Previously the apparent permeability coefficient (P(app)) of ibuprofen was observed to vary depending on the perfusion medium employed. The present work explores the possible contributions to these differences. Studies were undertaken using an in situ single pass rat gut technique. Lumenal drug concentrations and plasma drug levels were assayed by HPLC. Absorption rate constants (k(0)) were determined from fractions of drug unabsorbed from the intestineat steady state. Plasma data were fitted to a two compartment open model with zero-order input. Significant differences in net fluid flux were observed between the various buffered perfusion media, with fluxes varying from -0.044+/-0.006 ml min(-1) to +0.057+/-0.013 ml min(-1), the lower and negative values occurring for lower pH media and the larger positive values tending to occur with media of higher pH. A linear relationship was found between the P(app) of ibuprofen and net water flux (y=1.13+11.3x; r(2)=0.80). Apparent zero-order rate constants for ibuprofen appearance in plasma correlated well with absorption rate constants estimated from steady state lumenal drug concentration [k(0(gut))]. From the linear relationship between P(app) and fluid flux a normalized P(app) for ibuprofen (i.e. the P(app) in the absence of net fluid flux) of 1.1 x 10(-4)cms(-1) was determined Net lumenal fluid flux is dependent on perfusion medium composition and significantly alters ibuprofen absorption. The differences observed for P(app) were reflected in systemic drug absorption concentrations. The findings of these studies underline the importance of standardizing the osmolarity of experimental media used for the determination of intestinal permeability data.