Surveillance for multi-drug and rifampicin resistant tuberculosis and treatment outcomes among previously treated persons with tuberculosis in the era of GeneXpert in rural eastern Uganda

RATIONALE: Previously treated persons with bacteriologically confirmed pulmonary tuberculosis (BC-PTB) have increased risk of developing multi-drug resistant or rifampicin resistant tuberculosis (MDR/RR-TB). Surveillance for resistance is critical to identify and treat MDR/RR-TB to ensure cure and prevent transmission. There are limited studies conducted on this subject. OBJECTIVES: We examined the frequency and factors associated with MDR/RR-TB surveillance among previously treated persons with BC-PTB, and described their treatment outcomes in rural eastern Uganda. METHODS: We reviewed treatment records for BC-PTB between January 2015 and June 2018 at 10 clinics in eastern Uganda. We collected data on demographics, surveillance for MDR/RR, use of GeneXpert and treatment outcomes. We performed bivariate and multivariate analyses. For multivariate analysis, we used the modified Poisson regression analysis with robust standard errors and stated the results as adjusted risk ratio (aRR) with 95% confidence intervals (CI). All analyses were conducted in R version 3.5.2. MEASUREMENTS AND MAIN RESULTS: We obtained records for 135 previously treated persons with BC-PTB and of these, 41 (30.4%) had undergone surveillance for MDR/RR-TB. Treatment failures were less likely to have surveillance compared to relapses (aRR, 0.28; 95% CI, 0.08–0.95), and there was an increasing trend in the likelihood for surveillance between 2015 and 2018 (aRR, 1.77; 95% CI, 1.39–2.25). There was no difference in MDR/RR-TB surveillance rate between health facilities with and without GeneXpert on-site (aRR, 1.52; 95% CI, 0.81–2.86) and between male and female patients (aRR, 0.54; 95% CI, 0.21–1.37). Overall, 92 (68.1%) previously treated persons with BC-PTB were successfully treated for tuberculosis. CONCLUSIONS: MDR/RR-TB surveillance and treatment success rates among previously treated persons with BC-PTB in rural eastern Uganda are low. Tuberculosis programs should strengthen MDR/RR-TB surveillance and especially target those with treatment failure.