Prolonged prone positioning shows oxygenation benefit but inconclusive mortality effect in COVID-19 ARDS

This systematic review, meta-analysis, and trial-sequential analysis evaluated the clinical benefits of prolonged versus standard prone positioning in adult patients with coronavirus disease 2019 (COVID-19) and moderate-to-severe acute respiratory distress syndrome (ARDS). The analysis pooled data from 996 patients across seven studies, comprising 592 patients in the prolonged positioning group and 404 in the standard positioning group. The included studies were six observational studies and one randomized controlled trial (RCT). The specific clinical setting and follow-up duration were not reported. The intervention was defined as prolonged prone positioning with a mean session duration greater than 24 hours, compared to a standard protocol with mean session durations of 24 hours or less. The analysis did not provide further details on the specific dosing protocols, frequency of sessions, or criteria for initiation and cessation of prone positioning in either group.

The primary outcomes were mortality and the incidence of pressure injuries. For mortality, the analysis found a non-significant trend toward lower mortality with prolonged positioning. The relative risk (RR) was 0.81, with a 95% confidence interval (CI) of 0.60 to 1.09 (P = 0.16). In absolute terms, the mortality rate was 33.8% in the prolonged group versus 39.8% in the standard group. For pressure injuries, there was a borderline significant increase associated with prolonged positioning. The RR was 1.27 (95% CI: 1.00 to 1.62; P = 0.05), with an absolute incidence of 30.2% in the prolonged group compared to 26.2% in the standard group.

Key secondary outcomes included intensive care unit (ICU) length of stay, respiratory parameters, and oxygenation. There was no significant difference in ICU length of stay between groups, with a mean difference of 2.74 days favoring the standard group (P = 0.13). Specific absolute numbers for this outcome were not reported. The analysis found no significant differences in positive end-expiratory pressure or driving pressure between the two strategies; specific effect sizes and data were not reported. However, oxygenation showed significant improvements with prolonged positioning. During prone positioning, the partial pressure of arterial oxygen-to-fraction of inspired oxygen (PaO2/FiO2) ratio improved by a mean difference (MD) of 17.42 mmHg (P = 0.003). After prone positioning, the PaO2/FiO2 ratio improvement was even greater, with an MD of 23.83 mmHg (P = 0.008). Absolute numbers for these oxygenation outcomes were not reported.

Detailed safety and tolerability findings were not reported in the analysis. The provided data does not include rates of adverse events, serious adverse events, or treatment discontinuations specifically related to the prone positioning protocols. The borderline increase in pressure injuries (RR 1.27) represents the only quantifiable safety signal from the reported outcomes.

These results add to a complex body of evidence on prone positioning in ARDS. Prior landmark studies, such as the PROSEVA trial, established the mortality benefit of prone positioning (sessions of at least 16 hours) in severe ARDS, but primarily in non-COVID populations. This meta-analysis specifically in COVID-19 ARDS suggests a physiological benefit (improved oxygenation) with longer sessions, but it does not replicate the clear mortality benefit seen in earlier research. The findings highlight that the optimal duration and patient population for prone positioning in COVID-19-related ARDS may differ from general ARDS management.

The analysis has several key methodological limitations. The evidence base is predominantly observational (six of seven studies), which limits the ability to establish causality and increases susceptibility to confounding and selection bias. The single included RCT limits the strength of the conclusions. Furthermore, the trial-sequential analysis performed indicated that the current cumulative evidence is insufficient to confirm or refute a benefit or harm regarding mortality, meaning the sample size is inadequate for a definitive conclusion. Heterogeneity in patient management, prone positioning protocols, and definitions of outcomes across the seven included studies may also affect the consistency of the pooled results.

For clinical practice, these findings suggest that while prolonged prone positioning (>24 hours per session) in COVID-19 ARDS is associated with significantly better oxygenation parameters compared to standard duration, this physiological improvement did not translate into a statistically significant mortality reduction in this analysis. Clinicians should note the borderline increased risk of pressure injuries with prolonged sessions. Decisions on prone positioning duration should therefore balance the potential for improved gas exchange against the practical challenges and skin injury risks of very prolonged sessions, recognizing that the evidence for a net clinical benefit remains inconclusive.

Important questions remain unanswered. It is unclear if specific subgroups of COVID-19 patients with ARDS (e.g., defined by severity, comorbidities, or timing) might derive a mortality benefit from prolonged positioning. The optimal protocol for prolonged sessions—including exact duration, frequency, and criteria for cessation—requires definition through prospective RCTs. The impact of prolonged positioning on other important clinical outcomes, such as ventilator-free days, long-term functional status, and specific safety events beyond pressure injuries, is not addressed by this analysis and warrants further investigation.