Journal of Clinical Anesthesia (2015) xx, xxx–xxx
Original contribution
Respiratory gas exchange during robotic-assisted laparoscopic radical prostatectomy☆ Philip Lebowitz MD, MBA (Professor of Clinical Anesthesiology)⁎, Adam Yedlin MD (Assistant Professor of Clinical Anesthesiology), A. Ari Hakimi MD (Fellow, Urology), Christopher Bryan-Brown MD (Emeritus Professor of Anesthesiology), Mahesan Richards MD (Assistant Professor of Clinical Anesthesiology), Reza Ghavamian MD (Professor of Urology) Albert Einstein College of Medicine, Montefiore Medical Center, Bronx, NY, USA Received 17 September 2014; revised 17 March 2015; accepted 1 June 2015
Keywords: Laparoscopic prostatectomy; Total respiratory compliance; Ventilation/perfusion mismatching; Pulmonary shunting; Respiratory gas effects; Complications during recovery
Abstract Study Objective: Robotic-assisted laparoscopic prostatectomy requires patients to be secured in a steep Trendelenburg position for several hours. Added to the CO2 pneumoperitoneum that is created, this positioning invariably restricts diaphragmatic and chest wall excursion, which can adversely affect respiratory gas exchange. This study sought to measure the extent of respiratory gas change during this procedure. Design: Retrospective, institutional review board approved. Setting: Operating room. Patients: N = 186 males, American Society of Anesthesiologists 2-3, with prostatic carcinoma undergoing robotic-assisted laparoscopic radical prostatectomy. Interventions: Arterial blood gases and noninvasive respiratory measurements were recorded for those patients (n = 32) in whom a radial arterial catheter had been inserted intraoperatively, specifically timed to different phases of the procedure: supine lithotomy, steep Trendelenburg, and return to supine. Ventilatory parameters were standardized. Measurements: Systemic blood pressure, heart rate, respiratory rate, PaO2, PaCO2, oxygen saturation as measured by pulse oximetry, and end-tidal carbon dioxide pressure. Main Results: Although no patients developed perioperative respiratory complications, the PaO2 invariably fell (395 vs 316 mm Hg; P = .001) while the patients were in steep Trendelenburg, and the PaCO2–end-tidal carbon dioxide pressure rose (10.0 vs 13.4 mm Hg; P b .0001). Upon return to supine, patients' respiratory measurements promptly returned to within 15% of baseline. Subgroup analysis for high-BMI vs low-BMI patients as well as for patients with pulmonary disease and/or a smoking history showed similar individual effects and only small, although significant, respiratory gas exchange aberrations.
☆
Disclosures: There were no grants, sponsors, or funding sources that provided direct financial support to the research work contained in the manuscript. ⁎ Correspondence: Philip Lebowitz, MD, MBA, Montefiore Medical Center, 111 E. 210 St, Bronx, NY 10467, USA. Tel.: + 1 718 920 6183. E-mail addresses: [email protected] (P. Lebowitz), [email protected] (A. Yedlin), [email protected] (A.A. Hakimi), [email protected] (C. Bryan-Brown), [email protected] (M. Richards), [email protected] (R. Ghavamian). http://dx.doi.org/10.1016/j.jclinane.2015.06.001 0952-8180/© 2015 Elsevier Inc. All rights reserved.
2
P. Lebowitz et al. Conclusions: Positioning patients with a CO2 pneumoperitoneum in steep Trendelenburg for several hours imposes restriction of diaphragmatic and chest wall movement sufficient for respiratory gas exchange to be adversely affected. Return of function to within 15% of baseline occurred within minutes after return to supine and release of the CO2 pneumoperitoneum. No patients during the study period developed pulmonary complications that required alteration in their level of care. © 2015 Elsevier Inc. All rights reserved.
1. Introduction Robotic-assisted laparoscopic radical prostatectomy (RALP) is currently the most commonly used surgical technique for treatment of localized prostate cancer. It requires the patient to be positioned on the operating room table with the legs raised in a modified lithotomy position, the thorax tightly restrained with adhesive tape, and the head down 30° from supine in extreme Trendelenburg position. Coupled with the required surgical pneumoperitoneum, this positioning invariably compromises total respiratory compliance. Attempting to maintain adequate minute ventilation without causing barotrauma, the anesthesiologist frequently encounters impaired oxygenation and carbon dioxide elimination during the procedure. Trendelenburg positioning and CO2 pneumoperitoneum adversely affect oxygenation and increase the alveolar to arterial oxygen gradient by decreasing functional residual capacity and increasing pulmonary shunt fraction. What is less frequently observed is that the reduction in functional residual capacity also adversely affects CO2 elimination and increases the arterial to end-tidal carbon dioxide pressure gradient (Pa-ETCO2) by increasing dead space ventilation [1-3]. Increasing age also contributes to an increase in Pa-ETCO2 [2,4], which is relevant because RALP is typically performed on an aging male population due to the nature of prostate cancer. Although several studies have looked at various parameters including pulmonary blood flow, pulmonary gas exchange, lung compliance, and shunt during different combinations of steep Trendelenburg and CO2 pneumoperitoneum [1,5-7], the extent to which oxygenation and ventilation might be compromised intraoperatively during RALP remains poorly documented. This study sought to further quantify the extent of respiratory gas exchange impairment during this procedure.
2. Materials and methods After approval from the hospital's institutional review board, we reviewed the medical records of all patients at our hospital who had undergone RALP during 2008 and the first 6 months of 2009. All patients (n = 186) who underwent this procedure during the study period were included. All patients had been deemed suitable to undergo general anesthesia for
this elective surgery and were classified as American Society of Anesthesiologists I-III. Upon review of the medical records, we found 32 patients who had radial arterial catheters placed and blood gas data available from multiple key points of the surgical operation including anesthetized baseline (supine lithotomy), after CO2 insufflation with extreme Trendelenburg, and after return to supine. From each of these patients' medical records, we extracted their age, height, weight, and pulmonary status (smoking history, pulmonary function tests, and chronic obstructive pulmonary disease diagnosis) as well as intraoperative tidal volume, respiratory rate, positive end-expiratory pressure (PEEP) level, ratio of inspiratory time to expiratory time ratio, peak airway pressure, pulse oximetry, end-tidal CO2 (ETCO2), PaO2, PaCO2, blood pressure, and heart rate. Our institutional experience with these cases had led us to empirically determine that the optimal ventilation for these patients is achieved with pressure-controlled mechanical ventilation to an intended tidal volume of 6 mL/kg with a ventilator rate of 10 breaths per minute, an I/E ratio of 1:1, and PEEP of 5 cm H2O. We did not allow the peak airway pressure to exceed 40 cm H2O. Consequently, if maintaining a tidal volume of 6 mL/kg under this ventilator pattern caused the peak airway pressure to exceed 40 cm H2O, we would decrease the tidal volume so as to keep the peak airway pressure below this level and incrementally raise the respiratory rate to as high as 13 breaths per minute to approximate a constant minute ventilation. This protocol of mechanical ventilation was indeed used in all 32 of this subgroup of patients. Anesthesia was maintained with sevoflurane or desflurane in 100% oxygen as well as nondepolarizing neuromuscular blockade for all patients; fentanyl was added in 50-μg increments to maintain heart rate in the 60-100 beats per minute range. Collected data were used to compare changes in pulmonary function during the procedure. Continuous data were analyzed using a 2-tailed Student t test. In comparing individual patients throughout different phases of the surgical procedure, we analyzed the data with a paired t test. Data obtained from arterial blood gas measurements were also analyzed using analysis of variance to detect differences for the group as a whole among the stages of the surgical procedure. P b .05 was considered statistically significant. We also performed subgroup analyses to determine if there were any differences in respiratory changes between
3 patients with high body mass index (BMI) (N 25 kg/m2) and low BMI (b 25 kg/m2) as well as in patients with a history of either pulmonary dysfunction or smoking and those patients without any such history. In comparing data between different groups, the data were analyzed using an unpaired t test. In considering outcomes of the procedure, including postoperative pulmonary complications, we used χ2 analysis. P b .05 was considered statistically significant.
3. Results For the 32 medical records reviewed, the mean age was 60 years; the mean weight, 81.4 kg; and the mean BMI, 26 kg/m2. Eighteen patients had BMI N 25 kg/m2, whereas 14 patients had BMI b 25 kg/m2. Fourteen patients had a history of pulmonary dysfunction or extensive smoking, and 18 patients did not. The average total anesthesia time of 4 hours included surgery time of approximately 3 hours, which itself included intraperitoneal insufflation time of approximately 2 hours 15 minutes. Pneumoperitoneum with CO2 was maintained at 15 cm H2O throughout the insufflation period. Anesthetic induction, patient positioning, and anesthetic emergence comprised 1 hour of the total anesthesia time. Not surprisingly, patients experienced lower tidal volumes, increased peak airway pressures, decreased PaO2, increased PaCO2, increased ETCO2, and worsened alveolar-arterial partial pressure of oxygen gradient and Pa-ETCO2 gradients after undergoing intraperitoneal insufflation and extreme Trendelenburg positioning, compared with baseline (Table 1). These differences were all statistically significant. Tidal volumes decreased from 530 ± 102 mL (mean ± SD) at baseline to 473 ± 75 mL after insufflation and extreme Trendelenburg (P = .0056). Peak airway pressure increased from 16.8 ± 2.8 mm Hg to 34.2 ± 4.8 mm Hg (P b .0001). PaO2 decreased from 395 ± 62 mm Hg to 315.7 ± 93.8 mm Hg (P = .001); PA-PaO2 gradient correspondingly increased. PaCO2 increased from 38.3 ± 5.8 to 46.8 ± 5.5 (P b .0001). ETCO2 increased from 28.4 ± 3.7 to 33.4 ± 3.8 (P b .0001). Finally, Pa-ETCO2 gradient increased from baseline of 10 ± 4.5 to 13.4 ± 4.3 (P b .0001). Upon repositioning back to supine, all tidal volumes and peak airway pressures returned to near-baseline values. Oxygenation, as determined by PaO2, was still significantly decreased (P b .001) but was trending upward and returned to within 15% of baseline values. PaCO2, ETCO2, and Pa-ETCO2 gradients all remained significantly elevated during the initial recovery period shortly after repositioning to supine. The subgroup analyses did not show any significant respiratory changes between high (n = 18) vs low (n = 14) BMI groups (Table 2) or between patients with underlying pulmonary dysfunction (n = 14) vs those without (n = 18) such dysfunction (Table 3). Although there were some minor statistically significant differences between the subgroups at baseline, percent changes with insufflation and extreme
Trendelenburg were not significant for any parameter. The most notable finding for the BMI subgroups was that patients with BMI N 25 kg/m2 had significantly increased baseline peak airway pressures when compared with those with BMI b 25 kg/m2 (P = .001), whereas the most notable finding for the pulmonary function subgroups was that patients with pulmonary dysfunction had a significantly increased baseline PaCO2 (P = .02) and Pa-ETCO2 gradient (P = .001). No patients, regardless of risk group, had any pulmonary sequelae despite the significant respiratory perturbations noted above.
4. Discussion This study demonstrates that, although respiratory perturbations during RALP are significant, the lungs seem to have the capacity to withstand the effects of a CO2 pneumoperitoneum and steep Trendelenburg position inasmuch as the measured values, for the most part, promptly returned toward baseline with release of the pneumoperitoneum and repositioning supine, and no major pulmonary sequelae occurred. Although peak airway pressures and alveolar to arterial gradients certainly worsened with insufflation and extreme Trendelenburg positioning, oxygenation as measured by pulse oximetry and intermittent arterial blood gas remained satisfactory throughout the procedure. The worsened total respiratory compliance, however, resulted in decreased tidal volumes after insufflation. The combination of decreased minute ventilation (to keep the peak airway pressure b 40 cm H2O) and the vascular absorption of insufflated CO2 led to a rise in PaCO2 and ETCO2. Although PaCO2, ETCO2, and Pa-ETCO2 gradient remained elevated during the initial return to supine position, these findings are not unexpected after multiple hours of CO2 pneumoperitoneum. The values might well have fully normalized after a longer recovery period. No patient became hypoxemic during the procedure, nor did any patient develop clinical or radiologic atelectasis or pneumonia afterwards. Our institutional ventilatory practice for these patients was corroborated by Choi et al [8] who concluded that pressure-controlled ventilation offered greater dynamic compliance and lower peak airway pressure than volumecontrolled ventilation during RALP. A study by Kalmar et al [1] concluded that the combination of prolonged steep Trendelenburg position and CO2 pneumoperitoneum during RALP was well tolerated with hemodynamic and pulmonary variables remaining within safe limits. To that point, Schrijvers et al [6] wrote that pulmonary gas exchange is well preserved during RALP in steep Trendelenburg position by noting that venous admixture did not change and dead space ventilation increased but returned to baseline after patients resumed supine position. Similarly, Lestar et al [5] prospectively studied patients undergoing robotic
4
P. Lebowitz et al. Table 1
Oxygenation and ventilatory changes at various stages of RALP (n = 32).
Tidal volume (mL) Peak airway pressure (cm H2O) PaO2 (mm Hg) PaCO2 (mm Hg) ETCO2 (mm Hg) Pa-ETCO2 gradient (mm Hg)
Baseline
Trendelenburg + insufflation
P value (compared with baseline)
Recovery
P value (compared with baseline)
530 ± 102 (495-565) 16.8 ± 2.8 (16-18) 395 ± 62 (373-416) 38.3 ± 5.8 (36-40) 28.4 ± 3.7 (27-30) 10.0 ± 4.5 (8-11)
473 ± 75 (447-499) 34.2 ± 4.8 (33-36) 316 ± 94 (283-348) 46.8 ± 5.5 (45-49) 33.4 ± 3.8 (32-35) 13.4 ± 4.3 (12-15)
.0056 b .0001 .001 b .0001 b .0001 b .0001
533 ± 96 (500-566) 18.3 ± 4.4 (17-20) 340 ± 70 (316-364) 46.3 ± 9.2 (43-49) 33.8 ± 5.8 (32-36) 12.5 ± 6.8 (10-15)
.50 .060 b .001 b .001 b .001 .046
Data are expressed as mean ± SD with 95% confidence intervals.
prostatectomy and concluded that, although lung compliance was halved, gas exchange was overall unaffected with ventilation-perfusion distribution, shunt, and dead space being unaltered during the study. In contrast, we found that ventilation-perfusion imbalance, shunt, and dead space ventilation were very much in evidence during RALP. Just the same, we conclude that, although perturbations in respiratory function during RALP are certainly noteworthy, they should not be an obstacle to providing optimal surgical exposure during RALP. Even so, an individual RALP patient might experience so much ventilation/perfusion (V/Q) mismatching that, to prevent hypoxemia, he would require a raised fraction of inspired oxygen (if not already receiving 100% O2), an alveolar recruitment maneuver with or without the application of PEEP, or an increase in minute ventilation (notably an increased tidal volume) [9]. Steep head-down tilt has been shown to increase pulmonary interstitial pressures and lung fluid content in various animal models. In humans, steep head-down tilt has been shown to alter ventilatory homogeneity. In a study by Henderson et al [7], 30° head-down tilt resulted in a greater heterogeneity of pulmonary blood flow (relative dispersion) during and for up to 1 hour afterwards, suggesting that steep head-down tilt increases pulmonary capillary pressures and Table 2
fluid efflux in the lung, possibly leading to the development of interstitial pulmonary edema and a more heterogeneous distribution of ventilation and perfusion. Regardless of the etiology, abnormal increases in PaCO2 indicate that CO2 removal by ventilation is not keeping up with CO2 production due to metabolism or iatrogenic introduction. Up to a point, PaCO2 and ETCO2 can be normalized by increasing minute ventilation. Under the unphysiological conditions imposed by RALP, however, the normalization process may result in high airway pressure and potential barotrauma. Abnormally increased Pa-ETCO2 gradients (greater than the normal 3-5 mm Hg) reflect V/Q abnormalities suggestive of shunt and/or increased dead space ventilation. The gradient trend can be used as an effective clinical tool to evaluate or gauge the efficacy of treatment, whereby an increase in the gradient could indicate the therapy is not improving the patient's condition and a decrease of the gradient can indicate the therapy is successful. An increase in the gradient from baseline along with other clinical symptoms can indicate deterioration in the patient's ventilatory efficiency [10]. Choi et al [4] further showed that the Pa-ETCO2 gradient increases with age during RALP, making this measurement that much more useful in older patients.
Oxygenation and ventilatory changes in low vs high BMI.
Age (y) BMI (kg/m2) Baseline TV (mL) Insufflation + Trendelenburg TV (mL) Baseline peak airway pressure (cm H2O) Insufflation + Trendelenburg peak airway pressure (cm H2O) Baseline PaO2 (mm Hg) Insufflation + Trendelenburg PaO2 (mm Hg) Baseline PaCO2 (mm Hg) Insufflation + Trendelenburg PaCO2 (mm Hg) Baseline ETCO2 (mm Hg) Insufflation + Trendelenburg ETCO2 (mm Hg) Baseline Pa-ETCO2 gradient (mm Hg) Insufflation + Trendelenburg Pa-ETCO2 gradient (mm Hg) Data are expressed as mean ± SD with 95% confidence intervals.
Low BMI, n = 14
High BMI, n = 18
P
61.6 ± 7.4 (58-65) 23.2 ± 1.5 (22-24) 483 ± 71 (446-520) 436 ± 62 (403-468) 15.3 ± 2.0 (14-16) 30.6 ± 4.2 (28-33) 428 ± 50 (402-454) 367 ± 72 (329-405) 39.1 ± 7.0 (35-43) 47.7 ± 7.1 (44-51) 30.0 ± 3.5 (28-32) 34.4 ± 4.0 (32-36) 9.6 ± 3.8 (8-11) 13.1 ± 4.2 (11-15)
58.9 ± 7.6 (55-62) 29.1 ± 3.4 (28-31) 567 ± 101 (520-614) 526 ± 96 (482-570) 18.3 ± 2.7 (17-19) 37.3 ± 3.2 (36-39) 359 ± 67 (328-390) 284 ± 99 (238-330) 37.3 ± 4.7 (35-39) 45.7 ± 5.5 (43-48) 27.3 ± 3.2 (26-29) 32.7 ± 4.1 (31-34) 10.7 ± 4.0 (9-12) 13.9 ± 4.4 (12-16)
.3 b .0001 .012 .024 .001 b .0001 .002 .010 .370 .360 .024 .210 .012 .02
5 Table 3
Oxygenation and ventilatory changes in patients with and without pulmonary dysfunction.
Age (y) BMI (kg/m2) Baseline TV (mL) Insufflation + Trendelenburg TV (mL) Baseline peak airway pressure (cm H2O) Insufflation + Trendelenburg peak airway pressure (cm H2O) Baseline PaO2 (mm Hg) Insufflation + Trendelenburg PaO2 (mm Hg) Baseline PaCO2 (mm Hg) Insufflation + Trendelenburg PaCO2 (mm Hg) Baseline ETCO2 (mm Hg) Insufflation + Trendelenburg ETCO2 (mm Hg) Baseline Pa-ETCO2 gradient (mm Hg) Insufflation + Trendelenburg Pa-ETCO2 gradient (mm Hg)
No pulmonary disease, n = 18
Pulmonary disease, n = 14
P
58.4 ± 7.7 (55-62) 26.6 ± 4.8 (24-29) 529 ± 116 (475-582) 474 ± 77 (438-509) 16.4 ± 2.9 (15-18) 33.9 ± 5.3 (31-36) 402 ± 67 (371-433) 317 ± 112 (265-369) 36 ± 5.4 (33-38) 45.3 ± 6.2 (42-48) 28.4 ± 4.1 (26-30) 34 ± 4.7 (31-36) 8 ± 3.7 (6-10) 11 ± 3.7 (9-13)
62.1 ± 7.1 (58-66) 26.9 ± 2.8 (25-28) 539 ± 72 (501-577) 471 ± 74 (432-510) 18.1 ± 2.5 (17-19) 35.7 ± 4.1 (33-39) 366 ± 69 (330-402) 317 ± 76 (277-357) 40.6 ± 5.1 (38-43) 47.9 ± 6.1 (45-51) 28.3 ± 2.7 (27-30) 32.5 ± 2.9 (31-34) 13 ± 4.0 (11-15) 16 ± 3.8 (14-18)
.15 .88 .77 .64 .09 .28 .13 .99 .02 .23 .96 .3 .001 .001
Data are expressed as mean ± SD with 95% confidence intervals.
Our study involved patients who had radial arterial catheters placed to assist in the anesthetic management. Inserting an arterial line is reasonable to monitor PaCO2 in high-risk RALP patients for 3 reasons: (1) ETCO2 does not reliably reflect PaCO2; (2) the normal gradient of 3-5 mm Hg between PaCO2 and ETCO2 is increased as a result of V/Q mismatching during RALP; and (3) even with normal ETCO2, achieved by increasing minute ventilation during RALP, PaCO2 may exceed 50 mm Hg, resulting in respiratory acidosis [11]. On the other hand, RALP patients in our institution did well, with or without intraarterial blood pressure and intraarterial respiratory gas monitoring. If the anesthesiologist remains cognizant of the respiratory effects wrought by the unphysiological trespasses imposed by the surgical requirements for this procedure, the anesthesiologist can adjust intraoperative ventilator settings so as to minimize the adverse consequences of restrained diaphragmatic excursion, chest expansion, and intraalveolar fluid accumulation. Although atelectasis and pulmonary edema can occur during RALP as a result of restricted ventilation, the anesthesiologist can use pressure-controlled ventilation, physiologic PEEP (5 cm H2O), an I/E ratio approaching 1:1, and intravenous fluid restriction (b 2 L of crystalloid for the entire procedure) to attenuate these unwanted complications. Likewise, fluid restriction serves to minimize soft tissue edema of the head and neck resulting from gravitational dependency during steep Trendelenburg positioning. All patients were tracheally extubated in the operating room shortly after the conclusion of surgery. Limitations of this study include the small number of patients overall and the smaller number in the subgroups. The retrospective nature of the study is also a major limitation, as blood gases were not uniformly drawn at precise points during anesthesia and could only be broadly categorized into baseline, after insufflation and extreme Trendelenburg positioning, and return to supine position. In addition, it would have been useful to see an additional blood gas later in the recovery period to determine if, in fact, the
PaO2, PaCO2, ETCO2, and Pa-ETCO2 gradients fully returned to baseline and within what period. Our study also did not involve any patients older than 72 years or with BMI greater than 38 kg/m2, and further study needs to be done to determine how patients at the extremes of age and BMI tolerate the unphysiological conditions imposed by RALP. This study confirmed that adverse effects on respiratory gas exchange from steep Trendelenburg position and CO2 pneumoperitoneum take place during RALP. However, if the anesthesiologist adjusts ventilatory parameters so as to optimize oxygenation and CO2 elimination during these surgically imposed circumstances, attentive anesthetic management can avert postoperative pulmonary complications. The authors would like to acknowledge the contributions of Singh Nair, PhD, in the statistical analysis of the study results.
References [1] Kalmar AF, Foubert L, Hendrickx JF, Mottrie A, Absalom A, Mortier EP, et al. Influence of steep Trendelenburg position and CO2 pneumoperitoneum on cardiovascular, cerebrovascular, and respiratory homeostasis during robotic prostatectomy. Br J Anaesth 2010 Apr;104(4):433-9. http:// dx.doi.org/10.1093/bja/aeq018 [Epub 2010 Feb 18]. [2] Casati A, Salvo I, Torri G, Calderini E. Arterial to end-tidal carbon dioxide gradient and physiological dead space monitoring during general anaesthesia: effects of patients' position. Minerva Anestesiol 1997;63:177-82. [3] Hirvonen EA, Nuutinen LS, Kauko M. Ventilatory effects, blood gas changes, and oxygen consumption during laparoscopic hysterectomy. Anesth Analg 1995;80:961-6. [4] Choi DK, Lee IG, Hwang JH. Arterial to End-tidal carbon dioxide pressure gradient increases with age in the steep Trendelenburg position with pneumoperitoneum. Korean J Anesthesiol 2012;63(3): 209-15. http://dx.doi.org/10.4097/kjae.2012.63.3.209 [Published online Sep 14, 2012]. [5] Lestar M, Gunnarsson L, Lagerstrand L, Wiklund P, OdebergWernerman S. Hemodynamic perturbations during robot-assisted laparoscopic radical prostatectomy in 45° Trendelenburg position. Anesth Analg 2011;113(5):1069-75. http://dx.doi.org/10.1213/ANE. 0b013e3182075d1f [Epub 2011 Jan 13].
6 [6] Schrijvers D, Mottrie A, Traen K, De Wolf AM, Vandermeersch E, Kalmar AF, et al. Pulmonary gas exchange is well preserved during robot assisted surgery in steep Trendelenburg position. Acta Anaesthesiol Belg 2009;60(4):229-33. [7] Henderson AC, Levin DL, Hopkins SR, Olfert IM, Buxton RB, Prisk GK. Steep head-down tilt has persisting effects on the distribution of pulmonary blood flow. J Appl Physiol 2006;101(2): 583-9. [8] Choi EM, Na S, Choi SH, An J, Rha KH, Oh YJ. Comparison of volumecontrolled and pressure-controlled ventilation in steep Trendelenburg position for robot-assisted laparoscopic radical prostatectomy. J Clin
P. Lebowitz et al. Anesth 2011;23(3):183-8. http://dx.doi.org/10.1016/j.jclinane.2010.08. 006 [Epub 2011 Mar 4]. [9] Kalmar AF, De Wolf AM, Hendrickx JFA. Anesthetic considerations for robotic surgery in the steep Trendelenburg position. Adv Anesth 2012;30(1):75-96. [10] AARC Clinical Practice Guideline Capnography/Capnometry during Mechanical Ventilation 2003 Revision & Update. Respir Care 2003; 48(5):534-9. [11] Capnography—a comprehensive educational website designed, produced, and maintained by Bhavani-Shankar Kodali. Available at: http://www. capnography.com/lapscopy/lapscopy.htm. [Accessed March 2013].
Original contribution
Respiratory gas exchange during robotic-assisted laparoscopic radical prostatectomy☆ Philip Lebowitz MD, MBA (Professor of Clinical Anesthesiology)⁎, Adam Yedlin MD (Assistant Professor of Clinical Anesthesiology), A. Ari Hakimi MD (Fellow, Urology), Christopher Bryan-Brown MD (Emeritus Professor of Anesthesiology), Mahesan Richards MD (Assistant Professor of Clinical Anesthesiology), Reza Ghavamian MD (Professor of Urology) Albert Einstein College of Medicine, Montefiore Medical Center, Bronx, NY, USA Received 17 September 2014; revised 17 March 2015; accepted 1 June 2015
Keywords: Laparoscopic prostatectomy; Total respiratory compliance; Ventilation/perfusion mismatching; Pulmonary shunting; Respiratory gas effects; Complications during recovery
Abstract Study Objective: Robotic-assisted laparoscopic prostatectomy requires patients to be secured in a steep Trendelenburg position for several hours. Added to the CO2 pneumoperitoneum that is created, this positioning invariably restricts diaphragmatic and chest wall excursion, which can adversely affect respiratory gas exchange. This study sought to measure the extent of respiratory gas change during this procedure. Design: Retrospective, institutional review board approved. Setting: Operating room. Patients: N = 186 males, American Society of Anesthesiologists 2-3, with prostatic carcinoma undergoing robotic-assisted laparoscopic radical prostatectomy. Interventions: Arterial blood gases and noninvasive respiratory measurements were recorded for those patients (n = 32) in whom a radial arterial catheter had been inserted intraoperatively, specifically timed to different phases of the procedure: supine lithotomy, steep Trendelenburg, and return to supine. Ventilatory parameters were standardized. Measurements: Systemic blood pressure, heart rate, respiratory rate, PaO2, PaCO2, oxygen saturation as measured by pulse oximetry, and end-tidal carbon dioxide pressure. Main Results: Although no patients developed perioperative respiratory complications, the PaO2 invariably fell (395 vs 316 mm Hg; P = .001) while the patients were in steep Trendelenburg, and the PaCO2–end-tidal carbon dioxide pressure rose (10.0 vs 13.4 mm Hg; P b .0001). Upon return to supine, patients' respiratory measurements promptly returned to within 15% of baseline. Subgroup analysis for high-BMI vs low-BMI patients as well as for patients with pulmonary disease and/or a smoking history showed similar individual effects and only small, although significant, respiratory gas exchange aberrations.
☆
Disclosures: There were no grants, sponsors, or funding sources that provided direct financial support to the research work contained in the manuscript. ⁎ Correspondence: Philip Lebowitz, MD, MBA, Montefiore Medical Center, 111 E. 210 St, Bronx, NY 10467, USA. Tel.: + 1 718 920 6183. E-mail addresses: [email protected] (P. Lebowitz), [email protected] (A. Yedlin), [email protected] (A.A. Hakimi), [email protected] (C. Bryan-Brown), [email protected] (M. Richards), [email protected] (R. Ghavamian). http://dx.doi.org/10.1016/j.jclinane.2015.06.001 0952-8180/© 2015 Elsevier Inc. All rights reserved.
2
P. Lebowitz et al. Conclusions: Positioning patients with a CO2 pneumoperitoneum in steep Trendelenburg for several hours imposes restriction of diaphragmatic and chest wall movement sufficient for respiratory gas exchange to be adversely affected. Return of function to within 15% of baseline occurred within minutes after return to supine and release of the CO2 pneumoperitoneum. No patients during the study period developed pulmonary complications that required alteration in their level of care. © 2015 Elsevier Inc. All rights reserved.
1. Introduction Robotic-assisted laparoscopic radical prostatectomy (RALP) is currently the most commonly used surgical technique for treatment of localized prostate cancer. It requires the patient to be positioned on the operating room table with the legs raised in a modified lithotomy position, the thorax tightly restrained with adhesive tape, and the head down 30° from supine in extreme Trendelenburg position. Coupled with the required surgical pneumoperitoneum, this positioning invariably compromises total respiratory compliance. Attempting to maintain adequate minute ventilation without causing barotrauma, the anesthesiologist frequently encounters impaired oxygenation and carbon dioxide elimination during the procedure. Trendelenburg positioning and CO2 pneumoperitoneum adversely affect oxygenation and increase the alveolar to arterial oxygen gradient by decreasing functional residual capacity and increasing pulmonary shunt fraction. What is less frequently observed is that the reduction in functional residual capacity also adversely affects CO2 elimination and increases the arterial to end-tidal carbon dioxide pressure gradient (Pa-ETCO2) by increasing dead space ventilation [1-3]. Increasing age also contributes to an increase in Pa-ETCO2 [2,4], which is relevant because RALP is typically performed on an aging male population due to the nature of prostate cancer. Although several studies have looked at various parameters including pulmonary blood flow, pulmonary gas exchange, lung compliance, and shunt during different combinations of steep Trendelenburg and CO2 pneumoperitoneum [1,5-7], the extent to which oxygenation and ventilation might be compromised intraoperatively during RALP remains poorly documented. This study sought to further quantify the extent of respiratory gas exchange impairment during this procedure.
2. Materials and methods After approval from the hospital's institutional review board, we reviewed the medical records of all patients at our hospital who had undergone RALP during 2008 and the first 6 months of 2009. All patients (n = 186) who underwent this procedure during the study period were included. All patients had been deemed suitable to undergo general anesthesia for
this elective surgery and were classified as American Society of Anesthesiologists I-III. Upon review of the medical records, we found 32 patients who had radial arterial catheters placed and blood gas data available from multiple key points of the surgical operation including anesthetized baseline (supine lithotomy), after CO2 insufflation with extreme Trendelenburg, and after return to supine. From each of these patients' medical records, we extracted their age, height, weight, and pulmonary status (smoking history, pulmonary function tests, and chronic obstructive pulmonary disease diagnosis) as well as intraoperative tidal volume, respiratory rate, positive end-expiratory pressure (PEEP) level, ratio of inspiratory time to expiratory time ratio, peak airway pressure, pulse oximetry, end-tidal CO2 (ETCO2), PaO2, PaCO2, blood pressure, and heart rate. Our institutional experience with these cases had led us to empirically determine that the optimal ventilation for these patients is achieved with pressure-controlled mechanical ventilation to an intended tidal volume of 6 mL/kg with a ventilator rate of 10 breaths per minute, an I/E ratio of 1:1, and PEEP of 5 cm H2O. We did not allow the peak airway pressure to exceed 40 cm H2O. Consequently, if maintaining a tidal volume of 6 mL/kg under this ventilator pattern caused the peak airway pressure to exceed 40 cm H2O, we would decrease the tidal volume so as to keep the peak airway pressure below this level and incrementally raise the respiratory rate to as high as 13 breaths per minute to approximate a constant minute ventilation. This protocol of mechanical ventilation was indeed used in all 32 of this subgroup of patients. Anesthesia was maintained with sevoflurane or desflurane in 100% oxygen as well as nondepolarizing neuromuscular blockade for all patients; fentanyl was added in 50-μg increments to maintain heart rate in the 60-100 beats per minute range. Collected data were used to compare changes in pulmonary function during the procedure. Continuous data were analyzed using a 2-tailed Student t test. In comparing individual patients throughout different phases of the surgical procedure, we analyzed the data with a paired t test. Data obtained from arterial blood gas measurements were also analyzed using analysis of variance to detect differences for the group as a whole among the stages of the surgical procedure. P b .05 was considered statistically significant. We also performed subgroup analyses to determine if there were any differences in respiratory changes between
3 patients with high body mass index (BMI) (N 25 kg/m2) and low BMI (b 25 kg/m2) as well as in patients with a history of either pulmonary dysfunction or smoking and those patients without any such history. In comparing data between different groups, the data were analyzed using an unpaired t test. In considering outcomes of the procedure, including postoperative pulmonary complications, we used χ2 analysis. P b .05 was considered statistically significant.
3. Results For the 32 medical records reviewed, the mean age was 60 years; the mean weight, 81.4 kg; and the mean BMI, 26 kg/m2. Eighteen patients had BMI N 25 kg/m2, whereas 14 patients had BMI b 25 kg/m2. Fourteen patients had a history of pulmonary dysfunction or extensive smoking, and 18 patients did not. The average total anesthesia time of 4 hours included surgery time of approximately 3 hours, which itself included intraperitoneal insufflation time of approximately 2 hours 15 minutes. Pneumoperitoneum with CO2 was maintained at 15 cm H2O throughout the insufflation period. Anesthetic induction, patient positioning, and anesthetic emergence comprised 1 hour of the total anesthesia time. Not surprisingly, patients experienced lower tidal volumes, increased peak airway pressures, decreased PaO2, increased PaCO2, increased ETCO2, and worsened alveolar-arterial partial pressure of oxygen gradient and Pa-ETCO2 gradients after undergoing intraperitoneal insufflation and extreme Trendelenburg positioning, compared with baseline (Table 1). These differences were all statistically significant. Tidal volumes decreased from 530 ± 102 mL (mean ± SD) at baseline to 473 ± 75 mL after insufflation and extreme Trendelenburg (P = .0056). Peak airway pressure increased from 16.8 ± 2.8 mm Hg to 34.2 ± 4.8 mm Hg (P b .0001). PaO2 decreased from 395 ± 62 mm Hg to 315.7 ± 93.8 mm Hg (P = .001); PA-PaO2 gradient correspondingly increased. PaCO2 increased from 38.3 ± 5.8 to 46.8 ± 5.5 (P b .0001). ETCO2 increased from 28.4 ± 3.7 to 33.4 ± 3.8 (P b .0001). Finally, Pa-ETCO2 gradient increased from baseline of 10 ± 4.5 to 13.4 ± 4.3 (P b .0001). Upon repositioning back to supine, all tidal volumes and peak airway pressures returned to near-baseline values. Oxygenation, as determined by PaO2, was still significantly decreased (P b .001) but was trending upward and returned to within 15% of baseline values. PaCO2, ETCO2, and Pa-ETCO2 gradients all remained significantly elevated during the initial recovery period shortly after repositioning to supine. The subgroup analyses did not show any significant respiratory changes between high (n = 18) vs low (n = 14) BMI groups (Table 2) or between patients with underlying pulmonary dysfunction (n = 14) vs those without (n = 18) such dysfunction (Table 3). Although there were some minor statistically significant differences between the subgroups at baseline, percent changes with insufflation and extreme
Trendelenburg were not significant for any parameter. The most notable finding for the BMI subgroups was that patients with BMI N 25 kg/m2 had significantly increased baseline peak airway pressures when compared with those with BMI b 25 kg/m2 (P = .001), whereas the most notable finding for the pulmonary function subgroups was that patients with pulmonary dysfunction had a significantly increased baseline PaCO2 (P = .02) and Pa-ETCO2 gradient (P = .001). No patients, regardless of risk group, had any pulmonary sequelae despite the significant respiratory perturbations noted above.
4. Discussion This study demonstrates that, although respiratory perturbations during RALP are significant, the lungs seem to have the capacity to withstand the effects of a CO2 pneumoperitoneum and steep Trendelenburg position inasmuch as the measured values, for the most part, promptly returned toward baseline with release of the pneumoperitoneum and repositioning supine, and no major pulmonary sequelae occurred. Although peak airway pressures and alveolar to arterial gradients certainly worsened with insufflation and extreme Trendelenburg positioning, oxygenation as measured by pulse oximetry and intermittent arterial blood gas remained satisfactory throughout the procedure. The worsened total respiratory compliance, however, resulted in decreased tidal volumes after insufflation. The combination of decreased minute ventilation (to keep the peak airway pressure b 40 cm H2O) and the vascular absorption of insufflated CO2 led to a rise in PaCO2 and ETCO2. Although PaCO2, ETCO2, and Pa-ETCO2 gradient remained elevated during the initial return to supine position, these findings are not unexpected after multiple hours of CO2 pneumoperitoneum. The values might well have fully normalized after a longer recovery period. No patient became hypoxemic during the procedure, nor did any patient develop clinical or radiologic atelectasis or pneumonia afterwards. Our institutional ventilatory practice for these patients was corroborated by Choi et al [8] who concluded that pressure-controlled ventilation offered greater dynamic compliance and lower peak airway pressure than volumecontrolled ventilation during RALP. A study by Kalmar et al [1] concluded that the combination of prolonged steep Trendelenburg position and CO2 pneumoperitoneum during RALP was well tolerated with hemodynamic and pulmonary variables remaining within safe limits. To that point, Schrijvers et al [6] wrote that pulmonary gas exchange is well preserved during RALP in steep Trendelenburg position by noting that venous admixture did not change and dead space ventilation increased but returned to baseline after patients resumed supine position. Similarly, Lestar et al [5] prospectively studied patients undergoing robotic
4
P. Lebowitz et al. Table 1
Oxygenation and ventilatory changes at various stages of RALP (n = 32).
Tidal volume (mL) Peak airway pressure (cm H2O) PaO2 (mm Hg) PaCO2 (mm Hg) ETCO2 (mm Hg) Pa-ETCO2 gradient (mm Hg)
Baseline
Trendelenburg + insufflation
P value (compared with baseline)
Recovery
P value (compared with baseline)
530 ± 102 (495-565) 16.8 ± 2.8 (16-18) 395 ± 62 (373-416) 38.3 ± 5.8 (36-40) 28.4 ± 3.7 (27-30) 10.0 ± 4.5 (8-11)
473 ± 75 (447-499) 34.2 ± 4.8 (33-36) 316 ± 94 (283-348) 46.8 ± 5.5 (45-49) 33.4 ± 3.8 (32-35) 13.4 ± 4.3 (12-15)
.0056 b .0001 .001 b .0001 b .0001 b .0001
533 ± 96 (500-566) 18.3 ± 4.4 (17-20) 340 ± 70 (316-364) 46.3 ± 9.2 (43-49) 33.8 ± 5.8 (32-36) 12.5 ± 6.8 (10-15)
.50 .060 b .001 b .001 b .001 .046
Data are expressed as mean ± SD with 95% confidence intervals.
prostatectomy and concluded that, although lung compliance was halved, gas exchange was overall unaffected with ventilation-perfusion distribution, shunt, and dead space being unaltered during the study. In contrast, we found that ventilation-perfusion imbalance, shunt, and dead space ventilation were very much in evidence during RALP. Just the same, we conclude that, although perturbations in respiratory function during RALP are certainly noteworthy, they should not be an obstacle to providing optimal surgical exposure during RALP. Even so, an individual RALP patient might experience so much ventilation/perfusion (V/Q) mismatching that, to prevent hypoxemia, he would require a raised fraction of inspired oxygen (if not already receiving 100% O2), an alveolar recruitment maneuver with or without the application of PEEP, or an increase in minute ventilation (notably an increased tidal volume) [9]. Steep head-down tilt has been shown to increase pulmonary interstitial pressures and lung fluid content in various animal models. In humans, steep head-down tilt has been shown to alter ventilatory homogeneity. In a study by Henderson et al [7], 30° head-down tilt resulted in a greater heterogeneity of pulmonary blood flow (relative dispersion) during and for up to 1 hour afterwards, suggesting that steep head-down tilt increases pulmonary capillary pressures and Table 2
fluid efflux in the lung, possibly leading to the development of interstitial pulmonary edema and a more heterogeneous distribution of ventilation and perfusion. Regardless of the etiology, abnormal increases in PaCO2 indicate that CO2 removal by ventilation is not keeping up with CO2 production due to metabolism or iatrogenic introduction. Up to a point, PaCO2 and ETCO2 can be normalized by increasing minute ventilation. Under the unphysiological conditions imposed by RALP, however, the normalization process may result in high airway pressure and potential barotrauma. Abnormally increased Pa-ETCO2 gradients (greater than the normal 3-5 mm Hg) reflect V/Q abnormalities suggestive of shunt and/or increased dead space ventilation. The gradient trend can be used as an effective clinical tool to evaluate or gauge the efficacy of treatment, whereby an increase in the gradient could indicate the therapy is not improving the patient's condition and a decrease of the gradient can indicate the therapy is successful. An increase in the gradient from baseline along with other clinical symptoms can indicate deterioration in the patient's ventilatory efficiency [10]. Choi et al [4] further showed that the Pa-ETCO2 gradient increases with age during RALP, making this measurement that much more useful in older patients.
Oxygenation and ventilatory changes in low vs high BMI.
Age (y) BMI (kg/m2) Baseline TV (mL) Insufflation + Trendelenburg TV (mL) Baseline peak airway pressure (cm H2O) Insufflation + Trendelenburg peak airway pressure (cm H2O) Baseline PaO2 (mm Hg) Insufflation + Trendelenburg PaO2 (mm Hg) Baseline PaCO2 (mm Hg) Insufflation + Trendelenburg PaCO2 (mm Hg) Baseline ETCO2 (mm Hg) Insufflation + Trendelenburg ETCO2 (mm Hg) Baseline Pa-ETCO2 gradient (mm Hg) Insufflation + Trendelenburg Pa-ETCO2 gradient (mm Hg) Data are expressed as mean ± SD with 95% confidence intervals.
Low BMI, n = 14
High BMI, n = 18
P
61.6 ± 7.4 (58-65) 23.2 ± 1.5 (22-24) 483 ± 71 (446-520) 436 ± 62 (403-468) 15.3 ± 2.0 (14-16) 30.6 ± 4.2 (28-33) 428 ± 50 (402-454) 367 ± 72 (329-405) 39.1 ± 7.0 (35-43) 47.7 ± 7.1 (44-51) 30.0 ± 3.5 (28-32) 34.4 ± 4.0 (32-36) 9.6 ± 3.8 (8-11) 13.1 ± 4.2 (11-15)
58.9 ± 7.6 (55-62) 29.1 ± 3.4 (28-31) 567 ± 101 (520-614) 526 ± 96 (482-570) 18.3 ± 2.7 (17-19) 37.3 ± 3.2 (36-39) 359 ± 67 (328-390) 284 ± 99 (238-330) 37.3 ± 4.7 (35-39) 45.7 ± 5.5 (43-48) 27.3 ± 3.2 (26-29) 32.7 ± 4.1 (31-34) 10.7 ± 4.0 (9-12) 13.9 ± 4.4 (12-16)
.3 b .0001 .012 .024 .001 b .0001 .002 .010 .370 .360 .024 .210 .012 .02
5 Table 3
Oxygenation and ventilatory changes in patients with and without pulmonary dysfunction.
Age (y) BMI (kg/m2) Baseline TV (mL) Insufflation + Trendelenburg TV (mL) Baseline peak airway pressure (cm H2O) Insufflation + Trendelenburg peak airway pressure (cm H2O) Baseline PaO2 (mm Hg) Insufflation + Trendelenburg PaO2 (mm Hg) Baseline PaCO2 (mm Hg) Insufflation + Trendelenburg PaCO2 (mm Hg) Baseline ETCO2 (mm Hg) Insufflation + Trendelenburg ETCO2 (mm Hg) Baseline Pa-ETCO2 gradient (mm Hg) Insufflation + Trendelenburg Pa-ETCO2 gradient (mm Hg)
No pulmonary disease, n = 18
Pulmonary disease, n = 14
P
58.4 ± 7.7 (55-62) 26.6 ± 4.8 (24-29) 529 ± 116 (475-582) 474 ± 77 (438-509) 16.4 ± 2.9 (15-18) 33.9 ± 5.3 (31-36) 402 ± 67 (371-433) 317 ± 112 (265-369) 36 ± 5.4 (33-38) 45.3 ± 6.2 (42-48) 28.4 ± 4.1 (26-30) 34 ± 4.7 (31-36) 8 ± 3.7 (6-10) 11 ± 3.7 (9-13)
62.1 ± 7.1 (58-66) 26.9 ± 2.8 (25-28) 539 ± 72 (501-577) 471 ± 74 (432-510) 18.1 ± 2.5 (17-19) 35.7 ± 4.1 (33-39) 366 ± 69 (330-402) 317 ± 76 (277-357) 40.6 ± 5.1 (38-43) 47.9 ± 6.1 (45-51) 28.3 ± 2.7 (27-30) 32.5 ± 2.9 (31-34) 13 ± 4.0 (11-15) 16 ± 3.8 (14-18)
.15 .88 .77 .64 .09 .28 .13 .99 .02 .23 .96 .3 .001 .001
Data are expressed as mean ± SD with 95% confidence intervals.
Our study involved patients who had radial arterial catheters placed to assist in the anesthetic management. Inserting an arterial line is reasonable to monitor PaCO2 in high-risk RALP patients for 3 reasons: (1) ETCO2 does not reliably reflect PaCO2; (2) the normal gradient of 3-5 mm Hg between PaCO2 and ETCO2 is increased as a result of V/Q mismatching during RALP; and (3) even with normal ETCO2, achieved by increasing minute ventilation during RALP, PaCO2 may exceed 50 mm Hg, resulting in respiratory acidosis [11]. On the other hand, RALP patients in our institution did well, with or without intraarterial blood pressure and intraarterial respiratory gas monitoring. If the anesthesiologist remains cognizant of the respiratory effects wrought by the unphysiological trespasses imposed by the surgical requirements for this procedure, the anesthesiologist can adjust intraoperative ventilator settings so as to minimize the adverse consequences of restrained diaphragmatic excursion, chest expansion, and intraalveolar fluid accumulation. Although atelectasis and pulmonary edema can occur during RALP as a result of restricted ventilation, the anesthesiologist can use pressure-controlled ventilation, physiologic PEEP (5 cm H2O), an I/E ratio approaching 1:1, and intravenous fluid restriction (b 2 L of crystalloid for the entire procedure) to attenuate these unwanted complications. Likewise, fluid restriction serves to minimize soft tissue edema of the head and neck resulting from gravitational dependency during steep Trendelenburg positioning. All patients were tracheally extubated in the operating room shortly after the conclusion of surgery. Limitations of this study include the small number of patients overall and the smaller number in the subgroups. The retrospective nature of the study is also a major limitation, as blood gases were not uniformly drawn at precise points during anesthesia and could only be broadly categorized into baseline, after insufflation and extreme Trendelenburg positioning, and return to supine position. In addition, it would have been useful to see an additional blood gas later in the recovery period to determine if, in fact, the
PaO2, PaCO2, ETCO2, and Pa-ETCO2 gradients fully returned to baseline and within what period. Our study also did not involve any patients older than 72 years or with BMI greater than 38 kg/m2, and further study needs to be done to determine how patients at the extremes of age and BMI tolerate the unphysiological conditions imposed by RALP. This study confirmed that adverse effects on respiratory gas exchange from steep Trendelenburg position and CO2 pneumoperitoneum take place during RALP. However, if the anesthesiologist adjusts ventilatory parameters so as to optimize oxygenation and CO2 elimination during these surgically imposed circumstances, attentive anesthetic management can avert postoperative pulmonary complications. The authors would like to acknowledge the contributions of Singh Nair, PhD, in the statistical analysis of the study results.
References [1] Kalmar AF, Foubert L, Hendrickx JF, Mottrie A, Absalom A, Mortier EP, et al. Influence of steep Trendelenburg position and CO2 pneumoperitoneum on cardiovascular, cerebrovascular, and respiratory homeostasis during robotic prostatectomy. Br J Anaesth 2010 Apr;104(4):433-9. http:// dx.doi.org/10.1093/bja/aeq018 [Epub 2010 Feb 18]. [2] Casati A, Salvo I, Torri G, Calderini E. Arterial to end-tidal carbon dioxide gradient and physiological dead space monitoring during general anaesthesia: effects of patients' position. Minerva Anestesiol 1997;63:177-82. [3] Hirvonen EA, Nuutinen LS, Kauko M. Ventilatory effects, blood gas changes, and oxygen consumption during laparoscopic hysterectomy. Anesth Analg 1995;80:961-6. [4] Choi DK, Lee IG, Hwang JH. Arterial to End-tidal carbon dioxide pressure gradient increases with age in the steep Trendelenburg position with pneumoperitoneum. Korean J Anesthesiol 2012;63(3): 209-15. http://dx.doi.org/10.4097/kjae.2012.63.3.209 [Published online Sep 14, 2012]. [5] Lestar M, Gunnarsson L, Lagerstrand L, Wiklund P, OdebergWernerman S. Hemodynamic perturbations during robot-assisted laparoscopic radical prostatectomy in 45° Trendelenburg position. Anesth Analg 2011;113(5):1069-75. http://dx.doi.org/10.1213/ANE. 0b013e3182075d1f [Epub 2011 Jan 13].
6 [6] Schrijvers D, Mottrie A, Traen K, De Wolf AM, Vandermeersch E, Kalmar AF, et al. Pulmonary gas exchange is well preserved during robot assisted surgery in steep Trendelenburg position. Acta Anaesthesiol Belg 2009;60(4):229-33. [7] Henderson AC, Levin DL, Hopkins SR, Olfert IM, Buxton RB, Prisk GK. Steep head-down tilt has persisting effects on the distribution of pulmonary blood flow. J Appl Physiol 2006;101(2): 583-9. [8] Choi EM, Na S, Choi SH, An J, Rha KH, Oh YJ. Comparison of volumecontrolled and pressure-controlled ventilation in steep Trendelenburg position for robot-assisted laparoscopic radical prostatectomy. J Clin
P. Lebowitz et al. Anesth 2011;23(3):183-8. http://dx.doi.org/10.1016/j.jclinane.2010.08. 006 [Epub 2011 Mar 4]. [9] Kalmar AF, De Wolf AM, Hendrickx JFA. Anesthetic considerations for robotic surgery in the steep Trendelenburg position. Adv Anesth 2012;30(1):75-96. [10] AARC Clinical Practice Guideline Capnography/Capnometry during Mechanical Ventilation 2003 Revision & Update. Respir Care 2003; 48(5):534-9. [11] Capnography—a comprehensive educational website designed, produced, and maintained by Bhavani-Shankar Kodali. Available at: http://www. capnography.com/lapscopy/lapscopy.htm. [Accessed March 2013].
