Orogastric and intravenous indomethacin administration to very premature neonates with patent ductus arteriosus: population pharmacokinetics, absolute bioavailability, and treatment outcome.

A population pharmacokinetic model was developed after administration of orogastric and/or intravenous indomethacin for the treatment of patent ductus arteriosus in preterm infants. Plasma indomethacin concentrations (n=227) were obtained from 90 preterm infants of median gestational age 27 weeks, mean postnatal age of 12 days, and a mean current weight (WT) of 1010 g. Infants received one to three courses of indomethacin (0.1 mg/kg per day for 6 days). A one-compartment model was fitted to the data to obtain estimates of clearance (CL), volume of distribution (V), absorption rate constant (Ka) and orogastric bioavailability (F), using NONMEM. Model robustness was assessed by bootstrapping with replacement (500 samples). The structural model was: CL (L/h)=0.0166 (WT / 0.936)1.54; V (L)=0.484 (WT / 0.936)1.41; F=0.986; Ka (h(-1))=0.786. The interindividual variability for CL and V was 57.7% and 45.6%, respectively. There remained considerable residual unexplained variability (45.4%). Mean (range) conditional estimates from individual infants for CL, V, and elimination half-life were 18.9 (4.7-45.5) mL/h/kg, 509 (191-1120) mL/kg, and 20.0 (12.0-37.3) hours, respectively. Complete ductus closure occurred in 67% of patients. Infants of lower gestational age or birth weight had less chance of successful ductal closure. There was no obvious dose-response relationship between systemic exposure to varying plasma indomethacin concentrations and ductus closure, which does not support individualized indomethacin dosing based on monitoring to a target plasma concentration.