Treatment outcomes and long-term relapse-free survival after multidrug-resistant tuberculosis treatment in Latvia: a retrospective national cohort study.
retrospective_cohort · Level III
Where this comes from
- Record sourced from PubMed, PMID 42006570.
- Also identified by DOI 10.1016/j.lanepe.2026.101676 and PMC identifier 13091304.
- Licence recorded as CC BY.
- The licence permits redistribution, so the abstract is shown in full and the full text is available from the publisher.
Abstract
Treatment success rates of multidrug-resistant and rifampicin-resistant tuberculosis (MDR/RR-TB) lag behind those for drug-susceptible TB. However, outcome definitions censoring follow-up at treatment completion may underestimate effectiveness and fail to capture relapse-free survival. We conducted a retrospective national cohort study including all adults initiating individualized MDR/RR-TB treatment in the Republic of Latvia between 2005 and 2021. Demographic, clinical, and microbiological data were linked to long-term follow-up on relapse and vital status. Treatment outcomes were classified according to World Health Organization (WHO), TBnet (Tuberculosis Network European Trials Group), expert consilium, and long-term outcome definitions. Predictors of cure were assessed using Firth logistic regression. A landmark analysis at 18 months evaluated the association between treatment duration (≤9 months, 10-17 months, and ≥18 months) and relapse-free survival. Among 1299 patients (median age 44 years; 74.9% male), cure rates were 4.8% (n = 62) under WHO definitions, 53.1% (n = 690) under TBnet definitions, and 60.8% (n = 790) under the consilium-based classification. Under long-term follow-up outcome definitions, 56.5% (n = 734) achieved cure and 76.9% (n = 999) achieved relapse-free survival. Receiving ≥3 susceptible drugs independently predicted WHO-defined treatment success (adjusted OR 6.53, 95% CI 2.22-31.69; p < 0.001). In the landmark analysis, treatment duration ≤9 months was associated with higher hazard of relapse or death compared with ≥18 months (HR 1.76, 95% CI 1.03-3.00; p = 0.038), whereas outcomes were similar for 10-17 months vs. ≥18 months (HR 0.71, 95% CI 0.42-1.22; p = 0.22). In this national cohort treated with individualized MDR/RR-TB regimens, long-term relapse-free outcomes substantially exceeded treatment success defined at treatment completion. These findings support relapse-free survival as complement to end-of-treatment metrics. This study received no external funding.