Source: Masters Abstracts International, Volume: 57-05.

Advisers: Lucian Davis; Malabika Sarker.

Introduction: Diabetes mellitus (DM) comorbidity with tuberculosis (TB) patients has been shown to increase both the risk of development of tuberculosis and the risk of death from TB. In a world where one third of the population is a latent carrier of TB and the DM burden is constantly growing, comorbidity poses a serious risk. In Bangladesh there are no studies on the effect of DM TB comorbidity on treatment outcomes.

Objectives: The primary objective of this study was to determine the impact of DM on TB treatment non-response (positive sputum culture at month 2).

Methods: We conducted a retrospective cohort study of patients with sputum confirmed TB from six urban and eleven rural areas throughout Bangladesh. We measured fasting glucose, and conducted an interview where anthropometric, demographic and socioeconomic data were collected. We recorded TB status (determined by sputum culture) at months 2,3,4-5 and 6 from treatment cards. A multivariate Logistic regression was conducted using sputum conversion status at 2-months.

Results: Adjusting for monthly income, sex, marital status and family history of diabetes, prediabetic patients had 2.07 (95% CI: 1.17 - 3.68) times the risk of treatment non-response (TB did not clear after two months). Diabetics not taking medication and diabetics taking medication had 0.90 (95% CI: 0.38 - 2.18) and 1.04 (95% CI: 0.38 - 2.86) times the risk, respectively, of treatment non-response compared to those without diabetes, however these findings were not significant. Conclusion: Prediabetes was a risk factor for treatment non-response in Bangladesh. In light of the increasing burden of diabetes and prediabetes worldwide and the implications of tuberculosis treatment, national programmes need to establish a coordinated response at the preventive and treatment levels.