Prognostic value of presepsin in sepsis and septic shock: a meta-analysis.

Background: Presepsin, an innate immune activation biomarker, shows potential for predicting the prognosis of sepsis, but its predictive accuracy remains unclear. This meta-analysis aims to evaluate its efficiency for predicting the risk of mortality in sepsis and septic shock. Methods: Per PRISMA guidelines, four databases were searched until January 2025. Sixteen observational studies with 2,066 patients were included. Pooled sensitivity, specificity, DOR, and AUC were calculated using bivariate random-effects models. The sources of heterogeneity were explored via subgroup analyses and meta-regression. Study quality was assessed with QUADAS-2. Results: Presepsin showed moderate accuracy for predicting the risk of mortality (pooled AUC 0.80, 95% CI: 0.76-0.83). The pooled sensitivity and specificity were 76% (95% CI: 69-82%) and 70% (95% CI: 59-78%), respectively. Sensitivity was notably higher in patients with septic shock (90%) compared to those with sepsis (75%), whereas specificity was comparatively lower (50% vs. 77%). Substantial heterogeneity stemmed from threshold variability and geographic differences (particularly in Korean cohorts). Prospective studies had higher sensitivity (80%) than retrospective studies (60%). No publication bias was detected. Conclusions: Presepsin is a pragmatic biomarker for predicting the risk of mortality, with enhanced sensitivity in septic shock. Presepsin may be integrated into clinical decision-making for early intervention. Systematic Review Registration: PROSPERO, identifier CRD42025639268.