Population pharmacokinetics of midazolam administered by target controlled infusion for sedation following coronary artery bypass grafting.

BACKGROUND

Midazolam is commonly used for short-term postoperative sedation of patients undergoing cardiac surgery. The purpose of this multicenter study was to characterize the pharmacokinetics and intersubject variability of midazolam in patients undergoing coronary artery bypass grafting.

METHODS

With institutional review board approval, 90 consenting patients undergoing coronary artery bypass grafting were enrolled at three study centers. All subjects received sufentanil and midazolam via target-controlled infusions. After operation, midazolam was titrated to maintain deep sedation for at least 2 h. It was then titrated downward to decrease sedation for a minimum of 4 h more and was discontinued before tracheal extubation. Arterial blood samples were taken throughout the study and were assayed for midazolam and 1-hydroxymidazolam. Midazolam population pharmacokinetic parameters were estimated using NONMEM. Cross-validation was used to estimate the performance of the model.

RESULTS

The pharmacokinetics of midazolam were best described by a simple three-compartment mammillary model. Typical pharmacokinetic parameters were V1 = 32.2 l, V2 = 53 l, V3 = 245 l, Cl1 = 0.43 l/min, Cl2 = 0.56 l/min, and Cl3 = 0.39 l/min. The calculated elimination half-life was 15 h. The median absolute prediction error was 25%, with a bias of 1.4%. The performance in the cross-validation was similar. Midazolam metabolites were clinically insignificant in all patients.

CONCLUSIONS

The intersubject variability and predictability of the three-compartment pharmacokinetic model are similar to those of other intravenous anesthetic drugs. This multicenter study did not confirm previous studies of exceptionally large variability of midazolam pharmacokinetics when used for sedation in intensive care settings.