Evaluation of risk factors associated with the development of MDR-and XDR-TB in a tertiary care hospital: A retrospective cohort study

Background: Drug resistant tuberculosis (DR-TB) infringes substantial burden in terms of longer treatment duration, morbidity and mortality. Timely identification of patients at risks of DR-TB will aid individualized treatment. Current study was aimed to ascertain several factors associated with DR-TB among patients attending a tertiary care hospital. Methods: This retrospective study was conducted among patients with confirmed diagnosis of DR-TB and drug susceptible TB (DS-TB) seeking medical care from a tertiary care hospital during 2014-2019. The types of DR-TB included were rifampicin resistant tuberculosis (RR-TB), Multidrug resistant tuberculosis (MDR-TB) and extensively drug resistant tuberculosis (XDR-TB). Appropriate statistical methods were implied to evaluate the factors associated with DR-TB. Results: Out of 580 patients, DS-TB was diagnosed in 198 (34.1%) patients while DR-TB was present in 382 patients. Of resistance cases, RR-TB, MDR-TB and XDR-TB were diagnosed in 176 (30.3%), 195 (33.6%) and 11 (1.9%) patients, respectively. Significant differences (P< 0.05) in demographics and clinicolaboratory characteristics were observed between patients with DS-TB and DR-TB. Logistic regression analysis revealed age ?38 years (OR: 2.5), single marital status (OR: 11.1), tobacco use (OR: 2.9), previous treatment (OR: 19.2), treatment failure (OR: 9.2) and cavity on chest X-ray (OR: 30.1) as independent risk factors for MDR-TB. However, XDR-TB was independently associated with age group of ?38 years (OR: 13.6), students (OR: 13.0), previous treatment (OR: 12.5), cavity on chest X-ray (OR: 59.6). The independent risk factors associated with RR-TB are age ?38 years (OR: 2.8), females (OR: 5.7), unemployed (OR: 41.5), treatment failure (OR: 4.9), previous treatment (OR: 38.2) and cavity on chest X-ray (OR: 4.3). ROC curve analysis accentuate the excellent predictive accuracy of all logistic regression models as shown by AUC (0.968, P< 0.001) for MDR-TB, AUC (0.941, P< 0.001) for XDR-TB and AUC (0.962, P< 0.001) for RR-TB. Conclusions: Current study demonstrates a sizeable extent of resistant cases among pulmonary TB patients. This study presaged significant risk of DR-TB among females, young adults, unemployed, smokers, patients with previous treatment failure and cavitation on chest X-ray. Timely identification of high risk patients will give pronounced advantages regarding appropriate choices of prevention, treatment and disease control.