We read with expectations the study “Individualized Positive End-expiratory Pressure on Postoperative Atelectasis in Patients with Obesity: A Randomized Controlled Clinical Trial,” by Li et al.  We congratulate the authors for succeeding in completing this interesting study despite the COVID-19 pandemic. The study is one of a few that have managed to use computed tomography to assess the amount of early postoperative atelectasis after using an intraoperative protective ventilation strategy. In the study, there was a small but statistically significant reduction in postoperative atelectasis for an intervention group treated intraoperatively with individualized positive end-expiratory pressure (PEEP), guided by optimal dynamic compliance, compared with a control group receiving a fixed PEEP of 8 cm H2O. However, the postoperative Pao2/fraction of inspired oxygen (Fio2) ratio as a measure of oxygenation did not differ between groups, nor the rate of postoperative pulmonary complications.

The knowledge of how to optimally ventilate healthy lungs during surgery has come a long way in recent years. Unfortunately, it has been notoriously difficult to translate intraoperative protective ventilation strategies into a reduced incidence of postoperative pulmonary complications.  Among other reasons for this, one contributing explanation may be an insufficient understanding of how the emergence procedure contributes to postoperative atelectasis formation. The main factors explaining atelectasis are a high alveolar oxygen concentration in combination with airway closure, as shown in a great number of studies. Both are present shortly after induction of anesthesia, but also after awakening when the PEEP is discontinued. Thus, after extubation, the lungs are at risk of developing atelectasis irrespective of the intraoperative ventilatory strategy. In the study by Li et al., it is therefore surprising to find that the primary outcome variable, postoperative atelectasis, is assumed to be a consequence of only the intraoperative ventilatory strategy, without any comments on the possible importance of emergence preoxygenation or any measures taken to counteract the effect of airway closure after extubation. Without knowledge of the oxygen level used during extubation, or the use of any pressure support facemask ventilation after extubation, it is difficult to conclude what most likely could explain the reduced postoperative atelectasis found in the study. The authors refer to a study by Pereira et al. that demonstrated reduced postoperative atelectasis with individualized PEEP in nonobese patients, but, in fact, the results of that study might have an important complementary or even sole explanation: weaning was performed under pressure-support mode, keeping Fio2 at 0.5 and maintaining PEEP according to the patient´s randomization.

It is furthermore noteworthy that the intervention group in the study by Li et al. has a considerable amount of postoperative atelectasis of mean 9.5% of the total lung volume. This corresponds to a four to five times larger aerated lung volume being collapsed. Thus, despite being treated with an ambitious strategy to keep the lung open intraoperatively, shortly after awakening large atelectases are present. This further suggests that the emergence procedure is an important and often neglected phase of anesthesia with possible importance for postoperative patient outcomes. Therefore, to be able to learn as much as possible from this and future studies, we believe it is important to fully describe and comment on the exact procedure during emergence and extubation.

Our final concern regards the use of Pao2/Fio2 ratio without considering other potential factors that affect oxygenation, such as hypoventilation and ventilation to perfusion (V̇/Q̇ ) mismatch. It would be interesting to know the Paco2 and at which Pao2 and Fio2 the Pao2/Fio2 ratio was calculated at 30 min after extubation. More specifically, how was the exact Fio2 ascertained during spontaneous breathing, and can the authors provide a reference that validates the usefulness of Pao2/Fio2 ratio in awake patients breathing supplemental oxygen?