Time to Multidrug-Resistant Tuberculosis Treatment Initiation in Association with Treatment Outcomes in Shanghai, China

In high tuberculosis (TB)-burden countries such as China, the diagnosis of multidrug-resistant tuberculosis (MDR-TB) using conventional drug susceptibility testing (DST) takes months, making treatment delay inevitable. Poor outcomes of MDR-TB might be associated with delayed, even inappropriate, treatment. The purposes of this study were to investigate the time to MDR-TB treatment initiation and to assess the association between early treatment and treatment outcomes. Between April 2011 and December 2014, this population-based retrospective cohort study collected the demographic and clinical characteristics and the drug susceptibility profiles of all registered MDR-TB patients in Shanghai, China. The dates of TB and MDR-TB diagnoses, DST performance, and treatment initiation were extracted to calculate the times to treatment. In total, 284 of 346 MDR-TB patients were eligible for analysis, and 68.3% (194/284) had favored outcomes. The median time to treatment initiation from TB diagnosis was 172 days among those with favored outcomes and 190 days among those with poor outcomes. Treatments initiated within 60 days after performing DST (odds ratio [OR], 2.56; 95% confidence interval [CI], 1.22 to 5.36) and empirical treatments (OR, 2.09; 95% CI, 1.01 to 4.32) were positively associated with favored outcomes. Substantial delays to MDR-TB treatment were observed when conventional DST was used. Early treatment predicted favored outcomes. Rapid diagnostic methods should be scaled up and improvements should be made in patient management and information linkage to reduce treatment delay.