Biomedical Department, Mycobacteriology Unit, Institute of Tropical Medicine, Antwerp, Belgium.
Department of Clinical Sciences, Institute of Tropical Medicine, Antwerp, Belgium.
PLoS One. 2020 May 18;15(5):e0233500. doi: 10.1371/journal.pone.0233500. eCollection 2020.
Meta-analyses on impact of isoniazid-resistant tuberculosis informed the World Health Organization recommendation of a levofloxacin-strengthened rifampicin-based regimen. We estimated the effect of initial rifampicin resistance (Rr) and/or isoniazid resistance (Hr) on treatment failure or relapse. We also determined the frequency of missed initial and acquired Rr to estimate the impact of true Hr.
Retrospective analysis of 7291 treatment episodes with known initial isoniazid and rifampicin status obtained from individual patient databases maintained by the Damien Foundation Bangladesh over 20 years. Drug susceptibility test results were confirmed by the programme's designated supra-national tuberculosis laboratory. To detect missed Rr among isolates routinely classified as Hr, rpoB gene sequencing was done randomly and on a sample selected for suspected missed Rr.
Initial Hr caused a large recurrence excess after the 8-month regimen for new cases (rifampicin for two months), but had little impact on rifampicin-throughout regimens: (6 months, new cases; 3.8%; OR 0.8, 95%CI:0.3,2.8; 8 months, retreatment cases: 7.3%, OR 1.8; 95%CI:1.3,2.6). Rr was missed in 7.6% of randomly selected "Hr" strains. Acquired Rr was frequent among recurrences on rifampicin-throughout regimens, particularly after the retreatment regimen (31.9%). It was higher in mono-Hr (29.3%; aOR 3.5, 95%CI:1.5,8.5) and poly-Hr (53.3%; aOR 10.2, 95%CI 4.4,23.7) than in susceptible tuberculosis, but virtually absent after the 8-month new case regimen. Comparing Bangladesh (low Rr prevalence) with a high Rr prevalence setting,true Hr corrected for missed Rr caused only 2-3 treatment failures per 1000 TB cases (of whom 27% were retreatments) in both.
Our analysis reveals a non-negligible extent of misclassifying as isoniazid resistance of what is actually missed multidrug-resistant tuberculosis. Recommending for such cases a "strengthened" regimen containing a fluoroquinolone provokes a direct route to extensive resistance while offering little benefit against the minor role of true Hr tuberculosis in rifampicin-throughout first-line regimen.
异烟肼耐药结核病的荟萃分析为世界卫生组织推荐左氧氟沙星强化利福平方案提供了依据。我们估计初始利福平耐药（Rr）和/或异烟肼耐药（Hr）对治疗失败或复发的影响。我们还确定了初始和获得性 Rr 的漏检频率，以估计真正的 Hr 的影响。
对 20 年来 Damien 基金会孟加拉国通过个体患者数据库获得的 7291 例已知初始异烟肼和利福平状况的治疗情况进行回顾性分析。药物敏感性试验结果由该项目指定的超国家结核病实验室确认。为了在常规分类为 Hr 的分离物中检测到漏检的 Rr，随机进行 rpoB 基因测序，并对疑似漏检 Rr 的分离物进行选择。
初始 Hr 在新病例（两个月利福平）的 8 个月方案后导致复发过多，但对利福平全程方案影响不大：（6 个月，新病例；3.8%；OR0.8，95%CI：0.3，2.8；8 个月，复治病例：7.3%；OR1.8；95%CI：1.3，2.6）。在随机选择的“Hr”菌株中，有 7.6%漏检 Rr。利福平全程方案的复发病例中，获得性 Rr 很常见，特别是在复治方案后（31.9%）。在单-Hr（29.3%；aOR3.5，95%CI：1.5，8.5）和多-Hr（53.3%；aOR10.2，95%CI4.4，23.7）中比敏感结核病更高，但在 8 个月新病例方案后几乎不存在。将孟加拉国（低 Rr 流行率）与高 Rr 流行率环境进行比较，校正漏检 Rr 后的真正 Hr 仅导致每 1000 例结核病病例中有 2-3 例治疗失败（其中 27%为复治）。
我们的分析显示，将实际上漏检的耐多药结核病误诊为异烟肼耐药的程度不可忽视。推荐此类病例使用包含氟喹诺酮的“强化”方案会直接导致广泛耐药，而对利福平全程一线方案中真正的 Hr 结核病的次要作用几乎没有益处。
