Methotrexate and Liver Fibrosis in People With Psoriasis: A Systematic Review of Observational Studies

C.M. Maybury; Z.K. Jabbar-Lopez; T. Wong; A.P. Dhillon; J.N. Barker; C.H. Smith


The British Journal of Dermatology. 2014;171(1):17-29. 

In This Article


There was a 22% increased risk of 'any fibrosis' on biopsy following use of MTX. There was a trend towards MTX use and progression to significant fibrosis being associated, but this was not statistically significant. Of note, cumulative dose and duration of MTX therapy were not associated with fibrosis/cirrhosis on biopsy (as independent variables). From these data we can infer that MTX use did increase the risk of developing fibrosis in the study cohort; however, we must qualify this: MTX increases the risk of fibrosis but not in everyone and/or not on its own. The lack of a clear dose-dependent relationship indicates that other factors may influence the development of pathology.

Studies included in our meta-analysis were old and of low quality. Selection bias was universal. Detection bias was another major weakness: the variety of histological scoring systems employed introduced systematic differences in the primary outcome, i.e. whether a liver biopsy was reported as normal or fibrotic. An agreement on the most appropriate system for future reporting of biopsies that is validated in this cohort would strengthen future research.

We were unable to assess the impact of psoriasis disease severity, alcohol, obesity and diabetes because of underreporting and/or inconsistent reporting. We were also unable to evaluate the potentially preventative impact of folic acid supplementation. When considering drug-induced liver injury, it is recognized that a small proportion of patients may develop a 'hypersensitivity' reaction. One study[25] did report that three patients who had previously received MTX and who had significant fibrosis were withdrawn from the drug and their biopsies improved, but when MTX was reinstituted, an 'accelerated progression to significant fibrosis' was reported.

This is the first systematic review to focus solely on the risk attributable to MTX in the development of fibrosis within the population with psoriasis. By including only studies with at least two biopsies, and presenting only pooled data from pre/postbiopsy studies, we emphasized the importance of temporality when considering causation (the requirement that exposure must precede effect). This paper is also the first to evaluate systematically the quality of studies and role of potential confounders.

We acknowledge that this systematic review was limited by consideration of only published work and only English-language papers. Restricting our analysis to studies with at least two biopsies did inevitably result in the exclusion of literature reporting on histology findings from 'single' or nonsequential biopsy findings, although for the reasons above, we felt it was justified.

Whiting-O'Keefe et al.[19] evaluated a mixed population of patients with rheumatoid arthritis and psoriasis. They reported that 28% of patients demonstrated disease progression associated with MTX use. Patients with psoriasis were more likely to progress and have advanced fibrosis/cirrhosis than patients with rheumatoid arthritis. These findings are consistent with our results. However, in contrast they demonstrated that alcohol use and average cumulative dose did influence progression of disease, although data were taken from 'incomplete alcohol consumption' histories.

Montaudie et al.[20] performed the only other systematic review to address liver disease in the population with psoriasis. They reported that due to variability the risk of fibrosis associated with MTX use could not be quantified and that obesity and type 2 diabetes do influence the development of fibrosis whereas alcohol does not. We would interpret these results with caution. Their evaluation of comorbidities considered only small numbers of patients, and one of the studies (Rosenberg et al.)[21] presented data that did not allow for the assessment of temporality: the ability to compare the results of two biopsies over time and assess the risk produced by various exposures (MTX, hyperglycaemia, alcohol) on the development of disease.

In at least a subset of people with psoriasis, MTX appears to increase the risk of developing fibrosis. What we do not know, and what this meta-analysis failed to answer, is which patients make up this subset. Are they the heavy drinkers? Are these the people with metabolic syndrome? Or are genetic factors in the metabolism of MTX significant in determining who develops drug toxicity?

According to the World Health Organization ( worldwide obesity has nearly doubled since 1980, with approximately 35% of the world's adult population now overweight or obese. If these studies were performed today, the number of obese study participants might be far greater. Therefore, it is more important than ever before that we determine the risk associated with being obese (in the context of having psoriasis and taking MTX) in the development of fibrosis.

Another important point is that although fibrosis/cirrhosis was reported on biopsies, there was very little data on clinical outcomes, such as liver failure. It is hoped that through the widespread participation of many dermatologists reporting adverse events (including incident liver pathology) to the British Association of Dermatologists Biologic Interventions Register (, which is collecting data on MTX, the long-term side-effects of drugs used to treat chronic conditions such as psoriasis may be better evaluated.

Until doctors are presented with persuasive evidence that exposure to an outcome (be it alcohol or MTX) causes disease, they will not change their practice. Causality needs to be proven before preventative medicine can begin. This meta-analysis does not provide persuasive enough evidence. The studies included do not represent the general population with psoriasis and key confounders such as obesity and alcohol use were not systematically reported. However, the data we report do show an association between MTX use and disease progression according to liver biopsy data. This warrants further evaluation in a prospectively recruited, larger group of patients, with consideration given to the role of psoriasis disease severity as well as alcohol and comorbid disease.