Abstract
BACKGROUND
Methotrexate (MTX) is one of the most frequently prescribed drugs in contemporary medicine with a well-recognised hepatotoxic potential, for which stringent laboratory and histological surveillance has long been advocated.
AIM
To estimate the population burden of end-stage methotrexate-related liver disease (MTX-LD) in the United States and identify independent host risk factors for this disease entity.
METHODS
We analysed the records of all individuals who had been listed for, and/or received, liver transplantation in the United States, as reported to the Organ Procurement and Transplantation Network between 1 October 1987 and 31 December 2011, and identified those whose liver disease was attributed, wholly or partly, to MTX therapy. We also compared the demographic and clinical characteristics of adult individuals with MTX-LD with those listed and/or transplanted for alcoholic liver disease (ALD, n = 43,285), non-alcoholic steatohepatitis (NASH, n = 7569) and primary sclerosing cholangitis (PSC, n = 8526) using the adjusted odds ratios (AORs) derived from multi-variable logistic regression models.
RESULTS
Of 158 904 adults who had been listed for, and/or received, liver transplantation during the study period, only 117 (0.07%) had MTX-LD. Compared with individuals with ALD and PSC, those with MTX-LD were more likely to be older (AORs per 5-year increase: 1.27, P < 0.001 and 1.33, P < 0.001 respectively); female (AORs: 1.78, P = 0.003 and 3.87, P < 0.001); Caucasian (AORs: 3.03, P = 0.001 and 2.05, P = 0.04); and diabetic (AORs: 2.76, P < 0.001 and 4.12, P < 0.001). With the exception of Caucasian ethnicity (AOR: 1.94, P = 0.05), the odds of these characteristics did not differ from individuals with NASH. The odds of elevated body mass index among MTX-LD individuals were higher than those with PSC (AOR per 5 kg/m(2) : 1.51, P < 0.001); similar to those with ALD (AOR per 5 kg/m(2) :1.15, P = 0.1); and lower than those with NASH (AOR per 5 kg/m(2) : 0.66, P < 0.001).
CONCLUSIONS
The United States population burden of end-stage methotrexate-related liver disease is likely to be exceedingly small, suggesting the need for reappraisal of current hepatotoxicity surveillance guidelines. The risk factor profile of methotrexate-related liver disease supports the notion that it may share a common pathogenesis with NASH.
Collapse