Ethics of drug studies in infants: how many samples are required for accurate estimation of pharmacokinetic parameters in neonates?

Our study aimed to determine the least number of samples that are required to obtain accurate pharmacokinetic parameters in neonates. Patients treated with either netilmicin or ceftazidime had between five and eight samples drawn for drug concentration measurement after the first or the last dose of the drug. Pharmacokinetic parameters were calculated using all the available points as a reference and then recalculated with 2, 3, or 4 points. Systemic clearance and volume of distribution were significantly different from the reference value when 2 points were used for netilmicin after the first dose and ceftazidime after the last dose. Had parameters obtained from 2 points been used, mean dosage would have been underestimated by 15% for ceftazidime and 11% for netilmicin, and some patients would have received only 65% of the dose calculated from all available points. When 3 points were used, dosage would have been underestimated by a mean of only 1% for ceftazidime and 5% for netilmicin when compared with the dosage estimated from the reference parameters. We conclude that 3 concentration time points may be all that are required for estimation of pharmacokinetic parameters sufficiently accurate for practical purposes in neonates.