Article Text


Aminotransferase levels during treatment of rheumatoid arthritis with leflunomide in clinical practice
  1. A Hoi,
  2. G O Littlejohn
  1. Department of Rheumatology, Monash Medical Centre, 246 Clayton Road, Clayton Victoria 3168, Australia
  1. Correspondence to:
    Professor G O Littlejohn;

Statistics from

Leflunomide is a disease modifying drug with proven efficacy in the treatment of rheumatoid arthritis (RA).1–6 Clinical trials have indicated that low level, reversible, asymptomatic increases in serum transaminases may occur, although serious hepatotoxicity is uncommon.6–8 It remains unclear as to whether the characteristics of leflunomide transaminitis, reported in clinical trials, occur in a less vigorously monitored and clinically heterogeneous normal care RA population.


Between November 1999 and September 2001 we audited all non-hospital clinic patients with RA, privately managed with leflunomide by one physician (GOL), for the frequency and severity of transaminitis. These patients received usual care and were felt to be more likely to have other comorbidities, polypharmacy, and poorer review compliance than clinical trial patients.

Serum alanine aminotransferase (ALT) levels are reported, as they were the only transaminase determined by the each of the four regional laboratories which took part in monitoring the geographically widespread patients as part of their liver function test panel. One hundred consecutive patients were followed up prospectively for a mean of nine months. The mean number of previous disease modifying antirheumatic drugs (DMARDs) was 3.65, the patients’ mean age was 56 years and mean disease duration was 14 years. Sixty three per cent of patients received leflunomide added to methotrexate. Blood tests were usually performed monthly to three monthly. When ALT levels were significantly abnormal, management depended on which combination of DMARDs the patient was taking, the timing of the abnormality, the pre-leflunomide ALT characteristics, and the clinical response to leflunomide.

Ninety five per cent of increases occurred in the first 270 days of observation. Thirty per cent of patients had at least one episode of a raised ALT level above the individual laboratory upper limit of normal (ULN). Of these, 21/30 (79%) had levels less than twice the ULN, while 4/30 (13%) had levels more than three times the ULN. Six patients had abnormal ALT levels on more than three occasions. All patients with severe or frequent ALT abnormality were receiving a methotrexate/leflunomide combination, with one discontinuing leflunomide. Of the rest, all ALT levels returned to normal with observation or reduction in the dose of methotrexate (if used in combination) or leflunomide. No patient had symptoms of liver toxicity or required a washout procedure.

The mean level of serum ALT over the next three months did not vary significantly whether a leflunomide loading dose was used or not. The mean ALT level in 35 randomly selected patients receiving a methotrexate/leflunomide combination was 78.6% of the ULN compared with 51.2% when receiving methotrexate monotherapy in the two previous years (p<0.0001), although both were still within the normal range.


This group of patients had severe recalcitrant RA as shown by the long duration of the disease and the high number of previous DMARDs. In this study we report ALT levels, which are generally more sensitive to change than aspartate aminotransferase (AST), to show minor changes in these patients with RA treated with leflunomide in routine clinical practice similar to those seen under more rigorous clinical trial conditions.1–8 The combination of leflunomide and methotrexate led to all of the significant rises in ALT. However, the low level of the highest increases and the return to normal with observation or dose modification, particularly of methotrexate if used in combination, is reassuring to the clinician, who is used to minor fluctuations in ALT/AST in methotrexate monitoring programmes. Currently in patients with RA taking methotrexate, serum transaminase levels, particularly AST, are the best surrogate marker and predictor of clinically significant liver disease.9–11

Although this study reflects “real world” rather than rigid clinical trial patient characteristics and monitoring, it is short term and further observation over time is required to evaluate clinically significant liver disease in this setting.


View Abstract

Request permissions

If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.