Printer Friendly

Interference with testing for lysergic acid diethylamide.

Testing for drugs of abuse (DOA) is an important diagnostic tool that may contribute to the establishment of the diagnosis in the intoxicated patient. [3] Several types of DOA tests have been used in the clinical laboratory, such as chromatographic methods and antibody-based assays. More recently, immunoassays have become a popular tool because they are easy to perform and guarantee short turnaround times. Unfortunately, immunoassays are associated with such caveats as lack of sensitivity or lack of specificity [1]. Results that are positive by immunoassay may be confirmed by more complicated and time-consuming techniques such as thin-layer chromatography or gas chromatography-mass spectrometry (GC-MS). Frequently, clinical decisions have to be made before preliminary results by immunoassay can be confirmed by another method. Therefore, the interpretation of test results requires the knowledge of substances that may interfere with the testing [1]. We have evaluated the new homogenous Emit immunoassay for lysergic acid diethylamide (LSD) from Behring Diagnostics. The assay was associated with a high false-positive rate when we tested urine samples that were submitted by patients from our hospital population. Further investigation revealed that certain prescribed medications interfered with the LSD Emit test in vitro, but not with Emit tests for other DOA.

Materials and Methods

Procedures followed were in accordance with the ethical standards of our institutional review board. Consecutive urine specimens that were submitted for DOA testing had been collected over 6 weeks. Samples were stored at 4 [degrees]C and shielded from light. LSD was measured in the urine with the Emit II[R]assay from Behring Diagnostics (Cupertino, CA) on a Syva-30R automated assay analyzer (Behring Diagnostics). LSD was also analyzed with the RIA from Diagnostics Products Corp. (Coat-A-Count; Los Angeles, CA) and the RIA from Roche Diagnostic Systems (Abuscreen; Branchburg, NY). Randomly selected urine samples were sent to a reference laboratory for the detection of LSD parent compound by GC-MS. Ions were monitored at m/z ratios of 253, 293, and 395 on a Hewlett-Packard 5890 GC-5970 MSD instrument (Palo Alto, CA) as described previously [2]. The manufacturers of the two RIAs and the Emit assay reported a cutoff concentration of 0.5 [micro]g/L for a presumptive positive LSD result. The threshold concentrations for positive LSD results by the immunoassays were determined in our laboratory by serial dilution of a LSD calibrator (Radian Corp., Austin, TX). The cutoff concentration for LSD by GC-MS was reported to be 0.2 [micro]g/L. Current therapeutic medications from patients who tested positive for LSD by Emit were identified by review of the medical charts. These drugs were obtained from the hospital pharmacy and dissolved in normal urine samples that had previously tested negative for LSD.

Results and Discussion

During the 6 weeks, 1898 urine samples were submitted for DOA testing. Specimens originated from various psychiatric clinics, from a methadone maintenance program, from psychiatric inpatients, and from the psychiatric emergency room (ER). DOA testing was also requested on specimens from ER patients, on specimens from patients who underwent treatment for acute surgical or medical conditions, and on urine samples that were received from a shelter program for homeless veterans. Seventy-nine of the 1898 submitted urine samples (4.2%) tested positive for LSD by Emit (Table 1).

The surprisingly high rate of positive LSD Emit test results prompted us to compare the test with other commercially available assays. Of the 79 LSD-positive specimens, 8 were of insufficient quantity for further testing and 71 were reanalyzed with two RIAs. All of the 71 urine samples (3.7% of all screened urine specimens) that had tested positive for LSD by Emit subsequently tested negative with both RIAs.

Several factors could have accounted for the discrepant results that were obtained with the three assays. First, a lower threshold concentration for LSD by Emit could have resulted in the detection of LSD by Emit, whereas LSD may have been missed by the RIAs. We determined the minimal concentrations resulting in positive LSD test results for the Emit, Coat-a-Count RIA, and Abuscreen RIA to be 0.43, 0.31, and 0.58 [micro]g/L, respectively. Furthermore, LSD was not detected in the urine samples with the Abuscreen RIA even at a lower threshold concentration of 0.2 [micro]g/L. These results demonstrate that the positive test results by Emit were not due to different threshold concentrations of the assays.

The discrepant LSD results between Emit and RIA may also have been caused by interfering substances, resulting in false-negative measurements by RIA or false-positive results by Emit. Visual inspection of urine samples and determination of urine pH and creatinine failed to indicate that the samples were adulterated [3, 4]. To evaluate the accuracy of the immunoassays, we tried to detect LSD in urine samples with a chromatographic method. Ten of the 71 urine samples with discrepant LSD results between Emit and RIA were randomly selected and submitted for detection of LSD by GC-MS. LSD was not found in any of the 10 urine samples. On the basis of these results, a negative LSD result by GC-MS would have been expected in at least 53 of the 71 urine samples that tested positive for LSD by Emit (one-sided 95% exact confidence interval: 53-71). Because urine samples tested negative for LSD by three alternative assays, we concluded that the positive LSD results by Emit were due to unidentified interfering substances.

Because the highest rate of false-positive LSD results were obtained from patients who had been seen by psychiatric, medical, or surgical services, we suspected the patients' prescribed medications as a cause of the drug interference. Most of the prescribed drugs have previously been shown to be excreted into the urine as metabolites rather than as unchanged compounds [5]. Since drug metabolites were not readily available, we evaluated whether the parent compounds of therapeutic drugs would interfere with the LSD Emit test in vitro by supplementing normal urine with 47 prescription drugs. The following 26 medications did not have an effect on LSD measurements by Emit at the tested concentrations of one mg/mL: acetaminophen, albuterol, allopurinol, alprazolam, atenolol, caffeine, carbamazepine, clonazepam, diazepam, flunitrazepam, glyburide, hydrochlorothiazide, ibuprofen, indomethacin, librium, lidocaine, lorazepam, methaqualone, methocarbamol, nicotine, nifedipine, perphenazine, phenobarbital, procainamide, ranitidine, or temazepam. In contrast to these 26 medications, 21 drugs caused the reading of the LSD Emit to exceed the threshold value (Table 2). These interfering drugs were diluted to determine the minimal interfering concentration. During the serial dilution the interfering drugs showed a semilogarithmic dose-signal relation. Some of the implicated prescription medications may not be detected in urine at the concentrations that have been found to interfere with the assay in vitro (Table 2). These medications cannot be excluded as a cause of interference because metabolites rather than parent drugs may have interfered with the assay. To evaluate the specificity of the interference, we tested most of these supplemented urine samples for other DOA by Emit. All of the treated urine samples tested negative for amphetamines, barbiturates, benzodiazepines, cannabinoids, cocaine metabolites, methadone, opiates, phencyclidine, or propoxyphene (Table 2). These results demonstrate that the interfering drugs are highly specific for the Emit assay that detects LSD.

We estimated the number of interfering therapeutic medications that were prescribed for patients with positive LSD results by Emit. The 79 urine samples that tested positive for LSD by Emit had been submitted by 48 patients. Two of the 48 patients (4%) were not taking any known prescription drugs at the time of testing. Three of the 48 patients (6%) were prescribed only medications that did not interfere with the LSD Emit assay in vitro at the tested concentrations. The 43 remaining patients (90%) were taking one or several of the therapeutic drugs that were identified to interfere with the Emit LSD assay in vitro (Table 3). These results suggest that the patients' prescribed medications or their metabolites may have been responsible for most of the false-positive LSD assay results by Emit. Other medications may also have interfered with the LSD Emit assay, since drug metabolites were not tested.

Our study demonstrates that the new LSD Emit assay was associated with a high rate of false-positive results (3.7%) overall, with a false-positive rate that exceeded 10% in patients from psychiatric clinics or medical clinics. On the other hand, a lower false-positive rate of 0.5% was found in subjects from the veterans shelter, and the false-positive rate of the assay could be acceptable in a more general population. Although it is stated in the LSD Emit package insert that samples from patients taking chlorpromazine may produce positive results with the assay, we found that many different medications resulted in false-positive assay results in vitro, including antipsychotic drugs, antidepressants, anxiolytic drugs, antiemetic medication, analgesics, cardiovascular medications, and antibiotics. Our results suggest that therapeutic drugs or their metabolites may be considered a cause for the observed interference with the LSD assay. In conclusion, positive LSD test results by Emit should be confirmed by an alternative method similar to the two-step protocols used in employment-based DOA testing [6, 7]. In clinical situations where confirmatory tests are not readily available, such as in the ER, the current version of the LSD Emit assay cannot be recommended as a screening assay for intoxicated patients because positive results seem to be inconclusive.

We thank Joseph Hoffmann for his critical review of the manuscript. We also are indebted to Behring Diagnostics and Roche Diagnostic Systems for providing us with a free test kit for LSD.

References

[1.] Liu RH. Evaluation of common immunoassay kits for effective workplace drug testing. In: Liu RH, Goldberger BA, eds. Handbook of workplace drug testing, 1st ed. Washington, DC: AACC Press, 1995:67-129.

[2.] Paul BD, Mitchell JIM, Burbage R, Moy M, Sroka R. Gas chromatographic-electron-impact mass fragmentometric determination of lysergic acid diethylamide in urine. J Chromatogr 1990;529:10312.

[3.] Mikkelsen SL, Ash K0. Adulterants causing false negatives in illicit drug testing. Clin Chem 1988;34:2333-6.

[4.] Warner A. Interference of common household chemicals in immunoassay methods for drugs of abuse. Clin Chem 1989;35:64851.

[5.] Goodman and Gilman's The pharmacological basis of therapeutics, 8th ed. New York: Pergamon Press, 1990:1655-735.

[6.] Hoyt DW, Finnigan RE, Nee T, Shults TF, Butler T.I. Drug testing in the workplace-are methods legally defensible? A survey of experts, arbitrators, and testing laboratories. JAMA 1987;258: 504-9.

[7.] Ferrara SD, Tedeschi L, Frison G. Drug-of-abuse testing in urine: statistical approach and experimental comparison of immunochemical and chromatographic techniques. J Anal Toxicol 1994; 18:278-91.

DETLEF RITTER, (1,2) * CHERISE M. CORTESE, (2) LINDA C. EDWARDS, (1) ,JUDITH L. BARR, (2) HYUNG D. CHUNG, (1,2) and CHRISTOPHER LONG (2)

(1) Pathology and Laboratory Medicine, 113JC, John Cochran Veterans Affairs Medical Center, 915 North Grand Blvd., Saint Louis, MO 63106.

(2) Department of Pathology, Saint Louis University Health Sciences Center, 3635 Vista at Grand, P.O. Box 15250, Saint Louis, MO 63110-0250.

(3) Nonstandard abbreviations: DOA, drugs of abuse; GC-MS, gas chromatography-mass spectrometry; LSD, lysergic acid diethylamide; and ER, emergency room.

* Author for correspondence. Fax 314-268-5110; e-mail ritterdg@ sluvca.slu.edu.

Received October 2, 1996; revised December 11, 1996; accepted December 20, 1996.
Table 1. Origin of LSD test requests.

 No. specimens No. LSD positives
Source tested (% of group)

Psychiatric clinics 467 15 (3.2%)
Methadone clinic 845 29 (3.4%)
Psychiatric wards 139 22 (15.8%)
Psychiatric ER 35 6 (17.1%)
Medical ER 6 0 (0%)
Surgery/medical services 39 5 (12.8%)
Veterans shelter 367 2 (0.5%)
Total 1898 79 (4.2%)

Table 2. Therapeutic drugs that interfere with LSD
measurements in vitro.

 Minimal concn. with positive
Drug LSD test result, mg/L

Amitriptyline (a) 5
Benzphetamine <1000 (b)
Buproprion (a) 35
Buspirone <1000 (b)
Cephradine 580
Chlorpromazine <4 (b)
Desipramine (a) 75
Diltiazem (a) 900
Doxepin (a) 11
Fentanyl (a) 0.1
Fluoxetine (a) 6
Haloperidol (a) 0.07
Imipramine (a) 50
Labetolol (a) 60
Metoclopramide <0.8 (b)
Prochlorperazine 10
Risperidone (a) 4
Sertraline (a) 0.2
Thioridazine (a) 2
Trazodone (a) 15
Verapamil (a) 5

(a) Negative for other DOA tests by Emit (see text).

(b) Lower drug concentrations were not tested.

Table 3. Interfering medications that were prescribed for
43 patients who tested positive with the LSD Emit assay.

Drug No. of prescriptions Prescribed dosages

Sertraline 13 50 mg qd-100 mg bid
Haloperidol 11 5 mg qd-20 mg bid
Trazodone 10 10 mg qd-100 mg qd
Verapamil 7 180 mg qd-240 mg qd
Chlorpromazine 4 100 mg bid-100 mg tid
Fluoxetine 4 20 mg qd
Risperidone 4 3 mg bid
Doxepin 3 50 mg qd-150 mg qd
Labetolol 2 100 mg bid-200 mg bid
Thioridazine 2 50 mg qid
Amitriptyline 1 100 mg qd
Buspirone 1 10 mg tid
Cephradine 1 500 mg qid
Diltiazem 1 120 mg qd
Metoclopramide 1 10 mg qd
Prochlorperazine 1 10 mg bid

Bid, twice daily; tid, three times daily; qid, four times daily;
qd, daily.
COPYRIGHT 1997 American Association for Clinical Chemistry, Inc.
No portion of this article can be reproduced without the express written permission from the copyright holder.
Copyright 1997 Gale, Cengage Learning. All rights reserved.

Article Details
Printer friendly Cite/link Email Feedback
Title Annotation:Drug Monitoring and Toxicology
Author:Ritter, Detlef; Cortese, Cherise M.; Edwards, Linda C.; Barr, Judith L.; Chung, Hyung D.; Long, Chri
Publication:Clinical Chemistry
Date:Apr 1, 1997
Words:2157
Previous Article:Quantitative colorimetric determination of urinary p-aminophenol with an automated analyzer.
Next Article:Development of a urodilatin-specific antibody and radioimmunoassay for urodilatin in human urine.
Topics:

Terms of use | Copyright © 2017 Farlex, Inc. | Feedback | For webmasters