Recurrent tuberculosis in an urban area in China: Relapse or exogenous reinfection?

Abstract

Recurrent tuberculosis is an important indicator of the effectiveness of tuberculosis control and can occur by relapse or exogenous reinfection. We conducted a retrospective cohort study on all bacteriologically confirmed tuberculosis cases that were successfully treated between 2000 and 2012 in Shanghai, an urban area with a high number but a low prevalence rate of tuberculosis cases and a low prevalence of HIV infection. Genotyping the Mycobacterium tuberculosis from clinical isolates was used to distinguish between relapse and reinfection. In total, 5.3% (710/13,417) of successfully treated cases had a recurrence, a rate of 7.55 (95% CI 7.01-8.13) episodes per 1000 person-years, more than 18 times the rate of tuberculosis in the general population. Patients who were male, age 30-59, retreatment cases, had cavitation, diabetes, drug-resistant or multidrug-resistant tuberculosis in their initial episode of tuberculosis, were at high risk for a recurrence. Among 141 recurrent cases that had paired isolates, 59 (41.8%) had different genotypes, indicating reinfection with a different strain. Patients who completed treatment were still at high risk of another episode of tuberculosis and exogenous reinfection contributed a significant proportion of the recurrent tuberculosis cases. Targeted control strategies are needed to prevent new tuberculosis infections in this setting.

Keywords: China; Exogenous reinfection; Recurrence; Tuberculosis.

Figure 1

Flow chart of tuberculosis cases included and excluded from the study. TB, tuberculosis.

Figure 2

Kaplan-Meier survival estimates for recurrent tuberculosis due to relapse (n=82, solid line) and exogenous reinfection (n=59, dashed line). Only the patient with a known relapse or reinfection were included. (Kolmogorov-Smirnov test for difference in relapse versus exogenous reinfection, P = 0.02).