Evidence Hub

Association of preadmission metformin use and prognosis in patients with sepsis with diabetes: a systematic review and meta-analysis

Zhang M, Lin Z, Chen C, Zhong X, Liu W, Chen Z, Chen T, Song C et al.

Frontiers in endocrinology, 2026

12,687 participants

Abstract

Background: Preadmission metformin may lower mortality in diabetic sepsis patients, but evidence is conflicting, necessitating a systematic review and meta-analysis for confirmation.

Methods: We systematically searched MEDLINE (via PubMed), EMBASE, and Cochrane CENTRAL from inception to September 1, 2025, for cohort studies evaluating metformin use in septic patients with diabetes. Study quality was assessed using the Newcastle-Ottawa Scale. Two reviewers independently screened studies, extracted data, and evaluated methodological quality. Meta-analysis was conducted using STATA statistical software and Review Manager software, calculating pooled odds ratios with 95% confidence intervals via the inverse variance random-effects model. The MET group included diabetic sepsis patients with preadmission metformin exposure, and the NM group included those without.

Results: This meta-analysis of 14 studies (12,687 patients), all with low bias risk, demonstrated that preadmission metformin use in sepsis-diabetes patients was associated with reduced overall mortality (OR 0.58, 95% CI 0.44-0.75, P < 0.00001). Significant reductions were observed in 28-day (OR 0.61, P = 0.002), 90-day (OR 0.48, P = 0.001), 365-day (OR 0.33, P = 0.0005), and in-hospital mortality (OR 0.43, P < 0.02). However, 30-day (OR 0.71, P = 0.06), 60-day (OR 0.72, P = 0.22), and ICU mortality (OR 0.76, P = 0.25) showed no significant differences. Notably, metformin also significantly improved serum creatinine (MD…

How strong is the evidence?

Evidence Hub reads this as a meta-analysis, done in people. Its evidence level is 1 of 5 (Highest · pooled evidence). Pooled evidence from several studies sits at the top of the evidence pyramid. Reported size: 12,687 participants or records.

How the five levels of evidence work. The level is assigned automatically from the abstract and can be wrong, so check the paper.

Read the paper

Related studies

Go deeper in Evidence Hub

Evidence Hub adds a study data sheet (design, sample size, evidence level, key numbers), an AI summary and a chat grounded in the paper's text.

Open this study Sign up free