Objective
To determine the efficacy and safety of awake prone positioning versus usual care in non-intubated adults with hypoxemic respiratory failure due to covid-19.
Design
Systematic review with frequentist and bayesian meta-analyses.
Study eligibility
Randomized trials comparing awake prone positioning versus usual care in adults with covid-19 related hypoxemic respiratory failure. Information sources were Medline, Embase, and the Cochrane Central Register of Controlled Trials from inception to 4 March 2022.
Data extraction and synthesis
Two reviewers independently extracted data and assessed risk of bias. Random effects meta-analyses were performed for the primary and secondary outcomes. Bayesian meta-analyses were performed for endotracheal intubation and mortality outcomes. GRADE certainty of evidence was assessed for outcomes.
Main outcome measures
The primary outcome was endotracheal intubation. Secondary outcomes were mortality, ventilator-free days, intensive care unit (ICU) and hospital length of stay, escalation of oxygen modality, change in oxygenation and respiratory rate, and adverse events.
Results
17 trials (2931 patients) met the eligibility criteria. 12 trials were at low risk of bias, three had some concerns, and two were at high risk. Awake prone positioning reduced the risk of endotracheal intubation compared with usual care (crude average 24.2%
v
29.8%, relative risk 0.83, 95% confidence interval 0.73 to 0.94; high certainty). This translates to 55 fewer intubations per 1000 patients (95% confidence interval 87 to 19 fewer intubations). Awake prone positioning did not significantly affect secondary outcomes, including mortality (15.6%
v
17.2%, relative risk 0.90, 0.76 to 1.07; high certainty), ventilator-free days (mean difference 0.97 days, 95% confidence interval −0.5 to 3.4; low certainty), ICU length of stay (−2.1 days, −4.5 to 0.4; low certainty), hospital length of stay (−0.09 days, −0.69 to 0.51; moderate certainty), and escalation of oxygen modality (21.4%
v
23.0%, relative risk 1.04, 0.74 to 1.44; low certainty). Adverse events related to awake prone positioning were uncommon. Bayesian meta-analysis showed a high probability of benefit with awake prone positioning for endotracheal intubation (non-informative prior, mean relative risk 0.83, 95% credible interval 0.70 to 0.97; posterior probability for relative risk <0.95=96%) but lower probability for mortality (0.90, 0.73 to 1.13; <0.95=68%).
Conclusions
Awake prone positioning compared with usual care reduces the risk of endotracheal intubation in adults with hypoxemic respiratory failure due to covid-19 but probably has little to no effect on mortality or other outcomes.
Systematic review registration
PROSPERO CRD42022314856.