Diagnostic efficiency of different amphetamine screening tests--the search for an optimal cutoff

Increased use of designer drugs (amphetamines and amphetamine-like substances) raises the need for fast screening tests in urine in clinical settings, workplace and drug rehabilitation. Immunological assays currently used are subject to unwanted crossreactivities, partly depending on the cutoff concentrations used. The values recommended in Europe and the USA are 500 and 1000 ng/ml, respectively. In Switzerland, the recommended concentration of 300 ng/ml results in a high rate of false-positive urine samples and expensive, time-consuming confirmation testing. Using the Abbott Axsym analyzer, we found numerous false positives from patients in rehabilitation centers due to concomitant medication. Therefore, the diagnostic sensitivity and specificity of the Abbott test at different cutoff concentrations and the sensitivity of the Roche Cobas Integra, Beckman Synchron and Biosite Triage point-of-care test were examined. HPLC BioRad Remedi was chosen as the method of higher hierarchical order. The specificity of the Axsym analyzer (300 ng/ml) was 86%. At 500 ng/ml or 1000 ng/ml the specificity was increased to 99 or 100%, respectively, while the sensitivity only decreased from 97 to 91 or 81%, respectively. In summary, the cutoff concentration for amphetamine screening tests should not be below 500 ng/ml to avoid a high rate of false-positive results.