Published online August 31, 2007
PEDIATRICS Vol. 120 No. 3 September 2007, pp. e504-e509 (doi:10.1542/peds.2006-2199)
This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Right arrow P3Rs: Submit a response
Right arrow Alert me when this article is cited
Right arrow Alert me when P3Rs are posted
Right arrow Alert me if a correction is posted
Right arrow Citation Map
Services
Right arrow E-mail this article to a friend
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Add to My File Cabinet
Right arrow Download to citation manager
Right arrow reprints & permissions
Citing Articles
Right arrow Citing Articles via CrossRef
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Saidinejad, M.
Right arrow Articles by Ewald, M. B.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Saidinejad, M.
Right arrow Articles by Ewald, M. B.
Related Collections
Right arrow Therapeutics & Toxicology

ARTICLE

Interference by Carbamazepine and Oxcarbazepine With Serum- and Urine-Screening Assays for Tricyclic Antidepressants

Mohsen Saidinejad, MD, MSa, Terence Law, BSb and Michele Burns Ewald, MDa,c

a Division of Emergency Medicine
c Program in Medical Toxicology
b Department of Laboratory Medicine, Children's Hospital Boston, Harvard Medical School, Boston, Massachusetts


   ABSTRACT
TOP
 ABSTRACT
METHODS
 RESULTS
 DISCUSSION
 CONCLUSIONS
 REFERENCES
 
OBJECTIVE. The purpose of this work was to evaluate the potential cross-reactivity of 2 antiepileptic medications containing 3-ringed structures, namely, carbamazepine and oxcarbazepine, with screening assays for tricyclic antidepressants.

METHODS. A cross-sectional study of 52 patients between 3 and 19 years of age who had been prescribed either carbamazepine or oxcarbazepine was conducted. A serum fluorescence-polarized immunoassay and a urine enzyme-linked immunoassay were used. The serum carbamazepine or oxcarbazepine level was measured. Gas chromatography/mass spectrometry, a confirmatory test for tricyclic antidepressant detection, was subsequently performed on the serum specimen.

RESULTS. A linear dependency on medication level was observed with the serum fluorescence-polarized immunoassay assay. This relationship was stronger for carbamazepine (4.2 µg/L tricyclic antidepressant detected per microgram/liter of carbamazepine) than for oxcarbazepine (0.7 µg/L tricyclic antidepressant detected per milligram/liter). At higher carbamazepine levels (8.0–11.6 mg/L), 12 of 13 patients had a positive serum fluorescence-polarized immunoassay result; at lower levels (0.1–7.9 mg/L), only 1 of 20 had a positive result. None of the patients who were receiving oxcarbazepine showed significant tricyclic antidepressant activity on either assay.

CONCLUSIONS. Carbamazepine interferes at a statistically significant level with serum fluorescence-polarized immunoassay assay and in a dose-dependent fashion. Neither carbamazepine nor oxcarbazepine exhibit significant tricyclic antidepressant activity on urine enzyme-linked immunoassay assay.

Key Words: tricyclic antidepressants • toxicology screening assays • assay cross-reactivity • false-positive • 3-ringed structure

Abbreviations: TCA—tricyclic antidepressant • ED—emergency department • MHD—10-monohydroxy derivative • FPIA—fluorescence-polarized immunoassay • GC/MS—gas chromatography/mass spectrometry • EIA—enzyme-linked immunoassay

Drug-screening assays may be used when a patient is suspected of a toxic ingestion. Their high sensitivity, rapid turnaround, and low cost allow them to be routinely used. Furthermore, specimens can be easily obtained from blood or urine for testing with these assays. On the other hand, certain limitations of drug-screening tests, such as low specificity, dependence on laboratory detection cutoffs, and potential interference from other substances, must be recognized. Prescription medications with similar molecular structures to the compound being screened for can react with toxicology assays and create a false-positive test result, leading to possible unnecessary interventions and treatment. For this reason, clinical correlation is indicated; a confirmatory test for any positive results obtained from the screen assay may be required. We sought to determine the extent of interference of carbamazepine or oxcarbazepine, 2 anticonvulsants containing a 3-ringed moiety in their molecular structures, with the tricyclic antidepressant (TCA) assays with the Abbott TDx assay for detection of TCA at our institution.


   METHODS
TOP
 ABSTRACT
 METHODS
RESULTS
 DISCUSSION
 CONCLUSIONS
 REFERENCES
 
Study Design
A cross-sectional study of eligible patients between the ages of 3 and 18 years, who were prescribed either carbamazepine or oxcarbazepine, was conducted over a period that began in November 2003. The study was concluded April 2004, when we reached our recruitment goal of 50 to 55 patients.

Patient Recruitment
With approval from the committee on human investigations, as well as informed consent from the participants or their legal guardians, eligible patients were enrolled in the study. Patients were identified and recruited in the study during a routinely scheduled visit to either the epilepsy center or during an emergency department (ED) visit at a tertiary care urban children's hospital. For the patients who were recruited from the epilepsy center, consecutive eligible patients were identified from days chosen based on investigator availability. All of the patients who were recruited were presumed to be on prescription doses of the anticonvulsant of interest and were not symptomatic. Once enrolled in the study, laboratory samples of blood and urine were collected during the same clinic visit. No further patient involvement was required. Patients were excluded from participation in the study if they had ingested a TCA or any medication known to interfere with the TCA assay within the last 7 days.

Specimen Collection
Blood (3 mL) and urine (3 mL) were collected from each patient. Only patients who required blood collection as part of their clinical evaluation were included in the study. Blood remaining after the completion of required clinical diagnostic testing was then used for the study. Urine collection was contingent on the patient's ability to provide a urine specimen during the time of the visit. Patients were not required to stay beyond their routine appointment to provide a urine sample if they were not able to void or not toilet trained.

Statistical Analysis
A regression analysis was performed with a measure of association between serum levels of carbamazepine or oxcarbazepine, the latter of which was performed on the basis of the 10-monohydroxy derivative (MHD), a metabolite of oxcarbazepine. Using the Pearson correlation and Fisher's exact test, the association of interest was measured for linearity and statistical significance. For the study to have an 80% power to detect ≥10% to 20% of interassay differences, we required ≥50 patients.

Specimen Analysis
Each blood specimen was tested with the hospital's standard TCA-screening assay (cutoff for positive: 50 µg/L TCA), the serum fluorescence-polarized immunoassay (FPIA) by Abbott TDx analyzer (Abbott Laboratories, Abbott Park, IL). Our institution uses a lower detection cutoff standard compared with the manufacturer recommendation to avoid missing any potential exposure to TCA in our pediatric population. All of the blood samples, whether positive or negative on the TDx, were subsequently sent to an off-site laboratory for confirmatory testing with the gas chromatography/mass spectrometry (GC/MS) method for detection of TCA. Each patient also had a serum anticonvulsant drug level (carbamazepine or oxcarbazepine) measured as part of routine clinical care (therapeutic level for carbamazepine: 4–12 mg/L; oxcarbazepine level: 10–30 mg/L for the MHD, an active metabolite of oxcarbazepine). Anticonvulsant quantification by FPIA uses the Roche Integra 700 analyzer (Roche Diagnostics Corp, Indianapolis, IN). The instrument is used to quantify serum carbamazepine level.

Urine specimens were tested with the enzyme-linked immunoassay (EIA; Microgenics Corporation, Fremont, CA) urine assay (recommended manufacturer detection cutoff: 150 µg/L TCA), a screening method that is currently being evaluated at our institution for its inclusion as a potential confirmatory test for the identification of TCAs. Our study samples were requisitioned to the clinical laboratory separately from the other routine specimens. Urine samples were not sent for confirmatory GC/MS because of the fact that the urine-screening assay itself was evaluated as a confirmatory test against the serum FPIA for TCA detection.

Data Management
Study samples of blood and urine were processed in the laboratory on a weekly basis and were kept in a freezer before testing; all of the study samples were deidentified. All of the data were kept on a Microsoft Excel (Microsoft Corp, Redmond, WA) database in accordance with the institution's confidentiality in record maintenance and clinical investigational protocol.


   RESULTS
TOP
 ABSTRACT
 METHODS
 RESULTS
DISCUSSION
 CONCLUSIONS
 REFERENCES
 
We identified 75 patients for possible eligibility by screening upcoming patient visits to the epilepsy center and from a convenience sample of ED patients based on medical history. Twenty two of the 75 eligible patients were not enrolled after it was determined that they did not require routine laboratory investigation. One patient's family was unable to participate because of time constraints. Fifty-two eligible patients were enrolled beginning in November 2003. Our sample size goal of 50 to 55 (based on the statistical power analysis) was reached in April 2004. Thirty three of the 52 patients were receiving carbamazepine, whereas 19 patients had been prescribed oxcarbazepine. Of the 33 patients receiving carbamazepine, 13 (39.4%) tested positive on the serum FPIA. Interestingly, 0 of the 19 patients receiving oxcarbazepine (0.0%) tested positive on the serum FPIA. Of the 52 total enrolled patients, 43 patients were also able to provide a urine specimen for testing; none (0 of 43) of the urine specimens obtained tested positive with the urine EIA. The remaining 9 patients were unable to void during the visit time interval. Absence of a true TCA exposure was validated by negative GC/MS tests on the serum samples (a gold standard for detection of TCA). None of the participants experienced any adverse events, such as dizziness, ataxia, vomiting, central nervous system depression, or cardiac arrhythmias, during the study period.

Results from the serum FPIA testing analyzed by using a simple linear regression model and the Pearson's correlation coefficient revealed a significant linear dependency on medication level for carbamazepine (4.2 mg/L TCA detected per milligram/liter of carbamazepine; R2 = 0.92; P < .0001; Fig 1) and for the MHD metabolite (0.7 mg/L TCA detected per milligram/liter of oxcarbazepine; R2 = 0.46; P = .0014; Fig 2). A positive serum FPIA occurred almost invariably in patients with carbamazepine levels ≥8 mg/L. At and above this level (8.0–11.6 mg/L), 12 of 13 patients had a positive FPIA; below this level (0.1–7.9 mg/L), only 1 of 20 tested positive for TCA (P < .0001 by Fisher's exact test). The linear correlation between serum carbamazepine/oxcarbazepine level and serum TCA was maintained even at levels that were below detection threshold for a positive result (Fig 3).


Figure 1
View larger version (11K):
[in this window]
[in a new window]

  		FIGURE 1 Linear regression of carbamazepine levels on serum TCA levels (in patients who received therapeutic carbamazepine doses) using Abbott TDx FPIA. R2 = 0.9174; P < .0001.

 

Figure 2
View larger version (10K):
[in this window]
[in a new window]

  		FIGURE 2 Linear regression of MHD levels on serum TCA levels (in patients who received therapeutic oxcarbazepine doses) using Abbott TDx FPIA. R2 = 0.4594; P < .0014.

 

Figure 3
View larger version (23K):
[in this window]
[in a new window]

  		FIGURE 3 Comparison of positive serum TCA-assay activity (on the basis of the threshold for positive result at 50 µg/L) for carbamazepine and oxcarbazepine (MHD) using the Abbott TDx FPIA.

 

   DISCUSSION
TOP
 ABSTRACT
 METHODS
 RESULTS
 DISCUSSION
CONCLUSIONS
 REFERENCES
 
TCA ingestions are associated with a high morbidity and mortality risk.1 Prompt recognition and management of a patient who is suspected of ingesting a potentially toxic dose of this class of antidepressants is critical. Toxicology screening assays for TCA provide a method for confirming exposure; however, a strong clinical suspicion is of critical importance in making the diagnosis. Certain medications, such as carbamazepine and oxcarbazepine, have a 3-ringed molecular structure similar to the TCAs and can potentially cause a false-positive TCA result as reported in a previous case series.2

Carbamazepine is commonly used to treat focal onset seizures with or without secondary generalization, as well as trigeminal neuralgia.3 The mechanism of action of carbamazepine is thought to be related to its action on voltage-sensitive sodium channels in the neuron's cell membrane.3 The therapeutic range for carbamazepine is 4 to 12 mg/L.4 It has a tricyclic molecular structure, which strongly resembles a TCA and is shown to have some antidepressant properties (Fig 4). 5, 6 Metabolized in the liver by the microsomal enzyme CYP3A4, the 10,11-epoxy metabolite of carbamazepine is at least as pharmacologically active as carbamazepine itself.7


Figure 4
View larger version (8K):
[in this window]
[in a new window]

  		FIGURE 4 Molecular structures of carbamazepine (A), amitriptyline (a TCA) (B), and oxcarbazepine (C).

 
In a recent case series, 3 patients presented to the ED of an urban children's hospital with intentional overdoses of an unknown substance. The FPIA for TCA was positive in each case, despite the fact that none of the patients had been prescribed or had a known ingestion of a TCA, and their electrocardiograms showed no evidence of TCA toxicity. In each case, a further review of the medication history revealed that they had all been prescribed carbamazepine. A subsequent carbamazepine level was obtained, and in each case was found to be in the toxic range.2

Oxcarbazepinearbazepine, another widely used anticonvulsant medication, has a similar molecular structure to carbamazepine and the TCA (Fig 4). It is the 10-keto analog of carbamazepine and is metabolized in the liver into MHD, a metabolically active compound.8 The metabolism of oxcarbazepine is a reduction reaction as compared with the oxidative reaction involved in the metabolism of carbamazepine.9 The majority of antiepileptic properties of oxcarbazepine can be attributed to MHD. The half-life of oxcarbazepine is 2 hours, compared with its metabolite MHD, which is 9 hours. Oxcarbazepine is 40% protein bound and is excreted through the kidneys.10 Oxcarbazepine inhibits the microsomal enzyme CYP2C19 and induces the enzymes CYP3A4/5.11

The structural similarities of carbamazepine and oxcarbazepine with TCAs can affect the FPIA-screening assays designated to detect the presence of these antidepressants. This assay is used by many laboratories to detect the amount of fluorescence light emitted by the specimen.12 The information obtained from the amount of fluorescence is then converted into micrograms per liter of TCA present in the specimen.

It is important to note that several other drugs have been reported to interfere with and produce a false-positive result on the TCA assay.2, 13 These medications include diphenhydramine,14 propoxyphene,15 quetiapine,16 thioridazine,17 cyproheptadine,15 and cyclobenzaprine.18 All of the medications contain a multiple-ringed structure resembling the TCA (diphenhydramine and propoxyphene only have 2 rings but contain an alkyl group long enough to loop around and result in TCA-screen reactivity).

The urine EIA uses a glucose-6-phosphate dehydrogenase enzyme-labeled drug.19 Drug in the sample competes for a fixed number of antibody binding sites with this enzyme-labeled drug. When drug is absent from the specimen, the antibody will bind exclusively to the enzyme-labeled drug, and the enzyme activity is inhibited. When drug is present, it binds to the antibody, and enzyme-labeled drug remains unbound, allowing for its activity. A direct relationship is present between drug concentrations in the specimen and this enzyme activity.19

All of the results were verified by a negative GC/MS test, which is the gold-standard test for detection of TCA.20, 21 One of the unique findings in our study was that false-positive results were seen at therapeutic drug levels. None of our patients had a known overdose of carbamazepine or oxcarbazepine by history or physical examination findings, nor did any have symptoms. Although screening assays, such as the ones used in our study, are designed to detect toxic levels of a medication, we were able to demonstrate that a significant rate of false-positive results could be found in patients who were prescribed therapeutic doses of these medications. Our current institutional cutoff of 50 µg/L TCA is used in pediatric patients to detect any significant exposure, as mentioned previously. Even at this detection threshold, a false-positive result would be considered statistically significant. This relationship, however, is not as nearly pronounced in the case of oxcarbazepine. In the case of the EIA urine assay, with the cutoff threshold of 150 µg/L that is recommended by the manufacturer, no false-positive results were found in any of our patients.


   CONCLUSIONS
TOP
 ABSTRACT
 METHODS
 RESULTS
 DISCUSSION
 CONCLUSIONS
REFERENCES
 
Carbamazepine interferes with the FPIA serum assay at a statistically significant level and in a dose-dependent fashion. A strong correlation exists between serum carbamazepine levels and positive TCA-screening results. At our institution's current cutoff point for a positive TCA result on the FPIA, even a therapeutic serum level of carbamazepine can produce a false-positive result. A positive serum FPIA occurred almost invariably in patients with carbamazepine levels ≥8 mg/L. Oxcarbazepine interferes with the FPIA serum assay, but this interference not statistically significant at our current institutional cutoff threshold. EIAs show no statistically significant interference from either carbamazepine or oxcarbazepine with the TCA assay at the cutoff threshold recommended by the manufacturer.


   ACKNOWLEDGMENTS
 
This work was funded and supported by the institutional General Clinical Research Center.

We thank Drs Bourgeois, Riviello, Bergin, and Takeoka from the Epilepsy Center at Children's Hospital Boston for their inspiration and generosity and Michael W. Shannon, MD, MPH, for guidance and advice in preparation of this manuscript.


   FOOTNOTES
 
Accepted Mar 7, 2007.

Address correspondence to Mohsen Saidinejad, MD, MS, Children's Hospital Boston, Ida C. Smith Building, 300 Longwood Ave, Boston, MA 02115. E-mail: mohsen.saidinejad{at}childrens.harvard.edu

The authors have indicated they have no financial relationships relevant to this article to disclose.


   REFERENCES
TOP
 ABSTRACT
 METHODS
 RESULTS
 DISCUSSION
 CONCLUSIONS
 REFERENCES
 

  1. Henry J, Alexander C, Sener EK. Relative mortality from overdose of antidepressants [published correction appears in BMJ. 1995;310:911]. BMJ. 1995;310 :221 –224[Abstract/Free Full Text]
  2. Matos M, Burns MM, Shannon MW. False-positive tricyclic antidepressant drug screen results leading to the diagnosis of carbamazepine intoxication. Pediatrics. 2000;105(5) . Available at: www.pediatrics.org/cgi/content/full/105/5/e66
  3. Carbatrol tablets. In: Physicians Desk Reference [electronic library]. Greenwood Village, CO: Thomson Micromedex; 2005
  4. Kerr BM, Thummel KE, Klein SM, et al. Human liver carbamazepine metabolism: role of CYP3A4 and CYP2C8 in 10,11-epoxide formation. Biochem Pharmacol. 1994;47 :1969 –1979[CrossRef][ISI][Medline]
  5. Post RM, Uhde TW, Ballenger JC, Chatterji DC, Greene RF, Bunney WE Jr. Carbamazepine and its -10,11-epoxide metabolite in plasma and CSF: relationship to antidepressant response. Arch Gen Psychiatry. 1983;40 :673 –676[Abstract]
  6. Ernst CL, Goldberg JF. Antidepressant properties of anticonvulsant drugs for bipolar disorder. J Clin Psychopharm. 2003;23 :182 –192[CrossRef][ISI][Medline]
  7. Bourgeois BFD, Wad N. Individual and combined antiepileptic and neurotoxic activity of carbamazepine and carbamazepine-10, 11-epoxide in mice. J Pharmacol Exp Ther. 1984;231 :411 –415[Abstract/Free Full Text]
  8. Trileptal tablets. In: Physicians Desk Reference [electronic library]. Greenwood Village, CO: Thomson Micromedex; 2005
  9. Grant SM, Faulds D. Oxcarbazepine: a review of its pharmacology and therapeutic potential in epilepsy, trigeminal neuralgia and affective disorders. Drugs. 1992;43 :873 –888[ISI][Medline]
  10. Baruzzi A, Albani F, Riva R. Oxcarbazepine: pharmacokinetic interaction and their clinical relevance. Epilepsia. 1994;35(suppl 3) :S14 –S19
  11. Bang L, Goa K. Oxcarbazepine: a review of its use in children with epilepsy. Paediatr Drugs. 2003;5 :557 –573[CrossRef][Medline]
  12. Fluorescent Polarized Immunoassay (FPIA) [package insert]. Abbott Park, IL: Abbott Laboratory; 2001
  13. Sorisky A, Watson DC. Positive diphenhydramine interference in the EMIT-st assay for tricyclic antidepressants in serum. Clin Chem. 1986;32 :715[Free Full Text]
  14. Chattergoon DS, Verjee Z, Anderson M, Carbamazepine interference with an immune assay for tricyclic antidepressants in plasma. J Toxicol Clin Toxicol. 1998;36 :109 –113[ISI][Medline]
  15. Wians FH Jr, Norton JT, Wirebaugh SR. False-positive serum tricyclic antidepressant screen with cyproheptadine. Clin Chem. 1993;39 :1355 –1356[Free Full Text]
  16. Hendrickson RG, Morocco AP. Quetiapine cross-reactivity among tricyclic antidepressant immunoassays. Clin Toxicol. 2003;41 :105 –108[CrossRef]
  17. Ryder KW, Glick MR. The effect of thioridazine on the automatic clinical analyzer serum tricyclic anti-depressant screen. Am J Clin Pathol. 1986;86 :248 –249[ISI][Medline]
  18. Wong EC, Koenig J, Turk J. Potential interference of cyclobenzaprine and norcyclobenzaprine with HPLC measurement of amitriptyline and nortriptyline: resolution by GC-MS analysis. J Anal Toxicol. 1995;19 :218 –224[ISI][Medline]
  19. Urine Enzyme-Linked Assay [package insert]. Fremont, CA: Microgenix Laboratories; 2003
  20. Way BA, Stickle D, Mitchell ME, Koenig JW, Turk J. Isotope dilution gas chromatographic-mass spectrometric measurement of tricyclic antidepressant drugs: utility of the 4-carbethoxyhexafluorobutyryl derivatives of secondary amines. J Anal Toxicol. 1998;22 :374 –382[ISI][Medline]
  21. Vandemark FL, Adams RF, Schmidt GJ. Liquid-chromatographic procedure for tricyclic drugs and their metabolites in plasma. Clin Chem. 1978;24 :87 –91[Abstract/Free Full Text]
PEDIATRICS (ISSN 1098-4275). ©2007 by the American Academy of Pediatrics




This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Right arrow P3Rs: Submit a response
Right arrow Alert me when this article is cited
Right arrow Alert me when P3Rs are posted
Right arrow Alert me if a correction is posted
Right arrow Citation Map
Services
Right arrow E-mail this article to a friend
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Add to My File Cabinet
Right arrow Download to citation manager
Right arrow reprints & permissions
Citing Articles
Right arrow Citing Articles via CrossRef
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Saidinejad, M.
Right arrow Articles by Ewald, M. B.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Saidinejad, M.
Right arrow Articles by Ewald, M. B.
Related Collections
Right arrow Therapeutics & Toxicology