Tox and Hound – U(ds) and I

by Howard Greller

Recently, I’ve moved offices, and came across a number of old letters. A lot of it was just regular business stuff (e.g., credentialing packets, requests for privileges, a metric ton of menus, etc.), but there was a significant amount of correspondence as well, including some letters of a personal nature. In lieu of a regular post, I decided to share one of these old letters in hopes that others can learn from my experiences, both good and bad. The content of the letter is unchanged, but I’ve added commentary to help cut through some of the more deeply personal, and painful, sections.

&Howard

I’m just going to rip the Band-Aid off here, Urine Drug Screen, we’re done. It’s over.

I know that this is going to come as a surprise to you. We’ve been together for so long, but that’s part of the problem. We’ve fallen into a routine, a routine that doesn’t give me what I need or want, and hasn’t for so, so long. I don’t even know why I call on you anymore, other than out of a very bad habit. I’ve had a little time to sit and reflect and I realize that, on the balance sheet of our relationship, there is far more bad than good. I understand that you’ll want to fight for the relationship, but I just don’t see how it can work.

People working in healthcare have relied for years on the urine drug screen (UDS) to diagnose patients with a substance-related condition. They’re relatively inexpensive, use something that is readily available (urine), most can be automated, and typically they give results quickly. Its ease of use has ingrained it into the DNA of practice, to the detriment of both patients and practitioners.

I know you don’t understand our relationship. Which is not surprising considering that you don’t understand your job, either. You brag about your ability to “screen” people, and yet you continue to go a step beyond anything that you’ve ever been trained to do, and try and “diagnose” people. What is that? Did you learn that from your old relationships (that you talk about constantly, which is super creepy, btw), like when you were with federally regulated transportation? Or that other one where people worked with large industrial equipment?

The most commonly used UDS methodology is an immunoassay that consists of a panel of five tests, The panels have changed over time, but most commonly test for amphetamine, cocaine, THC, opioids and one of either benzodiazepine, barbiturate (much less common) or phencyclidine (PCP).^​1​ They are used as sensitive screens for the use of substances that would be considered harmful or dangerous to use while doing things like flying an airplane or driving a truck.^​2,3​

You are definitely not an ER doctor. It’s my job to take people with acute signs and symptoms and come up with a diagnosis. You don’t know how to answer the question as to whether or not this patient’s symptoms are caused by a drug, and you can’t. I know it gives you a sense of importance and power, but it really just frustrates me and creates confusion. I can’t afford to have you lead me down the wrong path anymore.

Clinicians rely on diagnostic testing to answer questions. For example, is this patient altered because their sodium is abnormal? (Accurate, specific, reliable test) The UDS was never designed or intended for this and has been co-opted for a purpose that it was never meant for. It is not a diagnostic tool to tell you that the patient in front of you is that way because of a drug.

The UDS uses an antibody directed at a drug. A single agent is often used as a “representative” of a class of drugs, or a common metabolite is used to capture agents of a family that share a common metabolic pathway. However, not every drug within a class is structurally similar or has common metabolites, so many things can be missed, leading to false negative results. Cross-reactivity of the antibody with unrelated drugs can produce false positive results, further contributing to the inappropriate application of these tests. Additionally, the tests in the panel can only detect known substances (i.e., there is a defined methodology) which can be picked up by the immunoassay. The standard UDS does not detect synthetic (e.g., synthetic cannabinoids) or designer drugs (e.g., “bath salts”), emerging novel psychoactive substances (e.g., MDMA, tryptamines, other substituted amphetamines), most plant-derived agents (e.g., ayahuasca), or combinations of products.^​4​

You’re always “yes” or “no”. I never get anything else from you. Just yes. Or no. Yes. No. What’s worse is that I can’t even trust you to answer that question correctly most of the time. You’ve been falsely positive to me. And falsely negative as well. And the lies. The falsehoods. Even when you were positive, you were wrong. And sometimes, even the opposite was true.

Sure, you can tell that someone used marijuana at some point in the past. Great. I’ll put on Dark Side of the Moon and grab them a bag of Doritos.

And yeah, as you like to constantly remind me, you’re pretty good at saying whether someone has used cocaine, [but not when it matters].

Oh? You think that that patient was using an opioid? Really? You mean the one that was hypoventilating that I already gave naloxone, and is now awake and talking to me? Thank you, next.

Benzodiazepines. Sure. Tell me that they used benzos. You know what I’m going to do, right? Exactly. Give them more benzos. . .

And then there are amphetamines. What is it with you and amphetamines? You scream “AMPHETAMINE” the same way that Dug, from the movie “Up” , shouts “SQUIRREL!”

Additionally, each of the tests on the UDS, although reported as “positive” or “negative”, has a threshold value associated with the assay. There are cutoff limits that were established for workplace testing by the DHHS. These are in place in order to try and limit false-positive results (e.g., to prevent a positive opioid screen from eating too many poppy seeds, or a positive THC screen from second-hand smoke – really overly inclusive positives, but you get the idea). Because of the nature of the tests themselves, they are subject to numerous false results, both positive and negative, that can have unexpected consequences for the patient being tested, both medically and socially. Most UDS are not further confirmed through a different method (i.e., GC/MS). The depth and breadth of cross-reacting agents are astounding and is commonly not even considered when these tests are ordered.

Amphetamine screens typically react to anything that has a phenylethylamine backbone, which includes amphetamine and methamphetamine. But the ubiquity of this chemical core also gives false positive results from things like bupropion, ephedrine, pseudoephedrine, ranitidine, methylphenidate, trazodone, etc. The synthetic amphetamines (e.g., “bath salts”, “2C” drugs, etc.) that don’t have the phenylethylamine core give false negative results. MDMA is also commonly missed.^{​1,2,5–7​}

Benzodiazepine screens often are designed to detect oxazepam, which is its own agent (Serax) as well as being the common metabolite of diazepam (Valium) and chlordiazepoxide (Librium). Some commonly used benzos, such as alprazolam (Xanax) and clonazepam (Klonopin) are not metabolized through oxazepam and thus can give a false negative result. Lorazepam (Ativan) is hepatically metabolized through glucuronidation (to lorazepam-glucuronide, which does not trigger the assay), and detection can be increased if the sample is first hydrolyzed with a β-glucuronidase to liberate the lorazepam.^​8​ On the other side of the detection coin, sertraline and efavirenz have been reported to cause a false positive result.^{​1,2,5–7​}

The cocaine screen is a good assay, with very few, if any, false positives (possibly coca tea or mate de coca) and negatives (some reports of fluconazole, but that usually messes with confirmatory testing with GC/MS). It screens for the presence of the major metabolite, benzoylecgonine. Multiple investigations reinforce the practice that asking the patient about cocaine use is easier and more reliable.^{​1,2,5–7,9​}

The opioid screen detects commonly encountered opioids (things derived from the opium poppy itself), such as heroin, codeine, and morphine, targeted against morphine. To distinguish heroin from other opioids, it’s unique metabolite, 6-monoacetylmorphine (6-MAM), can be tested for separately. The opioid screen is subject to false positives, such as the aforementioned poppy seeds, but additionally some common pharmaceuticals such as imipramine, naltrexone, naloxone, quinolones (ofloxacin but not ciprofloxacin) and rifampin. False negatives include all of the synthetic opiates (the broader class that includes synthetics, semi-synthetics, and the opioids) such as hydrocodone, hydromorphone, oxycodone, oxymorphone, fentanyl, methadone, tramadol, buprenorphine, and meperidine. Specialized screens exist for the detection of these agents, and are commonly employed in pain management and treatment of opioid use disorder. And even those, more specialized screens have issues. The methadone screen can be falsely positive from diphenhydramine, doxylamine, olanzapine, quetiapine, and verapamil.^{​1,2,5–7,10​}

If your lab has a PCP assay, given the persistence of this compound in the world, you are also subject to numerous false positive results. The PCP assay has been reported to produce a false positive with dextromethorphan, diphenhydramine, ibuprofen, lamotrigine, ketamine, methylenedioxypyrovalerone (MDPV or a “bath salt”), thioridazine, tramadol, and venlafaxine. And while there are no known false negatives, you probably don’t need this assay anyway.^{​1,2,5–7​}

THC, like cocaine, is a good assay targeting the major metabolite of marijuana found in urine, 9-carboxy-THC (11-nor-9-carboxy-delta-9-tetrahydrocannabinol, if you want to be specific). Although this is a good assay, there are a number of reported false positive screens with efavirenz, NSAIDs (including ibuprofen), promethazine, riboflavin, and even baby soap. . . Notably and very important, the synthetic cannabinoids share no similarity and are not detected by this screen (which is fine, since they are NOT POT).^{​1,2,5–7​}

It’s like I ask you what the weather is like outside, and you tell me what happened last week, in a different city, two hours after I’ve already walked into the rain without an umbrella. My friend @toxicologist12 said you were trouble, and I didn’t listen.

Beyond the ability of the UDS to actually detect what you think you are testing for, there is the problem of detection times. This is the window during which a substance can still make the assay positive. There are many things that play into this, but the most important are the myriad of pharmacokinetic variables associated with each drug. Where is it distributed (e.g., fat, muscle, etc)? Is it metabolized and how (e.g. liver, plasma, kidney)? Are the metabolites active, or are they detected? Are there patient-specific factors that play a role (e.g., genetic variability, body mass, comorbid illness, urine pH, etc.)? What about the pattern of usage (e.g., chronic use, recent use, heavy use, etc.)? Detection times can vary widely, ranging from days to months. Use of a drug in the past, once the urine has collected all of the detectable metabolic detritus, does not correlate with the clinical picture or impairment at the time of testing. What is past is prologue. And the problem here is that if you see a patient with a clinical syndrome AND a positive test, you are much more likely to ascribe that syndrome to that test. Even though the data that you are using can be days to weeks old. It’s like saying that the chest pain patient in front of you is having an MI, based on the labs and ECG that were obtained when they were last in the hospital with an MI a year ago . . . And that is all predicated on the fact that the test is correct . . .

Did we have some good times? Yes. Of course. All relationships do. Sure, I remember that one time you told me that guy was using cocaine before I gave him a beta-blocker? Thanks, . . . again. Remember that family that we put through the wringer when you accused them of giving their kid PCP? Ya. They gave her diphenhydramine. That’s on you.

The important thing for most practitioners is that urine drug screens rarely change management, which holds true as well in children, even when a comprehensive screen is used.^{​11–13​} Relying on a test with all of the issues described above leads to the use of additional resources, sometimes invasive or painful procedures and involvement of additional agencies, such as social work or child protective services, that can have detrimental consequences outside of the clinical encounter. The testing characteristics and the prevalence of false positive and false negative results further reinforce that these assays should not be used for this purpose.

Yes. We’ve tried therapy. It doesn’t work. I don’t trust our psychiatrist and how often she sides with you. Always refusing to accept patients until she gets your opinion. You say that she really needs the information you had. Which turned out not to be true. She just ignores you. Time and time and time again. And we just waste money. Thousands and thousands of dollars, and hours and hours and hours, for nothing.

There are multiple studies that have both retrospectively^​14​ and prospectively explored whether or not the UDS had any clinical utility in the management of psychiatry patients in the ED. Clinical guidelines from ACEP and the UK clinical toxicologists say that they do not.^​15,16​ They don’t change management, they don’t impact treatment, they cost money and they waste time. Time and time again, the most helpful intervention was taking a history and performing a physical examination.^{​14,17–21​} Even the psychiatrists tend to ignore the results^​17​ . . .

You gave me confidence where it wasn’t warranted and broke me down when it was. We’re through.

Good luck with your new job at the IOC.

&Howard