One of the rare, but potentially serious, side effects of MTX is hepatotoxicity. Many studies have attempted to quantify this risk in RA patients on MTX and identify predisposing risk factors. However, this effort has been limited by (1) inconsistent monitoring of liver enzymes and screening for hepatitis B (HBV) and hepatitis C (HCV), (2) failure to account for the effects of NSAIDs, corticosteroids, and alcohol on the liver, and (3) the lack of routine pre-treatment biopsies to exclude pre-existing liver disease, including mild histologic abnormalities that have been described in RA patients in the absence of MTX.
In a case-controlled study by Walker et al, the cumulative risk at 5 years of developing clinically significant liver disease (defined as histologic evidence of cirrhosis, evidence of clinical liver disease, bilirubin >5, or abnormal liver scan or CT) was 1 in 1000. Duration of therapy and older age at time of initiation of MTX were associated with an increased risk of developing liver disease. However, previously proposed risk factors, such as alcohol ingestion, obesity, smoking and diabetes, as well as therapy with gold, hydroxychloroquine, cyclophosphamide, penicillamine or sulfasalazine were not associated with increased risk. In a meta-analysis of 15 studies regarding the progression of histopathologic changes of the liver in patients receiving MTX for psoriasis, psoriatic arthritis, RA and other arthropathies, Whiting-O'Keefe et al. concluded that the risk of liver toxicity in patients on long-term methotrexate therapy was significant, with an incidence of 5% of advanced pathology on liver biopsy. The risk of liver toxicity was associated with total cumulative dose of MTX and heavy alcohol consumption. However, as pointed out by Kremer et al, when this data was reanalyzed specifically looking at RA patients and the development of clinically significant fibrosis or cirrhosis, the frequency of liver disease was 1.2%, assuming that all patients had normal histology at baseline.
Because it has been suggested that exposure to hepatotoxic agents in the setting of hepatitis B infection may have a synergistic effect on hepatocellular injury when compared to each risk factor alone, DMARDs such as MTX are seldom used in patients with hepatitis B or C infection. Mok et al published the only study specifically designed to examine the safety of DMARDs in RA patients with chronic viral hepatitis. In this case-controlled study, changes in ALT levels, taken to reflect hepatotoxicity, were compared between two groups on continuous DMARD therapy for a minimum of 6 months. The first group consisted of patients with both RA and chronic viral hepatitis, while the second group consisted of patients with either chronic viral hepatitis or RA alone. Liver enzymes were recorded at baseline and at three-month intervals during DMARD therapy. Of the 29 patients with RA with chronic hepatitis, 23 were HBsAg positive and 6 were anti-HCV Ab positive. 16/29 (55.2%) of patients with RA and chronic hepatitis developed abnormal ALT levels after a mean of 1.9 years of DMARD use, compared to 13/62 (21%) of patients with RA alone and 128/623 (20.5%) of patients with chronic viral hepatitis. Subanalysis of the effects of individual DMARDs, used either alone or in combination, demonstrated that hydroxychloroquine and sulfasalazine were associated with abnormal ALT levels over 50% of the time in patients with RA and chronic viral hepatitis. In contrast, the two patients on MTX and the 3 patients on gold maintained normal ALT levels throughout the study period, although the mean cumulative dose of MTX was only 650mg. No definite conclusions regarding the safety of these drugs could be drawn given the small numbers in each group. However, these results suggest that chronic hepatitis and DMARD use may have a synergistic toxic effect on the liver, and that DMARDs such as hydroxychloroquine that are generally considered less hepatotoxic than MTX may not be completely safe in patients with chronic hepatitis.
Interestingly, several case reports have described the occurrence of acute liver decompensation in previously well HBV carriers after the withdrawal of low-dose MTX, suggesting that discontinuation of MTX in these patients could potentially be hazardous. The first case, reported by Flowers et al, described a 57 year old female with RA and HBV, normal liver enzymes, positive hepatitis B surface antigen (HBsAg), and active hepatitis on liver biopsy, who developed transient elevations in her serum transaminases on two different occasions when her MTX was stopped for bone marrow and pulmonary toxicity. The patient then developed fulminant hepatic failure (FHF) requiring orthotopic liver transplantation 41 days after discontinuation of MTX. Similarly, Narvaez et al described a case of a 67 year old male with RA and undiagnosed hepatitis B who was started on MTX, and remained well until MTX was discontinued for pulmonary interstitial fibrosis. Three weeks after discontinuation of MTX, he had a progressive increase in his liver enzymes and prolongation of the prothrombin time. Hepatitis B serologies at that time revealed the presence of HBsAg, IgM anti-HBc, serum HBV DNA and anti-HBe, consistent with disease reactivation. The patient died 6 months later, and the autopsy revealed severe chronic active hepatitis with extensive bridging necrosis.
Finally, Ito et al described a 75 year old female with RA and HBV (+HBsAg and anti-HBe) who did well on MTX 7.5mg weekly over three years until MTX was discontinued for mildly elevated transaminases. Within two weeks, the patient developed progressive hepatic dysfunction leading to death from fulminant hepatitis. Again, hepatitis serologies were consistent with reactivation of HBV. The mechanism behind the above described events can be explained by the fact that liver injury in hepatitis B (HBV) infection relates primarily to the elicited immune response, rather than to the cytopathic effects of the virus itself. Immunosuppressive medications like MTX allow for ongoing viral replication and widespread infection of hepatocytes. When therapy is stopped, the immune response is restored, allowing for a T cell-mediated immune response to HBV, destruction of infected hepatocytes, and potentially severe liver damage.
In 1994, the ACR published official guidelines on monitoring for hepatotoxicity in RA patients on MTX. Recommendations include obtaining baseline liver tests (AST, ALT, alkaline phosphatase, bilirubin, and albumin) and hepatitis B and C serologies prior to initiating therapy, and serial monitoring of AST, ALT and albumin levels at 4-8 week intervals. Pre-treatment liver biopsies were recommended only in patients with a prior history of excessive alcohol consumption, persistent hepatitis B or C antigenemia, or unexplained, persistently abnormal baseline liver tests. A liver biopsy should also be performed if a patient has elevated AST or ALT values in 5 of 9 screening labs within one year, or a below normal albumin level in the setting of well-controlled RA. The presence of hepatitis B or C infection was not listed as a direct contraindication to MTX therapy. Moreover, in a recent review regarding the use of potentially hepatotoxic medications in patients with underlying liver disease, only alcoholic liver disease and steatohepatitis were listed as conditions increasing the risk of hepatotoxicity with MTX.
In summary, there is extremely limited data on the safety of DMARDs in RA patients with concomitant chronic viral hepatitis, but the published data suggests that the combination of DMARDs and chronic hepatitis may have a synergistic hepatotoxic effect. Thus, DMARDs should be used judiciously in such patients, with preferential use of traditionally less hepatotoxic DMARDs, like sulfasalazine or gold, in patients whose disease mandates further therapy.
If MTX is to be used in patients with HBV or HCV infection, it should probably not be started without prior consultation with a gastroenterologist to obtain a baseline liver biopsy. Once MTX is started, liver tests should be strictly monitored every 4-8 weeks, and repeat liver biopsy done if liver tests become persistently abnormal while on therapy. If MTX is to be withdrawn, it should also be done with close monitoring, particularly in patients with HBV infection, since acute hepatic decompensation has been described following methotrexate withdrawal in this setting.
Anti-TNF agents may also represent a viable alternative to traditional DMARDs in this patient population, as there is increasing literature suggesting that biologic agents such as etanercept and infliximab may be safe in patients with chronic viral hepatitis, particularly hepatitis C (See Ask the Expert - "In patients with rheumatoid arthritis who have evidence of hepatitis B or C infection, can TNF-a antagonists or IL-1 antagonists be safely used?"). Unfortunately, the costs of such therapy are often prohibitive. However, patient assistance programs are in existence for all the anti-TNF agents, although they are limited primarily to patients living within the United States. Adalimumab is currently the only anti-TNF agent with a program for qualifying patients outside the United States, with assistance programs in Puerto Rico and Guam.
 Clegg DO, Furst DE, Tolman KG, Pogue R. Acute, reversible hepatic failure associated with methotrexate treatment of rheumatoid arthritis. J Rheumatol. 1989 Aug;16(8):1123-6.
 Walker AM, Funch D, Dreyer NA, Tolman KG, Kremer JM, Alarcon GS, et al. Determinants of serious liver disease among patients receiving low-dose methotrexate for rheumatoid arthritis. Arthritis Rheum. 1993 Mar;36(3):329-35.
 Whiting-O'Keefe QE, Fye KH, Sack KD. Methotrexate and histologic hepatic abnormalities: a meta-analysis. Am J Med. 1991 Jun;90(6):711-6.
 Kremer JM, Alarcon GS, Lightfoot RW Jr, Willkens RF, Furst DE, Williams HJ, Dent PB, Weinblatt ME. Methotrexate for rheumatoid arthritis. Suggested guidelines for monitoring liver toxicity. American College of Rheumatology. Arthritis Rheum. 1994 Mar;37(3):316-28.
 de la Monte SM, Hutchins GM, Moore GW. Risk factors for development of lethal sequelae after hepatitis B virus infection in humans. Am J Med. 1984 Sep;77(3):482-8.
 Mok MY, Ng WL, Yuen MF, Wong RW, Lau, CS. Safety of disease modifying anti-rheumatic agents in rheumatoid arthritis patients with chronic viral hepatitis. Clin Exp Rheumatol. 2000 May-Jun;18(3):363-8.
 Flowers MA, Heathcote J, Wanless IR, Sherman M, Reynolds WJ, Cameron RG, Levy GA et al. Fulminant Hepatitis as a Consequence of Reactivation of Hepatitis B Virus Infection after Discontinuation of Low-Dose Methotrexate Therapy. Annals of Internal Medicine 1990, 112 (5) 381-382.
 Narvaez J, Rodriguez-Moreno J, Martinez-Aguila M, Clavaguera MT. Severe Hepatitis Linked to B Virus Infection after Withdrawal of Low Dose Methotrexate Therapy. J Rheum 1998; 25(10): 2037-2038. Letter.
 Ito S, Nakazono K, Murasawa A, Mita Y, Hata K, Saito N, et al. Development of fulminant hepatitis B (precore variant mutant type) after the discontinuation of low-dose methotrexate therapy in a rheumatoid arthritis patient. Arthritis Rheum. 2001 Feb;44(2):339-42.
 Lewis, JH. The rational use of potentially hepatotoxic medications in patients with underlying liver disease. Expert Opin Drug Saf. 2002 Jul;1(2):159-72. Review.
 Wu, DD. "Ask the Expert" - "In patients with rheumatoid arthritis who have evidence of hepatitis B or C infection, can TNF-a antagonists or IL-1 antagonists be safely used?". Rheumatology.HSS.edu, 2003 November 6.