To better understand mannitol pharmacokinetics, the authors constructed and compared population models for high- versus low-dose bolus infusions in humans. Patients (aged 18-75, American Society of Anesthesiologists physical status 1-3) scheduled for elective craniotomy with an anticipated need for intraoperative mannitol were randomly assigned to receive either 0.5 (n = 10) or 1.0 (n = 12) g/kg of 20% mannitol over 15 minutes. Serial blood samples were collected at the predetermined intervals over 12 hours. Plasma mannitol concentrations were measured by gas chromatography and subjected to pharmacokinetic analysis; a 3-compartment model best described mannitol disposition characteristics. Weight and dose were the important covariates for rapid peripheral volume of distribution (V2) and central clearance (CL1), respectively. Estimated population means were 2.80, 8.86, and 12.0 L for central (V1), rapid (V2), and slow (V3) volumes of distribution, respectively. Clearances of the central compartments (CL1) were 0.07 versus 0.04 L/min in the high- versus low-dose group, respectively. Thus, mannitol kinetics can be considered as nonlinear. Clearances of the rapid peripheral (CL2) and slow peripheral compartments (CL3) were identical (2.07 and 0.16 L/min) in both. The current weight-based dosing guidelines yielded greater than expected plasma drug concentrations in obese patients.
- Nonlinear mixed effect modeling