These investigators only counted elevations of transaminase enzymes of 3x as "significant". THey did not performe liver biopsies. When we performed our studies of patients on long-term MTX which were used in the Guidlines (36. Kremer JM, Alarcon GS, Lightfoot RW, Wilkens RF, Furst DE, Williams HJ, Weinblatt ME. Methotrexate for rheumatoid arthritis: suggested guidelines for monitoring liver toxicity. Arthritis and Rheumatism, 37:3:316-328, 1994) we used prospective transaminase enzymes monitored every 4 weeks for 11 years. We found that when baseline and annual liver biopsies were performed (avg of 11 biopsies per participant) that there were highly significant correlations between the number of any elevation of transaminase enzymes into the abnormal range and worsening hepatic pathology measured by light and EM. So the British study definition of an abnormal transaminase enzyme is misleading. The differences in methodology (definition of abnormality and correlation of transaminase elevations with annual liver histology grade) are enormous in the new study. I'm afraid that the blog has not performed a service by featuring this article. Joel Kremer