Diagnostic Accuracy of Serum and Urinary Neutrophil Gelatinase-associated Lipocalin for Predicting Sepsis-associated Acute Kidney Injury: A Systematic Review and Meta-Analysis

Introduction. Sepsis‑associated acute kidney injury (SA‑AKI) is a frequent and serious complication among critically ill patients and is associated with substantial morbidity and mortality. Conventional diagnostic markers such as serum creatinine and urine output often detect kidney injury only after functional decline and may be confounded in septic states. Neutrophil gelatinase‑associated lipocalin (NGAL) has emerged as a potential early biomarker of tubular injury; however, its diagnostic performance in sepsis remains uncertain. This systematic review and meta‑analysis evaluated the diagnostic accuracy of serum and urinary NGAL for predicting SA‑AKI in adults with sepsis.Methods. A systematic search of the PubMed database was conducted from inception to 2025 to identify studies evaluating serum, plasma, or urinary NGAL in adult patients with sepsis. Eligible studies reported diagnostic accuracy for AKI defined according to KDIGO, AKIN, or RIFLE criteria. Data extraction was performed independently by two reviewers. Pooled sensitivity, specificity, diagnostic odds ratios (DORs), and summary receiver operating characteristic (SROC) curves were estimated using a bivariate random‑effects model. Subgroup analyses explored differences according to clinical setting and timing of biomarker measurement.Results. Seventeen studies met the inclusion criteria. For serum NGAL, the pooled sensitivity was 0.77 (95% CI: 0.63 to 0.86) and pooled specificity was 0.68 (95% CI: 0.52 to 0.80), with an AUC of 0.773 and a diagnostic odds ratio of 6.15 (95% CI: 4.38 to 8.64). Urinary NGAL demonstrated comparable but slightly higher diagnostic performance, with pooled sensitivity of 0.75 (95% CI: 0.66 to 0.82), specificity of 0.71 (95% CI: 0.64 to 0.78), an AUC of 0.782, and a diagnostic odds ratio of 7.12 (95% CI: 4.17 to 12.16). Subgroup analyses suggested modestly improved diagnostic performance in ICU populations.Conclusions. Both serum and urinary NGAL demonstrate moderate diagnostic accuracy for predicting SA‑AKI in adult patients with sepsis. Urinary NGAL showed slightly better discriminatory performance in several clinical contexts. Rather than replacing established KDIGO‑based diagnostic criteria, NGAL may serve as a complementary biomarker to support early risk stratification. Larger prospective studies with standardized assay methods and diagnostic thresholds are needed before routine clinical implementation.