Is Isoniazid Treatment For LTBI Safe For RA Patients With TNF Inhibitor Therapy?

Abstract

Background/Purpose: Isoniazid (INH) is increasingly used in RA patients with evidence of latent tuberculosis infection (LTBI) for whom TNF inhibitor therapy is planned. The potential for hepatotoxicity of INH has been concerned in RA patients treated with DMARDs including methotrexate (MTX). However, the safety and impact of INH for RA patients on persistency of TNF inhibitors has not been known. We aimed to study the safety of INH treatment for LTBI and its impact on the persistency of TNF inhibitors

Methods: Data were extracted from medical records of RA patients who had treated with TNF inhibitor at our university hospital from December 2000 to November 2011 (REtrospective study for Safety and Efficacy of Anti-RA treatment with biologiCs, RESEARCh). Comprehensive chart reviews were undertaken on all patients, and the data for each patient consists of two components: demographic and clinical characteristics (age, sex, disease duration, disease activity, comorbidity and laboratory finding) and information related with TNF inhibitors (type of drug, history of DMARDs before starting TNF inhibitor, concomitant medication andtreatment period).

A total 312 patients aged 18 years old or older were enrolled in this study, and among them 96 patients (30.9%) experienced INH treatment for LTBI. The occurrences of LFT abnormality were evaluated in both patients with and without INH treatment.Then Kaplan-Meier curve andCox proportional hazard analysis were used to evaluate its impact on persistency of TNF inhibitors

Results: There was no difference in demographic and clinical features in both groups except the frequency of MTX use (83.3% in INH treatment group and 72.2% in INH non-user group, P=0.049). The LFT abnormality was more commonly happened in patients treated with INH than those without INH treatment (20.8% vs. 8.8%, P=0.005). This increased risk for hepatotoxicity by INH treatment was persistent after adjusting covariates including MTX use in multivariate regression analysis (OR 3.18, 95% CI 1.48-6.84). However, the persistent rate of TNF inhibitors during 5 years using Kaplan-Meier curve was not different between two groups with and without INH treatment (49.4% vs. 54.6%, p=0.79 in log-rank test). INH treatment for LTBI was not associated with discontinuation of TNF inhibitors in Cox proportional hazard model (HR 1.02, 95%CI 0.66-1.58).

Conclusion: INH treatment for LTBI in RA patients who started TNF inhibitors is associated with the occurrence of LFT abnormality. However, it does not affect the persistency of TNF inhibitors.