Abstract
OBJECTIVE: This study aimed to investigate the impact of diabetes mellitus (DM) on tuberculosis (TB) treatment response using bacterial clearance as a surrogate marker. METHOD: We compared smear microscopy, culture, and tuberculosis molecular bacterial load assay (TB-MBLA) for treatment monitoring. Following that, bacterial clearance was longitudinally monitored among TB-only (TB without DM) and TB-diabetes (TBDM) patients using TB-MBLA. RESULTS: Ninety-three participants, including 59 TB-only and 34 TBDM patients, were enrolled. TB-only patients exhibited higher upper zone infiltrations (32/35 vs 16/22, p = 0.059) suggesting a trend towards significance, and significantly more cavitation in the same zone (16/18 vs 7/13, p = 0.028). There was a high proportion of Mycobacterium africanum (Maf) among the TBDM cohort (p = 0.0044).At baseline, TB-only patients exhibited a higher average bacterial burden (4.49 logeCFU/mL) compared to the TBDM group (3.91 logeCFU/mL) (p = 0.042). The bacterial load in the TB-only group decreased significantly during treatment but the TBDM group experienced delayed clearance throughout the intensive phase of anti-TB treatment even at day 56 (p = 0.028). The TB-only group demonstrated a shorter median time to TB-MBLA conversion to negative (57 days) compared to the TBDM group (62 days) (p = 0.022). CONCLUSION: These findings underscore the urgent call for understanding the interplay between diabetes and TB, emphasizing the need for tailored interventions in optimizing TB care for individuals comorbid with diabetes.