Methotrexate-associated hepatotoxicity: retrospective analysis of 210 patients with rheumatoid arthritis

Abstract

Beginning in the 1980s, methotrexate has been used successfully to treat rheumatoid arthritis. The magnitude and severity of short- and long-term methotrexate toxicity, however, have not been adequately investigated. Our study was performed to determine the prevalence of hepatotoxicity in patients with rheumatoid arthritis receiving long-term methotrexate therapy. We conducted a retrospective, computer-assisted review of all Duke University Medical Center patients undergoing liver biopsy for methotrexate monitoring from January 1979 to January 1988. A total of 538 biopsies were performed in 399 patients, 259 of whom had inflammatory arthritis (210 with rheumatoid arthritis, 47 with psoriatic arthritis, and two with seronegative spondyloarthropathy). No evidence of cirrhosis was defined in the cohort with rheumatoid arthritis; however, six patients with rheumatoid arthritis had histologic changes of fibrotic liver disease (prevalence of 2.9 percent in the group with rheumatoid arthritis) while taking methotrexate. Five of the six patients were obese and three had glucose intolerance or overt diabetes mellitus, and one person admitted to alcohol usage. Only one patient with fibrotic liver disease had elevated liver function test results, and no person showed a declining serum albumin level at the time of biopsy. Sixty-one patients with rheumatoid arthritis underwent multiple samplings (44 with two, 13 with three, and four with four biopsies). Fourteen of these patients showed progressive hepatic disease, whereas four patients improved. Although the prevalence of methotrexate hepatotoxicity in this large cohort of patients with rheumatoid arthritis was low, a small but definite risk of hepatic fibrosis, not predictable by laboratory screening, still exists.