Diagnostic & Prognostic Value of Point-of-Care Lactate (POC Lactate) Measured in ED Triage for Sepsis Outcomes in Resource-Limited Hospitals. A Systematic Review

Abstract

Background: Delays in laboratory lactate at emergency department (ED) triage hinder early sepsis recognition in resource-limited hospitals. Point-of-care (POC) lactate offers rapid bedside assessment of hypoperfusion, but triage-time diagnostic accuracy, prognostic value, and operational impact remain variably reported across devices, sampling methods, and settings. Objective: To evaluate the diagnostic test accuracy (DTA) and prognostic associations of triage POC lactate for sepsis outcomes, identify optimal thresholds, examine moderators (LMIC vs HIC, capillary vs venous, device family, sepsis definition), and summarize feasibility/process effects. Methods: Systematic review and meta-analysis (PRISMA/PRISMA-DTA). Adults with suspected infection undergoing POC lactate at ED triage (≤2 h) were included. Bivariate random-effects models synthesized sensitivity/specificity with HSROC; random-effects meta-analyses pooled adjusted mortality, ICU admission, and mechanical ventilation (MV) estimates; meta-regression assessed prespecified moderators; feasibility outcomes were narratively synthesized; risk of bias used QUADAS-2/QUIPS; certainty was graded with GRADE. Results: Thirty-seven studies met criteria. At 3.5–4.0 mmol/L (k=16), pooled sensitivity was 0.72 (95% CI 0.66–0.77) and specificity 0.78 (0.72–0.83) with HSROC AUC 0.82; at 2.0 mmol/L (k=13), sensitivity was 0.84 (0.78–0.89) and specificity 0.58 (0.50–0.65). Elevated triage lactate predicted mortality (adjusted HR 1.88; adjusted OR 2.61), ICU admission (OR 2.10), and MV (OR 1.95). Adding lactate to qSOFA improved discrimination (ΔAUC ≈ +0.05–0.12). POC turnaround was ~60–120 seconds versus ~25–40 minutes for laboratory testing. Conclusion: Triage POC lactate provides moderate diagnostic accuracy and meaningful prognostic value; a 3.5–4.0 mmol/L threshold offers balanced performance, while 2.0 mmol/L maximizes sensitivity. Integration into nurse-led triage with basic QA is warranted.