C-reactive protein as a predictor of mortality in tuberculosis: systematic review and meta-analysis.

BACKGROUND: Tuberculosis (TB) remains a leading cause of global mortality, underscoring the need for accessible prognostic tools. C-reactive protein (CRP) is a widely available acute-phase biomarker that may predict outcomes in TB. This systematic review and meta-analysis assessed the prognostic value of baseline CRP in predicting mortality among adults with TB.

METHODS: Eligible studies included cohort studies, observational studies, and control arms of randomized controlled trials involving adults (≥ 18 years) with microbiologically confirmed TB. The exposure of interest was baseline CRP level, and the primary outcome was mortality, reported as adjusted hazard ratios (aHRs) or odds ratios (aORs). Data extraction followed the CHARMS-PF checklist. Risk of bias was assessed using the QUIPS, while certainty of evidence was evaluated using GRADE.

RESULTS: From 1,277 records, nine studies met the inclusion criteria (three retrospective cohorts, three prospective cohorts, and three case-control studies) conducted in Japan, China, South Korea, Chinese Taipei, and South Africa. Five studies reported significant associations between elevated baseline CRP and increased mortality. In pooled analyses, three studies reporting aORs showed a modest but statistically significant association between higher baseline CRP and mortality (aOR 1.07, 95% CI 1.03-1.11; I²=0%). In contrast, pooled aHRs from four studies (aHR 1.02, 95% CI 0.99-1.05; I²=60%) and pooled cHRs from two studies (cHR 1.75, 95% CI 0.58-5.29; I²=91%) were not statistically significant.

CONCLUSIONS: Baseline CRP showed a modest association with mortality in pooled aOR analyses, but not in pooled hazard ratio analyses. Nevertheless, its low cost and wide availability suggest potential utility as part of multimodal prognostic models, especially in high-risk populations and resource-limited settings. High-quality prospective studies with standardized CRP protocols are needed to clarify its prognostic role.

REGISTRATION: PROSPERO (CRD420251101984).