, 41, Ventral tegmental area (VTA) ...................................... 33–36
43, 50, 52, 186 Vertebra .............................................107–109, 120, 126, 137
Sub-occipital puncture...................................................... 238 Vertebra’s pedicles ............................................................. 125
Substance P ........................................................................ 37 Vertebral bodies ........................................................ 115, 146
Substantia nigra (SN) ....................................... 19, 20, 32–34 Vestibulocochlear nerve ............................ 152–154, 156, 160
Suction ............................................................ 154, 160, 166, Vibrating saw ................................................................... 246
169, 171, 239
Suicide gene therapy (SGT) ............................................... 17 W
Surgery ......................................... 5, 6, 31, 36, 40, 43, 49, 50, Waterjet dissection ................................... 152, 153, 157–161
56, 58, 59, 67, 70, 74, 76–77, 83, 86, 88–91, 93, 95, 97, Weight drop injury (WDI) ................................ 4, 6, 7, 9–11
100, 102, 103, 119, 120, 123, 124, 126–128, 130, Williger raspatory ............................................. 205, 210, 211
136–137, 146, 153, 154, 157, 159–161, 166–172, 182,
197, 200, 202–206, 208–212, 221–222, 231, 232, 239, X
243, 245, 247
X-ray ....................................... 22–24, 27, 28, 85, 86, 88, 245
Surgical ............................................................................. 6, 7
Xylazine ........................................ 40, 41, 142, 143, 160, 179,
Surgical clipping ............................................................... 228
183, 187, 200, 203, 208, 221, 234, 239
Sutura coronalis .................................................................. 79
Sutura lamboidea ................................................................ 79 Z
Sutura sagittalis .................................................................. 79
Swine ................................................. 165, 166, 170, 171, 196 Zygoma .............................................................................. 59
Sylvian fissure ....................................239–241, 243, 244, 247 Zygomatic arch ............................................. 59–61, 239, 240

Experimental Neurosurgery in Animal Models

Uploaded by

Copyright:

Available Formats

Experimental Neurosurgery in Animal Models

Uploaded by

Document Information

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Experimental Neurosurgery in Animal Models

Uploaded by

Copyright:

Available Formats

Neuromethods 116

Miroslaw Janowski Editor

For further volumes:

ISSN 0893-2336 ISSN 1940-6045 (electronic)

Library of Congress Control Number: 2016939947

© Springer Science+Business Media New York 2016

Printed on acid-free paper

This Humana Press imprint is published by Springer Nature

Saskatoon, Canada Wolfgang Walz

Baltimore, MD, USA Miroslaw Janowski

Preface to the Series. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . v

1 Animal Models of Traumatic Brain Injury . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1

14 Real-Time Convection Delivery of Therapeutics to the Primate Brain . . . . . . . 175

JOHN BACHER • Division of Veterinary Resources, Office of Research Services,

MIROSLAW JANOWSKI • The Russell H. Morgan Department of Radiology and Radiological

GARNETTE R. SUTHERLAND • Division of Neurosurgery, Seaman Family MR Research

Animal Models of Traumatic Brain Injury

Animal welfare is a big issue in several countries and it is of

extended apnea or death. CFPI is used when a more diffuse axonal

This is the most heterogeneous model of the three discussed in

1.3 General To be able to perform the craniotomy and other procedures on

countries due to animal welfare concerns. Interestingly, pentobar-

1.3.4 Physiological In shorter experiments (less than 30 min anesthesia), physiological

Fig. 4 Schematic drawing of the exposed rodent skull

Fig. 5 Most common placement of the craniotomy for FPI or CCI

Fig. 6 Possible locations for the anchoring screws

4. Use the midline and bregma sutures as reference points to

3.3 Weight 1. Sedate the animal.

4. Remove connective tissue and periost from the skull bone.

trauma coupling then has to be removed and cleaned or discarded,

4.3 Weight To make the experiments as reproducible as possible it is important

Experimental Radiosurgery in Animal Models

Radiosurgery is a surgical technique which is designed to produce

study the biological effects of radiosurgery in both normal and

3.5 weeks. Kamiryo et al. studied the radiosensitivity of brain by

[15–20]. In addition, locally produced TNF-α has been reported to

1.5 Functional Radiosurgery is rapidly expanding beyond its use as a treatment of

different from controls. Liscak et al. [26] evaluated the effects of

Vincent et al. performed hypothalamic radiosurgery on geneti-

2 Materials and Methods

6. Target is defined based on its predetermined distance from the

animals. This frame is widely used clinically for stereotactic

9. The images are transferred to Gammaplan computer via ether-

1. Anesthesia is a critical component of small animal radiosurgery.

last up to 3 h. While higher doses of anesthetic agents can kill

4 Future Experimental Radiosurgery

Rapid developments in computer hardware and software, imaging

1. Larsson B, Leksell L, Rexex B, Sourander P, Biological effects of stereotactic radiosurgery

Stereotactic Surgery in Rats

Stereotactic surgery has been an invaluable tool applied since early

Rat Brain in Stereotaxic Coordinates [3]. The coordinates describe

Human idiopathic PD is a progressive neurodegenerative disorder,

(AP) and ML coordinates and the dura as reference for the DV

3 Injection of 6-OHDA into the CPu

6-OHDA injection into the CPu, referred to as “partial” or “ter-

the dorsal part of the CPu is characterized by a mixed DA innerva-

4 Injection of 6-OHDA in the MFB

Unilateral 6-OHDA injection into the MFB leads to an almost

5 Injection of 6-OHDA into the SNc