Back

Impact of Diabetes Mellitus and Viral Hepatitis on the Toxicity Profile of Pre-Extensively Drug-Resistant Tuberculosis Chemotherapy: Focus on Neurotoxicity and Ototoxicity

Pavlov, S. A.; Tikhonov, A. M.; Gordeeva, O. M.

2025-12-06 respiratory medicine
10.64898/2025.12.02.25341277 medRxiv
Show abstract

ObjectiveTo analyse the impact of comorbid conditions (diabetes mellitus, viral hepatitis) on the frequency and spectrum of adverse drug reactions (ADRs) during the treatment of pre-extensively drug-resistant (Pre-XDR) pulmonary tuberculosis, with a specific focus on mitochondrial toxicity. Materials and MethodsA retrospective cohort study of 247 HIV-negative patients with Pre-XDR tuberculosis was conducted. Patients received chemotherapy regimens containing linezolid, fluoroquinolones, and injectable agents (consistent with guidelines prior to the 2021 WHO update). Toxicity was graded according to CTCAE standards. The correlation between comorbidities and the incidence of neurotoxic and hepatotoxic ADRs was assessed using multivariate regression analysis. ResultsHigh comorbidity burden was identified as an independent predictor of treatment failure (OR=3.45; p=0.007). The cohort was characterized by a high prevalence of viral hepatitis B/C (up to 12.2%) and diabetes mellitus (up to 10.2%). The most frequent ADRs were eosinophilia (33.6%), ototoxicity (26.3%), and polyneuropathy (up to 15.2% in previously treated patients). Surprisingly, age < 40 years was associated with poor outcomes (OR=4.15; p=0.013), likely driven by lower adherence. The development of severe ADRs significantly increased the risk of treatment failure (OR=4.17; p=0.037). ConclusionThe combination of Pre-XDR tuberculosis with metabolic comorbidities creates a synergistic "mitochondrial hit" effect. This manifests as increased neurotoxicity (attributable to linezolid) and ototoxicity. Even in the era of all-oral regimens, the intersection of diabetes and linezolid-based therapy requires enhanced neuroprotective monitoring.

Matching journals

The top 6 journals account for 50% of the predicted probability mass.

50% of probability mass above

"Similar papers" are the closest papers from that journal in the model's embedding space. They show what the match is built on, but the ranking comes mostly from a classifier over the whole training set, not from these examples alone.