Two-versus three-sample method for estimating gentamicin pharmacokinetic values.

The performances of two-sample and three-sample methods for estimating gentamicin pharmacokinetic values were studied. The medical records of patients who had received a gentamicin dosage consultation at a Veterans Affairs medical center between June 1989 and May 1991 were reviewed. For each patient, the pharmacokinetics service had prospectively used three serum gentamicin concentrations (SGCs) determined from three blood samples taken during an initial gentamicin regimen to estimate gentamicin pharmacokinetic values and determine the alternative would achieve the desired SGCs. In the two-sample method, the authors retrospectively used the initial peak and trough SGCs determined from two blood samples to estimate pharmacokinetic values and establish an alternative regimen. The two methods were evaluated by comparing their results with each other and with the actual pharmacokinetic values determined from SGCs measured during the subsequent gentamicin regimen. A total of 27 patients (all men; mean age, 62 years) were included in the study. The two-sample and three-sample methods differed significantly in their estimates of elimination rate constant and half-life but not clearance, volume of distribution, or the daily dose needed to achieve the intended SGCs. Neither method produced estimates that differed significantly from the actual pharmacokinetic values. The three-sample method was less biased than the two-sample method in the prediction of trough SGCs. In middle-aged and elderly men, a two-sample method and a three-sample method of estimating gentamicin pharmacokinetics differed significantly only in the prediction of trough SGCs. The difference was probably not clinically important.