Differences in EBA in the first two days of standard anti-tuberculosis treatment in different geographic regions.

Abstract

In a recent article entitled, “Comparison of time to positive and colony counting in an early bactericidal activity study of anti-tuberculosis treatment” we found a relatively lower early bactericidal activity (EBA) of combination anti-TB treatment with isoniazid (INH), rifampicin, ethambutol and pyrazinamide between days 0 and 2 in HIV-uninfected Ugandan adults with newly-diagnosed, initial episodes of smear-positive pulmonary TB.1 The mean EBA 0-2, measured in colony forming units (CFU), from 40 patients receiving directly observed therapy (DOT) was 0.26 ± 0.24 log10 CFU/ml/day. This result was similar to an EBA 0-2 value of 0.28 ± 0.42 log10 CFU/ml/day we calculated using data from an earlier 2003 clinical trial conducted at the same site.2 In comparison, previous studies done mainly in South Africa have demonstrated a higher EBA 0-2 for standard 4-drug therapy of approximately 0.5 log10 CFU/ml/day, while the EBA 0-14 was similar to that found in our study.3,4 The reason for the difference in EBA during the first 2 days of therapy is not clear. Patients in our study received DOT on an inpatient basis with Good Manufacturing Practice grade medications. Medications were administered as loose tablets, and all patients received 300 mg of INH. The mean dose of INH was 5.8 ± 0.8 mg/kg (range 3.7 to 7.9), which should yield maximum EBA 0-2.5 CFU counts were done on 16-hour pooled overnight sputum collections using standardized methods. The mean baseline sputum colony count of 6.57 ± 0.58 log10 CFU/ml was similar to previous EBA studies. Baseline sputum colony count has been associated with EBA 0-2 in previous studies,5,6 and we did find a correlation in this study (rs = 0.34, P = 0.03), however the relatively high mean baseline count in our patients would again predict maximal EBA 0-2. Finally, recent exposure to INH was excluded by urine INH metabolite testing (BBL Taxo INH Test Strips, BD, Sparks, MD, USA) during screening and at the time of enrollment to the ward to begin the study. All Mycobacterium tuberculosis (MTB) isolates from the study subjects were susceptible to INH, rifampicin, ethambutol and pyrazinamide, and sputum collection and processing methods in our study were similar to those used in the earlier studies.