Clinical trials suggest less hepatotoxicity and better adherence with 4 months rifampin (4R) versus 9 months isoniazid (9H) for treating latent tuberculosis infection (LTBI). Our objectives were to compare frequencies of severe hepatic adverse events and treatment completion, and direct health system costs of LTBI regimens 4R and 9H, in the general population of the province of Quebec, Canada, using provincial health administrative data. Our retrospective cohort included all patients starting rifampin or isoniazid regimens between 2003 and 2007. We estimated hepatotoxicity from hospitalisation records, treatment completion from community pharmacy records and direct costs from billing records and fee schedules. We compared rifampin to isoniazid using logistic (hepatotoxicity), log-binomial (completion), and gamma (costs) regression, with adjustment for age, co-morbidities and other confounders. 10 559 individuals started LTBI treatment (9684 isoniazid; 875 rifampin). Rifampin patients were older with more baseline co-morbidities. Severe hepatotoxicity risk was higher with isoniazid (n=15) than rifampin (n=1), adjusted OR=2.3 (95% CI: 0.3–16.1); there were two liver transplants and one death with isoniazid and none with rifampin. Overall, patients without co-morbidities had lower hepatotoxicity risk (0.1% versus 1.0%). 4R completion (53.5%) was higher than 9H (36.9%), adjusted RR=1.5 (95% CI: 1.3–1.7). Mean costs per patient were lower for rifampin than isoniazid: adjusted cost ratio=0.7 (95% CI: 0.5–0.9). Risk of severe hepatotoxicity and direct costs were lower, and completion was higher, for 4R than 9H, after adjustment for age and co-morbidities. Severe hepatotoxicity resulted in death or liver transplant in three patients receiving 9H, compared with no patients receiving 4R. The use of 4 months rifampin (4R) instead of 9 months of isoniazid (9H) for treatment of latent tuberculosis infection is supported by our findings of lower risk of severe hepatotoxicity, better completion and lower adjusted direct costs with 4R http://bit.ly/3agWLh9

中文翻译：

---

异烟肼或利福平治疗潜伏性结核感染：基于人群的肝毒性、完成和成本研究

临床试验表明，与 9 个月的异烟肼 (9H) 相比，4 个月的利福平 (4R) 治疗潜伏性结核感染 (LTBI) 的肝毒性更低，依从性更好。我们的目标是使用省级卫生行政数据，在加拿大魁北克省的一般人群中比较严重肝脏不良事件和治疗完成的频率，以及 LTBI 方案 4R 和 9H 的直接卫生系统成本。我们的回顾性队列包括 2003 年至 2007 年间开始使用利福平或异烟肼方案的所有患者。我们根据住院记录、社区药房记录的治疗完成情况以及账单记录和收费表中的直接成本估计了肝毒性。我们使用逻辑（肝毒性）、对数二项式（完成）和伽马（成本）回归比较了利福平和异烟肼，调整年龄、合并症和其他混杂因素。10 559 人开始 LTBI 治疗（9684 异烟肼；875 利福平）。利福平患者年龄较大，基线合并症更多。异烟肼 (n=15) 的严重肝毒性风险高于利福平 (n=1)，调整后的 OR=2.3（95% CI：0.3–16.1）；使用异烟肼进行了两次肝移植手术，其中 1 人死亡，而使用利福平则没有。总体而言，没有合并症的患者的肝毒性风险较低（0.1% 对 1.0%）。4R 完成 (53.5%) 高于 9H (36.9%)，调整后的 RR=1.5（95% CI：1.3-1.7）。利福平每位患者的平均成本低于异烟肼：调整后的成本比 = 0.7（95% CI：0.5-0.9）。在调整年龄和合并症后，4R 的严重肝毒性风险和直接成本较低，完成率高于 9H。3 名接受 9H 的患者出现严重的肝毒性导致死亡或肝移植，而接受 4R 的患者则没有。使用 4 个月的利福平 (4R) 代替 9 个月的异烟肼 (9H) 治疗潜伏性结核病感染得到了我们的研究结果的支持，即严重肝毒性的风险较低、完成率更高且调整后的直接成本较低，4R http://bit .ly/3agWLh9