Predictors of In-Hospital Mortality among Patients with Pulmonary Tuberculosis: A Systematic Review and Meta-analysis

Background There is uncertainty regarding which factors are associated with in-hospital mortality among patients with pulmonary TB (PTB). The aim of this systematic review and meta-analysis is to identify predictors of in-hospital mortality among patients with PTB. Methods We searched MEDLINE, EMBASE, and Global Health, for cohort and case-control studies that reported risk factors for in-hospital mortality in PTB. We pooled all factors that were assessed for an association, and presented relative associations as pooled odds ratios (ORs). Results We identified 2,969 records, of which we retrieved 51 in full text; 11 cohort studies that evaluated 5,468 patients proved eligible. Moderate quality evidence suggested an association with co-morbid malignancy and in-hospital mortality (OR 1.85; 95% CI 1.01-3.40). Low quality evidence showed no association with positive sputum smear (OR 0.99; 95% CI 0.40-2.48), or male sex (OR 1.09, 95% CI 0.84-1.41), and very low quality evidence showed no association with diabetes mellitus (OR 1.31, 95% IC 0.38-4.46), and previous TB infection (OR 2.66, 95% CI 0.48-14.87). Conclusion Co-morbid malignancy was associated with increased risk of in-hospital death among pulmonary TB patients. There is insufficient evidence to confirm positive sputum smear, male sex, diabetes mellitus, and previous TB infection as predictors of in-hospital mortality in TB patients.