Assessment of urine drug screen utility at autopsy to predict laboratory postmortem blood toxicology

Abstract

When faced with increasing drug-related deaths and decline in practicing forensic pathologists, the need to quickly identify toxicology-related deaths is evident in order to appropriately triage cases and expedite turnaround times. Lateral flow immunoassays conducted pre-autopsy offer quick urine drug screen (UDS) results in minutes and are used to inform the need for autopsy. Over 1000 medicolegal cases were reviewed to compare UDS results to laboratory enzyme-linked immunosorbent assay (ELISA) blood results to evaluate how well autopsy UDS predicted laboratory findings. Mass spectral analysis was performed on ELISA-positive specimens and these data were used to investigate UDS false-negative (FN) results when possible. Five different UDS devices (STAT One Step Drug of Abuse dip card and cassette, Premiere Biotech multi-drug and fentanyl dip cards and ATTEST 6-acetylmorphine (6-AM) dip card) were tested encompassing 11 drug classes: 6-AM, amphetamine/methamphetamine, benzodiazepines, benzoylecgonine, fentanyl, methadone, opioids, phencyclidine, and delta-9-tetrahydrocannabinol. Sensitivity, specificity, efficiency, and positive and negative predictive values >80% indicated that UDS was useful for predicting cases involving benzoylecgonine, methadone, methamphetamine, and phencyclidine. UDS was unreliable in predicting amphetamine, benzodiazepines, fentanyl, and opiates-related cases due to a high percentage of FN (up to 11.2%, 8.0%, 12.4%, and 5.5%, respectively) when compared to ELISA blood results. For the later analytes, sensitivities were as low as 57.5%, 60.0%, 72.2%, and 66.7%, respectively. Overall results support that UDS cannot replace laboratory testing. Because UDS is subject to false-positive and FN results users must understand the limitations of using UDS for triage or decision-making purposes.