The pharmacokinetics of trimethoprim-sulfamethoxazole were studied in 12 healthy adult subjects receiving trimethoprim at 20 mg/kg of body weight per day and sulfamethoxazole at 100 mg/kg/day, which is the conventional dose for treating Pneumocystis carinii pneumonia (PCP). Daily doses were evenly divided and orally administered every 6 h for 3 days. Trimethoprim, sulfamethoxazole, and N4-acetylsulfamethoxazole concentrations in serum and urine were measured by high-performance liquid chromatography. Five subjects withdrew from the study because of intolerable gastrointestinal and central nervous system toxicities. In the seven subjects that completed the study, the mean maximum serum drug concentrations after the last dose were 13.6 +/- 2.0, 372 +/- 64, and 50.1 +/- 10.9 micrograms/ml for trimethoprim, sulfamethoxazole, and N4-acetylsulfamethoxazole, respectively. The mean half-lives were 13.6 +/- 3.5, 14.0 +/- 2.3, and 18.6 +/- 4.3 h, respectively. Changes in absolute neutrophil count were significantly correlated with the minimum concentrations of trimethoprim and sulfamethoxazole in serum and trimethoprim area under the concentration-time curve (for all three parameters, r2 = 0.6 and P less than 0.05). Our findings add to the evidence that serum drug concentrations in adults following the conventional dose of trimethoprim-sulfamethoxazole for PCP are excessive and contribute to certain adverse reactions. Further studies are indicated in patients to optimize the dosing regimen of trimethoprim-sulfamethoxazole in the treatment of PCP.