Risk factors for types of recurrent tuberculosis (reactivation versus reinfection): A global systematic review and meta-analysis

The purpose of this meta-analysis (PROSPERO number: CRD42021243204) is to perform extensive and penetrating analyses on the risk factors associated with reactivation or reinfection. We searched PubMed and Embase using search terms. Risk factors (including gender, length of time between first onset and recurrent diagnosis, extrapulmonary tuberculosis, sputum smear, pulmonary cavity, Beijing family strains, diabetes, HIV infection, history of imprisonment, and immigration) were analyzed. The pooled risk ratio (RR) and 95% confidence interval (CI) were calculated with STATA 15.1. Heterogeneity was evaluated by I2 and P values. The meta-analysis included 25 studies with a total of 1,477 recurrent patients. After subgroup analysis, sensitivity analysis, and testing for publication bias, it was concluded that time spanning less than two years (RR = 1.56, 95% CI: 1.33-1.85) was a risk factor for endogenous reactivation, whereas coinfection with HIV (RR = 0.72, 95% CI: 0.63-0.83), Beijing family genotype (RR = 0.46, 95% CI: 0.32-0.67), history of imprisonment (RR = 0.36, 95% CI: 0.16-0.81), and immigration (RR = 0.66, 95% CI: 0.53-0.82) were associated with exogenous reinfection. The recurrence interval is a risk factor for the endogenous reactivation of tuberculosis. Infection with Beijing family strains, coinfection with HIV, imprisonment, and immigration contribute to the risk of exogenous reinfection. •Time spanning less than 2 years was a risk factor for endogenous reactivation•Co-infection with HIV was related to exogenous reinfection•Beijing genotype isolate was related to exogenous reinfection•History of imprisonment or immigration was associated with exogenous reinfection.