Azzam Ali, Jiyad Zainab, O'Beirne James
Department of Hepatology, Addenbrooke's Hospital, Cambridge, UK.
Griffith University, Brisbane, Queensland, Australia.
Australas J Dermatol. 2021 May;62(2):130-140. doi: 10.1111/ajd.13546. Epub 2021 Mar 26.
BACKGROUND/OBJECTIVE: Methotrexate (MTX) is widely used in various medical specialties. However, hepatotoxicity is an ongoing concern and this is thought to be directly associated with cumulative dose. We sought to synthesise the published literature to evaluate the association between methotrexate hepatotoxicity and cumulative dose.
A systematic review of Medline (PubMed) EMBASE, CINAHL and The Cochrane Library was performed. Full texts of articles were examined, and excluded articles were recorded with reasons for exclusion. A meta-analysis of correlation coefficients was performed using Fisher's z-transformation and a random effects model. Cochran's Q-test and the I statistic were calculated to assess heterogeneity.
A total of 35 studies met inclusion criteria. Measures of hepatotoxicity were highly varied and included liver biopsy, elastography, FibroTest, biochemical tests and scoring systems (Fib-4, APRI, AST:ALT). Some studies analysed for the association with MTX cumulative dose using more than one modality. Overall, 38 analyses found no significant association between MTX cumulative dose and hepatoxicity vs eight that identified a significant association. The pooled correlation coefficient from five studies which utilised elastography was 0.18 (95% CI, -0.09 to 0.42), with significant heterogeneity between studies (P < 0.0001), I = 92%).
Our synthesis of a large volume of studies in this review found no significant association between MTX cumulative dose and hepatotoxicity, both in terms of vote counting and with regard to the meta-analysis of correlation coefficients from studies that utilised elastography. This challenges the long-held belief that liver injury is a direct result of drug accumulation.
背景/目的：甲氨蝶呤（MTX）广泛应用于各个医学专科。然而，肝毒性一直是人们关注的问题，并且认为这与累积剂量直接相关。我们试图综合已发表的文献，以评估甲氨蝶呤肝毒性与累积剂量之间的关联。
对医学文献数据库（Medline，即PubMed）、EMBASE、护理学与健康领域数据库（CINAHL）以及考克兰图书馆进行了系统综述。检查了文章的全文，并记录被排除文章及其排除原因。使用费舍尔z变换和随机效应模型对相关系数进行了荟萃分析。计算了 Cochr an Q检验和I统计量以评估异质性。
共有35项研究符合纳入标准。肝毒性的测量方法差异很大，包括肝活检、弹性成像、纤维检测、生化检测和评分系统（Fib-4、APRI、AST:ALT）。一些研究使用不止一种方法分析与MTX累积剂量的关联。总体而言，38项分析发现MTX累积剂量与肝毒性之间无显著关联，而8项分析发现存在显著关联。五项使用弹性成像的研究的合并相关系数为0.18（95%CI，-0.09至0.42），研究之间存在显著异质性（P<0.0001，I=92%）。
我们在本综述中对大量研究的综合分析发现，无论是在计数表决方面，还是在对使用弹性成像的研究的相关系数进行荟萃分析方面，MTX累积剂量与肝毒性之间均无显著关联。这对长期以来认为肝损伤是药物积累直接结果的观点提出了挑战。
