A randomized-blinded evaluation of analytical error associated with the enzyme immunoassay (EMIT) of gentamicin.

Analytical error is a potential source of inaccuracy and imprecision in the pharmacokinetic dosing of the aminoglycosides. The present study was undertaken to determine the analytical error (precision and accuracy) associated with the EMIT assay for gentamicin as it is routinely used in a clinical pharmacokinetics laboratory, and to assess what impact this error would have on estimates of pharmacokinetic model parameters. Serum samples spiked with known amounts of gentamicin were assayed over a 6-week period along with routine clinical samples by a medical technologist who was blinded with regard to the expected concentration or number of replicates of each sample. Under these conditions, the assay method had a coefficient of variation (CV) of 4.8%. When this analytical error was incorporated into a first-order, one-compartment pharmacokinetic model, the CV was 0.8% for the y intercept and 1.2% for the slope. When all measured concentrations were compared with actual values, the slope and intercept were not significantly different from the line of reference. As indicated by these data, the EMIT assay for gentamicin as it is routinely used in a clinical laboratory is both precise and accurate, indicating that analytical error is not the major source of inaccuracy in pharmacokinetic dosing of gentamicin.