Association Between Blood Glucose Variability and Clinical Outcomes in Patients With Sepsis: A Systematic Review and Meta-Analysis.

AIMS: Glycaemic variability (GV) has emerged as an important prognostic indicator in critical illness, yet its predictive value among patients with sepsis remains unclear. This systematic review and meta-analysis aimed to evaluate the association between GV metrics and mortality outcomes in adult patients with sepsis.

METHODS

Cohort studies enrolling septic patients and reporting in-hospital, 28-day, or 30-day mortality in relation to GV were identified through PubMed, Embase, Cochrane Library, Scopus, CNKI, and Wanfang databases. Pooled odds ratios (ORs) were calculated using a random-effects model. Sensitivity analyses were performed to assess the robustness of the findings.

RESULTS

Ten studies comprising 18,337 patients were included. For categorical analysis, high-GV patients had nearly twice the mortality risk (OR = 1.99, 95% CI: 1.66-2.40, p < 0.0001; I2 = 45%). For continuous analysis, all 4 GV metrics showed significant associations with mortality: CoV (OR = 1.050, I2 = 76.6%), SD (OR = 1.0037, I2 = 83.5%), GLI (OR = 1.0171, I2 = 0.0%), and MAGE (OR = 1.0062, I2 = 0.0%). High GV was associated with prolonged ICU stay (0.95 days, p = 0.0018). Sensitivity analyses confirmed the result robustness.

CONCLUSIONS

Elevated GV is independently linked to an increased risk of death among patients with sepsis. GLI and MAGE are the most reliable GV metrics for prognostic assessment, whereas CoV and SD are less consistent. Standardised GV measurement and prospective studies are warranted to evaluate whether interventions targeting GV can improve outcomes in this population.