Epidemiologic and Bacterial Factors Facilitating Long-term Transmission of Multidrug-Resistant Tuberculosis in Shanghai, China

BACKGROUND: Long-term transmission of multidrug-resistant tuberculosis (MDR-TB) challenges TB control by generating new cases and enabling the emergence of extensively resistant strains. We investigated its epidemiologic and bacterial drivers in Shanghai, China. METHODS: We conducted a retrospective study of Mycobacterium tuberculosis isolates and associated epidemiological data from individuals diagnosed with rifampicin-resistant TB in Shanghai over 14 years (2004-2018). Using whole-genome sequencing, Bayesian reconstruction of transmission trees, and multivariable regression analysis to identify epidemiological and bacterial factors associated with the transmission of MDR-TB. RESULTS: Between 2004 and 2018, 1456 individuals in Shanghai were diagnosed with MDR or rifampicin-resistant TB, with whole-genome sequences available for 1100 isolates. The overall genomic clustering rate was 55.3%, with large clusters (those containing ≥5 cases) accounting for 39.9% of the clustered strains. Risk factors for clustered MDR-TB transmission included local residents (adjusted odds ratio [aOR] 2.28, 95% confidence interval [CI]: 1.67-3.11), diagnostic delays ≥2 months (aOR 1.75, 95% CI: 1.24-2.47), specific M. tuberculosis sublineages (L2.3.3-L2.3.6), and the rpoB S450L mutation with compensatory mutations (aOR 2.14, 95% CI: 1.64-2.78). Large MDR-TB clusters were significantly associated with long-term transmission (>5 years, P < .001). Long-term transmission clusters correlated with same-street residence, local residents, and MDR-TB strains carrying katG315T, rpoBS450L and compensatory mutations. CONCLUSIONS: Despite ample healthcare resources, MDR-TB persists in urban areas due to both epidemiological and bacterial factors. The rpoB S450L mutation with compensatory mutations enhances transmission even in the absence of clear epidemiologic links. Effective control measures must address both epidemiological and bacterial factors.