No abstract

The objective of the study: to identify risk factors for relapses in patients after the effective course of chemotherapy for multiple drug resistant tuberculosis (MDR TB). Subjects and methods. Medical files of 346 adult MDR TB patients were analyzed, they all had the effective treatment as per regimen IV in Tomsk Phthisiopulmonology Medical Center in 2009-2011. They were divided into 2 groups: Group 1 included 35 patients who developed a relapse of tuberculosis over the next 5 years; Group 2 consisted of 311 patients who had no relapse of the disease over the next 5 years. Results. The relapse rate in effectively treated MDR TB patients made 10.1% (35/346 patients). The following social factors contributed to higher chances of the relapse: retirement age (p = 0.045; OR = 2.86 [1.14-7.20]) and disability due to somatic diseases (p < 0.001; OR = 4.82 [2.13-10.90]); while biomedical factors were the following: HIV infection (OR = 19.19 [5.29-69.56]), mental illness (OR = 5.85 [2.27-15.03]), tobacco smoking (OR = 3.16 [1.08-9.20]). People with history of tuberculosis relapses (OR = 12.17 [4.19-35.34]) have higher chances of relapse, as well as those having the following characteristics of the tuberculosis disease during the effective chemotherapy course: destruction of lung tissue (OR = 7.48 [1.76-31.80]), positive results of sputum smear microscopy (OR = 2.91 [1.28-6.61]), persisting bacterial excretion (by culture) after 2 months of chemotherapy (OR = 4.98 [2.41-10.29]).

The objective of the study: to assess the prevalence of drug resistance of tuberculous mycobacteria (MTB) to pyrazinamide, linezolid, cross-resistance to fluoroquinolones, and cross-resistance to injectable anti-tuberculosis drugs among tuberculosis patients in Tomsk Region.Subjects and methods. The data of 814 patients with tuberculosis were analyzed. Of them, drug susceptibility test to pyrazinamide was performed in 812 patients; ofloxacin, levofloxacin, moxifloxacin – in 475 patients; kanamycin, amikacin, capreomycin – in 301 patients, and linezolid – in 423 patients.Results. The frequency of resistance to pyrazinamide is the highest in patients who were treated with pyrazinamide in the past (relapses and chronic cases). In the patients resistant to ofloxacin, the frequency of susceptibility to levofloxacin is low (16.9%), while susceptibility to moxifloxacin is higher (41.3%). The cross-resistance of MTB to kanamycin and amikacin makes less than 50% in the cases resistant to one of these two drugs. 38.6% of patients with poly-resistance to kanamycin and capreomycin, remain sensitive to amikacin. The level of MTB resistance to linezolid was minimal – 2.8% of those examined.

The objective of the study: to evaluate the possibility of using machine learning algorithms for prediction of relapses in multiple drug resistant tuberculosis (MDR TB) patients.Subjects and Methods. Сlinical, epidemiological, gender, sex, social, biomedical parameters and chemotherapy parameters were analyzed in 346 cured MDR TB patients. The tools of the scikit-learn library, Version 0.24.2 in the Google Colaboratory interactive cloud environment were used to build forecasting models.Results. Analysis of the characteristics of relapse prediction models in cured MDR TB patients using machine learning algorithms including decision tree, random forest, gradient boosting, and logistic regression using K-block stratified validation revealed high sensitivity (0.74 ± 0.167; 0.91 ± 0.17; 0.91 ± 0.14; 0.91 ± 0.16, respectively) and specificity (0.97 ± 0.03; 0.98 ± 0.02; 0.98 ± 0.02; 0.98 ± 0.02, respectively).Five main predictors of relapse in cured MDR-TB patients were identified: repeated courses of chemotherapy; length of history of tuberculosis; destructive process in the lungs; total duration of treatment less than 22 months; and use of less than five effective anti-TB drugs in the regimen of chemotherapy.

No abstract