Lymphopenia and mortality in adult sepsis: A systematic review and focused meta-analysis of prognostic evidence.
TL;DR
Lymphopenia-related phenotypes are associated with mortality in adult sepsis, but the small quantitative evidence base and moderate-to-substantial heterogeneity limit the precision and generalizability of the pooled effect.
Abstract
Background
Lymphopenia is common in sepsis, but published studies use heterogeneous exposure definitions and time points. We evaluated its prognostic association with mortality using a focused meta-analysis of comparable lymphopenia phenotypes and used additional evidence to contextualize biologic and infectious-source interpretation.
Methods
PubMed and Embase were searched from inception to April 1, 2026. Studies directly estimating lymphopenia-related mortality associations were assessed for quantitative synthesis. Pooling was restricted to comparable threshold-based, persistent, or baseline definitions using a random-effects generic inverse-variance model; trajectory-based and continuous-exposure studies were displayed separately. Additional mechanistic, biomarker, modeling, therapeutic, and pathogen-context studies were synthesized qualitatively.
Results
Of 29 included studies, seven provided direct prognostic evidence and four were eligible for quantitative pooling. Adverse lymphopenia-related phenotypes were associated with mortality (pooled relative effect 1.94, 95% CI 1.39-2.70; I2 = 62.4%). Leave-one-out analyses retained the direction of association, although heterogeneity was sensitive to individual studies. Two trajectory-based studies and one continuous-exposure study were not pooled. Pathogen-related observations were insufficient for pathogen-specific quantitative inference.
Conclusions
Lymphopenia-related phenotypes are associated with mortality in adult sepsis, but the small quantitative evidence base and moderate-to-substantial heterogeneity limit the precision and generalizability of the pooled effect. Pathogen-context interpretations should be considered hypothesis-generating. Standardized definitions, repeated measurements, and prospective validation are needed before routine use for risk stratification or treatment selection. PROTOCOL REGISTRATION PROSPERO CRD420261413974.