Monitoring of serum gentamicin concentrations and one-compartment pharmacokinetic analysis were performed in 41 preterm low birth weight infants (20 with birth weight of less than 1,500 g and 21 with birth weight of greater than or equal to 1,500 g) in the first week of life. Our dosing regimens, which were 2.0 mg/kg every 24 hr for the less than 1,500 g group and 2.0 mg/kg every 12 hr for the greater than or equal to 1,500 g group, successfully achieved the desired peak (4-8 micrograms/ml; 87.8%) and trough (less than or equal to 3 micrograms/ml; 97.5%) concentrations on the 4th day of treatment. In a one-compartment pharmacokinetic analysis, a large intersubject variability of pharmacokinetic parameters were observed on the 1st day of treatment. When we compared the parameters of the 1st day with those of the 4th day, apparent decreases in Vd and TBC were observed. The mean values for TBC and T1/2 or Kd of the two birth weight groups were significantly different from each other on the 4th day of treatment, suggesting a less maturity of renal functions in the less than 1,500 g group. The modified method of Sawchuk and Zaske was proven impractical in predicting steady-state serum concentrations because of an underestimation probably caused by the dramatic alteration of Vd due to a diuresis soon after birth. Based on these results, we recommend the above-described dosing regimen and emphasize the importance of a close monitoring of serum gentamicin concentrations and toxicities, instead of the individualized dosing approach in low birth weight infants in the first week of life.