Emit assays for five major anticonvulsant drugs. An evaluation of adaptations to two discrete kinetic analyzers

Because of the complex reaction kinetics observed with the enzyme multiplied immunoassay technic (EMIT), each adaptation of these assays must be specifically evaluated. Evaluations were done of the totally automated adaptations of these immunoassays for diphenylhydantoin, phenobarbital, primidone, carbamazepine, and ethosuximide to two kinetic analyzers that use markedly different reaction conditions, the Gilford System 3500 and the Abbott Bichromatic Analyzer-100. Comparisons of data from EMIT assays on the Gilford 3500 with data from EMIT assays on the ABA-100 and with data from gas-liquid chromatographic analyses for the five anticonvulsant drugs showed good correlations (r greater than or equal to 0.96) and little or no additive or proportional biases, as determined by joint confidence intervals. Between-run precision for each EMIT anticonvulsant drug assay on both analyzers was less than 6%, better than that obtained with gas chromatography. No interference by more than 40 different drugs, by hemoglobin (6 g/l), or by bilirubin (200 mg/l), and no or minimal cross-reactivity with any of the five EMIT anticonvulsant drug assays on either analyzer were found.