Drug Exposure and Treatment Outcomes in Patients With Multidrug-Resistant Tuberculosis and Diabetes Mellitus: A Multicenter Prospective Cohort Study From China.

ABSTRACT

BACKGROUND: The management of multidrug-resistant tuberculosis (MDR-TB) remains challenging. Treatment outcome is influenced by multiple factors; the specific roles of diabetes and glycemic control remain uncertain. This study aims to assess the impact of glycemic control on drug exposure, to investigate the association between drug exposure and treatment outcomes, and to identify clinically significant thresholds predictive of treatment outcome, among patients with diabetes. METHODS: This multicenter prospective cohort study involved patients with confirmed MDR-TB and diabetes. Drug exposure level was estimated by noncompartmental analysis. The minimum inhibitory concentrations (MICs) were determined for the individual Mycobacterium tuberculosis isolates. The influence of poor glycemic control (glycated hemoglobin ≥7%) on drug exposure and the associations between drug exposure and treatment outcome were evaluated by univariate and multivariate analysis. Classification and regression tree analysis was used to identify the drug exposure/susceptibility thresholds. RESULTS: Among the 131 diabetic participants, 43 (32.8%) exhibited poor glycemic control. Poor glycemic control was independently associated with decreased exposure to moxifloxacin, linezolid, bedaquiline, and cycloserine, but not clofazimine. Additionally, a higher ratio of drug exposure to susceptibility was found to be associated with a favorable MDR-TB treatment outcome. Thresholds predictive of 6-month culture conversion and favorable outcome were bedaquiline area under the concentration-time curve (AUC)/MIC ≥245 and moxifloxacin AUC/MIC ≥67, demonstrating predictive accuracy in patients, regardless of their glycemic control status. CONCLUSIONS: Glycemic control and optimal TB drug exposure are associated with improved treatment outcomes. This dual management strategy should be further validated in randomized controlled trials of patients with MDR-TB and diabetes.