A gentamicin pharmacokinetic population model and once-daily dosing algorithm for neonates

To develop a gentamicin pharmacokinetic population model and once-daily dosing algorithm for neonates younger than 10 days.

Prospective, open-label study.

Neonatal intensive care unit.

One hundred thirty-nine neonates prescribed gentamicin.

Gentamicin peak and trough serum concentrations were collected from 139 neonates divided into three groups who were receiving one of the following intravenous 24-hour gentamicin regimens during the first 10 days of life, based on gestational age and birth weight (group 1, < 28 wks, 2.5 mg/kg; group 2, 28-34 wks, 3 mg/kg; and group 3, > 34 wks, 4 mg/kg). A structural model was developed in ADAPT II software using a MAP Bayesian approach. Final population parameter estimates were calculated using iterative two-stage analysis. The median (range) gestational age and birth weight, respectively, were 32 weeks (23-42 wks) and 1.92 kg (0.47-5.00 kg). The final one-compartmental linear model had a median (range) gentamicin total clearance, half-life, and volume of distribution of 0.0709 L/hour (0.0151-0.246 L/hr), 8.59 hours (4.88-16.9 hrs), and 0.262 L (0.0903-0.929 L), respectively. Total clearance increased as gestational age increased (p<0.001). Group 1 (10.2 hrs) had a significantly longer half-life than either group 2 (8.89 hrs, p<0.01) or group 3 (6.98 hrs, p<0.01). Total clearance was associated with gestational age and birth weight: clearance (L/hr) = (0.00504 + [0.00108 x gestational age]) x birth weight (coefficient of determination [r2] = 0.897), and volume of distribution was associated with birth weight (r2 = 0.700). The following dosing algorithm was designed to reach a therapeutic 24-hour area under the curve (87.5 mg/L x hr) in neonates during the first 10 days after birth: 24-hour gentamicin dose (mg) = (0.441 + [0.0945 x gestational age]) x birth weight.

This dosing algorithm provides a new approach for determining initial gentamicin dosing regimens in neonates; however, clinical validation is required.