Methotrexate (MTX) use in inflammatory bowel disease (IBD) is complicated by unpredictable efficacy and toxicity. In rheumatoid arthritis, total and individual red blood cell MTX polyglutamates (RBC MTXGlu1-5) correlate with disease activity and possibly toxicity, and therefore may be useful in guiding treatment. It is unclear if this applies in IBD. The aim of this pilot study was to measure total and individual RBC MTXGlu1-5 concentrations in patients with IBD to see if these relate to efficacy and adverse effects. RBC MTXGlu1--5 concentrations were measured on three occasions in 18 patients with IBD receiving MTX at a constant dose for 3 or more months. The intrapatient variability, expressed as the coefficient of variation, of RBC MTXGlu1-5 concentrations at steady state was determined, and disease activity and adverse effects were assessed against concentrations. The intrapatient coefficients of variation of individual MTXGlu1-5 varied from 12% to 27%. In Crohn's disease, higher RBC MTXGlu4&5 concentrations correlated with worse disease activity (r = 0.42 and 0.53, respectively, P </= 0.03). RBC MTXGlu5 concentrations were higher in patients experiencing at least one adverse effect as a result of MTX compared with those without adverse effects (mean, 26.7 versus 11.1 nmol/8 x 10 RBCs; mean difference, 15.6; 95% confidence interval, 3.5-27.7; P = 0.04). Similar trends were observed for RBC MTXGlu4. Furthermore, patients who specifically experienced gastrointestinal adverse effects had higher RBC MTXGlu4&5 concentrations than patients without (P = 0.03 and 0.04, respectively). In summary, RBC MTXGlu1-5 concentrations can be measured accurately with low intrapatient variation. Unexpectedly, RBC MTXGlu4&5 concentrations correlated inversely with efficacy in Crohn's disease. RBC MTXGlu4&5 concentrations were higher in patients experiencing adverse effects. These findings suggest that RBC MTXGlu1-5 concentration monitoring may be of value in IBD and could assist with planning of larger studies.