Review Article

Graft Tensioning During Knee

Ligament Reconstruction:
Principles and Practice

Abstract
Seth L. Sherman, MD Failure to correctly tension grafts may overconstrain or
Peter N. Chalmers, MD underconstrain the knee, potentially predisposing the patient to
deteriorating clinical and/or radiographic results over time. Knee
Adam B. Yanke, MD
ligament reconstruction requires a fundamental understanding of
Charles A. Bush-Joseph, MD native anatomy, ligament biomechanics, and principles of graft
Nikhil N. Verma, MD tensioning. A successful strategy for graft tensioning takes into
Brian J. Cole, MD account the specific biomechanics of the ligament or ligaments in
question, the mechanical properties of the graft selected, the
Bernard R. Bach, Jr, MD
chosen fixation method, the selected tensioning method (ie,
manual or mechanical), and the overall goal of the reconstruction
(ie, isometry versus anisometry).

K nee ligament reconstructions are

among the most commonly per-
formed orthopaedic surgeries.1 De-
mechanical properties of the selected
graft and fixation method and to de-
termine the biomechanical goals of
spite the widespread utility of these the reconstruction.
procedures, variation exists regard-
ing appropriate intraoperative graft
tensioning.2-9 Although clinical evi-
Biomechanical Basis of
dence is largely equivocal or lack-
ing,5,8 the theoretic implications of
Graft Tensioning
improper graft tensioning are sub-
stantial, potentially leading to sub- Anatomy
optimal clinical and radiographic A well-executed knee ligament re-
results over time. Graft underten- construction should restore the pa-
sioning could lead to residual laxity, tient’s native anatomy. Although the
with a graft that behaves biomechan- details of footprint and bundle anat-
ically similar to a ligamentously defi- omy are beyond the scope of this re-
From the University of Missouri, cient knee.7 Conversely, graft over- view, it must be understood that
Columbia, MO (Dr. Sherman) and tensioning can lead to flexion graft tensioning is immaterial if pri-
Rush University Medical Center, contracture, increased ligamentous mary fixation points are nonana-
Chicago, IL (Dr. Chalmers,
stress, graft breakdown, subluxation tomic. For example, nonanatomic
Dr. Yanke, Dr. Bush-Joseph,
Dr. Verma, Dr. Cole, and Dr. Bach). of the tibia, and increased articular vertical anterior cruciate ligament
contact pressures.2,10-15 (ACL) grafts placed superior or
J Am Acad Orthop Surg 2012;20:
633-645 To optimize graft tensioning for proximal to the femoral footprint
knee ligament reconstruction, the fail more frequently, predisposing the
http://dx.doi.org/10.5435/
JAAOS-20-10-633 surgeon must have a thorough ap- knee to greater anterior laxity and
preciation for native knee anatomy, loss of flexion.16 The surgeon must
Copyright 2012 by the American
Academy of Orthopaedic Surgeons. biomechanics, and kinematics. It is first choose the anatomic target
critical to understand the specific structure to restore. Subsequent graft

October 2012, Vol 20, No 10 633

Graft Tensioning During Knee Ligament Reconstruction: Principles and Practice

tensioning is dependent on replicat- ion at the time of ligamentous ten- ness is defined as force per unit area
ing the specific biomechanical func- sioning is extremely important and of lengthening. A graft with a higher
tion of that ligament or bundle to will affect the biomechanical behav- stiffness value requires more force to
best recreate native knee kinematics. ior of the reconstruction. For exam- create the same degree of lengthen-
ple, the posterolateral bundle of the ing. Of the commonly used grafts,
Isometry and Anisometry ACL is physiologically tense in ex- bone–patellar tendon–bone (BPTB)
To develop a rationale for appropri- tension and lax in flexion.22 If this grafts have the greatest stiffness (Fig-
ate graft tensioning, the surgeon ligament is tensioned in extension, ure 1). Because BPTB grafts are
must understand several key biome- when the origin and insertion sites stiffer, less tension is required to re-
chanical concepts. Knee ligaments are farther apart, then relative liga- create normal kinematics. With a less
are divided into isometric and aniso- mentous laxity will be experienced stiff graft, more stretch is experi-
metric structures. Isometric liga- when the knee is brought into flex- enced with regular knee motion. This
ments are of equal length and ten- ion, as occurs with native kinemat- may be perceived by the patient as
sion regardless of the angle of knee ics. However, if this ligament is ten- increased laxity. To avoid this prob-
flexion because the distance between sioned in flexion, when the native lem, greater tension must be applied
their origin and insertion sites does ligament is less tense and shorter, at the time of initial fixation. For ex-
not change with knee flexion. As- then graft tension when the knee is ample, in cadaver knees, the tension
suming anatomic positioning in an brought to extension will exceed na- requisite to recreate anatomic knee
isometric graft, the angle of knee tive ligament tension, which may ACL tension is 16 N for BPTB con-
flexion at the time of tensioning lead to stiffness, flexion contracture, structs and 38 N for the doubled,
should not affect the kinematics of or graft attenuation. We generally non–pre-tensioned semitendinosus
the reconstruction. recommend tensioning at the posi- tendons.29 In clinical study, a recent
In anisometric structures, such as tion of maximum physiologic ten- meta-analysis demonstrated postop-
the ACL and its bundles,11,17 the pos- sion to recreate native laxity. Under- erative laxity, defined as a KT-1000
terior cruciate ligament (PCL) and its standing anisometry allows the arthrometer (Medmetric, San Diego,
bundles,3 and the posterolateral cor- surgeon to choose the appropriate CA) side-to-side difference of >3
ner of the knee (PLC),18-20 the length tension angle, thus preserving range mm, to be less prevalent in patients
and tension of the construct changes of motion while maximizing knee who receive BPTB compared with
with knee flexion. For instance, stability. those who receive hamstring grafts.1
placement of the femoral footprint at This less prevalent laxity may be be-
the roof of the notch is not only Biomechanical cause of the combination of stiffness
nonanatomic, but it also results in Considerations in Graft and stress relaxation phenomena.
anisometric reconstruction.21 This Selection These results have led several au-
can be easily demonstrated during A variety of grafts has been used for thors to suggest that hamstring grafts
ACL reconstruction (ACLR) by flex- knee ligament reconstruction. These should be more highly tensioned
ing and extending the knee after fem- graft choices are differentiated by than BPTB grafts to avoid postopera-
oral fixation; graft recession in the their biomechanical properties, most tive excess laxity.5,30
tibial tunnel is noticeable during ex- notably their stiffness and viscoelas- Viscoelastic activity is another crit-
tension. Therefore, the angle of flex- tic activity. With tensile forces, stiff- ical concept in graft selection. Vis-

Dr. Bush-Joseph or an immediate family member serves as an unpaid consultant to The Foundry and as a board member, owner,
officer, or committee member of the American Orthopaedic Society for Sports Medicine. Dr. Verma or an immediate family member
serves as a paid consultant to or is an employee of Smith & Nephew; has stock or stock options held in Omeros; has received
research or institutional support from Arthrex, Athletico, ConMed Linvatec, DePuy Mitek, MioMed Orthopaedics, and Smith & Nephew;
has received royalties from Smith & Nephew; and serves as a board member, owner, officer, or committee member of the American
Orthopaedic Society for Sports Medicine and the Arthroscopy Association of North America. Dr. Cole or an immediate family member
is a member of a speakers’ bureau or has made paid presentations on behalf of Genzyme; serves as a paid consultant to or is an
employee of AlloSource, Arthrex, BioMimetic Therapeutics, Carticept Medical, DePuy, and Zimmer; has received research or
institutional support from Arthrex, DJ Orthopaedics, Regentis Biomaterials, and Smith & Nephew; and has received royalties from
Arthrex, DJ Orthopaedics, Elsevier, and Lippincott. Dr. Bach or an immediate family member has received nonincome support (such
as equipment or services), commercially derived honoraria, or other non–research-related funding (such as paid travel) from Arthrex,
ConMed Linvatec, Linvatec, Össur, and Smith & Nephew. None of the following authors or any immediate family member has
received anything of value from or has stock or stock options held in a commercial company or institution related directly or indirectly
to the subject of this article: Dr. Sherman, Dr. Chalmers, and Dr. Yanke.

634 Journal of the American Academy of Orthopaedic Surgeons

 Seth L. Sherman, MD, et al

coelastic activity is defined as the Figure 1

nonlinear, time-dependent change in
strain (ie, change in length over unit
length) in response to a constant
stress (ie, force per unit area). Creep
and stress relaxation are important
types of viscoelastic behavior. These
refer, respectively, to the permanent
and nonpermanent lengthening of a
ligament in response to axial tension.
Because of a ligament’s viscoelastic-
ity, the tension in a ligament at the
time of intraoperative tensioning
may differ substantially from the ten-
sion on that structure during physio-
logic load after equilibration. For ex-
ample, hamstring grafts exhibit
viscoelastic behavior and postopera-
tive graft stretching. Semitendinosus
and gracilis grafts loaded at 65 N for
Stiffness for various grafts used in anterior cruciate ligament (ACL)
15 minutes lose up to 50% of their reconstruction. Data sources: bone–patellar tendon–bone,23 quadrupled ham-
original tension, whereas primate pa- string,24 doubled semitendinosus,24 quadriceps tendon,25 doubled gracilis,24
tellar tendons loaded at an average doubled tibialis anterior,26 native ACL,23 Achilles tendon,27 semitendinosus,28
gracilis,28 and tibialis anterior.26
of 50 N for 10 minutes lost only a
maximum 30% of their original ten-
sion.28,31 In a cadaver ACLR study of
doubled hamstring tendons, more than tensioning of soft-tissue allografts N/mm for the interference screw/
half of the tendon tension was lost, and and autografts. Our practice is to use WasherLoc (Biomet, Warsaw, IN)
KT-1000 translation notably increased a graft tensioning board, tensioning combination.14 In clinical practice,
in response to intraoperative knee cy- the graft with maximum one-hand these variations may manifest as dif-
cling, with KT-1000 values of 5.8 mm pull. The graft remains on this board ferences in postoperative laxity. In
for the intact knee, 8.1 mm for the re- for the portion of the operation be- one series of 60 nonrandomized (ie,
constructed knee, and 10.5 mm for the tween harvest and graft passage. The by order of appointments instead of
post-cycled knee.32 rate of tension loss is lower in BPTB by prepared, sealed, opaque en-
Pre-tensioning the graft may pre- grafts, obviating the need for pre- velopes) patients who underwent
vent ligament lengthening, loss of tensioning. Alternatively, fixing the ACLR with hamstring grafts, there
tension, and development of postop- femoral side and manually overten- was a difference in objective radio-
erative laxity. An in vivo study of sioning the tibial side with repetitive graphic outcomes depending on fixa-
quadrupled semitendinosus/gracilis cycling of the knee through a com- tion with interference screw versus
grafts showed that pre-tensioning plete range of motion can dynami- transcondylar cross-pins.33 The use
with 50 intraoperative flexion/ cally pre-tension a graft.4,5,7,8,11,25,32,33 of aperture fixation (ie, BPTB with
extension cycles from zero degrees to intra-articular screw fixation) versus
110° decreased lengthening by 7.7 Biomechanical suspensory fixation (ie, cortical but-
mm.33 In vivo lengthening of >0.5 Considerations in Fixation ton for quadrupled hamstring grafts)
mm has been demonstrated in BPTB Method may increase overall construct stiff-
grafts intraoperatively with similar Graft fixation strength varies sub- ness, may allow for earlier and ag-
cycling.34 These findings suggest that stantially between fixation types gressive range of motion, and may
pre-tensioning may help reduce the (Table 1). Differences in fixation contribute to decreases in postopera-
development of postoperative laxity, stiffness can reach an order of mag- tive laxity.35 Suspensory fixation also
especially in hamstring grafts. Clini- nitude: 18 N/mm for the EndoButton may subject grafts to the so-called
cal trials are lacking, however. (Smith & Nephew, Memphis, TN)/ windshield wiper effect: because of
The literature supports pre- suture post combination to 269 greater distance between the fixation

October 2012, Vol 20, No 10 635

Graft Tensioning During Knee Ligament Reconstruction: Principles and Practice

Table 1
Fixation-type Stiffness in Anterior Cruciate Ligament Reconstruction (in N/mm)
No. 5 Sutures One 20-mm Two Tandem Metal Interfer-
Femoral Tied to Post at Double Staples Washer at Washers at ence Screw WasherLoca at
Fixation 70 N/mm at 174 N/mm 192 N/mm 318 N/mm at 340 N/mm 506 N/mm

EndoButtonb at 24 18 21 21 22 22 23
N/mm
Mitekc anchor at 19 23 23 24 24 25
26 N/mm
Bone mulch screw 62 134 144 205 214 269
with bone com-
paction at 575
N/mm

a
Biomet, Warsaw, IN
b
Smith & Nephew, Memphis, TN
c
DePuy Mitek, Raynham, MA
Adapted with permission from Karchin A, Hull ML, Howell SM: Initial tension and anterior load-displacement behavior of high-stiffness anterior
cruciate ligament graft constructs. J Bone Joint Surg Am 2004;86(8):1675-1683.

point and the joint, the graft can the Tension Isometer (Medmetric) itself.15,39,40 In a canine model, in-
move back and forth within the tun- Graft Tensioner (Arthrotek, Warsaw, creasing tension within the native
nels during motion cycles, poten- IN), and the Intrafix device (DePuy ACL from baseline tension to >20 N
tially increasing laxity over time. As Mitek, Norwood, MA). The maxi- baseline tension led to focal degener-
a rule, aperture fixation methods mum tension applied with a single ation, increased vacuolization, more
avoid problems related to the wind- hand pull by a sports medicine sur- coarse and less oriented collagen fi-
shield wiper effect from micromotion geon is 99 N.37 In a cadaver study, bers, and a significant decrease in
by providing secure intra-articular hand tensioning versus device ten-
tensile strength at 12 weeks com-
fixation at the anatomic origin and sioning to 110 N did not affect post-
pared with native tendons.15 In a ca-
insertion of the graft.36 fixation laxity.38 Clinical studies that
nine BPTB ACLR model comparing
An important caveat of tibial aper- compare these devices with hand ten-
ture fixation is the difficulty of hard- sioning are lacking. If a surgeon de- 1 N to 39 N of tension, increased
ware removal in a revision situation. sires to tension to >100 N, a device tension caused myxoid degeneration,
If a surgeon chooses to use a con- may be recommended. In addition, if poor vascularity, and nonstatistically
struct with less stiff fixation, then su- the surgeon wishes to reproducibly significant decreases in load-to-
praphysiologic tension may be re- tension to a submaximal single hand failure strength at 3 months.40 Con-
quired to reproduce physiologic pull—which may be difficult to reli- versely, in a rabbit BPTB ACLR
laxity. Use of high-stiffness graft and ably reproduce—a tensioner may be model comparing tensions of 1 N,
fixation methods requires lower in- recommended. Although these de- 7.5 N, and 17.5 N, increased tension
traoperative tension to reach the de- vices may improve the reproducibil- caused no difference in cellularity,
sired native knee ligament tension.14 ity of tension applied, no biomechan- cell nucleus volume, or vascularity
In general, time zero graft fixation ical or clinical evidence indicates that after 32 weeks.39 These results sug-
parameters with most available fixa- there is clinical benefit to the use of a gest that supraphysiologic tendon
tion devices exceed physiologic loads tensioning device. We prefer using a tension may lead to detrimental bio-
in the early postoperative period, single hand-pull in the place of a ten- logic tendon changes, whereas ten-
leading to a very low incidence of re- sioning device, even though clinical don tension within normal range
ported early clinical failures. evidence to support this preference is likely does not have similar effects.
lacking. The clinical correlation of these find-
Method of Tensioning ings in humans is unknown. Sur-
Several devices have been developed Biologic Response to Graft geons should be aware that over-
to aid surgeons in the application Tension tensioning may have a biologic
and measurement of graft tension at Several studies have examined the bi- consequence that is currently poorly
the time of fixation. These include ologic effects of tension on the graft understood.

636 Journal of the American Academy of Orthopaedic Surgeons

 Seth L. Sherman, MD, et al

Figure 2
Clinical Approach to Graft
Tensioning

Anterior Cruciate Ligament

Native Kinematics
Understanding the biomechanical
properties of the native ACL helps
guide reconstruction. The native
ACL has a stiffness of 182 N/mm
and a load-to-failure of 1,725 N.23 In
cadaver models, considering both
bundles as a single unit, the ACL is
Cadaver specimens demonstrating anterior cruciate ligament (ACL) anatomy.
physiologically lax at 10° to 40° of
A, Axial view. The area outlined in black at the top is the posterior cruciate
flexion and physiologically tense (36 ligament footprint. The area outlined at the bottom is the ACL footprint.
to 56 N) at full extension because B, Sagittal view demonstrating the origin and insertion of the ACL.
the femoral origin lies posterior to (Reproduced with permission from Amis AA, Jakob RP: Anterior cruciate
ligament graft positioning, tensioning and twisting. Knee Surg Sports
the femoral center of rotation.11,17 Traumatol Arthrosc 1998;6[suppl 1]:S2-S12.)
Thus, tensioning at midflexion could
overconstrain the knee in extension,
whereas tensioning at extension iofemoral contact pressures.11 Several sioning in full extension or using
could theoretically result in laxity at randomized clinical trials have com- submaximal loading at 30° of flex-
midflexion. pared the effects of tensioning re- ion helps avoid overconstraint.10,30 If
The anatomic origin/insertion of gimes on clinical outcomes (Table 2). soft-tissue grafts are used, we prefer
the ACL is depicted in Figure 2. The These results suggest that, if tension- to use autograft quadrupled ham-
ACL is an anisometric structure. The ing is performed at 30°, hamstring string tendons24 with suspensory fix-
ligament is functionally divided into tendons may require 80 N of tension ation on the femur, and cortical
anteromedial and posterolateral bun- but do not benefit from further ten- screw and back-up staple/anchor fix-
dles. Tension within the anterome- sion.5 BPTB grafts fixed at 30° of ation on the tibia,14,35,36 tensioned at
dial bundle varies significantly less flexion are unlikely to require >20 N full extension with a maximum sin-
than does tension in the posterolat- of tension, but if fixed at full exten- gle hand pull,5,37 after first pre-
eral bundle, reflecting its more iso- sion, tensioning to 90 N may de- tensioning on a graft preparation
metric origin; but the anteromedial crease postoperative excess laxity.7,8 board to minimize stress relax-
bundle is slightly more lax in exten- Surgeons should be careful about ap- ation.30 Tensioning patterns do not
sion and tense in flexion, with high- plying high tension in flexion be- differ for BPTB or soft-tissue al-
est tension at 60°.22 The posterolat- cause of the risk of flexion contrac- lograft.42
eral bundle exhibits notably greater ture and overconstraint.10 Cadaver
variation in tension with knee flex- models have also revealed extensive Double-bundle Repairs
ion, reflecting its less isometric ori- stress relaxation for hamstring Notably fewer cadaver studies, and
gin, with laxity in flexion and ten- grafts,30 suggesting that without ex- no clinical studies, have been per-
sion in extension greatest at zero tensive pre-tensioning, higher initial formed to examine tensioning proto-
degrees to 15°.22 tension must be applied to these cols for double-bundle repairs. Bio-
grafts to predictably achieve similar mechanical evidence suggests that
Single-bundle Repair final laxity to BPTB grafts. tensioning bundles separately at their
Numerous cadaver biomechanical Although unequivocal high-quality individual positions of laxity, in
studies have examined single-bundle evidence is lacking to make strongly order to maximally constrain the
ACLR with various tensioning pro- supported recommendations, we pre- knee, leads to high individual graft
tocols. These studies have shown fer to perform autograft BPTB tensions and poor “reciprocity,”
that tensioning at 30° of flexion ACLR with interference screw aper- whereas tensioning both grafts at
leads to increased risk of flexion con- ture fixation, tensioned at full exten- 20° of flexion may ameliorate these
tracture,10 supraphysiologic graft sion, with a maximal single hand problems.12,13,43 Three-dimensional
tensions in extension,30 and high tib- pull (~99 N).7,8,10,37 Maximum ten- model data based on dual orthogo-

October 2012, Vol 20, No 10 637

Graft Tensioning During Knee Ligament Reconstruction: Principles and Practice

Table 2
Randomized Clinical Trials Comparing the Effect of Tensioning on Outcomes Following Anterior Cruciate
Ligament Reconstruction
Tensions
Compared
No. of Minimum (Degrees of Graft Type/
Study Pts Follow-up Flexion) Fixation Pre-tensioning Results

van Kampen 38 1 yr 20 vs 40 N BPTB, interfer- Three cycles at No significant difference in

et al8 (20°) ence screw final tension patient-reported Lysholm,
KT-1000,a IKDC, or rates
of contracture between
groups
Kim et al5 48 1 yr 79, 118, and Pentupled ham- Three cycles at No significant difference in
147 N (30°) string, staple final tension average patient-reported
fixation visual analogue scale of
knee laxity, KT-2000a side-
to-side difference
Yasuda et al41 70 2 yr 20, 40, 80 N Doubled ham- Flexion cycling × Significantly greater KT-
(30°) string, staple 1 min at final 1000a side-to-side differ-
fixation tension ence in the lower tension
group (2.2 vs 0.6 mm)
Graft augmentation with the
Leeds-Keio prosthesis was
used
Nicholas et al7 49 20 mo 45 vs 90 N BPTB, interfer- Ten cycles at final Significantly greater tibial
(full ence screw tension displacement and signifi-
extension) cantly more patients with
“abnormal” tibial displace-
ment in the low tension
group
No significant difference in
patient-reported modified
Knee Outcome Survey
scores

BPTB = bone–patellar tendon–bone, IKDC = International Knee Documentation Committee

a
KT-1000, KT-2000 arthrometer (Medmetrics, San Diego, CA)

nal fluoroscopic imaging suggests Posterior Cruciate Ligament has maximal tension in 90° of flex-
that tensioning both bundles at a low ion, and is the main stabilizer to pos-
Native Kinematics
flexion angle may prevent overcon- terior stress.3,46 The posteromedial
The anatomic origin and insertion of
straint.44 No clinical studies compar- bundle tightens in extension and
ing tensioning protocols have been the PCL are shown in Figures 2 and early flexion.3 Sectioning studies
performed, but excellent clinical out- 3. The PCL is an anisometric struc- demonstrate that the posteromedial
comes have been obtained by ten- ture. Viewed as a single structure, bundle produces small but statisti-
sioning the anteromedial bundle at the PCL is physiologically lax at full cally significant increases in mean
60° of flexion and the posterolateral extension and tense (112 N) at 90° laxity at zero degrees (+1.06 mm)
bundle at zero degrees to 15° of flex- of flexion because the central femo- and 10° (+0.83 mm) of flexion, but
ion.45 Given their physiologic loads, ral footprint lies anterior to the fem- plays a minimal role at higher flex-
tensioning the posterolateral bundle oral center of rotation. The PCL is ion angles.46
in extension and the anteromedial functionally divided into the antero-
bundle in midflexion replicates phys- lateral and posteromedial bundles. Single-bundle Repair
iologic stability patterns while mini- The anterolateral bundle comprises Several cadaver biomechanical stud-
mizing overconstraint.22,45 65% of the substance of the PCL, ies of single-bundle PCL repairs have

638 Journal of the American Academy of Orthopaedic Surgeons

 Seth L. Sherman, MD, et al

Figure 3 PCL force profiles; however, laxities

were greater than normal between
zero degrees and 30° of knee flexion.
The addition of a second posterome-
dial graft tensioned to 10 N at 30° of
flexion reduced laxity in early flex-
ion, but it did so at the expense of
higher-than-normal forces in the pos-
teromedial graft.48 In performing a
double-bundle reconstruction, the
surgeon should take care to avoid
overconstraint and decreased knee
motion when adding the posterome-
dial bundle tensioned at a low flex-
ion angle.48

Medial Ligament Complex

The anatomy of the medial aspect of
the knee is depicted in Figure 4. Ma-
jor structures that undergo graft re-
construction to stabilize the medial
side of the knee are the superficial
Photographs of a sagittally sectioned cadaveric femur with a retained medial
femoral condyle demonstrating posterior cruciate ligament (PCL) anatomy. medial collateral ligament (sMCL)
A, The arrows indicate tension within the proximodistally oriented and the posterior oblique ligament
posteromedial bundle of the PCL with the knee in extension. B, With flexion, (POL).
the posteromedial bundle, indicated by the arrow, is oriented
anteroposteriorly. (Adapted with permission from Amis AA, Gupte CM, Bull
AMJ, Edwards A: Anatomy of the posterior cruciate ligament and the Functional Kinematics
meniscofemoral ligaments. Knee Surg Sports Traumatol Arthrosc Reconstruction of the sMCL is im-
2006;14[3]:257-263.) portant because sectioning studies
demonstrate it to be the primary
static stabilizer of the knee to valgus
been performed, many of which were To date, no clinical trials compar- stress.20,49 The medial collateral liga-
intended to reconstruct the anterolat- ing various tensioning regimes in ment (MCL) was traditionally
eral bundle. These studies suggest PCL reconstruction have been con- thought to be an isometric structure
that tensioning at 90° of flexion, ducted. Given the biomechanical and because its proximal origin lies near
with an anterior tibial force of 134 clinical evidence, the authors prefer the femoral center of rotation.20 In
to 156 N, replicates native knee ki- single-bundle PCL reconstruction vivo,50 ex vivo,9 and modeling51 stud-
nematics and hence decreases the with Achilles tendon allograft, ten- ies have confirmed isometry for the
risk of overconstraint and extension sioned in 90° of flexion,4,6,47 bone central third of the sMCL. However,
loss.3 When using the transtibial plug fixation with an interference the sMCL/POL complex is relatively
technique, tensioning in flexion al- screw, and a long biologic interfer- wide, and when one divides these
lows force propagation to the intra- ence screw and supplementary corti- structures into anterior and posterior
articular portion of the graft around cal staple fixation for the soft-tissue halves, their function is best under-
the so-called killer turn. This turn is component.14,35,36 stood as anisometric. The anterior
created by the sharp angulation of portion (sMCL) elongates slightly
the graft exiting the tibial tunnel Double-bundle Repair and tightens maximally in flexion (1
headed toward the femoral insertion, Few cadaver or clinical studies com- to 2 mm at 90°); the posterior seg-
thus creating an area of abrasion. paring tensioning regimes in double- ment (POL) elongates and tightens in
Tensioning a graft at full extension bundle PCL repairs have been extension (2 to 4 mm at zero de-
may propagate force only to the tun- performed.48 Markolf et al48 demon- grees), with an average change of 2.8
nel portion of the graft, resulting in strated that a single anterolateral mm. Warren et al20 demonstrated
residual laxity.4,6,47 graft best reproduced the normal that the sMCL was maximally elon-

October 2012, Vol 20, No 10 639

Graft Tensioning During Knee Ligament Reconstruction: Principles and Practice

Figure 4 Double-bundle Reconstruction

Double-bundle constructs restore
both the sMCL and the POL. Some
authors suggest tensioning the POL
component at 60° of flexion,56
whereas others describe equivalent
results with tensioning at zero de-
grees or 30°.58 Feeley et al54 com-
pared the ability to restore valgus
stability in a cadaver model using
both single- and double-bundle re-
construction techniques. Grafts were
tensioned to 44 N and fixed at 30°
of flexion. Although the single-
bundle reconstruction decreased
opening with valgus force, only
double-bundle repairs were able to
restore the knee’s ability to respond
to both valgus and internal rotation
forces.
A, Illustration of the anatomy of the medial collateral ligament (MCL) In our practice, we prefer a double-
complex. B, Illustration of the reconstruction of the MCL complex.
POL = posterior oblique ligament, sMCL = superficial medial collateral bundle reconstruction with an iso-
ligament. (Adapted with permission from Coobs BR, Wijdicks CA, Armitage metric central sMCL, as described
BM, et al: An in vitro analysis of an anatomical medial knee reconstruction. above, and an anisometric recon-
Am J Sports Med 2010;38[2]:339-347.)
struction of the POL, with hand ten-
sioning near full extension.58 There is
gated at 45° degrees of flexion and In theory, the angle of tension for no evidence to recommend for or
was 1 mm shorter at 30°. The POL is an isometric structure such as the against varus or internal rotation
tighter in extension and has most of central sMCL should not influence during the tensioning process.
its effect at zero degrees and 30° of kinematic outcome. However, most
flexion.52 The POL shares only 10% studies report tensioning of the graft Lateral Collateral Ligament
of valgus load and helps to stabilize between 30° and 60°, corresponding and Posterolateral Corner
against internal rotation at all flex- to the ligament’s maximum length The anatomy of the PLC is shown in
ion angles.53 and highest resistance to valgus Figure 5. The major stabilizers of the
force. Subtle variations mentioned in PLC that are surgically reconstructed
Single-bundle Reconstruction the literature include tensioning the include the lateral collateral ligament
The goal of most surgeons when per- (LCL), popliteus, and popliteofibular
MCL at 30° to allow for 1 mm of
forming single-bundle reconstruction ligament (Figure 6). There are few
creep at 45°,56 tensioning at 30° or
is an isometric reconstruction of the biomechanical and clinical studies
45° with 50 N of force,55 and ten-
sMCL, which requires surgical preci- with regard to graft tensioning for
sioning with a varus moment.57
sion (Figure 4). In a computer navi- PLC reconstruction.
gation model, Feeley et al54 demon- For sMCL reconstruction, we pre-
strated that 4 mm of deviation in any fer an isometric reconstruction at the
Functional Kinematics
direction from the center of the center of the femoral and tibial The LCL, popliteus, and popliteofib-
sMCL footprint causes a significant sMCL footprints, as judged by the ular ligament confer stability to the
decrease in isometry. Most authors Kirschner wire/suture technique.55 knee in response to varus, external
locate an isometric point with a su- We use an Achilles tendon allograft, rotation, and posterior forces.18,19
ture looped over Kirschner wires at with maximum manual tension at With isolated insufficiency of all PLC
the origin and the insertion of the 30°, with a slight varus force.57 The structures, the largest increase to ex-
sMCL, looking for length change of proximal bone block is fixed with in- ternal rotation is at 30° of flexion.19
<3 to 4 mm through a motion cy- terference screw fixation, and the The LCL is an isometric ligament
cle.55 graft is stapled distally.14,35,36 that is the primary restraint to varus

640 Journal of the American Academy of Orthopaedic Surgeons

 Seth L. Sherman, MD, et al

Figure 5 force, especially at low flexion an-

gles.19,59 Direct force measurements
of the LCL during an applied varus
moment demonstrate loading re-
sponses at all angles of knee flexion,
with the response at 30° of flexion
significantly higher than that at 90°
of flexion.60 The popliteus is aniso-
metric, made up of tendon and mus-
cle, thus allowing for dynamic con-
trol and balance of tibial neutral
rotation.61 Sectioning of the poplit-
eus alone elucidates its importance in
preventing external rotation at 90°
to 120° of flexion without contribu-
tions to varus rotation or posterior
translation.19,59 The popliteofibular
ligament is anisometric, takes up
maximal tension during an external
rotation force at 60° to 90°, and is
Photograph of a cadaver specimen (A) and illustration (B) demonstrating the
anatomy of the posterolateral corner. FCL = fibular collateral ligament, lax with internal tibial rotation.61
LGT = lateral gastrocnemius tendon, PL = popliteofibular ligament, Isolated sectioning does not cause in-
PT = popliteus tendon. (Adapted with permission from LaPrade RF, Ly TV, creased varus, rotational, or transla-
Wentorf FA, Engebretsen L: The posterolateral attachments of the knee: A tional changes.59 However, the popli-
qualitative and quantitative morphologic analysis of the fibular collateral
ligament, popliteus tendon, popliteofibular ligament, and lateral teofibular ligament assists maximally
gastrocnemius tendon. Am J Sports Med 2003;31[6]:854-860.) with varus force at 45° of flexion.62

Figure 6 LCL and PLC Reconstruction

The goal of LCL reconstruction is an
isometric reconstruction. Most au-
thors recommend hand tensioning of
the LCL component at 20° to 30° of
knee flexion, with mild valgus to
prevent lateral gapping.63-65 Tension-
ing for the popliteus and popliteofib-
ular ligament component is more
variable and differs by author and
construct type. Comparisons of
fibula-based popliteal reconstruc-
tions indicate that most were ten-
sioned at 30° of knee flexion.64,66,67
Tibia-based reconstructions of the
popliteus were also tensioned by
hand at 30° to 90° of flexion.63,65,67
Controversy exists regarding the use
of an internal rotation force, with
PA (A) and lateral (B) illustrations demonstrating reconstruction of the
posterolateral corner. FCL = fibular collateral ligament, PFL = popliteofibular conflicting data for and against its
ligament, PT = popliteus tendon. (Reproduced with permission from use secondary to concerns for over-
McCarthy M, Camarda L, Wijdicks CA, Johansen S, Engebretsen L, LaPrade constraint.65
RF: Anatomic posterolateral knee reconstructions require a popliteofibular There are many different methods
ligament reconstruction through a tibial tunnel. Am J Sports Med
2010;38[8]:1674-1681.) for PLC reconstruction. In general,
these are fibula-based with one or

October 2012, Vol 20, No 10 641

Graft Tensioning During Knee Ligament Reconstruction: Principles and Practice

Table 3
Ex Vivo Cadaver Studies of Posterolateral Corner Reconstructions

Study No. of Bundles Femoral Tunnels Distal Fixation Tensioning

Ho et al66 2 2 TF Manual
2 1 TF Manual
Rauh et al64 2 1 TF Manual

3 2 TF and T Manual
65
McCarthy et al 2 2 TF, T, and PFL Manual

2 2 TF and T Manual
67
Kim et al 2 2 T, CF, and AFL 57 N
2 1 PF and T 57 N
2 1 TF 57 N
Feeley et al68 2 1 TF AP 30 N

2 1 TF Obl 30 N
2 2 TF AP 30 N
2 2 TF Obl 30 N
Apsingi et al63 2 1 TF a

a
3 2 TF and T

AFL = anterior popliteofibular ligament, AP = straight anterior-posterior transfibular tunnels, CF = central fibular, IR = internal rotation,
LCL = lateral collateral ligament, Obl = 40° transfibular tunnels, PF = posterior fibula, PFL = posterior popliteofibular ligament, T = tibia,
TF = transfibular, Val = valgus
a
Authors tensioned until forces were within intact specimen

two femoral tunnels and may include tion, we prefer a fibula-based tech- should also be taken into account,
a tibia-based component.67 Cadaver nique with a soft-tissue allograft, including the mechanical properties
studies of fibula-based techniques usually semitendinosus.69 We prefer of the selected graft, the fixation
have demonstrated no significant dif- either a single isometric tunnel on method, and the specific biomechani-
ference between the intact and recon- the femur or a double femoral tunnel cal goal of the reconstruction. Al-
structed knee to varus load or to ex- recreating the insertion of both the though extensive biomechanical ca-
ternal torque at any flexion angle.62,68 popliteus and the LCL.64,66,67 For the daver evidence can guide graft
However, two recent biomechanical single femoral tunnel, we fix the tensioning, few clinical trials with
studies in which all three functional graft with a soft-tissue biointerfer- patient-oriented functional clinical
components were anatomically re- ence screw under hand tension at outcomes are available to strongly
constructed separately documented 30°, with a slight valgus and internal support one method of tensioning
overconstraint of internal and varus rotation force.64,66,67 For the double over another. Future investigators
rotation, respectively.63,64 To date, no femoral tunnel, we differentially ten- are encouraged to perform high-
randomized clinical studies have in- sion and fix the LCL and popliteus quality trials in order to accurately
vestigated the different procedures. at 30° and 90°, respectively.64,66,67 direct graft tensioning protocols.
However, most clinical case series re-
port reasonable outcomes regardless
of technique used.69 A summary of Summary References
pertinent biomechanical studies for
PLC reconstruction is given in Table Graft tensioning for knee ligament Evidence-based Medicine: Levels of
3. If reconstructing the LCL alone, reconstruction relies on a thorough evidence are described in the table of
we prefer hand tensioning at 30° of appreciation of native knee anatomy contents. In this article, reference 35
knee flexion.63-65 For PLC reconstruc- and kinematics. Many other factors is a level I study. References 1, 5, 7,

642 Journal of the American Academy of Orthopaedic Surgeons

 Seth L. Sherman, MD, et al

Table 3 (continued)
Ex Vivo Cadaver Studies of Posterolateral Corner Reconstructions
LCL Flexion Posterior Fibula Tibia-based Flexion Force During
(degrees) Flexion (degrees) (degrees) Fixation Results

30 30 — IR, Val Same as intact

30 30 — IR, Val
30 30 — IR, Val Tighter than intact except
tibia-based at 30°
30 30 30 Val (LCL), IR (T)
30 — 60 Val Improved IR kinematics
with PFL
30 — 60 Val
30 — 30 — Same as intact
— 30 30 —
30 30 — —
30 30 — — Both same as intact for
varus, but Obl improved
for rotation
30 30 — —
30 30 — —
30 30 — —
20 — 90 — Same as intact
20 90 90 —

AFL = anterior popliteofibular ligament, AP = straight anterior-posterior transfibular tunnels, CF = central fibular, IR = internal rotation,
LCL = lateral collateral ligament, Obl = 40° transfibular tunnels, PF = posterior fibula, PFL = posterior popliteofibular ligament, T = tibia,
TF = transfibular, Val = valgus
a
Authors tensioned until forces were within intact specimen

8, 33, and 41 are level II studies. 4. Harner CD, Janaushek MA, Ma CB, Heide HJ, Bakens HJ: The effect of
Kanamori A, Vogrin TM, Woo SL: The different graft tensioning in anterior
Reference 16 is a level III study. Ref- effect of knee flexion angle and cruciate ligament reconstruction: A
erences 45, 50, 51, and 69 are level application of an anterior tibial load at prospective randomized study.
IV studies. References 36, 37, 55, the time of graft fixation on the Arthroscopy 1998;14(8):845-850.
biomechanics of a posterior cruciate
and 56 are level V expert opinion. ligament-reconstructed knee. Am J 9. Victor J, Wong P, Witvrouw E, Sloten
Sports Med 2000;28(4):460-465. JV, Bellemans J: How isometric are the
References printed in bold type indi- medial patellofemoral, superficial medial
cate those published within the past 5. Kim SG, Kurosawa H, Sakuraba K, collateral, and lateral collateral ligaments
Ikeda H, Takazawa S: The effect of of the knee? Am J Sports Med 2009;
5 years. initial graft tension on postoperative
37(10):2028-2036.
clinical outcome in anterior cruciate
1. Freedman KB, D’Amato MJ, Nedeff DD,
ligament reconstruction with 10. Austin JC, Phornphutkul C, Wojtys EM:
Kaz A, Bach BR Jr: Arthroscopic
semitendinosus tendon. Arch Orthop Loss of knee extension after anterior
anterior cruciate ligament
Trauma Surg 2006;126(4):260-264. cruciate ligament reconstruction: Effects
reconstruction: A metaanalysis
comparing patellar tendon and 6. Ma CB, Kanamori A, Vogrin TM, Woo of knee position and graft tensioning.
hamstring tendon autografts. Am J SL, Harner CD: Measurement of J Bone Joint Surg Am 2007;89(7):1565-
Sports Med 2003;31(1):2-11. posterior tibial translation in the 1574.
posterior cruciate ligament-reconstructed
2. Carson EW, Deng XH, Allen A, 11. Brady MF, Bradley MP, Fleming BC,
knee: Significance of the shift in the
Wickiewicz T, Warren RF: Evaluation of Fadale PD, Hulstyn MJ, Banerjee R:
reference position. Am J Sports Med
in situ graft forces of a 2-bundle tibial Effects of initial graft tension on the
2003;31(6):843-848.
inlay posterior cruciate ligament tibiofemoral compressive forces and joint
reconstruction at various flexion angles. 7. Nicholas SJ, D’Amato MJ, Mullaney MJ, position after anterior cruciate ligament
Arthroscopy 2007;23(5):488-495. Tyler TF, Kolstad K, McHugh MP: A reconstruction. Am J Sports Med 2007;
prospectively randomized double-blind 35(3):395-403.
3. Fox RJ, Harner CD, Sakane M, Carlin study on the effect of initial graft tension
GJ, Woo SL: Determination of the in situ on knee stability after anterior cruciate 12. Cuomo P, Rama KR, Bull AM, Amis
forces in the human posterior cruciate ligament reconstruction. Am J Sports AA: The effects of different tensioning
ligament using robotic technology: A Med 2004;32(8):1881-1886. strategies on knee laxity and graft
cadaveric study. Am J Sports Med 1998; tension after double-bundle anterior
26(3):395-401. 8. van Kampen A, Wymenga AB, van der cruciate ligament reconstruction. Am J

October 2012, Vol 20, No 10 643

Graft Tensioning During Knee Ligament Reconstruction: Principles and Practice

Sports Med 2007;35(12):2083-2090. tendon grafts for reconstruction of the comparison of patellar tendon and
anterior cruciate ligament: hamstring tendon anterior cruciate
13. Hoshino Y, Kuroda R, Nagamune K, Biomechanical evaluation of the use of ligament reconstruction. Am J Sports
et al: The effect of graft tensioning in multiple strands and tensioning tech- Med 2003;31(4):564-573.
anatomic 2-bundle ACL reconstruction
niques. J Bone Joint Surg Am 1999;
on knee joint kinematics. Knee Surg 36. Brucker PU, Lorenz S, Imhoff AB:
81(4):549-557.
Sports Traumatol Arthrosc 2007;15(5): Aperture fixation in arthroscopic
508-514. 25. Schatzmann L, Brunner P, Stäubli HU: anterior cruciate ligament double-bundle
14. Karchin A, Hull ML, Howell SM: Initial Effect of cyclic preconditioning on the reconstruction. Arthroscopy 2006;
tension and anterior load-displacement tensile properties of human quadriceps 22(11):1250, e1-e6.
behavior of high-stiffness anterior tendons and patellar ligaments. Knee
37. Cunningham R, West JR, Greis PE,
cruciate ligament graft constructs. J Bone Surg Sports Traumatol Arthrosc 1998;
Burks RT: A survey of the tension
Joint Surg Am 2004;86(8):1675-1683. 6(suppl 1):S56-S61.
applied to a doubled hamstring tendon
15. Katsuragi R, Yasuda K, Tsujino J, Keira 26. Almqvist KF, Jan H, Vercruysse C, graft for reconstruction of the anterior
M, Kaneda K: The effect of Verbeeck R, Verdonk R: The tibialis cruciate ligament. Arthroscopy 2002;
nonphysiologically high initial tension on tendon as a valuable anterior cruciate 18(9):983-988.
the mechanical properties of in situ ligament allograft substitute: 38. Thompson DM, Hull ML, Howell SM:
frozen anterior cruciate ligament in a Biomechanical properties. Knee Surg Does a tensioning device pinned to the
canine model. Am J Sports Med 2000; Sports Traumatol Arthrosc 2007;15(11): tibia improve knee anterior-posterior
28(1):47-56. 1326-1330. load-displacement compared to manual
16. Howell SM, Gittins ME, Gottlieb JE, 27. Park DK, Fogel HA, Bhatia S, et al: tensioning of the graft following anterior
Traina SM, Zoellner TM: The Tibial fixation of anterior cruciate cruciate ligament reconstruction? A
relationship between the angle of the ligament allograft tendons: Comparison cadaveric study of two tibial fixation
tibial tunnel in the coronal plane and of 1-, 2-, and 4-stranded constructs. Am devices. J Orthop Res 2006;24(9):1832-
loss of flexion and anterior laxity after J Sports Med 2009;37(8):1531-1538. 1841.
anterior cruciate ligament reconstruc- 39. Labs K, Perka C, Schneider F: The
tion. Am J Sports Med 2001;29(5):567- 28. Ciccone WJ II, Bratton DR, Weinstein
DM, Elias JJ: Viscoelasticity and biological and biomechanical effect of
574. different graft tensioning in anterior
temperature variations decrease tension
17. Arnold MP, Verdonschot N, van and stiffness of hamstring tendon grafts cruciate ligament reconstruction: An
Kampen A: The normal anterior cruciate following anterior cruciate ligament experimental study. Arch Orthop
ligament as a model for tensioning reconstruction. J Bone Joint Surg Am Trauma Surg 2002;122(4):193-199.
strategies in anterior cruciate ligament 2006;88(5):1071-1078. 40. Yoshiya S, Andrish JT, Manley MT,
grafts. Am J Sports Med 2005;33(2):277- Bauer TW: Graft tension in anterior
283. 29. Burks RT, Leland R: Determination of
graft tension before fixation in anterior cruciate ligament reconstruction: An in
18. Gollehon DL, Torzilli PA, Warren RF: cruciate ligament reconstruction. vivo study in dogs. Am J Sports Med
The role of the posterolateral and Arthroscopy 1988;4(4):260-266. 1987;15(5):464-470.
cruciate ligaments in the stability of the 41. Yasuda K, Tsujino J, Tanabe Y, Kaneda
human knee: A biomechanical study. 30. Numazaki H, Tohyama H, Nakano H,
Kikuchi S, Yasuda K: The effect of initial K: Effects of initial graft tension on
J Bone Joint Surg Am 1987;69(2):233- clinical outcome after anterior cruciate
242. graft tension in anterior cruciate
ligament reconstruction on the ligament reconstruction: Autogenous
19. Grood ES, Stowers SF, Noyes FR: Limits mechanical behaviors of the femur-graft- doubled hamstring tendons connected in
of movement in the human knee: Effect tibia complex during cyclic loading. Am series with polyester tapes. Am J Sports
of sectioning the posterior cruciate J Sports Med 2002;30(6):800-805. Med 1997;25(1):99-106.
ligament and posterolateral structures. 42. Jackson DW, Grood ES, Goldstein JD,
J Bone Joint Surg Am 1988;70(1):88-97. 31. Graf BK, Vanderby R Jr, Ulm MJ, Ro-
galski RP, Thielke RJ: Effect of et al: A comparison of patellar tendon
20. Warren LA, Marshall JL, Girgis F: The preconditioning on the viscoelastic autograft and allograft used for anterior
prime static stabilizer of the medical side response of primate patellar tendon. cruciate ligament reconstruction in the
of the knee. J Bone Joint Surg Am 1974; Arthroscopy 1994;10(1):90-96. goat model. Am J Sports Med 1993;
56(4):665-674. 21(2):176-185.
32. Boylan D, Greis PE, West JR, Bachus
21. Sidles JA, Larson RV, Garbini JL, KN, Burks RT: Effects of initial graft 43. Murray PJ, Alexander JW, Gold JE,
Downey DJ, Matsen FA III: Ligament tension on knee stability after anterior Icenogle KD, Noble PC, Lowe WR:
length relationships in the moving knee. cruciate ligament reconstruction using Anatomic double-bundle anterior
J Orthop Res 1988;6(4):593-610. hamstring tendons: A cadaver study. cruciate ligament reconstruction:
Arthroscopy 2003;19(7):700-705. Kinematics and knee flexion angle-graft
22. Gabriel MT, Wong EK, Woo SL, Yagi M, tension relation. Arthroscopy 2010;
Debski RE: Distribution of in situ forces 33. Kawano CT, de Moraes Barros Fucs 26(2):202-213.
in the anterior cruciate ligament in PM, Severino NR: Pretensioning of
response to rotatory loads. J Orthop Res quadruple flexor tendon grafts in two 44. Jordan SS, DeFrate LE, Nha KW,
2004;22(1):85-89. types of femoral fixation: Quasi- Papannagari R, Gill TJ, Li G: The in vivo
randomised controlled pilot study. Int kinematics of the anteromedial and
23. Noyes FR, Butler DL, Grood ES, Orthop 2011;35(4):521-527. posterolateral bundles of the anterior
Zernicke RF, Hefzy MS: Biomechanical cruciate ligament during weightbearing
analysis of human ligament grafts used 34. Beynnon BD, Johnson RJ, Fleming BC, knee flexion. Am J Sports Med 2007;
in knee-ligament repairs and et al: The measurement of elongation of 35(4):547-554.
reconstructions. J Bone Joint Surg Am anterior cruciate-ligament grafts in vivo.
1984;66(3):344-352. J Bone Joint Surg Am 1994;76(4):520- 45. Fu FH, Shen W, Starman JS, Okeke N,
531. Irrgang JJ: Primary anatomic double-
24. Hamner DL, Brown CH Jr, Steiner ME, bundle anterior cruciate ligament
Hecker AT, Hayes WC: Hamstring 35. Feller JA, Webster KE: A randomized reconstruction: A preliminary 2-year

644 Journal of the American Academy of Orthopaedic Surgeons

 Seth L. Sherman, MD, et al

prospective study. Am J Sports Med 54. Feeley BT, Muller MS, Allen AA, tion. J Bone Joint Surg Am 2007;89(11):
2008;36(7):1263-1274. Granchi CC, Pearle AD: Biomechanical 2359-2368.
comparison of medial collateral ligament
46. Race A, Amis AA: PCL reconstruction: 63. Apsingi S, Nguyen T, Bull AM, Unwin
reconstructions using computer-assisted
In vitro biomechanical comparison of A, Deehan DJ, Amis AA: A comparison
navigation. Am J Sports Med 2009;
‘isometric’ versus single and double- of modified Larson and ‘anatomic’
37(6):1123-1130. posterolateral corner reconstructions in
bundled ‘anatomic’ grafts. J Bone Joint
Surg Br 1998;80(1):173-179. 55. Adachi N, Ochi M, Deie M, Izuta Y, knees with combined PCL and
Kazusa H: New hamstring fixation posterolateral corner deficiency. Knee
47. Burns WC II, Draganich LF, Pyevich M, technique for medial collateral ligament Surg Sports Traumatol Arthrosc 2009;
Reider B: The effect of femoral tunnel or posterolateral corner reconstruction 17(3):305-312.
position and graft tensioning technique using the mosaicplasty system. 64. Rauh PB, Clancy WG Jr, Jasper LE, Curl
on posterior laxity of the posterior Arthroscopy 2006;22(5):571, e1-e3. LA, Belkoff S, Moorman CT III:
cruciate ligament-reconstructed knee. Biomechanical evaluation of two
Am J Sports Med 1995;23(4):424-430. 56. Borden PS, Kantaras AT, Caborn DN:
Medial collateral ligament reconstruction reconstruction techniques for
48. Markolf KL, Feeley BT, Jackson SR, with allograft using a double-bundle posterolateral instability of the knee.
technique. Arthroscopy 2002;18(4):E19. J Bone Joint Surg Br 2010;92(10):1460-
McAllister DR: Biomechanical studies of
1465.
double-bundle posterior cruciate
57. Marchant MH Jr, Tibor LM, Sekiya JK,
ligament reconstructions. J Bone Joint 65. McCarthy M, Camarda L, Wijdicks CA,
Hardaker WT Jr, Garrett WE Jr, Taylor
Surg Am 2006;88(8):1788-1794. DC: Management of medial-sided knee Johansen S, Engebretsen L, Laprade RF:
injuries, part 1: Medial collateral Anatomic posterolateral knee
49. Kennedy JC, Fowler PJ: Medial and reconstructions require a popliteofibular
anterior instability of the knee: An ligament. Am J Sports Med 2011;39(5):
1102-1113. ligament reconstruction through a tibial
anatomical and clinical study using stress tunnel. Am J Sports Med 2010;38(8):
machines. J Bone Joint Surg Am 1971; 58. Coobs BR, Wijdicks CA, Armitage BM, 1674-1681.
53(7):1257-1270. et al: An in vitro analysis of an
anatomical medial knee reconstruction. 66. Ho EP, Lam M-H, Chung MM, et al:
50. Liu F, Gadikota HR, Kozánek M, et al: Am J Sports Med 2010;38(2):339-347. Comparison of 2 surgical techniques for
In vivo length patterns of the medial reconstructing posterolateral corner of
collateral ligament during the stance 59. Pasque C, Noyes FR, Gibbons M, Levy the knee: A cadaveric study evaluated by
phase of gait. Knee Surg Sports M, Grood E: The role of the navigation system. Arthroscopy 2011;
Traumatol Arthrosc 2011;19(5):719- popliteofibular ligament and the tendon 27(1):89-96.
727. of popliteus in providing stability in the
human knee. J Bone Joint Surg Br 2003; 67. Kim S-J, Kim H-S, Moon H-K, Chang
51. Park SE, DeFrate LE, Suggs JF, Gill TJ, 85(2):292-298. W-H, Kim S-G, Chun Y-M: A
Rubash HE, Li G: The change in length biomechanical comparison of 3
of the medial and lateral collateral 60. Nielsen S, Helmig P: Posterior instability reconstruction techniques for
ligaments during in vivo knee flexion. of the knee joint: An experimental study. posterolateral instability of the knee in a
Knee 2005;12(5):377-382. Arch Orthop Trauma Surg 1986;105(2): cadaveric model. Arthroscopy 2010;
121-125. 26(3):335-341.
52. Griffith CJ, LaPrade RF, Johansen S,
Armitage B, Wijdicks C, Engebretsen L: 61. Krudwig WK, Witzel U, Ullrich K: 68. Feeley BT, Muller MS, Sherman S, Allen
Medial knee injury: Part 1, static Posterolateral aspect and stability of the AA, Pearle AD: Comparison of
function of the individual components of knee joint: II. Posterolateral instability posterolateral corner reconstructions
the main medial knee structures. Am J and effect of isolated and combined using computer-assisted navigation.
Sports Med 2009;37(9):1762-1770. posterolateral reconstruction on knee Arthroscopy 2010;26(8):1088-1095.
stability: A biomechanical study. Knee
53. Griffith CJ, Wijdicks CA, LaPrade RF, Surg Sports Traumatol Arthrosc 2002; 69. Lee SH, Jung YB, Jung HJ, Song KS, Ko
Armitage BM, Johansen S, Engebretsen 10(2):91-95. YB: Combined reconstruction for
L: Force measurements on the posterior posterolateral rotatory instability with
oblique ligament and superficial medial 62. Sigward SM, Markolf KL, Graves BR, anterior cruciate ligament injuries of the
collateral ligament proximal and distal Chacko JM, Jackson SR, McAllister DR: knee. Knee Surg Sports Traumatol
divisions to applied loads. Am J Sports Femoral fixation sites for optimum Arthrosc 2010;18(9):1219-1225.
Med 2009;37(1):140-148. isometry of posterolateral reconstruc-

October 2012, Vol 20, No 10 645