Pharmacokinetics of sustained serum methotrexate concentrations secondary to gastrointestinal obstruction.

A physiological pharmacokinetic model for methotrexate was refined and used to simulate serum methotrexate concentrations after high dose (5000 mg/m2) intravenous infusions with fixed normal values for all model parameters except the GI transit rate. There was good agreement between simulated and measured values when model simulations with the normal GI transit rate were compared to values measured following 109 doses administered to 27 patients with normal GI function. When model simulations were performed using GI transit rates representing 75, 50 and 10% of normal, there was a marked prolongation of the terminal serum methotrexate half-life, which was directly related to the reduction in the transit rate. When simulations were performed with GI transit reduced by 50%, the maximum amount of methotrexate in the GI lumen was 25% higher and occurred 4 hr later. Model simulations of serum methotrexate concentrations, using a GI transit rate reduced by 50%, were also in good agreement with serum concentrations measured in two patients with partial GI obstruction. These data establish a pharmacokinetic basis for previous clinical observations indicating sustained serum methotrexate concentrations in patients with GI obstructions and exemplify the utility of physiological pharmacokinetic models in assessing the potential effects of clinical variables on drug disposition.