Three-point versus two-point method for early individualization of aminoglycoside doses.

An approach using two serum concentrations following the initial dose of an aminoglycoside was retrospectively evaluated using the first and third samples from all three-point studies conducted by the pharmacokinetic consult service in 1988. The same Sawchuk and Zaske method was used for the three-point and two-point studies. Although the predicted peaks and troughs from the two-point study were statistically different from those of the three-point study, it is the authors' opinion that the root mean squared error of 0.38 mg/L (95% confidence interval [CI] 0.31-0.45) for the peaks and 0.09 mg/L (95% CI 0.06-0.11) for troughs is well within the limits of acceptable clinical error for gentamicin and tobramycin. Eliminating one-third of initial aminoglycoside concentrations could result in significant time and cost savings.