The plasma pharmacokinetics of ranitidine (50 mg i.v.) have been studied in 17 critically ill patients in an intensive care unit. Measurements of gastric aspirate pH were also made in 16 of these patients. Ranitidine therapy was part of the patients' normal drug regimen. Ranitidine plasma concentration was measured by high performance liquid chromatography and appropriate polyexponential equations were fitted to concentration-time data to enable calculation of relevant pharmacokinetic parameters. Values of the volume of the initial dilution space (median = 89 ml kg-1) and volume of distribution at steady state (median = 1.54 l kg-1) were about 60% of corresponding mean literature values for healthy controls. Plasma clearance (median = 4.22 ml min-1 kg-1) and terminal half-life (median = 4.7 h) were about 2-3 fold less and 2-3 fold greater, respectively, than values for healthy controls. There was wide interpatient variation in all the pharmacokinetic parameters. Renal impairment was considered to be largely responsible for the low plasma clearance. Gastric aspirate pH was measured at 0, 1 and 7 h after ranitidine administration and 58% of samples were found to be above pH 4. Four patients had gastric pH values which were consistently below pH 4 despite average trough plasma ranitidine concentrations equal to or greater than those required for a 50% suppression of gastric acid secretion in normal volunteers.