Clinical outcomes of patients with rifampicin resistant other than multi-drug resistant tuberculosis in Botswana; a 2006-2014 retrospective cohort analysis

Abstract

Background Rifampicin resistant tuberculosis (RR-TB) cases were historically managed using different drug-regimen depending on the resistance patterns. RR-TB other than MDR-TB were treated using modified regimen (first-line regimen plus fluoroquinolone +/-amikacin). World Health Organization (WHO) recommended that all cases of RR-TB should be treated using standardized Multi-drug-resistance TB (MDR-TB) regimen since rifampicin resistance (RR) is always accompanied by isoniazid resistance (INH). However recent evidence has shown otherwise and WHO stated that country-specific data should be examined to determine the relationship between rifampicin and isoniazid resistance. The recommendation to treat all cases of RR-TB as MDR-TB cases might not be relevant in our setting since the former practice have not been evaluated. Aim To evaluate the clinical treatment strategies amongst patients with rifampicin resistance other than MDR and their impact on treatment outcomes from 2006-2014 Objectives - To determine the prevalence of RR-TB with concomitant INH resistance among RR-TB cases Objectives - To determine the prevalence of RR-TB with concomitant INH resistance among RR-TB cases - To determine the clinical outcomes of RR-TB other than MDR-TB patients based on different treatment regimens - To determine the risk factors for unfavorable outcomes of patients with RR-TB other than MDR-TB Methodology A retrospective cohort study was carried out involving the review of data of all RR-TB cases as per microbiologic confirmation from 2006 to 2014. Patients with resistance to second-line drugs and children (<15years old) were excluded. A proportion of RR-TB with concomitant INH resistance was calculated. Treatment outcomes were categorized as favorable and unfavorable. The former if patients were cured or completed treatment and unfavorable if they had treatment failure, loss to follow-up or death. Multivariate logistic regression model was used to determine predictors of unfavorable outcomes. Results One thousand one hundred and thirty six (1 136) cases of RR-TB were recorded from 2006 to 2014. The proportion of cases of RR with concomitant INH resistance varied by years, ranging from 61% to 90% across the years, the average being 79%. Out of two hundred and sixteen RR-TB other than MDR-TB patients, 79.6% (172/216) had the treatment outcome records and were included in the analysis. Of those, 66.3% (114/172) patients were initiated on first-line regimen, 20.3% (35/172) on modified regimen and 13.4% (23/172) on standardized MDR-TB regimen. The mean length of treatment was 222 (+/- 93) days) for first line regimen, 447 (+/- 177) for modified regimen and 568 (+/- 219) for MDR-TB regimen. There was no statistically significant difference in unfavorable outcomes across the three treatment groups; first-line regimen, MDR-TB and modified regimen with 27% (31/114), 22% (5/23) and 17% (6/35), respectively, Pearson chi square, 1.6, P = 0.456. However, 8% (9/114) treatment failure and 10% (11/114) relapse were found only among those treated with the first-line regimen. The study did not find any statistically significant predictors for unfavorable outcomes. Conclusions Rifampicin resistance may be a reliable proxy for MDR-TB in a significant number of cases in Botswana due to a high proportion of RR-TB with concomitant INH resistance. Though the overall treatment outcome was similar among the three regimens used, because of the potential risk of treatment failure and relapse, modified regimen and MDR-TB regimen appear to be treatment of choice in our setting.