Urine Drug Screening in the Era of Designer Benzodiazepines: Comparison of Three Immunoassay Platforms, LC-QTOF-MS and LC-MS-MS

Abstract

This study investigated the presence of designer benzodiazepines in 35 urine specimens obtained from emergency department patients undergoing urine drug screening. All specimens showed apparent false-positive benzodiazepine screening results (i.e., confirmatory testing using a 19-component liquid chromatography-tandem mass spectrometry (LC-MS-MS) panel showed no prescribed benzodiazepines at detectable levels). The primary aims were to identify the possible presence of designer benzodiazepines, characterize the reactivity of commercially available screening immunoassays with designer benzodiazepines and evaluate the risk of inappropriately ruling out designer benzodiazepine use when utilizing common urine drug screening and confirmatory tests. Specimens were obtained from emergency departments of a single US Health system. Following clinically ordered drug screening using Abbott ARCHITECT c assays and laboratory-developed LC-MS-MS confirmatory testing, additional characterization was performed for investigative purposes. Specifically, urine specimens were screened using two additional assays (Roche cobas c502 and Siemens Dimension Vista) and LC-quadrupole time-of-flight mass spectrometry (LC-QTOF-MS) to identify presumptively positive species, including benzodiazepines and non-benzodiazepines. Finally, targeted, qualitative LC-MS-MS was performed to confirm the presence of 12 designer benzodiazepines. Following benzodiazepine detection using the Abbott ARCHITECT, benzodiazepines were subsequently detected in 28/35 and 35/35 urine specimens using Siemens and Roche assays, respectively. LC-QTOF-MS showed the presumptive presence of at least one non-Food and Drug Administration (FDA)-approved benzodiazepine in 30/35 specimens: flubromazolam (12/35), flualprazolam (11/35), flubromazepam (2/35), clonazolam (4/35), etizolam (9/35), metizolam (5/35), nitrazepam (1/35) and pyrazolam (1/35). Two or three designer benzodiazepines were detected concurrently in 13/35 specimens. Qualitative LC-MS-MS confirmed the presence of at least one designer benzodiazepine or metabolite in 23/35 specimens, with three specimens unavailable for confirmatory testing. Urine benzodiazepine screening assays from three manufacturers were cross-reactive with multiple non-US FDA-approved benzodiazepines. Clinical and forensic toxicology laboratories using traditionally designed LC-MS-MS panels may fail to confirm the presence of non-US FDA-approved benzodiazepines detected by screening assays, risking inappropriate interpretation of screening results as false positives.