Guidelines for blood test monitoring of methotrexate toxicity in juvenile idiopathic arthritis

OBJECTIVE: To assess the utility of the American College of Rheumatology guidelines for monitoring methotrexate (MTX)-related toxicity in a cohort of children with juvenile idiopathic arthritis (JIA). METHODS: Eighty-nine patients with JIA treated with MTX were monitored prospectively: aspartate aminotransferase (AST), alanine aminotransferase (ALT), complete blood count (CBC), and differential blood count were measured prior to starting MTX, and then monthly. Significantly abnormal blood tests (SABT) were prospectively defined as (1) significantly elevated liver enzymes (SELE) greater than twice the upper limit of normal; (2) granulocyte count < 1.5 109/l; (3) lymphocyte count < 0.9 109/l; or (4) hemoglobin decreased by > 2 g/l from previous level. Clinical interventions, current and cumulative MTX dose, duration of treatment, comorbidity, and concurrent medications at the time of the first SABT identification were recorded. Independent t tests and chi-squared tests were used for comparisons, and the probability of developing a SABT was calculated by Kaplan-Meier survival analysis. RESULTS: Forty percent of patients had a SABT: 26% had hematological abnormalities and 14% had SELE. Ninety-five percent of patients with SABT had symptoms consistent with a viral infection when the SABT was drawn and MTX dose was withheld until results had normalized on repeat testing. SABT persisting beyond one month occurred in only 2 patients, and their abnormalities resolved by 6 months with no specific identified cause; they resumed MTX at a later time without recurrence of SABT. There were no differences between patients with and without SABT with respect to current or cumulative MTX dose, duration of treatment, and concurrent medications at the time of the SABT. The probability of developing a SABT was estimated to be 11% at 3 months, compared to 10% probability of having an abnormal blood test by chance alone. CONCLUSION: Routine blood tests every 4 to 8 weeks in children with JIA are unnecessarily frequent.