Mortality Effect of Peripheral Perfusion-Guided Resuscitation in Vasodilatory Shock: A Systematic Review and Dual Frequentist-Bayesian Meta-Analysis of Randomized Trials.

Abstract

To determine the effect of peripheral perfusion-guided resuscitation on 28-day mortality in adults with vasodilatory shock. We searched PubMed, Embase, Cochrane Central Register of Controlled Trials, and Scopus from inception to November 2025. We included randomized controlled trials (RCTs) comparing peripheral perfusion-guided resuscitation with standard care or lactate-guided resuscitation; all eligible trials enrolled patients with septic shock. Trial characteristics, intervention and comparator protocols, and outcomes were extracted. Risk of bias was assessed with the Cochrane Risk of Bias tool for randomized trials 2.0 tool and certainty of evidence with the Grading of Recommendations, Assessment, Development, and Evaluation approach. We conducted a prespecified dual-framework meta-analysis comprising a primary frequentist random-effects model and a Bayesian hierarchical model with alternative prior assumptions. The review followed the Preferred Reporting Items for Systematic Reviews and Meta-Analyses and was preregistered with PROSPERO (CRD420251229564). Seven RCTs (n = 2408) were included. In the primary frequentist analysis, the pooled risk ratio (RR) for 28-day mortality with peripheral perfusion-guided resuscitation was 0.87 (95% CI, 0.76-1.01; p = 0.06; moderate-certainty evidence), with the CI crossing the line of no effect. No heterogeneity was detected (I2 = 0%). The two largest and lowest-risk trials were individually nonsignificant. The prespecified Bayesian analysis estimated a 97.2% posterior probability that RR is less than 1 (P [RR < 1]), broadly consistent across alternative prior assumptions. Among adults with septic shock, peripheral perfusion-guided resuscitation yielded a lower 28-day mortality estimate that did not meet statistical significance in the frequentist analysis. The Bayesian component of the dual framework was directionally concordant, estimating a high posterior probability that RR is less than 1, although the magnitude of any effect remains uncertain. Adequately powered trials are needed to confirm whether this signal reflects a clinically meaningful benefit.