Treatment Outcomes and Risk Factors for Poor Outcome among Patients with Multidrug-Resistant Tuberculosis at ALERT and University of Gondar Hospitals, Ethiopia

Abstract

: Multidrug-resistant tuberculosis (MDR-TB) presents an increasing threat to the global tuberculosis control. MDR-TB emerged as one of priority public health problem in Ethiopia. There has been a study reporting on predictors for death only. In this study, default and treatment failure, while arguably distinct from death, was included in our definition of poor outcome to allow for a more complete program evaluation. Therefore, we assessed the treatment outcomes and risk factors associated with poor outcome among MDR-TB patients at two national MDR-TB treatment centers, ALERT and University of Gondar hospitals, Ethiopia. Methods: Hospital based retrospective general cohort study was conducted at ALERT and University of Gondar hospitals, Ethiopia, from December 2010 to May 2014. We reviewed medical records of confirmed MDR-TB patients treated with a standardized regimen. The data was analyzed using SPSS version 20 computer software. To identify the risk factors related to poor treatment outcome(failure, default and death), bivariate comparison and multiple logistic regressions was performed .we used P < 0.2 in bivariate analysis to include variables in the original multivariate logistic regression model, and P< 0.05 considered as cut off point for presence of statistical significance. Results: Of 113 MDR-TB patients assessed, 6 (5.3%) had been diagnosed with primary MDR-TB, 107(94.7%) as secondary MDR-TB, and there had been no patients treated with second-line anti-TB drugs for this disease previously. Assessment of treatment outcomes showed that 68 (60.2%) patients were cured or completed therapy, 29 (25.6%) died, 15 (13.5%) defaulted, and treatment failed in 1 (0.9%). In a multivariate logistic regression model of these patients, independent risk factors for poor outcome included having baseline weight ≤ 45kg (adjusted odds ratio [AOR], 4.99;95% confidence interval [CI],1.270-19.582), positive smear at treatment initiation (AOR, 4.62; 95% CI, 1.406- 15.185), and HIV co-infection (AOR, 3.77; 95% CI, 1.145-12.436). Conclusion: Our study showed lower success rate in treating MDR-TB patients using a standardized regimen compared with WHO target. HIV co-infection, baseline weight ≤45kg, and positive smear at treatment initiation were shown to be independent risk factors for poor outcome. To decrease the poor outcome, ensuring adherence and paying special attention to this risky group of patients in addition to use of early diagnosis and initiation tuberculosis treatment is warranted.