Journal of Analytical Toxicology 2014;38:513 –518 doi:10.1093/jat/bku068
Special Issue
Determining Zolpidem Compliance: Urinary Metabolite Detection and Prevalence in Chronic Pain Patients David M. Schwope1*, Anne DePriest1, David L. Black1,2, Yale H. Caplan3, Edward J. Cone4 and Rebecca Heltsley1 1 Aegis Sciences Corporation, 515 Great Circle Road, Nashville, TN 37228, USA, 2Department of Pathology, Microbiology and Immunology, Vanderbilt University, Nashville, TN, USA, 3School of Pharmacy, University of Maryland, Baltimore, MD, USA, and 4Department of Psychiatry and Behavioral Sciences, Johns Hopkins School of Medicine, Baltimore, MD, USA
*Author to whom correspondence should be addressed. Email: [email protected]
Zolpidem (Ambienw) is the most prescribed insomnia treatment in the USA; however, little is known about zolpidem metabolite excretion in chronic pain patients. As zolpidem is extensively metabolized in vivo to zolpidem 4-phenyl carboxylic acid (ZCA), metabolite detection may provide improved accuracy for compliance determinations, thereby improving clinical decisions. Zolpidem and ZCA were extracted from 1 mL human urine by mixed-mode solid-phase extraction. Samples were analyzed by LC – MS-MS using positive electrospray ionization with multiple reaction monitoring mode employed for detection and quantification. Gradient chromatographic separation was achieved with a reversed-phase column in a rapid 1.8 min analysis. The assay was linear from 4 to 1,000 mg/L for zolpidem and 4 to 10,000 mg/L for ZCA. Interday recovery (bias) and imprecision (n 5 20) were 100–107% of target and 2.4–3.7% relative standard deviation, respectively. Extraction efficiencies were 78–90%. Pain compliance samples (n 5 3,142) were de-identified and analyzed for zolpidem and ZCA. Zolpidem was detected greater than limit of quantification in 720 specimens (22.9%), while ZCA was detected in 1,579 specimens (50.3%). Only five specimens contained zolpidem alone. ZCA was observed without parent zolpidem in 864 specimens, thereby increasing population detection rates by 27.5%. Addition of a zolpidem metabolite to compliance determinations substantially improved detection for zolpidem intake and also should prove useful in clinical and forensic settings.
Introduction Sleep disorders abound, with an insomnia disorder prevalence rate of 12 – 20% of the general adult population (1, 2). Currently, zolpidem (Ambienw) is the most prescribed sedative – hypnotic for short-term treatment of insomnia and the 15th most prescribed drug in the USA (3). It is an imidazopyridine derivative that binds to the gamma-aminobutyric acid – benzodiazepine receptor complex with relative selectivity for the Type-1 (omega-1) benzodiazepine receptor subtype and shares many similar pharmacological properties to benzodiazepines (4). Zolpidem is subject to abuse and may potentially lead to physical dependence (5, 6), substantiating Schedule IV placement under the Controlled Substance Act. Monitoring zolpidem use in pain patients may be warranted, particularly due to the increased risk of adverse effects with concomitant use of multiple other central nervous system depressants (4). In addition to its prevalence in clinical settings, zolpidem is often implicated in drug-facilitated crimes due to its rapid onset of action, rapid clearance from biological specimens and amnesic properties (7, 8). Zolpidem is extensively metabolized by human cytochrome P450 (CYP) enzymes to inactive metabolites, primarily by CYP3A4 (61%). Additional CYP enzymes contribute to zolpidem
metabolism, including CYP2C9 (22%), CYP1A2 (14%), CYP2D6 (2.5%) and CYP2C19 (0.5%) (9). Less than 1% of a dose is excreted in urine as unchanged drug (10). The major urinary metabolite is zolpidem 4-phenyl carboxylic acid (ZCA), also known as zolpidem metabolite I (Figure 1). This metabolite represents 33– 48% of a dose in urine (11), although its abundance is subject to pharmacokinetic interactions with CYP3A4 substrates that often alter formation rates (12). Rapid zolpidem elimination (t1/2 2.6 h) complicates urinary compliance testing, as detection is limited to samples obtained soon after ingestion. Zolpidem is observed at low microgram per liter concentrations in urine, with peak concentrations of 5–25 mg/L reported 12 h following ingestion of a single 10-mg dose; urinary zolpidem concentrations approached detection limits following 36 h postadministration in this small cohort (7). Lewis and Vine (13) were unable to detect parent zolpidem even 8 h after ingestion of a single dose although ZCA was identified [2 mg/L limit of detection (LOD)] in urine for a period of 72 h following zolpidem administration. Strano Rossi et al. detected parent zolpidem for up to 36 h after ingestion [0.5 mg/L limit of quantification (LOQ)]; ZCA was again detectable for at least 72 h (unknown LOD) in this case (14). Published methods and studies have often focused solely on detection of parent zolpidem, potentially yielding false-negative intake determinations at existing cutoff concentrations. This is problematic for compliance testing in clinical settings as noncompliance often leads to patient censure. Also, in drug-facilitated sexual assault (DFSA) casework victims may not immediately present for evaluation, resulting in undetectable parent zolpidem concentrations at the time of testing. Furthermore, immunoassays currently employed in some laboratories may not cross-react with ZCA, further obfuscating proper determinations (15). For these reasons, inclusion of the primary metabolite in mass spectrometry-based zolpidem analyses likely would yield improved accuracy in detecting intake. To this end, we developed and validated a sensitive LC– MS-MS method for the simultaneous detection of zolpidem and ZCA in human urine. This method is specific and rapid, quantifying both analytes in 1.8 min. This method will be useful to toxicologists in clinical and forensic settings, where zolpidem intake determinations require a sensitive, rapid method for quantification of zolpidem over clinically relevant concentrations.
Experimental Samples Urine samples were obtained from chronic pain patients over the period of February – June 2012 from pain clinics located in
# The Author 2014. Published by Oxford University Press. All rights reserved. For Permissions, please email: [email protected]
Urine analyses Sample preparation Internal standard (zolpidem-d6 and ZCA-d6) was added to each sample to yield 100 mg/L in urine. Solid-phase extraction columns were conditioned with methanol and 0.1 M acetic acid. Specimens (diluted 1 : 1 with 0.1 M acetic acid) were loaded by gravity flow, followed by washing with deionized water, 200 mM hydrochloric acid, methanol and ethyl acetate. Columns were then dried under vacuum (SPE vacuum manifold) for 5 min, and samples were eluted with 5 mL dichloromethane : isopropanol : ammonium hydroxide (78 : 20 : 2). Samples were evaporated to dryness, reconstituted in mobile phase A and submitted for LC–MS-MS analysis.
Figure 1. Zolpidem (A) and ZCA (B).
20 states (AL, FL, GA, IL, KY, LA, MD, MS, NC, NJ, NV, NY, OH, PA, SC, TN, TX, UT, VA and WV). All patients were enrolled in pain management compliance monitoring that included drug testing for licit and illicit drugs. The study was approved by the Essex Institutional Review Board (Lebanon, NJ, USA). Urine specimens were collected into polypropylene containers that were capped, labeled and stored at room temperature until shipment to the laboratory for analysis. The specimens were analyzed at Aegis Sciences Corporation (Nashville, TN, USA).
Instrumentation All experiments were performed on an AB Sciex API 3200 triple quadrupole mass spectrometer with a TurboV electrospray ionization (ESI) source (AB Sciex, Foster City, CA, USA). The mass spectrometer was interfaced with a Shimadzu LC-20ADXR high-performance liquid chromatograph (HPLC) (Shimadzu Corporation, Columbia, MD, USA). Evaporation under nitrogen was completed using a TurboVap LV evaporator from Biotage (Charlotte, NC, USA). Powder standards were measured using an XP-6 analytical balance (Mettler-Toledo, Columbus, OH, USA).
Standards and reagents Zolpidem standards and deuterated internal standards were purchased from Cerilliant (Round Rock, TX, USA). ZCA and ZCA-d6 were from Toronto Research Chemicals (Toronto, ON, Canada). All reagents and solvents were from Sigma – Aldrich (St. Louis, MO, USA) and were HPLC grade or better. Clean-Screen ZSDAU020 solid-phase extraction (SPE) columns (United Chemical Technologies, Bristol, PA, USA) were used to prepare samples. Drug free human urine was evaluated for the absence of zolpidem and ZCA prior to use. 514 Schwope et al.
LC– MS-MS analysis Chromatographic separation was performed using a C18 HPLC column (Thermo Scientific Hypersil Gold, 3 mm, 100 2.1 mm). Mobile phase was 0.1% formic acid in HPLC water (A) and 0.1% formic acid in acetonitrile (B). Gradient conditions began at 20% B, increasing to 95% B by 0.3 min followed by a 0.5 min hold and re-equilibration. Chromatographic run time was 1.8 min; cycle time was 2.1 min (Figure 2). Mass spectrometric data were acquired with positive ESI in multiple reaction monitoring (MRM) mode. MS-MS parameter settings (Table I) were optimized via direct infusion of individual analytes (100 mg/L) in 50 : 50 mobile phase A : mobile phase B at 10 mL/min. Source parameters were as follows: curtain gas, 30 psi; collision-activated dissociation, 5 psi; IonSpray voltage, 1,500 V; source temperature, 4508C; gas 1 (zero air), 40 psi and gas 2 (zero air), 50 psi. Data analyses Peak area ratios of target analytes and their respective internal standards were calculated for each concentration. Urine concentrations were measured in microgram per liter and were normalized to refractometer-determined urine specific gravity (16). Statistical calculations were completed with Microsoft Excel 2007 for Windows (Microsoft Corporation, Redmond, WA, USA) and JMP 9.0 for Windows (SAS Software, Cary, NC, USA). Validation The method was validated for bias, inter- and intrabatch imprecision, specificity, sensitivity, linearity, extraction efficiency, matrix effect, carryover, interference (endogenous and exogenous) and analyte instability. Bias and imprecision were evaluated at three concentrations within the linear dynamic range of each analyte. Intrabatch imprecision (% relative standard deviation [RSD]) ranged from 0.3 to 2.7% for all analytes at all concentrations (n ¼ 5); interbatch imprecision ranged from 2.4 to 4.2% (n ¼ 20). Bias calculated as the percent of target concentrations at low, mid and high quality control (QC) concentrations for each analyte (50, 100 and 250 ng/ mL, respectively), ranged from 100 to 107% of target concentrations (n ¼ 20) and was considered clinically insignificant. Specificity was based on precursor mass/fragment ion and relative retention time. Transition peak area ratios for QC and authentic specimens were required to be within 20% of the calibrator for each respective analyte. Relative retention time (analyte to internal standard) difference for QC and authentic specimens
Figure 2. Representative chromatograms for the (A) drug free (negative) extracted urine, (B) LOQ (4 mg/L in negative urine) and (C) patient specimen containing 120 mg/L zolpidem (1.00 min) and 2,220 mg/L ZCA (0.63 min).
Table I LC –MS-MS MRM Parameters Analyte
Q1 (amu)
Q3 (amu)
Dwell (ms)
DP
EP
CEP
CE
CXP
Zolpidem 1 Zolpidem 2 ZCA 1 ZCA 2 Zolpidem-d6 ZCA-d6
308.3 308.3 338.2 338.2 314.4 344.3
235.2 263.2 265.2 293.2 235.2 265.1
75 75 75 75 75 75
61 61 60 60 61 64
6 6 5 5 6 5
19 19 22 22 39 20
43 27 45 32 49 45
6 6 2 2 6 2
Q1, quadrupole 1; Q2, quadrupole 2; DP, declustering potential; EP, entrance potential; CEP, cell entrance potential; CE, collision energy; CXP, cell exit potential; ZCA, zolpidem 4-phenyl carboxylic acid.
was required to be within +0.01 min compared with the relative retention time of the analyte to internal standard of the calibrator. Alternatively, retention time for each analyte could be within 3% of its respective retention time in the calibrator. Sensitivity was evaluated by determining LOD and LOQ. A series of decreasing concentrations of fortified urine were analyzed to empirically determine LOD and LOQ. LOD was determined as the concentrations with acceptable chromatographic retention time and peak shape, a signal-to-noise ratio of at least 3 : 1 and transition peak area ratios within 20% of the calibrator. LOQ
was the lowest concentrations with acceptable chromatographic retention time and peak shape, a signal to noise of at least 10 : 1, transition peak area ratios within 20% of the calibrator ratio and acceptable bias and imprecision (within at least 20% of target concentration and RSD within at least 10%, n ¼ 6). Analyte linearity was determined with 10 concentrations on 3 separate days. Linearity of the method was investigated by calculation of the least-squares regression line (1/x weighting) and expressed by the squared correlation coefficient (R 2). Linear ranges were 4 – 1,000 mg/L for zolpidem and 4 – 10,000 mg/L for ZCA; these ranges should prove useful for clinical and forensic casework. R 2 values (1/x weighting) were deemed acceptable (R 2 . 0.995) for all analytes. Following initial linearity determination, batches were analyzed using a single-point calibrator at a concentration of 100 mg/L (line forced through zero) due to efficiency with respect to time and resources (17). Precision and accuracy fulfilled a priori requirements across the linear range with the chosen calibrator concentration. The extraction efficiency ( percent) for each analyte was determined at two concentrations (4 and 40 ng/mL) and ranged from 90 to 94% for zolpidem and from 64 to 77% for ZCA. Similar results were obtained for the corresponding deuterated analog, and quantification was not adversely affected. Matrix effect was Zolpidem Metabolite Detection and Prevalence 515
investigated through a postcolumn infusion (‘T’ connector) of the analytes while injecting an extracted blank urine sample. No significant ion suppression or enhancement was observed at the retention time for ZCA. Significant ion enhancement was observed at the retention time of parent zolpidem. However, similar ion enhancement for the deuterated internal standard minimized impact on quantification. In all cases, ion ratios met a priori specifications and QC samples all quantified within acceptable limits. Carryover in a negative specimen following a specimen containing the upper limit of quantification (ULOQ) was assessed. No quantifiable carryover was observed for any analyte. In some cases, traces of zolpidem signal were observed in carryover specimens; however, ion transition ratios were not within 20% of the calibrator and any signal present was less than LOD. For assessment of possible matrix interferences, negative urine samples were collected from six different sources and tested for potential endogenous interferences; no interference was observed in any pool for any analyte. In addition, interferences from over 55 illicit and common therapeutic drugs, metabolites and related compounds were evaluated by adding potential interferents at various concentrations into urine aliquots fortified at the LOQ. Compounds examined included several classes of common licit and illicit drugs, including amphetamines, barbiturates, cocaine, delta-9-tetrahydrocannabinol, benzodiazepines, opioids (natural, semi-synthetic and synthetic compounds and metabolites), muscle relaxants, zopiclone and zaleplon. A compound did not interfere if the LOQ quantified within 20% of target and had stable retention times and correct transition ratios. Of the compounds tested, the methadone metabolite 2ethylidene-1,5-dimethyl-3,3-diphenylpyrrolidine (EDDP) was the only identified interferent. At EDDP concentrations .10,000 mg/L, zolpidem-d6 suppression was apparent; however, the analyte remained unaffected yielding zolpidem quantitation differences .20% from target at low QC concentration. ZCA was not affected by any tested interferent. Finally, 60 additional de-identified pain compliance specimens (containing various licit and illicit analytes) were extracted and analyzed for possible endogenous and exogenous interferences. Of these, one had interference with zolpidem (but not the metabolite); this is believed to be related to EDDP concentrations .10,000 mg/L in this specimen. Analyte stability in urine (n ¼ 5) was evaluated under three conditions: freeze – thaw, autosampler and processed sample stability. For freeze–thaw stability, concentration changes ranged from 22.8 to 7.6% for zolpidem and 21.2 to 10.6% for ZCA. For autosampler stability, concentration changes ranged from 24.7 to 3.9% for zolpidem and 24.4 to 0.4% for ZCA. For processed sample stability, concentration changes ranged from 25.8 to 22.1% for zolpidem and 22.5 to 20.1% for ZCA. Results and Discussion With over 43 million prescriptions filled per year in the USA alone, zolpidem is the most prescribed sleep aid and is widely available (3). However, to date, no study has examined use of zolpidem metabolites for compliance determinations. Zolpidem is extensively metabolized and exhibits short detection windows in urine, potentially yielding false-negative results in clinical and forensic settings. Thus, detection and quantification 516 Schwope et al.
of both parent and metabolite in a rapid, high-throughput assay is a significant advancement for this compound and here we report the impact of metabolite inclusion on compliance determinations.
Samples A total of 3,142 directed analysis urine samples were obtained from chronic pain patients over 5 months. Results were de-identified and examined for zolpidem and ZCA prevalence.
Prevalence of metabolite in urine Of 3,142 specimens, zolpidem was detected greater than LOQ in 720 (22.9%), whereas ZCA was detected in 1,579 (50.3%) (Table II). Two specimens (0.06%) contained zolpidem greater than ULOQ and 45 specimens (1.43%) contained ZCA greater than ULOQ. Normalized median (range) zolpidem and ZCA concentrations were 28.3 (4.08 – 805) mg/L and 2,038 (4.53 – 23,000) mg/L, respectively (Figure 3). Only five specimens (0.16%) contained zolpidem alone (median concentration 488 mg/L). As ZCA was observed without parent zolpidem in 864 samples, addition of this metabolite increased detection rates by 27.5% in this group. We previously reported (18, 19) that inclusion of metabolites for compliance determinations in chronic pain patients is crucial for maximizing accuracy. In the present report, zolpidem positive determinations were increased from 22.9 to 50.3%, representing 864 additional positive identifications of zolpidem use in 5 months when compared with analyzing for zolpidem alone at the same LOQ. Others have noted (13, 14) that zolpidem detection windows can be increased with inclusion of ZCA in routine analysis. While the present method improved detection rates specifically related to inclusion of the metabolite, this study cannot provide information regarding improvement in zolpidem intake detection windows. Further controlled administration studies would be beneficial in more precisely defining these windows under acute and chronic dosing regimens. Administration studies with immediate- and controlled-release zolpidem may also provide additional insight into ZCA production and excretion rates, although the similar elimination half-lives between the products [mean 2.6 and 2.8 h, respectively (4, 11)] suggest minor impact on detection periods or concentrations. Monitoring for zolpidem use or misuse may be prudent for pain practitioners during compliance assessment. According to data from the National Health and Nutrition Examination Survey, more than half of zolpidem users in the USA are using
Table II Pain Compliance Cohort Descriptive Statistics for Samples within the Linear Range (4 –1,000 mg/L for Zolpidem and 4 –10,000 mg/L for ZCA)
Count (.LOQ) Mean Standard deviation Median Minimum Maximum
Zolpidem
ZCA
720 56.2 82.0 28.3 4.08 805
1,532 2,680 2,790 2,040 4.53 23,000
All concentrations are microgram per liter and normalized to urine specific gravity.
cannot be used for impairment determination, detection of these compounds ( principally the metabolite) can provide evidence for zolpidem intake, be it known or unknown. This is particularly important in DFSA casework, where victims may not immediately present for evaluation. Based on these findings, it is recommended to include zolpidem metabolite to minimize false-negative potential in these scenarios. Although we identified numerous specimens containing ZCA with no zolpidem parent, only five samples in the present study contained detectable zolpidem with no corresponding ZCA. Upon further review, two of these samples were from a single individual with samples taken on two separate occasions. This individual was also positive for illicit drugs and had the highest concentrations of zolpidem in the entire cohort. The three remaining samples were investigated further, and we determined that these samples had no metabolites present for all other compounds detected. While substantial CYP enzyme inhibition could yield observed patterns, sample adulteration also should be considered as a possible explanation. Corrected zolpidem concentrations of 13.5 – 1,700 mg/L (4.1 – 1,440 mg/L uncorrected) indicate that some technical knowledge (or luck) would be required for adulteration as only 0.2 – 72 mg of parent zolpidem (0.004 – 1.4% of a typical 5 mg dose) would need to be added to a 50 mL urine sample to produce observed concentrations. Nevertheless, adulteration remains a plausible hypothesis and while metabolite testing identified 99% of zolpidem use in the present study, inclusion of both analytes is recommended during routine analysis for maximum accuracy in compliance determinations.
Conclusions Figure 3. Histograms detailing distribution of observed urine concentrations (mg/L) for (A) zolpidem and (B) ZCA in chronic pain patients with bars indicating frequency and black squares representing cumulative percent.
other sedating medications, of which opioids and benzodiazepines are most common (20). Zolpidem may cause adverse drug reactions including daytime drowsiness, dizziness, hallucinations, agitation or bizarre behavior, sleepwalking and ‘sleep driving’ (4). The high incidence of polypharmacy and significance of adverse effects related to zolpidem may pose increased risk for pain patients. Emergency department visits related to adverse effects of zolpidem have increased 220% from 2005 to 2010; more than half of visits involved concomitant use of other medications, particularly opioids and benzodiazepines. Women and elderly patients (.65 years old) may be at increased risk and have been involved in a disproportionate number of emergency department visits related to zolpidem intake (21). Sex-related pharmacokinetic differences with lower clearance observed in women (22) may be a factor; consequently, in January 2013, the US Food and Drug Administration required manufacturers to lower recommended dosages of zolpidem for females (23). In addition to legitimate widespread use as a sedative –hypnotic, zolpidem also has been implicated in both DFSA (7, 13) and driving under the influence of drugs (DUID) (24, 25) forensic casework. Although urinary zolpidem and ZCA concentrations
Inclusion of ZCA in zolpidem urine assays is useful for accurate determinations in compliance or DFSA/DUID scenarios. Improvement in detection of zolpidem intake is critical as noncompliance can lead to censure in monitoring scenarios; the present study provided a 27.5% increase in detection rates for a chronic pain patient population at the same LOQ. This improvement in detection may also be useful in DFSA/DUID casework where determinations based on zolpidem alone may yield poor detection of ingestion. The present method is sensitive, specific and can detect and quantify both parent zolpidem and its primary urinary metabolite ZCA. The rapid extraction and analysis are beneficial and should be easily adopted in both forensic and clinical laboratories. This new method provides novel insight into zolpidem urinary excretion and will improve our ability to detect zolpidem intake in clinical and forensic drug testing casework.
Acknowledgments The authors thank Elizabeth Baker for library services and the technical staff of Aegis Sciences Corporation for performing analysis of the patient urines.
Funding This work was supported by Aegis Sciences Corporation. Zolpidem Metabolite Detection and Prevalence 517
References 1. Morin, C.M., LeBlanc, M., Belanger, L., Ivers, H., Merette, C., Savard, J. (2011) Prevalence of insomnia and its treatment in Canada. Canadian Journal of Psychiatry, 56, 540–548. 2. Roth, T., Coulouvrat, C., Hajak, G., Lakoma, M.D., Sampson, N.A., Shahly, V. et al. (2011) Prevalence and perceived health associated with insomnia based on DSM-IV-TR: International Statistical Classification of Diseases and Related Health Problems, Tenth Revision; and Research Diagnostic Criteria/International Classification of Sleep Disorders, Second Edition criteria: Results from the America Insomnia Survey. Biological Psychiatry, 69, 592–600. 3. Aitken, M., Kleinrock, M. (2013) Declining medicine use and costs: for better or worse? A review of the use of medicines in the United States in 2012. IMS Institute for Healthcare Informatics: Parsippany, NJ, pp. 1– 50. 4. Highlights of prescribing information: Ambien cr. http://products. sanofi.us/ambien_cr/ambienCR.html (accessed March 2014). 5. Jaffe, J.H., Bloor, R., Crome, I., Carr, M., Alam, F., Simmons, A. et al. (2004) A postmarketing study of relative abuse liability of hypnotic sedative drugs. Addiction, 99, 165–173. 6. Licata, S.C., Mashhoon, Y., Maclean, R.R., Lukas, S.E. (2011) Modest abuse-related subjective effects of zolpidem in drug-naive volunteers. Behavioral Pharmacology, 22, 160– 166. 7. Villain, M., Cheze, M., Tracqui, A., Ludes, B., Kintz, P. (2004) Windows of detection of zolpidem in urine and hair: application to two drug facilitated sexual assaults. Forensic Science International, 143, 157–161. 8. Kintz, P., Villain, M., Dumestre-Toulet, V., Ludes, B. (2005) Drug-facilitated sexual assault and analytical toxicology: the role of LC-MS/MS in a case involving zolpidem. Journal of Clinical Forensic Medicine, 12, 36–41. 9. Von Moltke, L.L., Greenblatt, D.J., Granda, B.W., Duan, S.X., Grassi, J.M., Venkatakrishnan, K. et al. (1999) Zolpidem metabolism in vitro: responsible cytochromes, chemical inhibitors, and in vivo correlations. British Journal of Clinical Pharmacology, 48, 89– 97. 10. Salva, P., Costa, J. (1995) Clinical pharmacokinetics and pharmacodynamics of zolpidem. Therapeutic implications. Clinical Pharmacokinetics, 29, 142–153. 11. Baselt, R.C. Disposition of Toxic Drugs and Chemicals in Man. 9th edition. Biomedical Publications: Ashland, OH, 2011; 1877 pp. 12. Hesse, L.M., von Moltke, L.L., Greenblatt, D.J. (2003) Clinically important drug interactions with zopiclone, zolpidem and zaleplon. CNS Drugs, 17, 513–532. 13. Lewis, J.H., Vine, J.H. (2007) A simple and rapid method for the identification of zolpidem carboxylic acid in urine. Journal of Analytical Toxicology, 31, 195–199.
518 Schwope et al.
14. Strano Rossi, S., Anzillotti, L., Castrignano, E., Frison, G., Zancanaro, F., Chiarotti, M. (2013) UHPLC-MS/MS and UHPLC-HRMS identification of zolpidem and zopiclone main urinary metabolites and method development for their toxicological determination. Drug Testing and Analysis (in press). doi:10.1002/dta.1470. 15. Reidy, L., Gennaro, W., Steele, B.W., Walls, H.C. (2008) The incidence of zolpidem use in suspected DUI drivers in Miami-Dade Florida: a comparative study using immunalysis zolpidem ELISA kit and gas chromatography-mass spectrometry screening. Journal of Analytical Toxicology, 32, 688–694. 16. Cone, E.J., Caplan, Y.H., Moser, F., Robert, T., Shelby, M.K., Black, D.L. (2009) Normalization of urinary drug concentrations with specific gravity and creatinine. Journal of Analytical Toxicology, 33, 1– 7. 17. Peters, F.T., Maurer, H.H. (2007) Systematic comparison of bias and precision data obtained with multiple-point and one-point calibration in six validated multianalyte assays for quantification of drugs in human plasma. Analytical Chemistry, 79, 4967–4976. 18. Cone, E.J., Zichterman, A., Heltsley, R., Black, D.L., Cawthon, B., Robert, T. et al. (2010) Urine testing for norcodeine, norhydrocodone, and noroxycodone facilitates interpretation and reduces false negatives. Forensic Science International, 198, 58 –61. 19. Heltsley, R., Zichterman, A., Black, D.L., Cawthon, B., Robert, T., Moser, F. et al. (2010) Urine drug testing of chronic pain patients. II. Prevalence patterns of prescription opiates and metabolites. Journal of Analytical Toxicology, 34, 32– 38. 20. Bertisch, S.M., Herzig, S.J., Winkelman, J.W., Buettner, C. (2014) National use of prescription medications for insomnia: NHANES 1999– 2010. Sleep, 37, 343– 349. 21. Substance Abuse and Mental Health Services Administration, Center for Behavioral Health Statistics and Quality. (2013) Emergency department visits for adverse reactions involving the insomnia medication zolpidem. The DAWN Report. SAMHSA: Rockville, MD, pp. 1–4. 22. Greenblatt, D.J., Harmatz, J.S., Singh, N.N., Steinberg, F., Roth, T., Moline, M.L. et al. (2014) Gender differences in pharmacokinetics and pharmacodynamics of zolpidem following sublingual administration. Journal of Clinical Pharmacology, 54, 282– 290. 23. Farkas, R.H., Unger, E.F., Temple, R. (2013) Zolpidem and driving impairment – identifying persons at risk. New England Journal of Medicine, 369, 689– 691. 24. Logan, B.K., Couper, F.J. (2001) Zolpidem and driving impairment. Journal of Forensic Sciences, 46, 105–110. 25. Coleman, D.E., Ota, K. (2004) Hallucinations with zolpidem and fluoxetine in an impaired driver. Journal of Forensic Sciences, 49, 392– 393.
Special Issue
Determining Zolpidem Compliance: Urinary Metabolite Detection and Prevalence in Chronic Pain Patients David M. Schwope1*, Anne DePriest1, David L. Black1,2, Yale H. Caplan3, Edward J. Cone4 and Rebecca Heltsley1 1 Aegis Sciences Corporation, 515 Great Circle Road, Nashville, TN 37228, USA, 2Department of Pathology, Microbiology and Immunology, Vanderbilt University, Nashville, TN, USA, 3School of Pharmacy, University of Maryland, Baltimore, MD, USA, and 4Department of Psychiatry and Behavioral Sciences, Johns Hopkins School of Medicine, Baltimore, MD, USA
*Author to whom correspondence should be addressed. Email: [email protected]
Zolpidem (Ambienw) is the most prescribed insomnia treatment in the USA; however, little is known about zolpidem metabolite excretion in chronic pain patients. As zolpidem is extensively metabolized in vivo to zolpidem 4-phenyl carboxylic acid (ZCA), metabolite detection may provide improved accuracy for compliance determinations, thereby improving clinical decisions. Zolpidem and ZCA were extracted from 1 mL human urine by mixed-mode solid-phase extraction. Samples were analyzed by LC – MS-MS using positive electrospray ionization with multiple reaction monitoring mode employed for detection and quantification. Gradient chromatographic separation was achieved with a reversed-phase column in a rapid 1.8 min analysis. The assay was linear from 4 to 1,000 mg/L for zolpidem and 4 to 10,000 mg/L for ZCA. Interday recovery (bias) and imprecision (n 5 20) were 100–107% of target and 2.4–3.7% relative standard deviation, respectively. Extraction efficiencies were 78–90%. Pain compliance samples (n 5 3,142) were de-identified and analyzed for zolpidem and ZCA. Zolpidem was detected greater than limit of quantification in 720 specimens (22.9%), while ZCA was detected in 1,579 specimens (50.3%). Only five specimens contained zolpidem alone. ZCA was observed without parent zolpidem in 864 specimens, thereby increasing population detection rates by 27.5%. Addition of a zolpidem metabolite to compliance determinations substantially improved detection for zolpidem intake and also should prove useful in clinical and forensic settings.
Introduction Sleep disorders abound, with an insomnia disorder prevalence rate of 12 – 20% of the general adult population (1, 2). Currently, zolpidem (Ambienw) is the most prescribed sedative – hypnotic for short-term treatment of insomnia and the 15th most prescribed drug in the USA (3). It is an imidazopyridine derivative that binds to the gamma-aminobutyric acid – benzodiazepine receptor complex with relative selectivity for the Type-1 (omega-1) benzodiazepine receptor subtype and shares many similar pharmacological properties to benzodiazepines (4). Zolpidem is subject to abuse and may potentially lead to physical dependence (5, 6), substantiating Schedule IV placement under the Controlled Substance Act. Monitoring zolpidem use in pain patients may be warranted, particularly due to the increased risk of adverse effects with concomitant use of multiple other central nervous system depressants (4). In addition to its prevalence in clinical settings, zolpidem is often implicated in drug-facilitated crimes due to its rapid onset of action, rapid clearance from biological specimens and amnesic properties (7, 8). Zolpidem is extensively metabolized by human cytochrome P450 (CYP) enzymes to inactive metabolites, primarily by CYP3A4 (61%). Additional CYP enzymes contribute to zolpidem
metabolism, including CYP2C9 (22%), CYP1A2 (14%), CYP2D6 (2.5%) and CYP2C19 (0.5%) (9). Less than 1% of a dose is excreted in urine as unchanged drug (10). The major urinary metabolite is zolpidem 4-phenyl carboxylic acid (ZCA), also known as zolpidem metabolite I (Figure 1). This metabolite represents 33– 48% of a dose in urine (11), although its abundance is subject to pharmacokinetic interactions with CYP3A4 substrates that often alter formation rates (12). Rapid zolpidem elimination (t1/2 2.6 h) complicates urinary compliance testing, as detection is limited to samples obtained soon after ingestion. Zolpidem is observed at low microgram per liter concentrations in urine, with peak concentrations of 5–25 mg/L reported 12 h following ingestion of a single 10-mg dose; urinary zolpidem concentrations approached detection limits following 36 h postadministration in this small cohort (7). Lewis and Vine (13) were unable to detect parent zolpidem even 8 h after ingestion of a single dose although ZCA was identified [2 mg/L limit of detection (LOD)] in urine for a period of 72 h following zolpidem administration. Strano Rossi et al. detected parent zolpidem for up to 36 h after ingestion [0.5 mg/L limit of quantification (LOQ)]; ZCA was again detectable for at least 72 h (unknown LOD) in this case (14). Published methods and studies have often focused solely on detection of parent zolpidem, potentially yielding false-negative intake determinations at existing cutoff concentrations. This is problematic for compliance testing in clinical settings as noncompliance often leads to patient censure. Also, in drug-facilitated sexual assault (DFSA) casework victims may not immediately present for evaluation, resulting in undetectable parent zolpidem concentrations at the time of testing. Furthermore, immunoassays currently employed in some laboratories may not cross-react with ZCA, further obfuscating proper determinations (15). For these reasons, inclusion of the primary metabolite in mass spectrometry-based zolpidem analyses likely would yield improved accuracy in detecting intake. To this end, we developed and validated a sensitive LC– MS-MS method for the simultaneous detection of zolpidem and ZCA in human urine. This method is specific and rapid, quantifying both analytes in 1.8 min. This method will be useful to toxicologists in clinical and forensic settings, where zolpidem intake determinations require a sensitive, rapid method for quantification of zolpidem over clinically relevant concentrations.
Experimental Samples Urine samples were obtained from chronic pain patients over the period of February – June 2012 from pain clinics located in
# The Author 2014. Published by Oxford University Press. All rights reserved. For Permissions, please email: [email protected]
Urine analyses Sample preparation Internal standard (zolpidem-d6 and ZCA-d6) was added to each sample to yield 100 mg/L in urine. Solid-phase extraction columns were conditioned with methanol and 0.1 M acetic acid. Specimens (diluted 1 : 1 with 0.1 M acetic acid) were loaded by gravity flow, followed by washing with deionized water, 200 mM hydrochloric acid, methanol and ethyl acetate. Columns were then dried under vacuum (SPE vacuum manifold) for 5 min, and samples were eluted with 5 mL dichloromethane : isopropanol : ammonium hydroxide (78 : 20 : 2). Samples were evaporated to dryness, reconstituted in mobile phase A and submitted for LC–MS-MS analysis.
Figure 1. Zolpidem (A) and ZCA (B).
20 states (AL, FL, GA, IL, KY, LA, MD, MS, NC, NJ, NV, NY, OH, PA, SC, TN, TX, UT, VA and WV). All patients were enrolled in pain management compliance monitoring that included drug testing for licit and illicit drugs. The study was approved by the Essex Institutional Review Board (Lebanon, NJ, USA). Urine specimens were collected into polypropylene containers that were capped, labeled and stored at room temperature until shipment to the laboratory for analysis. The specimens were analyzed at Aegis Sciences Corporation (Nashville, TN, USA).
Instrumentation All experiments were performed on an AB Sciex API 3200 triple quadrupole mass spectrometer with a TurboV electrospray ionization (ESI) source (AB Sciex, Foster City, CA, USA). The mass spectrometer was interfaced with a Shimadzu LC-20ADXR high-performance liquid chromatograph (HPLC) (Shimadzu Corporation, Columbia, MD, USA). Evaporation under nitrogen was completed using a TurboVap LV evaporator from Biotage (Charlotte, NC, USA). Powder standards were measured using an XP-6 analytical balance (Mettler-Toledo, Columbus, OH, USA).
Standards and reagents Zolpidem standards and deuterated internal standards were purchased from Cerilliant (Round Rock, TX, USA). ZCA and ZCA-d6 were from Toronto Research Chemicals (Toronto, ON, Canada). All reagents and solvents were from Sigma – Aldrich (St. Louis, MO, USA) and were HPLC grade or better. Clean-Screen ZSDAU020 solid-phase extraction (SPE) columns (United Chemical Technologies, Bristol, PA, USA) were used to prepare samples. Drug free human urine was evaluated for the absence of zolpidem and ZCA prior to use. 514 Schwope et al.
LC– MS-MS analysis Chromatographic separation was performed using a C18 HPLC column (Thermo Scientific Hypersil Gold, 3 mm, 100 2.1 mm). Mobile phase was 0.1% formic acid in HPLC water (A) and 0.1% formic acid in acetonitrile (B). Gradient conditions began at 20% B, increasing to 95% B by 0.3 min followed by a 0.5 min hold and re-equilibration. Chromatographic run time was 1.8 min; cycle time was 2.1 min (Figure 2). Mass spectrometric data were acquired with positive ESI in multiple reaction monitoring (MRM) mode. MS-MS parameter settings (Table I) were optimized via direct infusion of individual analytes (100 mg/L) in 50 : 50 mobile phase A : mobile phase B at 10 mL/min. Source parameters were as follows: curtain gas, 30 psi; collision-activated dissociation, 5 psi; IonSpray voltage, 1,500 V; source temperature, 4508C; gas 1 (zero air), 40 psi and gas 2 (zero air), 50 psi. Data analyses Peak area ratios of target analytes and their respective internal standards were calculated for each concentration. Urine concentrations were measured in microgram per liter and were normalized to refractometer-determined urine specific gravity (16). Statistical calculations were completed with Microsoft Excel 2007 for Windows (Microsoft Corporation, Redmond, WA, USA) and JMP 9.0 for Windows (SAS Software, Cary, NC, USA). Validation The method was validated for bias, inter- and intrabatch imprecision, specificity, sensitivity, linearity, extraction efficiency, matrix effect, carryover, interference (endogenous and exogenous) and analyte instability. Bias and imprecision were evaluated at three concentrations within the linear dynamic range of each analyte. Intrabatch imprecision (% relative standard deviation [RSD]) ranged from 0.3 to 2.7% for all analytes at all concentrations (n ¼ 5); interbatch imprecision ranged from 2.4 to 4.2% (n ¼ 20). Bias calculated as the percent of target concentrations at low, mid and high quality control (QC) concentrations for each analyte (50, 100 and 250 ng/ mL, respectively), ranged from 100 to 107% of target concentrations (n ¼ 20) and was considered clinically insignificant. Specificity was based on precursor mass/fragment ion and relative retention time. Transition peak area ratios for QC and authentic specimens were required to be within 20% of the calibrator for each respective analyte. Relative retention time (analyte to internal standard) difference for QC and authentic specimens
Figure 2. Representative chromatograms for the (A) drug free (negative) extracted urine, (B) LOQ (4 mg/L in negative urine) and (C) patient specimen containing 120 mg/L zolpidem (1.00 min) and 2,220 mg/L ZCA (0.63 min).
Table I LC –MS-MS MRM Parameters Analyte
Q1 (amu)
Q3 (amu)
Dwell (ms)
DP
EP
CEP
CE
CXP
Zolpidem 1 Zolpidem 2 ZCA 1 ZCA 2 Zolpidem-d6 ZCA-d6
308.3 308.3 338.2 338.2 314.4 344.3
235.2 263.2 265.2 293.2 235.2 265.1
75 75 75 75 75 75
61 61 60 60 61 64
6 6 5 5 6 5
19 19 22 22 39 20
43 27 45 32 49 45
6 6 2 2 6 2
Q1, quadrupole 1; Q2, quadrupole 2; DP, declustering potential; EP, entrance potential; CEP, cell entrance potential; CE, collision energy; CXP, cell exit potential; ZCA, zolpidem 4-phenyl carboxylic acid.
was required to be within +0.01 min compared with the relative retention time of the analyte to internal standard of the calibrator. Alternatively, retention time for each analyte could be within 3% of its respective retention time in the calibrator. Sensitivity was evaluated by determining LOD and LOQ. A series of decreasing concentrations of fortified urine were analyzed to empirically determine LOD and LOQ. LOD was determined as the concentrations with acceptable chromatographic retention time and peak shape, a signal-to-noise ratio of at least 3 : 1 and transition peak area ratios within 20% of the calibrator. LOQ
was the lowest concentrations with acceptable chromatographic retention time and peak shape, a signal to noise of at least 10 : 1, transition peak area ratios within 20% of the calibrator ratio and acceptable bias and imprecision (within at least 20% of target concentration and RSD within at least 10%, n ¼ 6). Analyte linearity was determined with 10 concentrations on 3 separate days. Linearity of the method was investigated by calculation of the least-squares regression line (1/x weighting) and expressed by the squared correlation coefficient (R 2). Linear ranges were 4 – 1,000 mg/L for zolpidem and 4 – 10,000 mg/L for ZCA; these ranges should prove useful for clinical and forensic casework. R 2 values (1/x weighting) were deemed acceptable (R 2 . 0.995) for all analytes. Following initial linearity determination, batches were analyzed using a single-point calibrator at a concentration of 100 mg/L (line forced through zero) due to efficiency with respect to time and resources (17). Precision and accuracy fulfilled a priori requirements across the linear range with the chosen calibrator concentration. The extraction efficiency ( percent) for each analyte was determined at two concentrations (4 and 40 ng/mL) and ranged from 90 to 94% for zolpidem and from 64 to 77% for ZCA. Similar results were obtained for the corresponding deuterated analog, and quantification was not adversely affected. Matrix effect was Zolpidem Metabolite Detection and Prevalence 515
investigated through a postcolumn infusion (‘T’ connector) of the analytes while injecting an extracted blank urine sample. No significant ion suppression or enhancement was observed at the retention time for ZCA. Significant ion enhancement was observed at the retention time of parent zolpidem. However, similar ion enhancement for the deuterated internal standard minimized impact on quantification. In all cases, ion ratios met a priori specifications and QC samples all quantified within acceptable limits. Carryover in a negative specimen following a specimen containing the upper limit of quantification (ULOQ) was assessed. No quantifiable carryover was observed for any analyte. In some cases, traces of zolpidem signal were observed in carryover specimens; however, ion transition ratios were not within 20% of the calibrator and any signal present was less than LOD. For assessment of possible matrix interferences, negative urine samples were collected from six different sources and tested for potential endogenous interferences; no interference was observed in any pool for any analyte. In addition, interferences from over 55 illicit and common therapeutic drugs, metabolites and related compounds were evaluated by adding potential interferents at various concentrations into urine aliquots fortified at the LOQ. Compounds examined included several classes of common licit and illicit drugs, including amphetamines, barbiturates, cocaine, delta-9-tetrahydrocannabinol, benzodiazepines, opioids (natural, semi-synthetic and synthetic compounds and metabolites), muscle relaxants, zopiclone and zaleplon. A compound did not interfere if the LOQ quantified within 20% of target and had stable retention times and correct transition ratios. Of the compounds tested, the methadone metabolite 2ethylidene-1,5-dimethyl-3,3-diphenylpyrrolidine (EDDP) was the only identified interferent. At EDDP concentrations .10,000 mg/L, zolpidem-d6 suppression was apparent; however, the analyte remained unaffected yielding zolpidem quantitation differences .20% from target at low QC concentration. ZCA was not affected by any tested interferent. Finally, 60 additional de-identified pain compliance specimens (containing various licit and illicit analytes) were extracted and analyzed for possible endogenous and exogenous interferences. Of these, one had interference with zolpidem (but not the metabolite); this is believed to be related to EDDP concentrations .10,000 mg/L in this specimen. Analyte stability in urine (n ¼ 5) was evaluated under three conditions: freeze – thaw, autosampler and processed sample stability. For freeze–thaw stability, concentration changes ranged from 22.8 to 7.6% for zolpidem and 21.2 to 10.6% for ZCA. For autosampler stability, concentration changes ranged from 24.7 to 3.9% for zolpidem and 24.4 to 0.4% for ZCA. For processed sample stability, concentration changes ranged from 25.8 to 22.1% for zolpidem and 22.5 to 20.1% for ZCA. Results and Discussion With over 43 million prescriptions filled per year in the USA alone, zolpidem is the most prescribed sleep aid and is widely available (3). However, to date, no study has examined use of zolpidem metabolites for compliance determinations. Zolpidem is extensively metabolized and exhibits short detection windows in urine, potentially yielding false-negative results in clinical and forensic settings. Thus, detection and quantification 516 Schwope et al.
of both parent and metabolite in a rapid, high-throughput assay is a significant advancement for this compound and here we report the impact of metabolite inclusion on compliance determinations.
Samples A total of 3,142 directed analysis urine samples were obtained from chronic pain patients over 5 months. Results were de-identified and examined for zolpidem and ZCA prevalence.
Prevalence of metabolite in urine Of 3,142 specimens, zolpidem was detected greater than LOQ in 720 (22.9%), whereas ZCA was detected in 1,579 (50.3%) (Table II). Two specimens (0.06%) contained zolpidem greater than ULOQ and 45 specimens (1.43%) contained ZCA greater than ULOQ. Normalized median (range) zolpidem and ZCA concentrations were 28.3 (4.08 – 805) mg/L and 2,038 (4.53 – 23,000) mg/L, respectively (Figure 3). Only five specimens (0.16%) contained zolpidem alone (median concentration 488 mg/L). As ZCA was observed without parent zolpidem in 864 samples, addition of this metabolite increased detection rates by 27.5% in this group. We previously reported (18, 19) that inclusion of metabolites for compliance determinations in chronic pain patients is crucial for maximizing accuracy. In the present report, zolpidem positive determinations were increased from 22.9 to 50.3%, representing 864 additional positive identifications of zolpidem use in 5 months when compared with analyzing for zolpidem alone at the same LOQ. Others have noted (13, 14) that zolpidem detection windows can be increased with inclusion of ZCA in routine analysis. While the present method improved detection rates specifically related to inclusion of the metabolite, this study cannot provide information regarding improvement in zolpidem intake detection windows. Further controlled administration studies would be beneficial in more precisely defining these windows under acute and chronic dosing regimens. Administration studies with immediate- and controlled-release zolpidem may also provide additional insight into ZCA production and excretion rates, although the similar elimination half-lives between the products [mean 2.6 and 2.8 h, respectively (4, 11)] suggest minor impact on detection periods or concentrations. Monitoring for zolpidem use or misuse may be prudent for pain practitioners during compliance assessment. According to data from the National Health and Nutrition Examination Survey, more than half of zolpidem users in the USA are using
Table II Pain Compliance Cohort Descriptive Statistics for Samples within the Linear Range (4 –1,000 mg/L for Zolpidem and 4 –10,000 mg/L for ZCA)
Count (.LOQ) Mean Standard deviation Median Minimum Maximum
Zolpidem
ZCA
720 56.2 82.0 28.3 4.08 805
1,532 2,680 2,790 2,040 4.53 23,000
All concentrations are microgram per liter and normalized to urine specific gravity.
cannot be used for impairment determination, detection of these compounds ( principally the metabolite) can provide evidence for zolpidem intake, be it known or unknown. This is particularly important in DFSA casework, where victims may not immediately present for evaluation. Based on these findings, it is recommended to include zolpidem metabolite to minimize false-negative potential in these scenarios. Although we identified numerous specimens containing ZCA with no zolpidem parent, only five samples in the present study contained detectable zolpidem with no corresponding ZCA. Upon further review, two of these samples were from a single individual with samples taken on two separate occasions. This individual was also positive for illicit drugs and had the highest concentrations of zolpidem in the entire cohort. The three remaining samples were investigated further, and we determined that these samples had no metabolites present for all other compounds detected. While substantial CYP enzyme inhibition could yield observed patterns, sample adulteration also should be considered as a possible explanation. Corrected zolpidem concentrations of 13.5 – 1,700 mg/L (4.1 – 1,440 mg/L uncorrected) indicate that some technical knowledge (or luck) would be required for adulteration as only 0.2 – 72 mg of parent zolpidem (0.004 – 1.4% of a typical 5 mg dose) would need to be added to a 50 mL urine sample to produce observed concentrations. Nevertheless, adulteration remains a plausible hypothesis and while metabolite testing identified 99% of zolpidem use in the present study, inclusion of both analytes is recommended during routine analysis for maximum accuracy in compliance determinations.
Conclusions Figure 3. Histograms detailing distribution of observed urine concentrations (mg/L) for (A) zolpidem and (B) ZCA in chronic pain patients with bars indicating frequency and black squares representing cumulative percent.
other sedating medications, of which opioids and benzodiazepines are most common (20). Zolpidem may cause adverse drug reactions including daytime drowsiness, dizziness, hallucinations, agitation or bizarre behavior, sleepwalking and ‘sleep driving’ (4). The high incidence of polypharmacy and significance of adverse effects related to zolpidem may pose increased risk for pain patients. Emergency department visits related to adverse effects of zolpidem have increased 220% from 2005 to 2010; more than half of visits involved concomitant use of other medications, particularly opioids and benzodiazepines. Women and elderly patients (.65 years old) may be at increased risk and have been involved in a disproportionate number of emergency department visits related to zolpidem intake (21). Sex-related pharmacokinetic differences with lower clearance observed in women (22) may be a factor; consequently, in January 2013, the US Food and Drug Administration required manufacturers to lower recommended dosages of zolpidem for females (23). In addition to legitimate widespread use as a sedative –hypnotic, zolpidem also has been implicated in both DFSA (7, 13) and driving under the influence of drugs (DUID) (24, 25) forensic casework. Although urinary zolpidem and ZCA concentrations
Inclusion of ZCA in zolpidem urine assays is useful for accurate determinations in compliance or DFSA/DUID scenarios. Improvement in detection of zolpidem intake is critical as noncompliance can lead to censure in monitoring scenarios; the present study provided a 27.5% increase in detection rates for a chronic pain patient population at the same LOQ. This improvement in detection may also be useful in DFSA/DUID casework where determinations based on zolpidem alone may yield poor detection of ingestion. The present method is sensitive, specific and can detect and quantify both parent zolpidem and its primary urinary metabolite ZCA. The rapid extraction and analysis are beneficial and should be easily adopted in both forensic and clinical laboratories. This new method provides novel insight into zolpidem urinary excretion and will improve our ability to detect zolpidem intake in clinical and forensic drug testing casework.
Acknowledgments The authors thank Elizabeth Baker for library services and the technical staff of Aegis Sciences Corporation for performing analysis of the patient urines.
Funding This work was supported by Aegis Sciences Corporation. Zolpidem Metabolite Detection and Prevalence 517
References 1. Morin, C.M., LeBlanc, M., Belanger, L., Ivers, H., Merette, C., Savard, J. (2011) Prevalence of insomnia and its treatment in Canada. Canadian Journal of Psychiatry, 56, 540–548. 2. Roth, T., Coulouvrat, C., Hajak, G., Lakoma, M.D., Sampson, N.A., Shahly, V. et al. (2011) Prevalence and perceived health associated with insomnia based on DSM-IV-TR: International Statistical Classification of Diseases and Related Health Problems, Tenth Revision; and Research Diagnostic Criteria/International Classification of Sleep Disorders, Second Edition criteria: Results from the America Insomnia Survey. Biological Psychiatry, 69, 592–600. 3. Aitken, M., Kleinrock, M. (2013) Declining medicine use and costs: for better or worse? A review of the use of medicines in the United States in 2012. IMS Institute for Healthcare Informatics: Parsippany, NJ, pp. 1– 50. 4. Highlights of prescribing information: Ambien cr. http://products. sanofi.us/ambien_cr/ambienCR.html (accessed March 2014). 5. Jaffe, J.H., Bloor, R., Crome, I., Carr, M., Alam, F., Simmons, A. et al. (2004) A postmarketing study of relative abuse liability of hypnotic sedative drugs. Addiction, 99, 165–173. 6. Licata, S.C., Mashhoon, Y., Maclean, R.R., Lukas, S.E. (2011) Modest abuse-related subjective effects of zolpidem in drug-naive volunteers. Behavioral Pharmacology, 22, 160– 166. 7. Villain, M., Cheze, M., Tracqui, A., Ludes, B., Kintz, P. (2004) Windows of detection of zolpidem in urine and hair: application to two drug facilitated sexual assaults. Forensic Science International, 143, 157–161. 8. Kintz, P., Villain, M., Dumestre-Toulet, V., Ludes, B. (2005) Drug-facilitated sexual assault and analytical toxicology: the role of LC-MS/MS in a case involving zolpidem. Journal of Clinical Forensic Medicine, 12, 36–41. 9. Von Moltke, L.L., Greenblatt, D.J., Granda, B.W., Duan, S.X., Grassi, J.M., Venkatakrishnan, K. et al. (1999) Zolpidem metabolism in vitro: responsible cytochromes, chemical inhibitors, and in vivo correlations. British Journal of Clinical Pharmacology, 48, 89– 97. 10. Salva, P., Costa, J. (1995) Clinical pharmacokinetics and pharmacodynamics of zolpidem. Therapeutic implications. Clinical Pharmacokinetics, 29, 142–153. 11. Baselt, R.C. Disposition of Toxic Drugs and Chemicals in Man. 9th edition. Biomedical Publications: Ashland, OH, 2011; 1877 pp. 12. Hesse, L.M., von Moltke, L.L., Greenblatt, D.J. (2003) Clinically important drug interactions with zopiclone, zolpidem and zaleplon. CNS Drugs, 17, 513–532. 13. Lewis, J.H., Vine, J.H. (2007) A simple and rapid method for the identification of zolpidem carboxylic acid in urine. Journal of Analytical Toxicology, 31, 195–199.
518 Schwope et al.
14. Strano Rossi, S., Anzillotti, L., Castrignano, E., Frison, G., Zancanaro, F., Chiarotti, M. (2013) UHPLC-MS/MS and UHPLC-HRMS identification of zolpidem and zopiclone main urinary metabolites and method development for their toxicological determination. Drug Testing and Analysis (in press). doi:10.1002/dta.1470. 15. Reidy, L., Gennaro, W., Steele, B.W., Walls, H.C. (2008) The incidence of zolpidem use in suspected DUI drivers in Miami-Dade Florida: a comparative study using immunalysis zolpidem ELISA kit and gas chromatography-mass spectrometry screening. Journal of Analytical Toxicology, 32, 688–694. 16. Cone, E.J., Caplan, Y.H., Moser, F., Robert, T., Shelby, M.K., Black, D.L. (2009) Normalization of urinary drug concentrations with specific gravity and creatinine. Journal of Analytical Toxicology, 33, 1– 7. 17. Peters, F.T., Maurer, H.H. (2007) Systematic comparison of bias and precision data obtained with multiple-point and one-point calibration in six validated multianalyte assays for quantification of drugs in human plasma. Analytical Chemistry, 79, 4967–4976. 18. Cone, E.J., Zichterman, A., Heltsley, R., Black, D.L., Cawthon, B., Robert, T. et al. (2010) Urine testing for norcodeine, norhydrocodone, and noroxycodone facilitates interpretation and reduces false negatives. Forensic Science International, 198, 58 –61. 19. Heltsley, R., Zichterman, A., Black, D.L., Cawthon, B., Robert, T., Moser, F. et al. (2010) Urine drug testing of chronic pain patients. II. Prevalence patterns of prescription opiates and metabolites. Journal of Analytical Toxicology, 34, 32– 38. 20. Bertisch, S.M., Herzig, S.J., Winkelman, J.W., Buettner, C. (2014) National use of prescription medications for insomnia: NHANES 1999– 2010. Sleep, 37, 343– 349. 21. Substance Abuse and Mental Health Services Administration, Center for Behavioral Health Statistics and Quality. (2013) Emergency department visits for adverse reactions involving the insomnia medication zolpidem. The DAWN Report. SAMHSA: Rockville, MD, pp. 1–4. 22. Greenblatt, D.J., Harmatz, J.S., Singh, N.N., Steinberg, F., Roth, T., Moline, M.L. et al. (2014) Gender differences in pharmacokinetics and pharmacodynamics of zolpidem following sublingual administration. Journal of Clinical Pharmacology, 54, 282– 290. 23. Farkas, R.H., Unger, E.F., Temple, R. (2013) Zolpidem and driving impairment – identifying persons at risk. New England Journal of Medicine, 369, 689– 691. 24. Logan, B.K., Couper, F.J. (2001) Zolpidem and driving impairment. Journal of Forensic Sciences, 46, 105–110. 25. Coleman, D.E., Ota, K. (2004) Hallucinations with zolpidem and fluoxetine in an impaired driver. Journal of Forensic Sciences, 49, 392– 393.
