The pharmacokinetics of cisplatin administered by continuous hyperthermic peritoneal perfusion (CHPP) was characterized in patients with peritoneal carcinomatosis. Cisplatin was added into the perfusate with escalating doses from 100 mg/m2 to 400 mg/m2. The hyperthermic perfusion was maintained for 90 minutes with a flow rate of 1.5 L/min and a target peritoneal temperature of 42.5 degrees C after a tumor debulking procedure. Samples of both the perfusate and blood were obtained during the perfusion and 30 minutes after the perfusion. Cisplatin plasma and perfusate concentrations were determined by flameless atomic absorption spectrometry with a lower limit of detection of 2 ng/ml and a coefficient of variation (CV) < 10%. Fifty-six patients were enrolled in the study. The mean (+/- SD) percentage of cisplatin present in the perfusate at the completion of perfusion was 27.8% +/- 20% of the total dose. The maximum cisplatin concentrations in the perfusate were 10 times higher than those in plasma. The area under the concentration-time curve (AUC) of the perfusate was 13 times higher than the AUC of plasma. A two-compartment model with an additional peritoneal cavity compartment fits to the data best based on the Akaike information criterion. However, the interpatient variability was considerably high (CV < 100%). In conclusion, cisplatin administered by hyperthermic peritoneal perfusion resulted in a pharmacological advantage by obtaining higher and direct drug exposure to the tumor in the peritoneal cavity while limiting systemic absorption and toxicity. Using a complex two-compartment model, the authors were able to characterize the pharmacokinetics of cisplatin given intraperitoneally via this technique.