Received on October 31, 2003
Accepted on January 8, 2004

Drug Monitoring and Toxicology

Analytic Performance of Immunoassays for Drugs of Abuse Below Established Cutoff Values

¹ Departments of Pediatrics, Pathology and Immunology, and Internal Medicine, Washington University School of Medicine, St. Louis, MO, and Pathology and Immunology
² Departments of Pediatrics, Pathology and Immunology, and Internal Medicine, Washington University School of Medicine, St. Louis, MO, and Drug Analysis Laboratory, Barnes-Jewish Hospital, St. Louis, MO
³ Departments of Pediatrics, Pathology and Immunology, and Internal Medicine, Washington University School of Medicine, St. Louis, MO, and Pathology and Immunology, and Internal Medicine, Washington University School of Medicine, St. Louis, MO
⁴ Departments of Pediatrics, Pathology and Immunology, and Internal Medicine, Washington University School of Medicine, St. Louis, MO, and Department of Pathology, Medical College of Wisconsin and Children’s Hospital of Wisconsin, Milwaukee, WI
⁵ Departments of Pediatrics, Pathology and Immunology, and Internal Medicine, Washington University School of Medicine, St. Louis, MO, and Children’s Mercy Hospital, Kansas City, MO
⁶ Departments of Pediatrics, Pathology and Immunology, and Internal Medicine, Washington University School of Medicine, St. Louis, MO, and Pediatrics, and Pathology and Immunology

^* To whom correspondence should be addressed. E-mail: Dietzen_D{at}kids.wustl.edu.

Background: The analytic performance and accuracy of drug detection below Substance Abuse and Mental Health Services Administration (SAMHSA) cutoffs is not well known. In some patient populations, clinically significant concentrations of abused drugs in urine may not be detected when current SAMHSA cutoffs are used. Our objectives were to define the precision profiles of three immunoassay systems for drugs of abuse and to evaluate the accuracy of testing at concentrations at which the CV was <20%.

Methods: Drug-free urine was supplemented with analytes to assess the precision in three commercial drugs-of-abuse immunoassay systems below the SAMHSA-dictated cutoffs for amphetamines, opiates, benzoylecgonine, phencyclidine, and cannabinoids. Consecutive urine samples with signals associated with a CV <20% by Emit^® immunoassay and below SAMHSA cutoffs were then subjected to confirmatory analysis.

Results: The CV of all immunoassay systems tested remained <20% to drug concentrations well below SAMHSA cutoffs. The accuracy of urine drug-screening results between the SAMHSA-specified cutoffs and the precision-based cutoffs was less than accuracy for specimens above the SAMHSA cutoffs, but the use of the precision-based cutoff produced a 15.6% increase in the number of screen-positive specimens and a 7.8% increase in the detection of specimens that yielded positive results on confirmatory testing.

Conclusion: The precision of three commercial immunoassay systems for drugs-of-abuse screening is adequate to detect drugs below SAMHSA cutoffs. Knowledge of the positive predictive values of screening immunoassays at lower cutoff concentrations could enable efficient use of confirmatory testing resources and improved detection of illicit drug use.