Optimal Duration of Anti-TB Treatment in Patients With Diabetes

BACKGROUND: Diabetes mellitus (DM) increases the risk of TB recurrence. Th is study investigated whether 9-month anti-TB treatment is associated with a lower risk of TB recurrence within 2 years aft er complete treatment than 6-month treatment in patients with DM with an emphasis on the impact of directly observed therapy, short course (DOTs). METHODS: Patients with pulmonary but not extrapulmonary TB receiving treatment of 173 to 277 days between 2002 and 2010 were identifi ed from the National Health Insurance Research Database of Taiwan. Patients with DM were then selected and classifi ed into two groups based on anti-TB treatment duration (9 months vs 6 months). Factors predicting 2-year TB recurrence were explored using Cox regression analysis. RESULTS: Among 12,688 patients with DM and 43,195 patients without DM, the 2-year TB recurrence rate was 2.20% and 1.38%, respectively ( P , .001). Of the patients with DM, recurrence rate decreased from 3.54% to 1.19% aft er implementation of DOTs ( P , .001). A total of 4,506 (35.5%) were classifi ed into 9-month anti-TB treatment group. Although a 9-month anti-TB treatment was associated with a lower recurrence rate (hazard ratio, 0.76 [95% CI, 0.59-0.97]), the benefit disappeared (hazard ratio, 0.69 [95% CI, 0.43-1.11]) under DOTs. Other predictors of recurrence included older age, male sex, malignancy, earlier TB diagnosis year, culture positivity aft er 2 months of anti-TB treatment, and anti-TB treatment being 80% consistent with standard regimen. CONCLUSIONS: Th e 2-year TB recurrence rate is higher in a diabetic population in Taiwan and can be reduced by treatment supervision. Extending the anti-TB treatment by 3 months may also decrease the recurrence rate when treatment is not supervised. CHEST 2015; 147(2): 520 528 [ Original Research Chest Infections ]