Drug Resistance and Comorbidities in the Treatment of Pulmonary Tuberculosis: A Multicenter Retrospective Cohort Study

Tuberculosis (TB) probably returned to being the world’s leading cause of death from a single infectious agent after three years during which it was replaced by COVID-19. Currently, there are two major, closely related challenges in TB treatment: a large number of cases of drug-resistant TB, as well as cases complicated by severe comorbidities. Materials and Methods: Our study included 219 patients with pulmonary multidrug-resistant TB (MDR-TB) who were treated in several clinics in St. Petersburg, Russian Federation. Of these patients, 47.0% had extensively drug-resistant TB (XDR-TB), and 48.4% had severe comorbidities. Univariate and multivariate exploratory analyses were performed to hypothesize factors affecting treatment success. Results: Both extensive drug resistance (XDR-TB) and the presence of comorbidity were significantly associated with a lower probability of successful treatment: OR = 0.56 (CI: 0.32–0.96, p = 0.033) and OR = 0.53 (CI: 0.30–0.91, p = 0.020), respectively. The use of bedaquiline was significantly associated with successful treatment in cases of XDR-TB: OR = 4.15 (CI: 1.32–16.20, p = 0.012). Only an insignificant opposite effect was identified for cases of non-XDR-TB: OR = 0.77 (p = 0.62). Resistance to thioamides was associated with unsuccessful treatment in cases complicated by comorbidity: OR = 0.46 (CI: 0.21–0.99, p = 0.044). Again, an only insignificant opposite effect was identified for cases without comorbidities: OR = 1.11 (p = 0.81). Almost all the patterns described above were replicated in the multivariate model. The following two differences with the univariate results were observed. First, the association between the use of bedaquiline and successful treatment became even more pronounced, and, as before, this was true only for XDR-TB: OR = 6.51 (CI: 1.98–26.04, p = 0.0036) for XDR-TB, and OR = 0.99 (p = 0.98) for non-XDR-TB. Second, the impact of comorbidities on treatment success remained significant only in conjunction with thioamide resistance. In addition, we found that the association between resistance to thioamides and unsuccessful treatment was especially pronounced in cases complicated by heart disease: OR = 0 (CI: 0–0.79, p = 0.0088). Conclusions: We confirmed that both XDR-TB and the presence of comorbidities are serious challenges in the treatment of tuberculosis. We also have reason to hypothesize that, first, bedaquiline can be a much more crucial component of therapy in cases of XDR-TB than in other cases of MDR-TB and, second, thioamides can play a positive role in cases complicated by comorbidities, especially by heart diseases. These findings should be considered as weak hypotheses that require further verification using independent data, as our analysis was exploratory rather than confirmatory.