In animal models, the antiinflammatory mechanism of action of methotrexate
has been attributed to elevation of the extracellular concentration of the
endogenous autocoid, adenosine. Our goal was to determine if methotrexate e
levates adenosine concentrations in plasma and at the site of disease in pa
tients with inflammatory bowel disease. In 10 patients with Crohn's disease
or ulcerative colitis, rectal adenosine and plasma adenosine concentration
s were measured before and immediately after a subcutaneous injection of me
thotrexate, 15 or 25 mg. The mean predose rectal adenosine concentration of
2.4 mu mol/l was not significantly different from the postdose concentrati
on of 2.1 mu mol/l, p = 0.17, (paired two-tailed t test). Rectal adenosine
concentration tended to correlate with rectal endoscopic disease activity,
r = 0.59, p = 0.067 (Spearman rank order correlation). After methotrexate i
njection, neither the mean daily plasma adenosine concentration, nor the pl
asma adenosine at any individual time point, were significantly different f
rom preinjection values. In patients with inflammatory bowel disease, an in
jection of methotrexate in the clinically effective dose range does not rai
se rectal or plasma adenosine concentrations. A role for adenosine as the m
ediator of the antiinflammatory action of methotrexate is not supported.