Higher PEEP for acute respiratory distress syndrome: a Bayesian meta-analysis of randomised clinical trials

Objective: Benefit or harm of higher positive end expiratory pressure (PEEP) for acute respiratory distress syndrome (ARDS) is controversial. We aimed to assess the impact of higher levels of PEEP in patients with ARDS under a Bayesian framework. Design: Systematic review and Bayesian meta-analysis of randomised clinical trials comparing higher to lower PEEP in adult patients with ARDS. Data sources: MEDLINE, EMBASE and Cochrane Central Register of Controlled Trials from 1996 to 1 March 2020. Review methods: We extracted data from high quality randomised clinical trials comparing higher to lower levels of PEEP in adult patients, using low tidal volume in both arms, and conducted a Bayesian meta-analysis using aggregate data from these studies. Results: Eight clinical trials including 3703 patients (n = 1833 for higher PEEP, n = 1870 for lower PEEP) were included. Under a minimally informative prior, the posterior probability of benefit with higher PEEP was 65% (relative risk, 0.97 [95% credible interval, 0.78-1.14]). In patients with moderate-to- severe ARDS, the posterior probability of benefit with higher PEEP was 77% (relative risk, 0.94 [95% credible interval, 0.77-1.13]). Down-weighting studies that employed a maximum recruitment strategy by 100% increased the posterior probability of benefit to 92% under a minimally informative prior. Conclusions: The probability of benefit or harm from routine use of higher PEEP for patients with ARDS ranges from 27% to 86%, and from 14% to 73% depending on one's prior, suggesting continued uncertainty and equipoise regarding the benefit of PEEP If data from trials using a maximum recruitment strategy is discounted to some extent because of uncertainty over the appropriateness of this approach, the available evidence suggests that higher PEEP could be beneficial for moderate-to-severe ARDS. However, well powered randomised clinical trials are needed to confirm these findings.