The population pharmacokinetics of panipenem was studied in 23 neonates. Th
eir postconceptional age (PCA) was 24.7-42.6 weeks and their body weight wa
s 530-4455 g at initiation of therapy. Panipenem was infused over a period
of 60 min in a dose of 10.2-34.7 mg/kg bd in 21 patients, tid in one patien
t and four times daily in one patient for a mean of 10.7 days. Blood sample
s were obtained just before the infusion and 1-2 h after and again 6 h afte
r the infusion. All the data for the 108 serum panipenem concentrations wer
e evaluated with a non-linear mixed-effect model (NONMEM with first-order m
ethod), a computer program designed for population pharmacokinetic analysis
. One- and two-compartment population pharmacokinetic parameters were measu
red. The two-compartment parameters were as follows: panipenem clearance CL
= 0.150 L/h, central volume of distribution = 0.54 L, intercompartmental c
learance = 0.014 Uh and peripheral volume of distribution = 0.28 L. The one
-compartment parameters were CL = 0.175 Uh and volume of distribution = 0.5
5 L. In the fitting process using the one-compartment model, significantly
fixed effects related to CL were PCA, postnatal age (PNA), gestational age
(GA), body weight (BW) and serum creatinine, and that for the distribution
volume (V) was BW. CL showed a logarithmic rise with PCA (CL = 0.00176 x ex
p(0.14 x PCA)). The CL levels in the patients with PCA < 33 weeks (0.098 Uh
) were significantly lower (P < 0.001) than those with PCA, 33 weeks (0.25
Uh). The final formulae for the population pharmacokinetic parameters are a
s follows: CL = 0.0832 (PCA < 33 weeks), CL = 0.179 x BW(PCA <greater than
or equal to> 33 weeks), V = 0.53 x BW (coefficient of variation; 23.9% for
CL, 28.5% for V). Based on these data, a simulated time-concentration curve
was compared with that for adult data in a clinical Phase I study. Our fin
dings suggest that the panipenem dosage regimen of 10-20 mg/kg every 12 h s
hould yield concentrations within the accepted therapeutic range.