Treatment outcomes among patients with multidrug-resistant tuberculosis – Authors' reply

In our reanalysis, we clarified definitions of the sample size used, including only those patients from each study with culture-confirmed multidrug-resistant tuberculosis who had started a second-line drug regimen. In the paper by Munsiff and colleagues,1 only 610 of 856 patients were treated with second-line drugs. [...]our recorded prevalence of HIV of 58% was on the basis of only the 352 patients infected with HIV from this smaller sample. Randomised clinical trials comparing programmatic and treatment regimen differences for patients with multidrug-resistant tuberculosis are needed and would provide a stronger evidence base for improving treatment outcomes in this population.