You are on page 1of 6

The Effects of the Reverse Trendelenburg Position on

Respiratory Mechanics and Blood Gases in Morbidly Obese

Patients During Bariatric Surgery
Valter Perilli, MD, Liliana Sollazzi, MD, Patrizia Bozza, MD, Cristina Modesti, MD,
Angelo Chierichini, MD, Roberto Maria Tacchino, MD, and Raffaela Ranieri, MD
Departments of *Anesthesiology and Surgery, Catholic University of Sacred Heart Rome, Rome, Italy

Anesthesia adversely affects respiratory function, particularly in morbidly obese patients. Although many
studies have been performed to determine the optimal
ventilatory settings in these patients, this question has
not been answered. The aim of this study was to evaluate the effect of reverse Trendelenburg position (RTP)
on gas exchange and respiratory mechanics in 15 obese
patients undergoing biliopancreatic diversion. A standardized anesthetic regimen was used and patients
were examined at standard times: 1) after tracheal intubation, 2) after laparotomy, 3) after positioning of subcostal retractors, 4) with retractors in RTP. The measurements of respiratory mechanics were repeated for a

major problem in anesthesia for morbidly obese

patients is the adequacy of pulmonary ventilation (1,2). Anesthesia adversely affects respiratory function, leads to a smaller functional residual
capacity (FRC), and promotes airway closure and atelectasis (2 4). In obese patients, FRC markedly decreases with possible hypoxemia in the perioperative
period (59). Although many studies have been performed to determine the optimal ventilatory settings
and posture in these patients, the question has not
been resolved (2). In particular, there are few reports
that deal with changes in respiratory mechanics and
gas exchange in obese patients placed in reverse Trendelenburg position during general anesthesia. Furthermore, Buchwald (10) claimed that the use of a
fixed-support retractor system and reverse Trendelenburg position is extremely useful in obese patients
undergoing surgery of the upper abdomen.
Therefore, the aim of this study was to evaluate the
effects of reverse Trendelenburg posture (RTP) on gas

Accepted for publication August 2, 2000.

Address correspondence and reprint requests to V. Perilli, MD,
Department of Anesthesiology, Catholic University of Sacred Heart
Rome, Largo A. Gemelli, 8, 00168 Roma, Italy.


Anesth Analg 2000;91:15205

wide range of tidal volumes by using the technique of

rapid occlusion during constant flow inflation. We
noted a wide alveolar-arterial oxygen difference
[P(Aa)o2] in all patients, particularly during Phase 3.
When the patients were placed in RTP, P(Aa)o2
showed a significant improvement and a return toward
baseline values. As for mechanics, total respiratory system compliance was significantly higher in RTP than in
the other phases. In conclusion, our data suggest that
RTP is an appropriate intraoperative posture for obese
subjects because it causes minimal arterial blood pressure changes and improves oxygenation.
(Anesth Analg 2000;91:1520 5)

exchange variables and respiratory mechanics in

obese patients undergoing abdominal surgery.

After institutional approval, informed consent was
obtained. We studied 15 otherwise healthy nonsmoking morbidly obese patients undergoing biliopancreatic diversion.
A standardized anesthetic regimen was used; after
5 min of breathing oxygen, anesthesia was induced
with thiopental (35 mg/kg) and tracheal intubation
was facilitated by succinylcholine (1 mg/kg IV). Anesthesia was maintained with incremental doses of
fentanyl (up to 5 g/kg), isoflurane, and N2O; neuromuscular blockade was induced with atracurium
(0.5 mg/kg plus 0.5 mg kg1 h1).
Patients lungs were mechanically ventilated (Servo
C; Siemens Elema AB, Berlin Germany) with constant
inspiratory flow, aiming at an ETco2 of 30 mm Hg;
respiratory rate was 12 2 breaths/min, tidal volume
(TV) ranged from 6 to 8 mL per kg of body weight and
was kept essentially constant throughout the surgical
procedure with the fraction of inspired oxygen (Fio2)
of 50%. No positive end-expiratory pressure was used
2000 by the International Anesthesia Research Society

2000;91:1520 5



Table 1. Respiratory Data

P(A a)o2 (mm Hg)
Pao2 (mm Hg)
Paco2 (mm Hg)
ETco2 (mm Hg)
PawPeak (cm H2O)
PawPlat (cm H2O)
Ctot (mL/cm H2O)

Phase 1

Phase 2

Phase 3

Phase 4

132 60 (79185)
177 68 (101290)
39 8 (2952)
33 6.4 (2546)
33 6 (2249)
23 4 (1536)
32 5 (2241)

162 69 (86183)
152 52 (79248)
34 5 (2947)
30 4.1 (2339)
30 5 (1943)
23 4 (1631)
31 6 (2140)

207 33 (155249)*
115 41 (78200)a
34 5 (2843)
29 5 (2437)
31 2 (3040)
22 4 (1929)
32 5.6 (2040)

159 63 (45262)
156 55 (85237)
32 4 (2640)
28 3.9 (2334)*
26 3 (2133)*
19 3 (1426)*
41 5 (3250)

Data are presented as mean sd (range).

PAWpeak peak airway pressure; PAWplat plateau airway pressure; Ctot total respiratory compliance; P(A a)o2 alveolar-arterial oxygen difference.
Statistical differences compared to Phase 1.
* P 0.05; P 0.01.

and airway plateau pressure was maintained at a level

of 35 cm H2O.
In addition to the usual multiparametric monitoring, direct arterial pressure monitoring (radial artery
catheter inserted after the induction) was used. The
average time of the surgical procedure was 131
36 min, and fluid administration consisted of 23 L
lactated Ringers solution (8 10 mL kg1 h1).
Respiratory mechanics and blood gases were examined at the following standard times: 1) after tracheal
intubation, 2) after laparotomy, 3) after positioning of
subcostal retractors, and 4) with retractors in RTP
(head up 30). The time interval between each phase
was never 20 min.
In these sets of measurements, end-expiratory and
end-inspiratory airway occlusions have been obtained
by using end-expiratory and end-inspiratory hold buttons of the Servo C for end-expiratory and endinspiratory occlusion, respectively. The occlusion at
the end of expiration provides measurement of intrinsic positive end-expiratory pressure. Specifically developed software provided online time-related trends
of airway pressures during inspiratory and expiratory
block. This system was based on a personal computer
equipped with a 12-bit analog-to-digital converter (20
samples/s per channel) and connected with a Servo
C37 pin analogic plug by using a short shielded cable.
The use of this software allows elaboration of data in
real time.
The measurement of respiratory mechanics was repeated for a wide range of TVs (6 8 different TVs for
each patient) to obtain a volume/pressure curve for
each patient in all phases. The different TVs were
assessed by changing respiratory frequency on the
ventilator, and after each measurement baseline ventilation was resumed. TVs ranged from 150 mL to
1200 mL and respiratory rates ranged from 6 to 30
breaths/min. During these maneuvers, Spo2 never decreased to 91%, and no muscular twitch was elicitable (Microstim Wellcome). No patient showed evidence of barotrauma on radiographs taken after
surgery, and there were no pulmonary complications
before hospital discharge.

The total respiratory system compliance (Ctot) was

conventionally obtained by dividing the deflation volume by the difference between the plateau pressure
measured during end-inspiratory airway occlusion
(breath holding for 2 4 s) and end-expiratory pressure.
In each phase before respiratory mechanics measurements, arterial blood samples were taken to evaluate
pulmonary gas exchange. Alveolar-arterial oxygen difference [P(A-a)o2] was obtained from the following

P(A a)o2 Fio2(PB PH2O)

Paco2/RQ Pao2 (1)
where PB is the actual barometric pressure, PH2O is
the water vapor tension at 37C, and RQ the respiratory quotient assumed to be 0.8.
A commercial software package (Statgraphics; SGS,
Rockville, MD) was used for analysis of the data.
Values were expressed as mean and sd unless otherwise stated; ranges are shown in Table 1. Statistical
analysis was performed as analysis of variance followed by post hoc comparisons of means (Tukey) when
the F values indicated significant differences among
groups. Regression analysis was used to determine
relationships (Pearsons correlation coefficients) between variables. P 0.05 was accepted as statistically

Anthropometric and spirometric data are in Table 2;
respiratory data are summarized in Table 1.
A wide P(Aa)o2 occurred in all patients throughout the operation; the higher values coincided with the
application of subcostal retractors. At this time (Phase
3), P(Aa)o2 showed a significant increase and Pao2
decreased. When the patients were placed in RTP,
oxygenation improved with a return toward baseline
values; P(Aa)o2 was ameliorated in 80% of the patients (12/15). As for respiratory mechanics, with the
same ventilatory setting, average Ctot was higher and



2000;91:1520 5

Table 2. Anthropometric and Spirometric Data

AGE (yr)
BMI (kg/m2)
FVC (% of predicted)
FEV1 (% of predicted)
FEV1/FVC (% of predicted)

32.8 9
130.5 6.5
163 6.3
46.7 9.8
95.4 6.
90.2 8
97.4 6.5

Values are n or mean sd.

BMI body mass index, FVC forced vital capacity, FEV1 forced
expiratory volume in 1 s.

Figure 2. Top, correlation between airways pressure (press) and

tidal volume (vol) in Phase 3. Top left, all data points are shown (n
141), r 0.90 P 0.001; top right, individual lines of correlation
are shown. Bottom, correlation between airway pressure and tidal
volume in Phase 4. Bottom left, all data points are shown (n 150),
r 0.91 P 0.001; bottom right, individual lines of correlation are

Figure 1. Top, correlation between airways pressure (press) and

tidal volume (vol) in Phase 1. Top left, all data points are shown (n
143), r 0.89 P 0.001; top right, individual lines of correlation
are shown. Bottom, correlation between airway pressure and tidal
volume in Phase 2. Bottom left, all data points are shown (n 150),
r 0.88 P 0.001; bottom right, individual lines of correlation are

compliance [P(Aa)o2 vs Ctot r 0.32, P 0.03],

whereas changes of P(Aa)o2 were more closely correlated with variations of Ctot (r 0.45, P 0.05).
Stepwise regression analysis showed that Fio2 and
Ctot accounted for 44% of change in Pao2. Other than
diastolic blood pressure, cardiovascular variables did
not change significantly (Table 4).

airway pressures were significantly lower in Phase 4
than in the other phases. Volume-pressure relationships were obtained in all phases for each patient;
individual lines of regression and pooled data (with
fitted lines) are shown in Figures 1 and 2. Correlations
between airway pressures and TVs were found highly
significant for both pooled and individual patients
(Table 3). The slopes of Phases 1, 2, and 3 are significantly lower than that of Phase 4 (P 0.01). Gas
exchange variables showed a weak correlation with

Measurement of respiratory mechanics can be useful

for examining patients whose lungs are mechanically
ventilated, and some techniques are suitable to evaluate anesthetized patients; we used one of these techniques, i.e., the rapid occlusion during constant flow
As previously mentioned, the maintenance of an
adequate pulmonary ventilation and oxygenation may
still be a major problem in anesthetized obese patients,
because anesthesia significantly affects respiratory

2000;91:1520 5



Table 3. Lines of Regression

Individual Lines



Pooled Lines









P Value



r Pearson coefficient of pooled lines; P significance level of pooled lines.

Significant difference compared to Phase 1: * P 0.01.

Table 4. Hemodynamic Data

SAP (mm Hg)

DAP (mm Hg)
HR (bpm)

Phase 1

Phase 2

Phase 3

Phase 4

121 8 (94165)
89 13 (5097)
75 12 (68115)

131 15 (98170)
85 15 (60100)
82 19 (62123)

129 27 (95165)
90 8 (65103)
80 11 (70108)

123 29 (99148)
75 24 (5696)*
87 15 (67114)

Data are presented as mean sd (range).

SAP systolic arterial pressure, DAP diastolic arterial pressure, HR heart rate.
Statistical differences compared to Phase 1 * P 0.05.

function. The decrease of FRC is one of the main side

effects of anesthesia on respiratory function, and this
change is particularly marked in morbidly obese patients (2,8,9,1113). Pelosi et al. (14) demonstrated that
the reduction of FRC is closely related to body mass
A cranial shift of the diaphragm has been identified
as an important factor causing decrease of FRC in
obese patients undergoing general anesthesia (8,14);
the loss of tone of this muscle may determine the
reduction in lung volume because of unopposed intraabdominal pressure (8,14). However, also atelectasis
seems related to a number of interacting factors that
include the shape of chest wall structures, volume,
and distribution of blood (14). In clinical practice,
some morbidly obese patients do not tolerate the supine posture, and it may even be fatal to them (2,7,15).
Large TVs (1520 mL/ideal body weight) are often
recommended for these patients to move tidal ventilation higher than the closing volume and consequently increase arterial oxygen tension (2,5,7,16).
This traditional approach to mechanical ventilation
intends to aggressively recruit and ventilate atelectatic
lung units, but may risk overdistention of the normal
lung units. Thus, large TVs may cause a decrease in
Paco2, respiratory alkalosis, cardiovascular impairment, and excessive stretch of nondependent lung
regions (2,16,17). Furthermore Bardoczky et al. (16)
demonstrated that very large TVs do not improve
oxygenation in obese patients.
However, even the use of positive end-expiratory
pressure to increase FRC and improve oxygenation in
obese patients is questionable (2,5). Although the use
of positive end-expiratory pressure is of proven value
for improving oxygenation in many situations involving respiratory failure, its role in anesthetized patients

is controversial. In normal subjects, positive endexpiratory pressure can reduce the atelectasis but not
necessarily the shunt (18); however, recently Pelosi et
al. (19) claimed that positive end-expiratory pressure
can improve oxygenation and respiratory mechanics
in obese patients. Nevertheless, a possible detrimental
effect of positive end-expiratory pressure on the oxygenation of obese patients has been described by Salem (20).
In this study performed during upper abdomen
surgery, we used the RTP to counteract the weight of
the abdominal contents and the effects of the retractors
on the diaphragm. A wide P(Aa)o2 was observed in
all of our patients in every phase, with a further significant increase during Phase 3; this worsening seems
related to the application of subcostal retractors,
which may cause a further decrease of FRC and a
more limited movement of the diaphragm during mechanical ventilation. In Phase 4, with the patients
placed in RTP without removing the retractors, oxygenation indexes improved, and P(Aa)o2 returned
toward the baseline values.
As for respiratory mechanics, RTP determined a
significant increase in the compliance that reached a
level higher than baseline; the slopes of volumepressure relationships in supine postures were less
than that in RTP (Table 3). The increased compliance
obtained by RTP and steeper volume/pressure regression lines suggest a recruitment of alveolar units and
therefore an increase of FRC; thus the operating compliance could be improved by increasing end expiratory volume toward the more compliant range.
A limitation of our study might be the lack of the
direct measure of FRC. However, because the reduction in FRC is most likely responsible for the decreased Ctot during general anesthesia, it is reasonable



Figure 3. Correlation between alveolar-arterial oxygen difference

in Phase 3 and changes of alveolar-arterial oxygen difference [Delta
P(Aa)o2] in Phase 4 is shown (y ax b; intercept loga a 14.25
(se 1.8) slope 9.52 (se 0.32); r 0.64; P 0.001).

to argue that the increase of Ctot obtained with RTP

can be related to an increased FRC (13,18). However,
despite this significant increase in compliance, well
above baseline values, P(A-a)o2 did not decrease
A weak correlation was found between compliance
and P(Aa)o2 (r 0.32; P 0.05; r 0.45; P
0.05), meaning that other factors play a significant
role in determining gas exchange efficiency.
As for hemodynamic factors, it is possible that RTP
may compromise cardiovascular function by reducing
venous return. Therefore, the beneficial effects of RTP
on Pao2 could be offset by a decreased cardiac output.
However, obese subjects show a reduced venous compliance and a smaller decrease of the central blood
volume and of the cardiac stroke volume during orthostatic stress (21). Moreover, even if extensive hemodynamic monitoring has not been performed, no
clinically relevant cardiovascular change was noted in
our study.
Another reasonable explanation for this weak relationship between compliance and P(Aa)o2 could be
the redistribution of pulmonary blood flow toward
less-ventilated dependent zones along a gravitational
gradient (22); so reversing the decrease in lung volume with RTP meets with limited success. In this
regard, Heneghan et al. (23) showed that there was no
improvement in oxygenation when lung volumes
were increased significantly by RTP (head up 30) in
normal anesthetized subjects.
However, in our series, other than P(Aa)o2 values
significantly wider than those found by Heneghan et
al. (23) in normal subjects (P 0.01), we found a
correlation between P(Aa)o2 values of Phase 3 and
the improvement in oxygenation obtained with RTP
(r 0.64; P 0.001) (Figure 3); so the improvement
in oxygenation obtained with RTP is greater when
P(Aa)o2 is wider. This relationship, as well as the

2000;91:1520 5

different type of patients, could explain the difference

with Heneghan et als (23) study.
Moreover, in upper abdomen surgery of obese subjects, the exposure of the operative field creates major
problems, and the use of a fixed-support retractor
system greatly facilitates the surgeon (10). Like any
procedure that increases the subdiaphragmatic pressure, it may make a further decrease in FRC and
pulmonary compliance leading to hypoxemia (2). In
this regard, the RTP offers potential advantages; it
ameliorates the oxygenation indexes, exposes at best
the subdiaphragmatic region, and allows mechanical
ventilation with safe levels of airway pressures.
In conclusion, our data suggest that RTP is a simple
and safe intraoperative posture for obese patients and
offers some cardiorespiratory advantages during upper abdominal surgery.
We express our thanks to Dr. A. Ceccarelli for his statistical analysis
of original data presented in this article.

1. Winer J, Brodsky JB, Merrell RC. Massive obesity and arterial
oxygenation. Anesth Analg 1981;60:6913.
2. berg B, Poulsen TD. Obesity: an anesthetic challenge. Acta
Anaesthesiol Scand 1996;40:191200.
3. Hickey RF, Visick WD, Fairley HB, Fourcade HE. Effects of
halothane on functional residual capacity and alveolar-arterial
oxygen tension difference. Anesthesiology 1973;38:20 4.
4. Rehder K, Marsh M. Respiratory mechanics during anesthesia
and mechanical ventilation. In: Macklem PT, Head J, eds. Handbook of physiology section 3, volume III: Mechanics of breathing part 2. Bethesda MD: American Physiological Society, 1986:
5. Shenkman Z, Shir Y, Brodsky JB. Perioperative management of
the obese patient. Br J Anaesth 1993;70:349 59.
6. Buckley FD. Anaesthesia for the morbidly obese patient. Can J
Anaesthesiol 1994;41:R94 100.
7. Dumont L, Mattis M, Mardirosoff C, et al. Changes in pulmonary mechanics during laparoscopic gastroplasty in morbidly
obese patients. Acta Anaesthesiol Scand 1997;41:408 13.
8. Pelosi P, Croci M, Ravagnan I, et al. Respiratory system mechanics in sedated, paralyzed, morbidly obese patients. J Appl
Physiol 1997;82:811 8.
9. Damia G, Mascheroni D, Croci M, Tarenzi L. Perioperative
changes in functional residual capacity in morbidly obese patients. Br J Anaesth 1988;60:574 8.
10. Buchwald H. Three helpful techniques for facilitating abdominal procedures, in particular for surgery in the obese. Am J Surg
1998;175:63 4.
11. Roothen HU, Sporre B, Engberg G, et al. Atelectasis and pulmonary shunting during induction of general anaesthesia: can
they be avoided? Acta Anaesthesiol Scand 1996;40:191200.
12. Nunn JF. Effects of anaesthesia on respiration. Br J Anaesth
1990;65:54 62.
13. Nunn JF. Respiratory aspects of anaesthesia. In: Nunn JF, eds.
Applied physiology. 4th ed. Oxford: Butterworth-Heinemann,
1993:384 417.
14. Pelosi P, Croci M, Ravagnan I, et al. The effects of body mass
index on lung volumes, respiratory mechanics, and gas exchange during general anesthesia. Anesth Analg 1998;87:
654 60.

2000;91:1520 5

15. Tsueda K, Debrand M, Zeok SS, et al. Obesity supine death

syndrome: reports of two morbidly obese patients. Anesth
Analg 1978;58:3457.
16. Bardoczky GI, Yernault JC, Houben JJ, dHollander AA. Large
tidal volume ventilation does not improve oxygenation in morbidly obese patients during anaesthesia. Anesth Analg 1995;81:
385 8.
17. MacIntyre NR. Minimizing alveolar stretch injury during mechanical ventilation.
18. Tockics L, Hedenstierna G, Strandberg , et al. Lung collapse
and gas exchange during general anesthesia: effects of spontaneous breathing, muscle paralysis, and positive end-expiratory
pressure. Anesthesiology 1987;66:157 67.
19. Pelosi P, Ravagnan I, Giurati G, et al. Positive end-expiratory
pressure improves respiratory function in obese but not in
normal subjects during anesthesia and paralysis. Anesthesiol



20. Salem MR. Does PEEP improve intraoperative arterial oxygenation in grossly obese patients? Anesthesiology 1978;48:280 2.
21. Stepniakowsky K, Egan BM. Additive effects of obesity and
hypertension to limit venous volume. Am J Physiol 1995;268:
R562 8.
22. Pelosi P, Croci M, Calappi E., et al. Prone positioning improves
pulmonary function in obese patients during general anesthesia.
Anesth Analg 1996;83:578 83.
23. Heneghan CPH, Bergman NA, Jones JG. Changes in lung volume and (PAO2-PaO2) during anaesthesia. Br J Anaesth
1984;56:437 45.