“Practical Guide” to Urine Drug Screening Clarified

To the Editor: Moeller et al recently provided a timely and important review of urine drug screening. Drug abuse is a serious medical and social problem in the United States. Urine drug testing (UDT) to detect abuse and diversion of prescription controlled medications, as well as abuse of illicit substances, is increasingly important in clinical medicine. Physicians' ability to accurately interpret UDT results, however, is poor.^{,  ,}  Education is critical; equally critical is the dissemination of accurate information. We would like to address several inaccuracies in the review.

Opioids. The authors correctly assert that fentanyl and oxycodone are undetectable by most urine screens for opiate drugs, but the reasons they provide are incorrect. Fentanyl is undetectable not because it has no metabolites (it does), but because the chemical structures of fentanyl and its metabolites differ radically from those of opiates (ie, morphine and codeine). Oxycodone is generally undetectable not because it is derived from thebaine—indeed, thebaine is a precursor to both codeine and morphine in the opium poppy—but chiefly because of a minor structural difference from opiates: a 14-hydroxyl group that prevents it from cross-reacting with opiate antibodies in screening assays. The authors state that semisynthetic derivatives of morphine are not used therapeutically because of their abuse potential. In fact, hydrocodone, oxycodone, and hydromorphone, all of which are listed in Table 4 as semisynthetic opiate derivatives, are among the most commonly prescribed opioids. The authors state, “Positive results for heroin abuse are caused by use of prescribed opiates, such as codeine and hydrocodone….” This is a misstatement. Codeine, morphine, (usually) hydrocodone, and heroin all yield positive opiate screens, but this means only that an individual has been exposed to an opiate. Exposure to heroin can be established only by demonstration of the 6-monoacetyl morphine metabolite by a specific confirmatory assay.

Cannabinoids. The authors perpetuate outdated information that nonsteroidal anti-inflammatory drugs (NSAIDs) can produce false-positive results for cannabinoids on the Syva EMIT and other immunoassay systems. While this was once true (their reference is nearly 20 years old), Syva has solved this problem by altering the formulation of EMIT.^,  This was never a problem for other immunoassays. The authors also state that hemp-containing foodstuffs can produce positive screens for cannabinoids. Again, while this was once true, a 2003 US Drug Enforcement Agency (DEA) ruling classified food and beverages containing any amount of tetrahydrocannabinol (THC) as Schedule I controlled substances, making it unlawful to manufacture, distribute, dispense, or import any such product without registration. This had an immediate effect on domestic manufacturers, whose hemp-containing products are now virtually free of THC.

Cocaine. Until several years ago, “Health Inca Tea,” a decocanized product that indeed contained detectable quantities of cocaine, was commercially available in the United States. The importation of this and similar products has since been banned by the DEA. Whereas these products are available over the Internet, they are classified as Schedule II drugs, making illegal their importation and possession.

Amphetamine and Methamphetamine. Screening for amphetamine—a term that ordinarily refers to both amphetamine and methamphetamine—by immunochemical methods has always been problematic because the phenylethylamine drug class, with its various controlled and over-the-counter derivatives, provides minimal antigenic character. Table 3 lists methamphetamine among the causes of false-positive results, but in fact amphetamine immunoassays are intended to detect both amphetamine and methamphetamine. The l- enantiomer of methamphetamine, desoxyephedrine, is not a controlled drug and can produce a positive screening result, although most modern immunoassays have a considerable degree of stereoselectivity and are much less reactive with the l-enantiomer. The authors correctly note that gas chromatography-mass spectrometry (GC-MS), without modification to separate chiral compounds, cannot distinguish between controlled (and frequently abused) d-methamphetamine and its less pharmacologically active enantiomer, desoxyephedrine. However, specifications for workplace drug testing from the Substance Abuse and Mental Health Services Administration have established a standard for GC-MS confirmation of d-methamphetamine that requires the presence (at >200 ng/mL) of the metabolically demethylated product, amphetamine, since the l-enantiomer undergoes only minimal demethylation. Although there is a reference to this requirement in a footnote to Table 1, it is not explained in the discussion, leaving the impression that confirmatory testing by nonchiral GC-MS does not rule out desoxyephedrine use, when in fact it does if performed and interpreted properly.

Appropriate interpretation of positive and negative results from urine drug screens often requires consideration of many variables, a point that Moeller and colleagues' review underscores. Misinformation, however, serves only to complicate the already difficult task that physicians face when using UDT as part of their clinical practice. We strongly urge clinicians to consult with a qualified professional (eg, certified Medical Review Officer or toxicologist) when questions arise about the meaning of results from UDT.