Detection of benzodiazepine intake in therapeutic doses by immunoanalysis of urine: two techniques evaluated and modified for improved performance.

We evaluated the EMIT (enzyme-multiplied immuno technique) and FPIA (fluorescence polarization immunoassay) urine screening systems for detection of benzodiazepine intake. Healthy male volunteers were given single oral therapeutic doses of alprazolam (2 mg), chlordiazepoxide (25 mg), flunitrazepam (1 mg), lorazepam (3.75 mg), nitrazepam (5 mg), and triazolam (0.25 mg), after which urine was collected for the next 32 h. The EMIT method failed to detect the intake of flunitrazepam, lorazepam, and nitrazepam. FPIA did not detect the intake of chlordiazepoxide, flunitrazepam, lorazepam, nitrazepam, and triazolam. Modification of the EMIT method to include enzymatic hydrolysis did not significantly alter the results obtained with this method. A modification of the FPIA method to include enzymatic hydrolysis and a lower cutoff value improved the results considerably, so that we reliably detected all studied substances but flunitrazepam. We conclude that (a) both EMIT and FPIA techniques, when used as intended by the manufacturers, are unreliable for the detection of intake of therapeutic doses of these benzodiazepines, and (b) the described modification of the FPIA should provide a much improved tool for detection of benzodiazepine intake.