Pressure Support Ventilation and Atelectasis: Comment

We read with great interest the article by Jeong et al.1 titled “Pressure Support versus Spontaneous Ventilation during Anesthetic Emergence—Effect on Postoperative Atelectasis: A Randomized Controlled Trial.” Although many studies have looked at the potential effects of various intraoperative open lung ventilation strategies on postoperative pulmonary outcomes, recent evidence suggests that their potential benefits may be limited if no action is taken to minimize lung derecruitment during the emergence period.2 Considering that postoperative atelectasis plays a central role in the development of postoperative pulmonary complications, and that maintaining positive pressure during emergence may help preserve lung aeration,3 the research question of Jeong et al. is of paramount importance. However, we have some concerns regarding key aspects of the study’s methodology.First, we were especially worried about elements used to define and measure the incidence of atelectasis, the study’s primary outcome. The authors’ definition (more than three lung sections with a non-zero atelectasis score) is not standard4 and has not been previously validated. Can the authors specify whether their definition was selected before conducting the study to reassure readers on the absence of data-driven threshold selection? Performing sensitivity analyses looking at different thresholds for the number of atelectatic lung sections necessary to classify the outcome would better assess the robustness of their findings.Second, we were puzzled to read that Jeong et al. not only used a modified and unvalidated echographic pulmonary aeration loss score5 but also introduced their own modifications, potentially further weakening the validity of their primary outcome classification. In particular, loss of lung sliding with lung pulse is not a sign of atelectasis but rather a sign of a well-aerated lung without ventilation. This finding could have indicated the presence of a mucous plug which may have been resolved after a simple coughing fit without causing any atelectasis. Including this sign in their atelectasis score seems problematic. We encourage the authors to use the lung ultrasound score, a validated echographic loss of aeration score, to report their results.6Third, their study was underpowered for their anticipated effect size. Using the same assumptions (an incidence of 53% in the control group and 37% in the intervention group for an absolute estimated effect of 16%), we calculated that a sample size of 302 patients would have been necessary even before considering a 15% dropout rate. Their greater-than-anticipated observed effect explains why their results achieved statistical significance. However, underpowered studies are prone to inflated results with positive results that are more likely to be false positives.7Fourth, the authors’ definition of hypoxemia, a secondary outcome, may lead to missing important clinical effects resulting from their intervention. A punctual event of oxygen saturation measured by pulse oximetry greater than 92% may be not be clinically significant in comparison with a prolonged postoperative need for high fractional inspired oxygen tension. Can the authors provide data on this secondary outcome using a time-weighted need for organ support, such as oxygen-free days or cumulative postoperative oxygen administration?The imaging study by Jeong et al. is an essential first step in clarifying the role of assisted ventilatory modes during anesthesia emergence. However, there is still a lot of work to be done to answer the salient question: Are assisted ventilatory modes an important part of an open lung strategy at emergence that may lead to a decreased incidence of postoperative pulmonary complications?Dr. Girard is a paid consultant for the point-of-care ultrasonography group of GE Healthcare (Milwaukee, Wisconsin). The other authors declare no competing interests.

Récupéré en direct depuis OpenAlex et désinversé. Les résumés ne sont pas conservés dans cette base de données : les index inversés représentent 8,6 Go des 9,3 Go de texte de la base, et le serveur dispose de 13 Go libres.