A pharmacokinetic study of apalcillin was performed in 12 patients in an intensive-care unit. All received a single dose of 30 mg/kg in a 30 min infusion followed by 90 mg/kg per day for 4 days; 6 patients (group A) were treated by intermittent administration for 2 days (30 mg/kg in 30 min infusion, 3 times daily), followed by continuous infusion for the next 2 days; 6 other patients (group B) first received continuous infusion followed by intermittent infusion. Serial serum specimens were collected after the first infusion and on the third and fifth days of treatment. Bronchial secretions were taken simultaneously via an endotracheal tube or the tracheostomy cannula, in order to study the penetration of the drug into the respiratory tract. Assays were performed by a microbiological method. A mean serum peak value of 87.1 .+-. 6.13 mg/l 5 min after the end of the first injection was followed by a slow decrease in serum levels and a residual value of 6.29 .+-. 3.21 mg/l (8 h). Intermittent administration resulted in a mean serum peak of 79.56 .+-. 12.35 mg/l, whereas after continuous infusion, a steady state of about 30 mg/l was obtained. No significant difference was found between pharmacokinetic parameters for the 2 groups. In bronchial secretions, a mean peak value of 5.8 mg/l was attained by the 2nd h. Decreased levels in bronchial secretions measured after 3 or 5 days treatment were possibly related to a decrease in inflammation.