OBJECTIVE Rheumatoid arthritis is associated with an increased cardiovascular risk, secondary to endothelial dysfunction. There is accumulating evidence that methotrexate reduces cardiovascular risk in rheumatoid arthritis, but the mechanisms involved are still unknown. In this study, we aimed to determine the effect of methotrexate on endothelial function and traditional cardiovascular risk factors in the adjuvant-induced arthritis (AIA) rat model. METHODS On the first signs of arthritis, methotrexate (1 mg/kg/week, s.c.) or saline (Vehicle) was administered to AIA for 3 weeks. Endothelial function was studied in aortic rings relaxed with acetylcholine in the presence or not of inhibitors of nitric oxide synthase, cyclooxygenase-2, arginase, EDHF and superoxide anions production. Arthritis and radiological scores, blood pressure and blood levels of cytokines, triglycerides, cholesterol, homocysteine and BMP-4 were measured. RESULTS Although methotrexate significantly reduced the arthritis score, it had no effect on Ach-induced relaxation. As regards mechanisms, methotrexate increased nitric oxide synthase activity and reduced the superoxide anions production but did not change arginase, cyclooxygenase-2 and EDHF pathways. Methotrexate did not change the radiological score or blood pressure, lipid, glucose and homocysteine levels. By contrast, methotrexate significantly reduced plasma IL-1β and TNF-α levels and increased serum BMP-4 level. CONCLUSIONS Despite a reduction of clinical and biological inflammation, methotrexate did not improve endothelial function in AIA rats. Overall data suggest that mechanisms other than the ED reduction are likely involved, and remain to be elucidated to better understand the cardiovascular benefits of methotrexate in rheumatoid arthritis.