See discussions, stats, and author profiles for this publication at: https://www.researchgate.

net/publication/330124778

Use of a modified passive leg-raising maneuver to predict fluid

responsiveness during experimental induction and correction of
hypovolemia in healthy isoflurane-anesthetized pigs

Article in American Journal of Veterinary Research · January 2019

DOI: 10.2460/ajvr.80.1.24

CITATIONS READS

4 28

3 authors, including:

Vaidehi Vinay Paranjape

Virginia Tech (Virginia Polytechnic Institute and State University)
20 PUBLICATIONS 50 CITATIONS

SEE PROFILE

All content following this page was uploaded by Vaidehi Vinay Paranjape on 31 January 2023.

The user has requested enhancement of the downloaded file.

Use of a modified passive leg-raising maneuver to
predict fluid responsiveness during experimental
induction and correction of hypovolemia in healthy
isoflurane-anesthetized pigs

Vaidehi V. Paranjape bvsc , mvsc , ms OBJECTIVE

To evaluate the use of a modified passive leg-raising maneuver (PLRM) to
Andre C. Shih dvm predict fluid responsiveness during experimental induction and correction
Fernando L. Garcia-Pereira dvm, ms of hypovolemia in isoflurane-anesthetized pigs.

Received March 6, 2018. ANIMALS

Accepted May 21, 2018. 6 healthy male Landrace pigs.
PROCEDURES
From the Departments of Comparative, Diagnostic,
and Population Medicine (Paranjape) and Large Animal Pigs were anesthetized with isoflurane, positioned in dorsal recumben-
Clinical Sciences (Garcia-Pereira), College of Veteri- cy, and instrumented. Following induction of a neuromuscular blockade,
nary Medicine, University of Florida, Gainesville, FL pigs were mechanically ventilated throughout 5 sequential experimental
32608; and Capital Veterinary Specialists, 3001 Hart- stages during which the blood volume was manipulated so that subjects
ley Rd, Jacksonville, FL 32257 (Shih).
transitioned from normovolemia (baseline) to hypovolemia (blood volume
depletion, 20% and 40%), back to normovolemia, and then to hypervolemia.
Address correspondence to Dr. Paranjape
(vparanjape24@gmail.com). During each stage, hemodynamic variables were measured before and 3 min-
utes after a PLRM and 1 minute after the pelvic limbs were returned to their
original position. The PLRM consisted of raising the pelvic limbs and caudal por-
tion of the abdomen to a 15° angle relative to the horizontal plane.
RESULTS
Hemodynamic variables did not vary in response to the PLRM when pigs
were normovolemic or hypervolemic. When pigs were hypovolemic, the
PLRM resulted in a significant increase in cardiac output and decrease in
plethysomographic variability index and pulse pressure variation. When the
pelvic limbs were returned to their original position, cardiac output and
pulse pressure variation rapidly returned to their pre-PLRM values, but the
plethysomographic variability index did not.
CONCLUSIONS AND CLINICAL RELEVANCE
Results suggested a modified PLRM might be useful for identification of
hemodynamically unstable animals that are likely to respond to fluid therapy.
Further research is necessary to validate the described PLRM for prediction
of fluid responsiveness in clinically ill animals. (Am J Vet Res 2019;80:24–32)

P erioperative hypovolemia is a common cause of

acute circulatory failure. The purpose of adminis-
tering fluid therapy to hemodynamically unstable pa-
tient outcomes, a decrease in morbidity, and shorter
duration of hospitalization.2,3 For hemodynamically
unstable patients, the cardiac preload status and the
tients is to increase the intravascular volume, thereby probability of whether the patient will respond to
improving tissue perfusion and oxygenation. How- a fluid challenge should be thoroughly assessed be-
ever, overzealous fluid therapy can cause pulmonary fore fluid therapy is initiated. According to the Frank-
and interstitial edema.1 In human medicine, goal- Starling law of the heart, administration of a fluid bo-
directed fluid therapy is associated with improved pa- lus will increase the ventricular preload and CO in
fluid responders but have minimal effect on the CO
ABBREVIATIONS of fluid nonresponders. Results of studies1,4 involving
CO Cardiac output
CVP Central venous pressure human patients indicate that < 50% of patients admit-
etISO End-tidal isoflurane concentration ted to critical care units are fluid responsive, which
MAP Mean arterial pressure is concerning because fluid nonresponders can easily
PAOP Pulmonary artery occlusion pressure become volume overloaded.
PI Perfusion index The PVI is derived from pulse oximetry and pro-
PLRM Passive leg-raising maneuver
PPV Pulse pressure variation vides an automated measurement of the percentage
PVI Plethysmographic variability index variation in the amplitude of the plethysmographic
TPP Total plasma protein waveform. It is calculated as the percentage varia-

24 AJVR • Vol 80 • No. 1 • January 2019

Unauthenticated | Downloaded 01/31/23 04:14 AM UTC

tion in the PI of mechanically ventilated individuals.5 to predict fluid responsiveness in human patients
Pulse pressure variation is derived from the direct ar- during anesthesia and in critical care settings, to our
terial pressure signal and is the percentage variation knowledge, its hemodynamic effects in clinical set-
in pulse pressure recorded during the respiratory cy- tings have not been evaluated for veterinary species.
cle of mechanically ventilated individuals.3,4 Both the The purpose of the study reported here was to
PVI and PPV reflect beat-to-beat oscillations in stroke describe the use of a modified PLRM to assess fluid
volume caused by changes in intrathoracic pres- responsiveness during experimental induction and
sure during mechanical ventilation. Measurement of correction of hypovolemia in isoflurane-anesthetized
ventilation-induced variation in pulse pressure and pigs. Our hypotheses were that the PLRM would in-
plethysmographic waveform allows clinicians to pre- crease CO and accurately predict fluid responsive-
dict a patient’s cardiovascular response to changes ness during hypovolemia in mechanically ventilated
in intravascular volume.3–5 In fact, PVI and PPV have pigs, dynamic cardiac preload indices such as PVI and
proven beneficial for assessment of fluid responsive- PPV could be used to accurately track variations in
ness during periods of hypovolemia in both human peripheral perfusion caused by the PLRM, and the
patients3–6 and dogs.7 PLRM would not affect CO, PVI, and PPV but would
In human medicine, a PLRM is commonly used accurately predict fluid nonresponsiveness during
as a simple bedside test to identify fluid responders volume replacement.
among septic and critically ill patients.8–10 This ma-
neuver involves passively raising a patient’s legs to Materials and Methods
angles of 30° to 60° from the horizontal plane of the
hospital bed, which transfers venous blood within Animals
the capacitance veins of the legs toward the heart All study procedures were reviewed and approved
and increases cardiac preload. This abrupt increase by the University of Florida Institutional Animal Care
in preload generally results in a > 10% increase in and Use Committee (protocol No. 201609273). Six
CO (measured with arterial pulse contour analysis or purpose-bred male Landrace pigs were used for the
echocardiography) in fluid responders.11,12 The pri- study. Prior to study initiation, all pigs were 8 to 10
mary advantage of the PLRM is that it is readily revers- months old, weighed approximately 20 kg, and were
determined to be healthy on the basis of results of
ible and can be repeatedly used to assess a patient’s
a physical examination, CBC, and serum biochemi-
response to an increase in cardiac preload without
cal evaluation. The pigs were housed and cared for in
the risks associated with fluid therapy (eg, pulmo-
accordance with Association for Assessment and Ac-
nary edema), particularly for fluid nonresponders.8,10
creditation of Laboratory Animal Care International
The PLRM can be easily and reliably performed in
guidelines.
spontaneously breathing patients, even those with
low tidal volume ventilation and low lung compliance Anesthetic induction and instrumentation
in which dynamic indices such as PPV and PVI can- For each pig, food but not water was withheld
not be reliably measured.1,13–16 Although PVI is not as for 15 hours before anesthesia induction. Pigs were
accurate in spontaneously breathing patients as it is premedicated with ketaminea (10 mg/kg, IM). Anes-
in mechanically ventilated patients, its measurement thesia was induced with 5% isofluraneb in oxygen de-
following the PLRM has been successfully used to livered via a face mask and circle system by an anes-
assess changes in left ventricular stroke volume in thesia machinec until the pig could be orotracheally
spontaneously breathing human patients.17–19 intubated. After successful intubation was achieved,
The standard PLRM used for human patients can- the endotracheal tube was secured, and anesthesia
not be used in veterinary species because of the wide was maintained with isoflurane.
variability in pelvic limb conformation, size, and A 22-gauge catheterd was aseptically placed in an
blood volume across species. In the field of biomed- auricular vein. The pig was then positioned in dorsal
ical research, there has been a recent trend to use recumbency, after which the end-tidal partial pres-
pigs instead of dogs in terminal studies. Pigs are an sure of CO2 and etISO were continuously monitored
excellent alternative for human and canine research, by an infrared gas analyzer,e which was attached to
especially with regard to cardiovascular function, a multiparameter patient monitor.f Isoflurane admin-
because of the numerous physiologic, anatomic, and istration was adjusted as necessary to maintain the
pathological similarities among those 3 species.20 Re- etISO at approximately 1.6%, which is approximately
sults of 1 study21 indicate that use of a PLRM during 1.2 times the minimum alveolar concentration of iso-
CPR significantly improves coronary perfusion and flurane reported for pigs.22 A spirometry sensor was
neurologic scores in pigs with experimentally in- placed between the endotracheal tube and y-piece
duced prolonged ventricular fibrillation. The investi- of the circle system and connected to a monitorg to
gators of that study21 concluded that use of the PLRM measure inspiratory and expiratory flow, tidal vol-
during CPR resulted in return of spontaneous circula- ume, minute volume, peak inspiratory pressure, pla-
tion and 24-hour survival rates that were comparable teau pressure, mean airway pressure, and dynamic
to those achieved with standard patient positioning compliance of the respiratory system. Dynamic com-
during CPR. Although the PLRM is commonly used pliance of the respiratory system was calculated as
AJVR • Vol 80 • No. 1 • January 2019 25

Unauthenticated | Downloaded 01/31/23 04:14 AM UTC

the expired tidal volume/(peak inspiratory pressure Experimental design
– positive end-expiratory pressure). The multiparam- Following completion of instrumentation, a neu-
eter monitor also continuously displayed the heart romuscular blockade was induced with a bolus of atra-
rate, lead II ECG tracing, noninvasive blood pressure curiump (1 mg/kg, IV). The efficiency of the blockade
readings, and esophageal temperature, which was was monitored by a supramaximal train-of-four elec-
maintained between 37° and 39°C by use of a forced- trical stimulusq of a common peroneal nerve. Atracu-
air warming device.h A clip-on pulse oximeter probei rium dosing was repeated as necessary on the basis of
was positioned on the thinnest portion of the tongue a return of twitches during the train-of-four function
and connected to a pulse oximeter monitor that con- on the nerve stimulator. Each pig underwent volume-
tinuously displayed the PVI, which was automatically controlled ventilation throughout the duration of the
calculated from the minimum and maximum PI over neuromuscular blockade. Tidal volume was held con-
each respiratory cycle. To enhance the PI signal and stant at 12 mL/kg, and the respiratory frequency was
increase the reliability of PVI values, the tongue was adjusted between 12 and 20 breaths/min to maintain
gently massaged and the probe was repositioned ap- Paco2 between 35 and 45 mm Hg.
proximately 5 minutes before each designated data While under the neuromuscular blockade, each
acquisition time. pig underwent 5 sequential nonrandom experimental
The left and right femoral arteries were dissect- stages. During each stage, cardiopulmonary and hemo-
ed, and a 5F catheterj was placed in each artery and dynamic variables were recorded immediately before
secured. The catheter in the right femoral artery was and 3 minutes after a modified PLRM was performed
used for collection of arterial blood samples for blood and 1 minute after the pelvic limbs were returned to
gas analyses,k measurement of PCV and TPP concen- their original position. The modified PLRM consisted
tration, and recording of direct systolic and diastolic of placement of a wooden plank between the caudal
blood pressures and MAP. A saline (0.9% NaCl) solu- portion of the pig’s body and the table surface in a cra-
tion–filled pressure transducer systeml was calibrated niocaudal orientation, with 1 end of the plank located
and used for direct blood pressure recordings. The at the level of the xiphoid process (Figure 1). At each
transducer was zeroed and placed at the level of the designated time, the free end of the plank was elevated
manubrium before each predetermined recording. until a 15° angle was achieved between the horizontal
The catheter in the left femoral artery was used for plane of the table surface and the pelvic limbs and ab-
withdrawal of blood during controlled hemorrhage domen. A wooden block was placed between the table
(ie, induction of hypovolemia). and plank to maintain the 15° angle for 5 minutes, af-
The left and right external jugular veins were ter which the block was removed and the pelvic limbs
carefully dissected, and an 8F catheter introducerm and abdomen were returned to their original posi-
was inserted in each vein and secured. A 7F thermo- tions. The head and thorax of the pig remained in the
dilution pulmonary artery catheter n was advanced same horizontal plane as the table surface throughout
through the introducer in the right jugular vein un- the PLRM. There was a 10-minute interval between
til its distal pressure sensing lumen was located in stages and a 10-minute interval between experimental
the pulmonary artery. Correct placement of the ther- manipulation of the circulation and the PLRM to allow
modilution catheter was determined on the basis of hemodynamic variables to stabilize.
observation of characteristic pressure waveforms Baseline measurements were obtained during
and pressure values after it was connected to the stage 1. After baseline measurements were recorded,
multiparameter monitor. The proximal port of the controlled hemorrhage was initiated. Blood was with-
pulmonary artery catheter was connected to a saline drawn through the catheter in the left femoral artery
solution–filled pressure transducer,l which was used over a period of 15 minutes until 20% (stage 2) and
to measure CVP. Cardiac output was measured with 40% (stage 3) of the estimated total circulating blood
a thermodilution technique, whereby a 10-mL bolus volume was removed, considering the total circulat-
of chilled (3° to 5°C) 5% dextrose solution was peri- ing blood volume for pigs is approximately 70 mL/
odically injected into the central venous port of the kg of body weight.23 Prior to anesthesia induction,
thermodilution catheter following the recording of each pig was individually weighed so that the exact
direct blood pressure values. The temperature of the amount of blood to be withdrawn during stages 2 and
injected dextrose solution was measured by an in-line 3 could be calculated. To ensure that blood was re-
thermistor that was located between the syringe and moved at a fairly constant rate during stages 2 and
injection port of the pulmonary artery catheter. At 3, two 60-mL syringes, which were flushed with an
each data acquisition time, the CO recorded repre- anticoagulant solution containing sodium citrate, cit-
sented the mean of 3 consecutive measurements. The ric acid, dextrose, and adenine, were used to transfer
catheter in the left jugular vein was used for adminis- blood from the femoral artery catheter to blood col-
tration of blood and colloids during volume replace- lection bags that contained the same anticoagulant
ment. All hemodynamic data were transferred to a solution. The bags were placed on scales and continu-
bioamplifier, which sent continuous real-time data to ously weighed as they were being filled to ensure that
a laptop computer equipped with softwareo to record only the calculated amount of blood was removed
it. during each stage.

26 AJVR • Vol 80 • No. 1 • January 2019

Unauthenticated | Downloaded 01/31/23 04:14 AM UTC

 During stage 4, the total volume of blood re- Response of hemodynamic variables to
moved during stages 2 and 3 was reinfused via the changes in blood volume and the PLRM
catheter in the left jugular vein over a period of 15 The mean ± SD values for hemodynamic variables
minutes to return the pig to a state of normovolemia. before and after the PLRM and after the pelvic limbs
During stage 5, a 500-mL bolus of a colloidr was ad- and abdomen were returned to their original positions
ministered through the catheter in the left jugular during each experimental stage were summarized
vein over a period of 15 minutes to achieve a state of (Table 1). Compared with baseline (stage 1) values,
hypervolemia. After the last data for stage 5 were ac- experimental induction of hypovolemia (stages 2 and
quired, each pig was euthanized by IV administration 3) was associated with a significant decrease in CO, PI,
of a pentobarbitals overdose. MAP, CVP, and PAOP and a significant increase in PVI,
During all experimental stages, fluid responsive- PPV, and heart rate before the PLRM was performed.
ness was defined as a > 15% increase in CO and > 15% The magnitude of those differences from baseline was
decrease in PVI between recordings obtained im- greater during stage 3 than during stage 2. In fact, all
mediately before and 3 minutes after the PLRM was of those variables differed significantly between stages
performed. Fluid nonresponsiveness was defined as a 2 and 3 at each of the 3 data acquisition times. All vari-
≤ 15% change in CO and PVI in response to the PLRM.

Statistical analysis
The distribution of each variable at each data ac-
quisition time (before and after PLRM and after the
pelvic limbs and abdomen were returned to their
original positions during each experimental stage)
was assessed for normality by means of the Shapiro-
Wilk test. All variables were found to be normally dis-
tributed; therefore, the results for all variables were
summarized as the mean ± SD. A 1-way ANOVA for
repeated measures was used to compare variables
among the 3 data acquisition times within each ex-
perimental stage and among the 5 experimental stag-
es. A post hoc Tukey adjustment was used when mul-
tiple pairwise comparisons were performed. Values
of P < 0.05 were considered significant. All analyses
were performed with commercially available statisti-
cal software.t

Results
Pigs
The 6 male study pigs had a median age of 9
months (range, 8 to 10 months) and mean ± SD body
weight of 20 ± 0.47 kg. Throughout the duration of
anesthesia, the mean ± SD etISO was 1.59 ± 0.31%. The
mean ± SD duration was 160.83 ± 17.3 minutes be-
tween premedication with ketamine and completion
of instrumentation, 177.45 ± 23.54 minutes for the
5 experimental stages, and 344.88 ± 13.45 minutes
between premedication and euthanasia. Those dura-
tions did not differ significantly (P = 0.14) among the
6 pigs. The mean ± SD core (esophageal) body tem- Figure 1—Illustrations depicting the positioning of a wooden
plank before (A) and during (B) a modified PLRM that was
perature was 37.9 ± 0.41°C throughout the duration used to assess fluid responsiveness in 6 healthy male Landrace
of the experiment and did not differ significantly (P pigs during states of normovolemia, hypovolemia, and hyper-
= 0.56) among pigs. The mean ± SD volume of blood volemia. The wooden plank was placed between the caudal
withdrawn from each pig during stage 2 was 280 ± portion of the pig’s body and the table surface in a craniocau-
dal orientation with 1 end of the plank located at the level of
12 mL, and the same volume was withdrawn from the xiphoid process. At each designated time, the free end of
each pig during stage 3. Thus, for each pig, approxi- the plank was elevated until a 15° angle was achieved between
mately 560 mL of blood was withdrawn to induce hy- the horizontal plane of the table surface and the pelvic limbs
povolemic shock. Following the PLRM, all pigs were and abdomen. A wooden block was placed between the table
classified as fluid responders when in a state of hy- and plank to maintain the 15° angle for 5 minutes, after which
the block was removed and the pelvic limbs and abdomen
povolemia (stages 2 and 3) and fluid nonresponders were returned to their original positions. The head and tho-
when in a state of normovolemia (stages 1 and 4) or rax of the pig remained in the same horizontal plane as the
hypervolemia (stage 5). table surface throughout the PLRM.

AJVR • Vol 80 • No. 1 • January 2019 27

Unauthenticated | Downloaded 01/31/23 04:14 AM UTC

Table 1—Mean ± SD values for hemodynamic variables for 6 healthy male isoflurane-anesthetized and mechanically ventilated
Landrace pigs immediately before and 3 minutes after a PLRM was performed and 1 minute after the pelvic limbs and abdomen
were returned to their original positions during each of 5 experimental stages.
Stage 1 Stage 2 Stage 3

Variable Before PLRM After PLRM After RTOP Before PLRM After PLRM After RTOP Before PLRM After PLRM After RTOP

CO (L/min) 3.4 ± 0.9 3.8 ± 0.6 3.6 ± 0.7 2.4 ±0.5*‡a 3.4 ±0.6‡b 2.5 ±0.4*‡a 1.6 ±0.6*†‡a 2.4 ±0.7*†‡b 1.7 ± 0.6*†‡a
PVI (%) 10 ± 2 9±4 10 ± 3 29 ± 4*‡a 17 ± 2*‡b 19 ± 2*‡b 37 ± 5*†‡a 23 ± 4*†‡b 21 ± 2*†‡b
PPV (%) 7±2 8±2 8±3 18 ± 2*‡a 9 ± 3‡b 17 ± 1*‡a 25 ± 3*†‡a 15 ± 1*†‡b 22 ± 2*†‡a
PI (%) 2.8 ± 1.3 2.7 ± 1.1 2.6 ± 0.3 1.5 ± 0.9* 1.6 ± 0.4* 1.5 ± 0.4* 0.7 ± 0.1*†‡ 0.9 ± 0.1*†‡ 0.8 ± 0.1*†‡
Spo2 (%) 99 ± 1 99 ± 1 99 ± 1 97 ± 2 97 ± 2 96 ± 3 95 ± 3 95 ± 3 95 ± 2
Heart rate (beats/min) 104 ± 12 101 ± 14 107 ± 15 127 ± 10*‡ 119 ± 13*‡ 123 ± 12*‡ 147 ± 14*†‡ 137 ± 15*†‡ 137 ± 15*†‡
MAP (mm Hg) 77 ± 8 82 ± 5 82 ± 6 54 ± 10*‡a 67 ± 5*‡b 57 ± 8*‡a 40 ± 5*†‡a 50 ± 10*†‡b 43 ± 10*†‡a
CVP (mm Hg) 6±2 7±2 6±2 3 ± 1*‡a 8 ± 1‡b 4 ± 1*‡a 1 ± 1*†‡a 5 ± 1*†‡b 2 ± 1*†‡a
PAOP (mm Hg) 9±2 9±2 9±2 5 ± 1*‡a 9 ± 2‡b 4 ± 2*‡a 2 ± 1*†‡a 8 ± 2†‡b 3 ± 1*†‡a
Respiratory rate 11 ± 1 11 ± 1 11 ± 1 10 ± 1 10 ± 1 10 ± 1 9±3 9±3 9±3
(breaths/min)
Petco 2 (mm Hg) 40 ± 5 40 ± 5 40 ± 5 34 ± 2 34 ± 2 34 ± 2 29 ± 2 28 ± 2 27 ± 2
PIP (cm H2O) 10 ± 2 10 ± 2 10 ± 2 10 ± 2 10 ± 2 10 ± 2 10 ± 1 10 ± 1 10 ± 1
Dynamic compliance 23 ± 2 23 ± 2 23 ± 2 23 ± 2 23 ± 2 23 ± 2 22 ± 2 22 ± 2 22 ± 2
(mL/cm H2O)

Stage 4 Stage 5

Variable Before PLRM After PLRM After RTOP Before PLRM After PLRM After RTOP

CO (L/min) 3.6 ± 0.8 3.9 ± 0.6 3.8 ± 0.7 3.9 ± 0.9* 3.8 ± 0.8 3.6 ± 0.7
PVI (%) 12 ± 4 10 ± 1 9±2 9±4 9±3 9±3
PPV (%) 8±2 9±1 8±1 8±1 9±1 8±1
PI (%) 1.3 ± 0.4 1.3 ± 0.3 1.4 ± 0.4 1.8 ± 0.6 1.7 ± 0.5 1.7 ± 0.6
Spo2 (%) 97 ± 2 98 ± 1 98 ± 1 99 ± 1 99 ± 1 99 ± 1
Heart rate (beats/min) 108 ± 8 105 ± 7 105 ± 10 109 ± 12 111 ± 12 110 ± 12
MAP (mm Hg) 80 ± 1 82 ± 14 79 ± 15 84 ± 16* 85 ± 15 83 ± 15
CVP (mm Hg) 9±1 10 ± 1 9±1 10 ± 1 9±1 10 ± 1
PAOP (mm Hg) 11 ± 2 12 ± 1 11 ± 1 12 ± 1 11 ± 1 11 ± 1
Respiratory rate 11 ± 1 11 ± 1 11 ± 1 11 ± 1 11 ± 1 11 ± 1
(breaths/min)
Petco2 (mm Hg) 37 ± 2 35 ± 1 36 ± 1 38 ± 1 37 ± 1 37 ± 1
PIP (cm H2O) 11 ± 1 11 ± 1 11 ± 1 10 ± 2 10 ± 2 10 ± 2
Dynamic compliance 21 ± 3 21 ± 3 21 ± 3 21 ± 3 21 ± 3 21 ± 3
(mL/cm H2O)

Pigs were positioned in dorsal recumbency throughout all 5 experimental stages. The PLRM consisted of raising the pelvic limbs and caudal portion of the abdomen until they
were at a 15° angle relative to the horizontal plane of the table surface. Each pig was in a state of normovolemia during stage 1 (baseline). During stage 2, a volume of blood equal
to 20% of the estimated total circulating blood volume was withdrawn over a period of 15 minutes. During stage 3, additional blood was withdrawn over a period of 15 minutes
until approximately 40% of the estimated total circulating blood volume was removed, and the pig was in a state of hypovolemic shock. During stage 4, the total volume of blood
removed during stages 2 and 3 was reinfused over a period of 15 minutes to restore the pig to a state of normovolemia. During stage 5, each pig received a 500-mL IV bolus of
a colloid over 15 minutes to achieve a hypervolemic state. There was a 10-minute interval between stages and a 10-minute interval between experimental manipulation of the
circulation and the PLRM to allow hemodynamic variables to stabilize. None of the mean values differed significantly between stages 4 and 5.
*Value differs significantly (P < 0.05) from the corresponding value at stage 1. †Value differs significantly (P < 0.05) from the corresponding value at stage 2. ‡Value
differs significantly (P < 0.05) from the corresponding values at stages 4 and 5.
Petco2 = End-tidal partial pressure of CO2. PIP = Peak inspiratory pressure. RTOP = Return to original position. Spo2 = Oxygen saturation as measured by pulse oximetry.
a,bWithin a stage, values with different superscript letters differ significantly (P < 0.05).

ables returned to their baseline values when normovole- Hematologic variables

mia was restored (stage 4). Experimental induction of The mean ± SD values for hematologic variables
hypervolemia caused a significant increase in CO and during each experimental stage were summarized (Ta-
MAP, compared with baseline values. ble 2). The effect of the PLRM on those variables was
The PLRM had no significant effect on any of the not evaluated. The mean PCV did not differ significantly
hemodynamic variables evaluated when pigs were in (P = 0.16) among the 5 stages. The mean TPP concentra-
a state of normovolemia (stages 1 and 4) or hyper- tion, Paco2, and arterial pH decreased significantly and
volemia (stage 5). When pigs were hypovolemic, the the mean lactate concentration increased significantly
PLRM resulted in a significant increase in CO and a from baseline values following experimental induction
significant decrease in PVI (Figure 2). During both of hypovolemia. When normovolemia was restored
stages 2 and 3, the CO increased by > 30% follow- (stage 4), all of those variables returned to values similar
ing the PLRM for all pigs. The PLRM also caused a to those at baseline, except for the mean lactate concen-
significant increase in MAP, CVP, and PAOP and a sig- tration, which remained significantly increased from
nificant decrease in PPV. After the pelvic limbs were baseline but was significantly decreased from that at
returned to their original position, all of those vari- stage 3. The mean lactate concentration differed signifi-
ables except PVI returned to values similar to those cantly (P = 0.03) between stages 2 and 3.
before the PLRM. The PVI remained unchanged from
that at 3 minutes after the PLRM. The PLRM had no Discussion
effect on the PI or dynamic compliance of the respira- Results of the present study indicated that a
tory system. modified PLRM was useful for assessing fluid respon-

28 AJVR • Vol 80 • No. 1 • January 2019

Unauthenticated | Downloaded 01/31/23 04:14 AM UTC

 siveness in hypovolemic pigs. When the PLRM was
applied to pigs that had undergone controlled hem-
orrhage such that the total circulating blood volume
was depleted by 20% and 40%, the CO increased by
approximately 30% and 33%, respectively. The PLRM
also caused a significant decrease in PVI and PPV
when applied to hypovolemic pigs; however, when
the PLRM was discontinued and the pelvic limbs
were returned to their original position, the PPV im-
mediately returned to its pre-PLRM value, whereas
the PVI was slower to respond. Interestingly, the PCV
did not vary significantly as pigs transitioned from a
state of normovolemia to severe hypovolemia back to
normovolemia and eventually to hypervolemia. We
attributed that to splenic contraction induced by the
physiologic stress associated with the experimental
procedures, which resulted in the release of RBCs
into the circulation, and the fact that blood withdraw-
al and reinfusion occurred over a fairly short period
of time, which precluded those changes from being
reflected in the PCV. Conversely, lactate concentra-
tion appeared to be acutely sensitive to changes in
the circulating blood volume.
In human patients, both the PLRM and
Trendelenburg maneuver are routinely used to
assess fluid responsiveness or as therapeutic
maneuvers for individuals awaiting fluid resuscita-
tion.24,25 The primary difference between the modi-
fied PLRM described in the present study and the
Trendelenburg maneuver is that, during the modified
PLRM, the head and thorax remain in a horizontal
plane while the pelvic limbs and caudal portion of
the abdomen are raised. During the Trendelenburg
maneuver, the whole body is tilted such that the
head is in a dependent position. The Trendelenburg
maneuver is associated with several undesirable ef-
Figure 2—Mean ± SD CO (A), PVI (B), and PPV (C) imme- fects that are not observed during the PLRM, such as
diately before (dotted line with circles) and 3 minutes after an increase in intraocular and intracranial pressures,
(solid line with triangles) the modified PLRM described in Fig- cerebral edema secondary to venous congestion, a de-
ure 1 was performed and 1 minute after the pelvic limbs and crease in respiratory expansion and vital lung capacity
abdomen were returned to their original positions (dashed
line with squares) during each of 5 sequential experimental resulting in atelectasis and altered ventilation-perfusion
stages for 6 healthy isoflurane-anesthetized Landrace pigs that ratios, and a high risk for regurgitation and aspiration of
were mechanically ventilated following induction of a neuro- gastric contents.24,25 Furthermore, Trendelenburg posi-
muscular blockade (1 mg of atracurium/kg, IV). Each pig was tioning will increase blood flow in the carotid arteries,
in a state of normovolemia during stage 1 (baseline). During which can cause physiologic changes that may mask flu-
stage 2, a volume of blood equal to 20% of the estimated total
circulating blood volume was withdrawn over a period of 15 id resuscitation–induced changes to the Frank-Starling
minutes. During stage 3, additional blood was withdrawn over curve, thereby making differentiation between hypo-
a period of 15 minutes until approximately 40% of the esti- perfused and hyperperfused states more challenging.26
mated total circulating blood volume was removed and the pig Regardless of whether patients undergo the
was in a state of hypovolemic shock. During stage 4, the total PLRM or Trendelenburg maneuver, the magnitude of
volume of blood removed during stages 2 and 3 was reinfused
over a period of 15 minutes to restore the pig to a state of the physiologic effect is dependent on the tilt angle
normovolemia. During stage 5, each pig received a 500-mL IV and duration of the maneuver. For the present study,
bolus of a colloid over 15 minutes to achieve a hypervolemic we modified the PLRM commonly used in human
state. There was a 10-minute interval between stages and a patients on the basis of our understanding of blood
10-minute interval between experimental manipulation of the
circulation and the PLRM to allow hemodynamic variables to volume differences between the legs of humans and
stabilize. None of the mean values differed significantly be- pelvic limbs of quadrupeds. We chose to raise the pel-
tween stages 4 and 5. *Value differs significantly (P < 0.05) vic limbs and caudal portion of the abdomen to a 15°
from the corresponding value at stage 1. †Value differs sig- angle relative to the horizontal plane of the table sur-
nificantly (P < 0.05) from the corresponding value at stage 2. face on the basis of results of a pilot study in which we
‡Value differs significantly (P < 0.05) from the corresponding
values at stages 4 and 5. a,bValues with different lowercase let- used a similar PLRM in dogs. That was a substantially
ters differ significantly (P < 0.05). smaller angle than the angle (45°) used by investiga-

AJVR • Vol 80 • No. 1 • January 2019 29

Unauthenticated | Downloaded 01/31/23 04:14 AM UTC

 Table 2—Mean ± SD hematologic variables during each of the 5 experimental stages for the pigs
of Table 1.
Variable Stage 1 Stage 2 Stage 3 Stage 4 Stage 5
PCV (%) 34.0 ± 3.9 31.8 ± 4.0 29.6 ± 3.3 30.0 ± 2.4 32.0 ±1.7
TPP (g/dL) 5.3 ± 0.2 4.5 ± 0.3* 3.4 ± 0.3*‡ 5.0 ± 0.4 4.9 ± 0.4
Lactate (mmol/L) 0.2 ± 0.4 2.4 ± 0.8*‡ 3.8 ± 0.9*†‡ 1.4 ± 1.5* 0.5 ± 1.2
Paco2 (mm Hg) 43.1 ± 1.2 36.2 ± 2.2*‡ 31.9 ± 1.1*†‡ 42.1 ± 3.1 42.9 ± 1.9
Pao2 (mm Hg) 358.6 ± 88.9 361.3 ± 91.2 351.0 ± 84.7 352.5 ± 94.0 343.1 ± 90.4
Arterial pH 7.27 ± 0.01 7.22 ± 0.01*‡ 7.17 ± 0.01*†‡ 7.27 ± 0.01 7.28 ± 0.0
Hematologic variables were recorded only once during each experimental stage.
See Table 1 for remainder of key.

tors of another study21 in which a PLRM was applied Complete control of the ventilation of the study pigs
to pigs. Nevertheless, the 15° angle was sufficient to was critical to assess the cyclic oscillations in the ple-
cause a > 30% increase in the CO of moderately to thysmographic waveform and percentage variation
severely hypovolemic pigs. Although the pigs of the in PI and calculate the PVI and PPV. Compliance of
present study were healthy, we do not see any reason the chest wall relative to compliance of the lungs is
why this inexpensive and easy-to-perform modified an important determinant for PVI.1,4 In the present
PLRM could not be used for clinical patients, particu- study, the PLRM did not significantly affect the dy-
larly those in which the Trendelenburg maneuver is namic compliance of the lungs and chest wall, which
contraindicated. Potential indications for use of the increased the sensitivity of the PVI for detection of
modified PLRM described in this study included de- PLRM-induced hemodynamic changes.
termination of the fluid responsiveness of a patient Traditionally, variables associated with cardiac
or as a therapeutic maneuver to aid in the temporary filling, such as CVP and PAOP, are used to assess fluid
resuscitation of patients in acute circulatory shock. responsiveness of patients. For the pigs of the present
The pigs of the present study were premedicated study, the changes observed in the CVP and PAOP in
with ketamine (10 mg/kg, IM) to provide sedation response to the controlled hemorrhage, PLRM, and vol-
and restraint during application of a face mask for de- ume replacement were similar to those observed for
livery of isoflurane during anesthesia induction. Ke- dogs of another study.7 The relationship between car-
tamine (5 to 20 mg/kg, IM) is commonly used in com- diac filling pressures, such as CVP and PAOP, and fluid
bination with other drugs to provide sedation and in- responsiveness of individuals is poor in various clinical
jectable anesthesia for short procedures in pigs.27 The settings,30–32 most likely because cardiac end-diastolic
mean ± SD duration between premedication with ket- volume is not solely dependent on cardiac filling pres-
amine and completion of instrumentation was 160.83 sure; it is also affected by compliance of the cardiac
± 17.3 minutes. Therefore, the systemic effects of ket- chambers, venous tone, and intrathoracic pressures.
amine should have dissipated before initiation of the The authors of a systematic review and meta-
5 experimental stages and measurement of hemody- analysis33 concluded that PVI, compared with CO,
namic variables. stroke volume, and PPV, was reasonably accurate for
Cyclic oscillations in stroke volume induced by identification of individuals likely to be responsive to
mechanical ventilation correspond to beat-to-beat fluid therapy when measured under controlled ven-
changes in the amplitude of pulse pressure measured tilation. Results of a study7 involving isoflurane-anes-
with an intra-arterial catheter (PPV) and in the ampli- thetized dogs that underwent controlled hemorrhage
tude of the plethysmographic waveform displayed by to induce hypotension followed by volume replace-
pulse oximetry (PVI). Changes in intrathoracic pres- ment indicate that PPV and PVI might be useful for
sure during an entire respiratory cycle of the ventila- identification of fluid-responsive dogs. Interestingly,
tor are used to predict the magnitude of cyclic oscil- when hypovolemia was induced in the pigs of the
lations in stroke volume, which further determines present study, the PVI, CO, and PPV readily respond-
whether an individual is likely to be responsive (ie, is ed to the PLRM, but when the pelvic limbs were re-
at a steep part of the Frank-Starling curve) or nonre- turned to their original position, the PVI was slow to
sponsive (ie, is at a flat portion of the Frank-Starling respond, whereas the CO and PPV rapidly returned to
curve) to fluid therapy.4,5 Hence, PVI and PPV are pre-PLRM values. The PVI is measured noninvasively
commonly referred to as dynamic preload variables by the pulse oximeter and is therefore dependent on
and need to be measured under strict ventilatory peripheral blood flow (ie, the pulse oximeter was
conditions (ie, controlled ventilation with a tidal vol- placed on the tongue). We postulated that the appar-
ume > 8 mL/kg)28,29 because spontaneous respiration ent lack of responsiveness in the PVI after the pelvic
makes their measurement less reliable.1 The pigs of limbs were returned to their original position was
the present study were administered atracurium to caused by interference in light absorption induced by
induce a neuromuscular blockade and were mechani- sudden modifications in blood flow as the position of
cally ventilated throughout the duration of the ex- the pelvic limbs was changed. Also, the pulse oxim-
periment to prevent patient-ventilator dyssynchrony. eter used in the present study has a built-in algorithm

30 AJVR • Vol 80 • No. 1 • January 2019

Unauthenticated | Downloaded 01/31/23 04:14 AM UTC

for calculation of the PVI, and there is a lag between we believe that the PLRM induced a steady hemo-
peripheral blood flow measurement and generation dynamic state for approximately 5 minutes, after
of the PVI readout. It is possible that recording the which the heart slowly adjusted to the new volume
PVI 1 minute after the pelvic limbs were returned status. Finally, experimental blood volume depletion
to their original position was an insufficient amount was set at 20% and 40% during stages 2 and 3, respec-
of time for the pulse oximeter to detect and register tively; therefore, receiver operating characteristic
changes in PVI associated with repositioning of the curve analysis could not be conducted to determine
pelvic limbs. the optimal blood volume loss necessary to accurate-
Perfusion index is an indicator of the pulse ly identify fluid responsiveness on the basis of PLRM
strength at the site where the pulse oximeter probe results.
is placed. To optimize measurement of PI, it is im- In the present study, a modified PLRM in which the
portant that the peripheral tissue where the probe pelvic limbs and caudal portion of the abdomen were
is placed has good blood perfusion and oxygen- raised at a 15° angle to the horizontal plane was used to
ation. Consequently, factors that affect perfusion and assess fluid responsiveness during experimental induc-
oxygenation of peripheral tissues, such as low CO, tion and correction of severe hypovolemia in isoflurane-
hypotension, hypothermia, vasoactive drug admin- anesthetized and mechanically ventilated pigs. Results
istration, vasoconstriction, and peripheral vascular indicated that, when pigs were in a state of moderate
disease, can adversely affect measurement of PI.34 (20% blood volume depletion) or severe (40% blood
The pigs of the present study were not administered volume depletion) hypovolemia, use of the PLRM re-
any drugs (eg, vasopressors or α2-adrenergic receptor sulted in a significant increase in CO and decrease in
agonists) that might cause peripheral vasoconstric- PVI and PPV. When the pelvic limbs and abdomen were
tion and were maintained in a state of normothermia returned to their original positions, the CO and PPV
throughout the experiment. Therefore, we believe rapidly returned to their pre-PLRM values; however, the
that the decrease in PI observed during stages 2 and PVI, which was noninvasively measured by means of
3 (hypovolemia) was the result of hypoperfusion of pulse oximetry, was slower to respond. Nevertheless,
the tongue secondary to substantial blood loss and the findings suggested that the modified PLRM might
compensatory vasoconstriction. It is also important be useful for identification of hemodynamically unsta-
to note that the pulse oximeter cannot distinguish ble animals that are likely to respond to fluid therapy.
whether a change in PI was caused by a change in Further research is necessary to validate the modified
intrathoracic pressure or peripheral tissue perfu- PLRM for prediction of fluid responsiveness in clinically
sion.33 Thus, anything that affects peripheral tissue ill veterinary patients.
perfusion (and by extension PI) will affect the accu-
racy of PVI for prediction of fluid responsiveness.34 Acknowledgments
In the present study, the PI values varied by < 1% Supported by University of Florida research funds. The fund-
throughout the duration of stage 3 (severe hypovole- ing source did not have any involvement in the study design, data
mia), and PVI values should be interpreted cautiously. analysis and interpretation, or writing and publication of the
manuscript.
The effects of hypovolemia and poor peripheral tis- The authors declare that there were no conflicts of interest.
sue perfusion on PVI warrant further investigation in The authors thank Dr. Luisito Pablo for creating the schematics
veterinary species. for Figure 1.
The present study had multiple limitations. The
sample size was small owing to the terminal nature Footnotes
of the study and other ethical considerations. The se- a. Zetamine, MWI Animal Health, Boise, Idaho.
quence of the experimental stages could not be ran- b. Isoflo, Abbott Laboratories, North Chicago, Ill.
c. Narkomed GS, Drager Inc, Telford, Pa.
domized because the transition of pigs from normo- d. Surflo Etfe IV catheters, Terumo Medical Corp, Somerset, NJ.
volemia to hypovolemia back to normovolemia and e. IntelliVue Gas Module G1, Philips Healthcare, Cambridge,
eventually to hypervolemia in that order was criti- Mass.
cal to evaluate whether the PLRM could be used to f. Intellivue MP50 Patient Monitor, Philips Healthcare, Cam-
predict fluid responsiveness and nonresponsiveness bridge, Mass.
g. Datex Ohmeda Cardiocap AS-5 with spirometry, GE Health-
among the study subjects. The data acquisition times care, Little Chalfont, England.
(immediately before and 3 minutes after the PLRM h. Bair Hugger Animal Health Blankets, 3M, Minneapolis, Minn.
and 1 minute after the pelvic limbs and abdomen i. Masimo Corp, Irvine, Calif.
were returned to their original positions) for hemo- j. ARROW Femoral Arterial Line Catheterization Kit, Teleflex
Inc, Morrisville, NC.
dynamic variables during each experimental stage k. i-STAT handheld blood gas analyzer, Abbott Laboratories,
were selected arbitrarily because we did not have East Windsor, NJ.
any pilot data and could not find any information in l. Deltran Pressure Transducer System DPT 324, Utah Medical
the scientific literature to help guide that decision. Products Inc, Midvale, Utah.
However, cardiovascular variables such as heart rate, m. ARROW Sheath Introducers, Teleflex Inc, Morrisville, NC.
n. Swan-Ganz Catheter Model 131HF7, Edwards Lifesciences
invasive blood pressure, CVP, PPV, and PVI were Corp, Irvine, Calif.
measured continuously. On the basis of evaluation of o. Chart Pro, version 7.3, ADInstruments Pty Ltd, Bella Vista,
the cardiovascular data collected during this study, NSW, Australia.

AJVR • Vol 80 • No. 1 • January 2019 31

Unauthenticated | Downloaded 01/31/23 04:14 AM UTC

 p. Atracurium besylate injection, Sagent Pharmaceuticals Inc, 16. Monnet X, Rienzo M, Osman D, et al. Passive leg raising pre-
Schaumburg, Ill. dicts fluid responsiveness in the critically ill. Crit Care Med
q. Innervator NS 252 Nerve Stimulator, Fisher and Paykel 2006;34:1402–1407.
Healthcare, Irvine, Calif. 17. Keller G, Cassar E, Desebbe O, et al. Ability of pleth variabili-
r. VetStarch, Zoetis Inc, Parsippany, NJ. ty index to detect hemodynamic changes induced by passive
s. VetCo, Albuquerque, NM. leg raising in spontaneously breathing volunteers. Crit Care
t. SAS, version 9.4, SAS Institute Inc, Cary, NC. 2008;12:R37.
18. Schoonjans A, Forget P, Labriola L, et al. Pleth variabil-
References ity index combined with passive leg raising-induced pulse
pressure variation to detect hypovolemia in spontaneously
1. Marik PE. Hemodynamic parameters to guide fluid therapy. breathing patients. Acta Anaesthesiol Belg 2010;61:147–150.
Transfus Altern Transfus Med 2010;1:102–112. 19. Delerme S, Renault R, Le Manach Y, et al. Variations in pulse
2. Rivers E, Nguyen B, Havstad S, et al. Early goal-directed thera- oximetry plethysmographic waveform amplitude induced
py in the treatment of severe sepsis and septic shock. N Engl by passive leg raising in spontaneously breathing volunteers.
J Med 2001;345:1368–1377. Am J Emerg Med 2007;25:637–642.
3. Lopes MR, Oliveira MA, Pereira VO, et al. Goal-directed fluid 20. Almond GW. Research applications using pigs. Vet Clin
management based on pulse pressure variation monitoring North Am Food Anim Pract 1996;12:707–716.
during high-risk surgery: a pilot randomized controlled trial. 21. Dragoumanos V, Iacovidou N, Chalkias A, et al. Passive leg
Crit Care 2007;11:R100. raising during cardiopulmonary resuscitation results in im-
4. Marik PE, Cavallazzi R, Vasu T, et al. Dynamic changes in ar- proved neurological outcome in a swine model of prolonged
terial waveform derived variables and fluid responsiveness in ventricular fibrillation. Am J Emerg Med 2012;30:1935–1942.
mechanically ventilated patients: a systematic review of the 22. Lundeen G, Manohar M, Parks C. Systemic distribution of
literature. Crit Care Med 2009;37:2642–2647. blood flow in swine while awake and during 1.0 and 1.5 MAC
5. Cannesson M, Desebbe O, Rosamel P, et al. Pleth variability isoflurane anesthesia with or without 50% nitrous oxide.
index to monitor the respiratory variations in the pulse ox- Anesth Analg 1983;62:499–512.
imeter plethysmographic waveform amplitude and predict 23. Hansard SL, Sauberlich HE, Comar CL. Blood volume of
fluid responsiveness in the operating theatre. Br J Anaesth swine. Proc Soc Exp Biol Med 1951;78:544–545.
2008;101:200–206. 24. Geerts BF, van den Bergh L, Stijnen T, et al. Comprehensive
6. Sandroni C, Cavallaro F, Marano C, et al. Accuracy of ple- review: is it better to use the Trendelenburg position or pas-
thysmographic indices as predictors of fluid responsiveness sive leg raising for the initial treatment of hypovolemia? J
in mechanically ventilated adults: a systematic review and Clin Anesth 2012;24:668–674.
meta-analysis. Intensive Care Med 2012;38:1429–1437. 25. Halm MA. Trendelenburg position: “put to bed” or angled
7. Klein AV, Teixeira-Neto FJ, Garofalo NA, et al. Changes in toward use in your unit? Am J Crit Care 2012;21:449–452.
pulse pressure variation and plethysmographic variability 26. Zadini F, Newton E, Abdi AA, et al. Use of the Trendelenburg
index caused by hypotension-inducing hemorrhage followed position in the porcine model improves carotid flow dur-
by volume replacement in isoflurane-anesthetized dogs. Am ing cardiopulmonary resuscitation. West J Emerg Med
J Vet Res 2016;77:280–287. 2008;9:206–211.
8. Monnet X, Teboul JL. Passive leg raising. Intensive Care Med 27. Hodgkinson O. Practical sedation and anesthesia in pigs. In
2008;34:659–663. Pract 2007;29:34–39.
9. Dong ZZ, Fang Q, Zheng X, et al. Passive leg raising as an in- 28. De Backer D, Heenen S, Piagnerelli M, et al. Pulse pressure
dicator of fluid responsiveness in patients with severe sepsis. variations to predict fluid responsiveness: influence of tidal
World J Emerg Med 2012;3:191–196. volume. Intensive Care Med 2005;31:517–523.
10. Monnet X, Teboul JL. Passive leg raising: five rules, not a 29. Renner J, Scholz J, Bein B. Monitoring fluid therapy. Best
drop of fluid! Crit Care 2015;19:18. Pract Res Clin Anaesthesiol 2009;23:159–171.
11. Cavallaro F, Sandroni C, Marano C, et al. Diagnostic accuracy 30. Kumar A, Anel R, Bunell E, et al. Pulmonary artery occlu-
of passive leg raising for prediction of fluid responsiveness in sion pressure and central venous pressure fail to predict
adults: systematic review and meta-analysis of clinical stud- ventricular filling volume, cardiac performance, or the re-
ies. Intensive Care Med 2010;36:1475–1483. sponse to volume infusion in normal subjects. Crit Care Med
12. Monnet X, Marik P, Teboul JL. Passive leg raising for predict- 2004;32:691–699.
ing fluid responsiveness: a systematic review and meta-anal- 31. Marik PE, Baram M, Vahid B. Does central venous pressure
ysis. Intensive Care Med 2016;42:1935–1947. predict fluid responsiveness? A systematic review of the lit-
13. Duus N, Shogilev DJ, Skibsted S, et al. The reliability and va- erature and the tale of seven mares. Chest 2008;134:172–178.
lidity of passive leg raise and fluid bolus to assess fluid re- 32. Osman D, Ridel C, Ray P, et al. Cardiac filling pressures are
sponsiveness in spontaneously breathing emergency depart- not appropriate to predict hemodynamic response to volume
ment patients. J Crit Care 2015;30:e1–e5. challenge. Crit Care Med 2007;35:64–68.
14. Biais M, Vidil L, Sarrabay P, et al. Changes in stroke volume 33. Chu H, Wang H, Sun Y, et al. Accuracy of pleth variability
induced by passive leg raising in spontaneously breathing pa- index to predict fluid responsiveness in mechanically venti-
tients: comparison between echocardiography and Vigileo/ lated patients: a systematic review and meta-analysis. J Clin
FloTrac device. Crit Care 2009;13:R195. Monit Comput 2016;30:265–274.
15. Préau S, Saulnier F, Dewavrin F, et al. Passive leg raising is 34. Broch O, Bein B, Gruenewald M, et al. Accuracy of the pleth
predictive of fluid responsiveness in spontaneously breath- variability index to predict fluid responsiveness depends on
ing patients with severe sepsis or acute pancreatitis. Crit the perfusion index. Acta Anaesthesiol Scand 2011;55:686–
Care Med 2010;38:819–825. 693.

32 AJVR • Vol 80 • No. 1 • January 2019

Unauthenticated | Downloaded 01/31/23 04:14 AM UTC

View publication stats