We have read with great interest the study by Pavlov et al1 where the authors conducted a systematic review of observational studies evaluating awake prone positioning for subjects with COVID-19 and hypoxemic respiratory failure. The results of this systematic review do not support a reduction in intubation rate associated with awake prone positioning despite improved oxygenation. We applaud the authors’ efforts in conducting this study, although we would like to point out a few considerations.
The prone position has been shown to improve oxygenation in patients with ARDS on mechanical ventilation.2 The interest in awake prone positioning grew rapidly with the COVID-19 pandemic since this is a low-cost intervention that can improve oxygenation through diverse physiological mechanisms in conscious patients with COVID-19 receiving oxygen therapy.3 However, its precise usefulness remains to be elucidated in well-designed randomized controlled trials (RCTs), and most evidence to date is drawn from observational studies.
We worry that the systematic review by Pavlov et al1 could have been outdated at publication since the authors included studies available up to August 15, 2020. In the recently published observational APRONOX study4 of awake prone positioning, a complementary systematic search of the literature (published and preprints) was conducted up to June 8, 2021, with a meta-analysis; observational studies of awake prone positioning were found to support a decreased intubation rate in subjects with COVID-19. Even when this outcome was not established through a formal systematic review like the one by Pavlov and colleagues,1 most of the studies meeting inclusion criteria (observational studies of awake prone positioning with a comparison group with enough data available to calculate intubation rates) were made available from late 2020 to 2021, reflecting that an important number of observational studies of awake prone positioning have been published after August 2020. In a different systematic review with meta-analysis, Chua et al5 found a lower mortality rate for subjects on awake prone positioning despite no apparent reduction in intubation rate during a similar study period. Even when the potential shortcoming of the time period in the study by Pavlov et al1 could be addressed by future systematic reviews or by reviewing multiple systematic reviews in an umbrella review, this should highlight the importance of conducting and publishing evermore “rapid living systematic reviews,”6 especially for rapidly changing areas of research like COVID-19.
Pavlov and colleagues1 should be recognized for reviewing abstracts in both English and Chinese. However, we noted that some of the authors are based at institutions in France, Canada, Mexico, and Spain. We wonder if the authors could have taken the effort to review also abstracts in French and Spanish, thereby taking advantage of their multinational group of authors to reach a more compelling and comprehensive study. By doing this, the authors would have addressed one of the main barriers of science: the language barrier.7
Evidence from RCTs evaluating awake prone positioning is urgently needed. However, researchers intending to investigate awake prone positioning should be warned that evaluating this intervention could be more complex than it seems. The recently published PROFLO trial8 is a good example since subjects in both the intervention and control group ultimately had at least some amount of exposition to awake prone positioning (median 9.0 and 3.4 h/d, respectively) with few subjects in the intervention group reaching the goal of ≥ 16 h/d in awake prone positioning (6%), which alongside the small sample size could explain why no differences in intubation rates were observed in this trial.
When enough evidence from RCTs assessing awake prone positioning is available, systematic review and meta-analysis will be helpful to picture the usefulness of awake prone positioning for patients with hypoxemic respiratory failure and COVID-19. It will be important to contrast the results of systematic review and meta-analysis of observational studies like the one performed by Pavlov et al1 (no difference in intubation rate), Chua et al5 (no difference in intubation rate but lower mortality rate), or the APRONOX group4 (decreased intubation rate) since associations from meta-analysis of observational studies and RCTs have been found to be opposite in direction in 37.1% of cases.9
Showing that prone positioning was useful for specific patients with ARDS on mechanical ventilation took several years and RCTs. We now know that prone positioning is useful when started early, for patients with moderate-to-severe ARDS, under lung protective ventilation strategies.10 Therefore, it could be too early to elaborate conclusions on awake prone positioning since there is still a long road ahead to identify patients and circumstances in whom awake prone positioning could be an effective intervention for hypoxemic respiratory failure.
Footnotes
- Correspondence: Javier Mancilla-Galindo MBBS, Neumología de Adultos, Instituto Nacional de Enfermedades Respiratorias, Ciudad de México, México. E-mail: javimangal{at}gmail.com
The authors have declared no conflicts of interest.
- Copyright © 2022 by Daedalus Enterprises