Obesity and Positive End-expiratory Pressure: Reply

We thank Suzuki et al.¹  for their interest in our recent work²  and would like to address their concerns. The challenge of combining patients of different study protocols spanning several years is a potential bias we noted ourselves.²  However, the single-center setting means that investigators and surgeons remained the same and the highly elective patients for bariatric surgery only were treated according to clinical standards which remained unchanged during the time. It seems unlikely that positive end-expiratory pressure (PEEP)–dependent physiologic effects were influenced by any minor change over time. Moreover, randomization guarantees that differences within each study are the result of chance alone, and the difference in ARISCAT (Assess Respiratory rIsk in Surgical patients in CATalonia) scores between the groups are both clearly presented and not in the least indicative of a meaningful imbalance in our opinion. In line with our clinical pathway for bariatric surgery and current anesthesia guidelines, lung function measurements were not performed before surgery.³  Patients with pulmonary disease, cardiac insufficiency, or increased intracranial pressure were not included in either study.

Indeed, duration of anesthesia and the operation differed significantly between groups. However, the individualized PEEP group was the one with the greatest duration but also the group with the best intraoperative lung mechanics and the highest oxygenation. Thus, even if the time of mechanical ventilation and capnoperitoneum time influenced our results, this emphasizes even more the necessity of an individual ventilation strategy.

As correctly noted by Suzuki et al.,¹  the intraoperative amount of fluid applied was higher in the individualized PEEP group. This arises from a coloading performed by the attending anesthesiologist during the recruitment maneuvers to intercept a drop in blood pressure and from PEEP titration, which increased the duration of anesthesia and thus the time during which the patient was administered fluids. Despite the measured differences in the groups, in all groups it is intraoperative restrictive fluid management.⁴  Applying more restrictive fluid infusion targets might further increase vasopressor requirements in obese patients, especially when using high PEEP, and this potential risk must be balanced with potential benefits of minimizing intraoperative atelectasis with higher PEEP.⁵  However, only patients scheduled for bariatric surgery were included in the studies in our center, where perioperative procedures (including preoperative oral intake) are highly standardized following the early recovery after surgery concept for bariatric surgery.³  The protocol was not changed during the study period without any systematic change in preoperative hydration or in dose of hypertensive drugs in the 4-yr interval in question.

Higher PEEP values may lead to cardiovascular instability as a result of impaired venous return, which was reflected in our study by the highest cumulative noradrenaline doses in the individualized PEEP group. However, mean arterial blood pressure did not differ between groups and overall norepinephrine doses were low, so we do not consider this to be a relevant issue in most patients. Excessive PEEP values should, however, be avoided in patients with significant right heart failure, and such patients were excluded in our study. Predefined rescue protocols were available if a PEEP level was not tolerated,^5,6  but none of our patients needed such a rescue protocol. Concerning the influence of PEEP on brain perfusion and intracranial pressure (ICP), an increase in thoracic pressure is partially transmitted to central venous pressure (CVP) and may thus increase venous downstream pressure of the brain. According to the vascular waterfall model of compressible tubes, cerebral venous outflow is only impaired if CVP is greater than ICP. Clinical data have shown that for patients with decreased chest wall compliance, as with our obese patients, higher PEEP had no effect on cerebral hemodynamics.⁷ 

Transpulmonary pressure was not included in the endpoints of the two original studies because its use as a correlate of lung stress has known limitations.⁸  Although electrical impedance tomography enables detection of regional information on overdistension and collapse, regional variations in lung expansion may not be adequately reflected by local pressure measurements in the esophagus. As noted in our article, in the context of predefined low tidal volume, information on regional heterogeneity might be more relevant to identify regions of increased stress as a substrate for postoperative pulmonary complications.

Suzuki and colleagues correctly note that postoperative outcomes differed from those of the original PROBESE (Effect of High PEEP vs. Low PEEP on Postoperative Pulmonary Complications in Obese Patients) study. In contrast to the PROBESE-study, our subanalysis was neither intended nor adequately powered to investigate postoperative pulmonary outcomes associated with an individualized ventilation strategy. The early recovery after surgery guidelines discussed the use of adequate PEEP with recruitment maneuvers to reduce postoperative pulmonary complications.³  Atelectasis plays a significant role in obese patients and should be avoided with regard to ventilator associated complications,⁹  The aim of the subanalysis was to investigate the effects of individualized PEEP on ventilation distribution and atlectasis formation with implications for lung function. Because there were no clear instructions when to stop oxygen therapy in the postanesthesia care unit in the single-center study, the duration of oxygen therapy would not be an adequate endpoint. Furthermore, to be able to better classify the results, the endpoints were based on previously published studies on individual ventilation, including one of the two studies included here.^5,10 

We highly appreciate the interest in our work and agree with Suzuki et al. that further research is necessary to determine whether the benefits of an individualized ventilation strategy lead to a lower incidence of postoperative pulmonary complications.

Dr. Wrigge has received research funding, lecture fees, and technical support from Dräger Medical, Lübeck, Germany; funding from Pfizer (Investigator Initiated Trial Program), Berlin, Germany; funding and lecture fees from InfectoPharm, Heppenheim, Germany; lecture fees from GE Healthcare, Freiburg, Germany, lecture fees from Maquet, Rastatt, Germany; lecture fees from MSD, Konstanz, Germany, advisory board honoraria from Liberate Medical, Kentucky; and technical support from Swisstom Corp., Landquart, Switzerland. Dr. Simon has received funding and lecture fees from InfectoPharm. The remaining authors declare no competing interests.