Evaluation of Immunoassay Performance for the Detection of Opioids in Urine

Le résumé fourni par la source

BACKGROUND: Immunoassay drug screens provide rapid analysis of urine for the presence of therapeutics and drugs of abuse. Compared to definitive (confirmatory) methods, immunoassays are prone to false-positive and -negative results. Laboratories generally rely on manufacturers' claims regarding method sensitivity and specificity; few have the resources to independently verify performance. In this study, we review the performance of our opioid immunoassay drug screens in comparison to a definitive method. METHODS: Results of 859 urine samples tested via opioid immunoassay screens for buprenorphine, fentanyl, methadone metabolite (2-ethylidene-1,5-dimethyl-3,3-diphenylpyrrolidine; EDDP), opiates, and oxycodone were compared to definitive results obtained via liquid chromatography-tandem mass spectrometry (LC-MS/MS). The data examined here included multiple samples from individual patients. Our quantitative LC-MS/MS method includes 19 opioid compounds (parent drugs plus metabolites). RESULTS: Immunoassay sensitivity and specificity ranged from 96% to 100% and 84% to 99%, respectively. The sensitivity and specificity of these screens were similar to manufacturers' claims with some exceptions. The opiates immunoassay had poor performance when limiting the comparison to its target compound, morphine, but improved when including all compounds listed in the manufacturer's instructions for use (IFU). While demonstrating good sensitivity, the buprenorphine immunoassay demonstrated lower specificity than stated in the IFU. CONCLUSIONS: The opioid immunoassay screens in use at our facility compared favorably to a definitive LC-MS/MS method. The urine fentanyl screen had the lowest sensitivity (96%) and had a specificity of 97%. The urine buprenorphine assay was the least specific (84%) and had a sensitivity of 99%.

Ce résumé expose les affirmations des auteurs. BNTIC ne l’interprète pas comme une validation indépendante des résultats.