We present a specific method for the determination of disodium clodronate in human plasma and urine using a gas-chromatographic system with nitrogen phosphorus detector (NPD). The compound was extracted from plasma and urine samples by an anion-exchange resin and derivatizated with bistrimethylsilyltrifluoroacetamide (BSTFA). Sodium bromobisphosphonate was used as internal standard. The calibration curves were linear in both plasma and urine, with a regression coefficient r > 0.9975 in plasma and r > 0.9977 in urine. The limit of quantitation was 0.3 microg/ml in plasma and 0.5 microg/ml in urine. The method was validated by intra-day assays at three concentration levels. During the study we carried out inter-day assays to confirm the feasibility of the method. The precision in plasma at 0.5, 15, and 45 microg/ml was 12.4, 0.2, and 6.5% (n = 40), respectively; in urine at 0.8, 8, and 40 microg/ml it was 8.6, 6.4, and 9.3% (n = 40), respectively. The method was accurate and reproducible, and was successfully applied to determine the pharmacokinetic parameters of clodronate in healthy volunteers after intravenous infusion and intramuscular injection of 200 mg of the compound. The Cmax after intravenous infusion and intramuscular injection was 16.1 and 12.8 microg/ml, respectively. AUC(0-48 h) after infusion administration and intramuscular injection was 44.2 +/- 18.0 and 47.5 +/- 12.4 h microg/ml, respectively. The elimination half-life in both administrations was 6.31 +/- 2.7 h.