I Case Report Two Cases of Fatal Amlodipine Overdose* Journal of Analytical Toxicology, Vol. 30, June 2006 Jason H. Sklerovl, ~, Barry Levinel, 2, Kathleen M. Ingwersen 3, Patricia A. Aronica-Pollack 2, and David Fowler 2 I Division of Forensic Toxicology, Office of the Armed Forces Medical Examiner, 1413 Research Blvd., Rockville, Maryland 20850; 20ffice of the Chef Medical Examiner, State of Maryland, 111 Penn St., Baltimore, Maryland 21201; and 3Armed Forces Medical Examiner, Landstuhl Regional Medical Center, Landstuhl, Germany [ Al~stract I Two fatal overdoses of the calcium channel blocker amlodipine are described. Postmortem samples were screened for volatiles and therapeutic and abused drugs. Amlodipine was measured by liquid chromatography-atmospheric pressure photoionization-mass spectrometry. The heart blood amlodipine concentrations for the two cases were 2.4 and 0.95 mg/l, and amlodipine was quantified in all other tissues. In the first case, venlafaxine and norvenlafaxine were also found, and the angiotensin receptor antagonist olmesartan was tentatively identified. The concentrations of amlodipine are compared with previously reported fatal and nonfatal overdoses. The medical examiners ruled in both cases that the manner of death was suicide and the causes of death were mixed drug intoxication and amlodipine intoxication. Introduction Amlodipine, R,S,2-[(2-aminoethoxy) methyl]-4-(2- chlorophenyl)-3-ethoxycarbonyl-5-methoxycarbonyl-6-methy]- 1,4-dihydropyridine, is a calcium channel blocker of the dihydropyridine class (Figure 1). Amlodipine is prescribed for the treatment of hypertension and angina pectoris and may have efficacy in the treatment of congestive heart failure (1). Its main site of action is vascular smooth muscle; although it also causes coronary vasodilation. The addition of the amino side A H3 C CH2OCH2CH2NH2 H3C I H 7 NO2 B OOOCH2CH2NCH20~H5 I~ CH3 Figure I. Chemical structure of amlodipine (A) and nicardipine (IS) (B). * Disclaimer: The opinions or assertions contained herein are the private views of the authors and are not to be construed as official or as reflecting the views of the Department of Defense or of the Army, Navy, or Air Force. ~ Author to whom correspondence should be addressed. E-mail: Jason.Sklerov@dc.gov. chain gives amlodipine a basic, hydrophilic nature that is unique to its class. This results in slower absorption (tm~ 6-9 h), a large volume of distribution (21 L/kg), and a longer elimination phase (ht2 = 35--45 h) compared with other dihydropyridines (1). The metabolism of amlodipine proceeds initially through oxidation to the pyridine form with subsequent oxidative deamination and aliphatic hydroxylation to multiple, inactive metabolites (2,3). Approximately 60% of an oral dose is excreted in the urine as metabolites, with less than 10% excreted unchanged. Doses of amlodipine range from 1.25 to 10 rag/daily, with steady-state concentrations reached in 3-7 days depending on the dose (4). Amlodipine is manufactured as a racemic mixture, but only the S-(-) enantiomer possesses the vasodilative effect (5). Maximum plasma concentrations of amlodipine have been reported at around 4 ]~g/l (6,7) for single 5-rag doses and 5-10 ~g/l (8-10) for single 10-rag doses. In 4 subjects variously taking 5-10-rag daily doses over a 4-8-month period, plasma concentrations measured 1 h after administration were between 8.7 and 13.7 l~g/l (11). In 9 angina pectoris patients treated for 7 days with a 5 rag/day amlodipine regimen, mean serum concentrations on the 7th and 56th days were 8.I ]~g/l and 9.0 Fg/L, respectively (4). The analysis of amlodipine has been reported by various techniques including immunoassay (12), high-performance thinlayer chromatography (8), gas chromatography (GC) with electrochemical detection (11) or mass spectrometry (MS) (3), liquid chromatography (LC) (5,9,13,14), LC-MS (3), or LC-MS-MS (6,15-17). This paper describes two fatal overdoses of amlodipine with tissue concentrations determined by an LC-MS method. Case Report 1 A 44-year-old caucasian male, who had a pre-existing heart condition, was found unresponsive in bed at approximately 10:00 p.m. Medical examination revealed no signs of life and apparent signs of lividity and rigor mortis. No resuscitative efforts were made. The subject was pronounced dead at the scene ap- 346 Reproduction (photocopying) of editorial content of this journal is prohibited without publisher's permission.
proximately 30 min after discovery. Five empty bottles of prescription medications were found: two bottles of Benicar (olmesartan medoxomi], 30 tablets per bottle at 40 mg each); one bottle of Norvasc (amlodipine besylate, 30 tablets per bottle at 10 mg each); a second bottle of Norvasc (90 tablets per bottle at 10 mg each); and one bottle of amlodipine (90 tablets at 10 mg each). Two empty bags of prescription medication were also found: one bag was hand-labeled "Amitriptyl', and the other was hand-labeled "Effexn XR". An investigative search of the scene revealed a personal journal with entries implying suicidal risk factors and intent using "pills". Medical records document that the decedent had been prescribed anti-hypertensive medications of differing types. There was no evidence of foul play. An autopsy revealed white foam in the mouth and upper airways, moderate pulmonary congestion and edema, cardiomegaly, moderate to focally severe coronary artery atherosclerosis, and multiple remote contusions with crusted abrasions on the prominent surfaces of the extremities. A large bolus of white tablet residues was recovered from the gastric contents. Case Report 2 A 66-year-old African-American female with a history of depression was found on the floor of her apartment at 11:30 p.m. after a family member had to forcibly enter the residence. Emergency services responded and pronounced the subject dead at 12:15 a.m. Approximately 2 weeks prior, the subject had been prescribed Norvasc for hypertension. The bottle that had originally contained 60 tablets (dose unspecified by investigator) was found empty. A partial autopsy was conducted on the day of the body's discovery. External examination revealed fixed, posterior lividity except for areas exposed to pressure. No evidence of significant recent injury or medical intervention was found. Internal examination of the liver and central nervous system was unremarkable except for the presence of a 2-cm calcified meningioma in the right frontal region of the brain. Methods Materials Amlodipine besylate was supplied by Pfizer (Sandwich Laboratories, Sandwich, U.K.) and had a purity of 99.8%. Nicardipine hydrochloride, ammonium formate, and sodium borate decahydrate were obtained from Sigma-Aldrich (St. Louis, MO). Formic acid was obtained from ICN Biomedicals (Aurora, OH). All solvents were high-performance liquid chromatography (HPLC) grade and purchased from Fisher Scientific (Pittsburgh, PA). Sample preparation Blood, urine, gastric contents, and tissues were collected at autopsy and stored, unpreserved, at -15~ A stock solution of amlodipine was prepared at a concentration of I mg/ml in methanol, and the nicardipine internal standard was prepared at a concentration of 0.01 mg/ml in methanol. Both standards were stored at -15~ in amber bottles. Standard curves were prepared in blood and urine at 0.1, 0.5, 1.0, 2.5, 5.0, and 10.0 mg/l for amlodipine. Two-hundred-microliter volumes of blood, urine, and bile were initially assayed, with the final quantitation of bile based on a dilution. Tissue samples and gastric contents (1.0 g) were homogenized in 25 ml of saturated sodium borate buffer using a Brinkmann (Westbury, NY) PT3000 tissue homogenizer. Twohundred-milligram aliquots of tissue homogenate were extracted. Two-hundred microliters of blood or urine calibrators and 200 IJL of the appropriately diluted specimens were added to clean, labeled 16- x 100-ram tubes, and I ml saturated sodium borate was added. Twenty microliters of the nicardipine internal standard (1.0 mg/l final concentration) were added to each tube along with 2 ml of ethyl acetate. The tubes were capped and mixed for 10 rain on an orbital mixer. After centrifuging the tubes for 5 min at 3000 rpm, the solvent was transferred to 10-mL conical tubes and evaporated to dryness under nitrogen at 40~ The sample residue was reconstituted with 50 IJL of HPLC mobile phase and transferred to autosampler vials. Instrumentation Biological extracts were analyzed using an Agilent 1100 LC-MS system (Palo Alto, CA) consisting of a vacuum degasser (G1379A), quaternary pump (G1311A), autosampler (G 1367A), and thermostatted column compartment (G1316A). The MS (G1956B) was equipped with an atmospheric pressure photoionization (APPI) interface (Syagen, Tustin, CA). Separation was performed using a Zorbax StableBond-C18, highthroughput cartridge column (30 x 2.1 mm, dp = 1.8 ljm, Agilent) held at 30~ The mobile phase consisted of ammonium formate buffer and acetonitrile [20mM, ph 4.5 (adjusted with 10% aqueous formic acid), 65:35 (v/v)]. The flow rate was 0.6 ml/min; the injection volume was 1 tjl The positive ions of amlodipine (rn/z 409, 294, and 238) and nicardipine (m/z 480 and 315) were formed by APPI. Pneumatic-assisted nebulization utilized nitrogen at 60 psi, and 350~ nitrogen drying gas was used at a flow of 6 L/min. The vaporizer temperature was 325~ and the APPI's krypton lamp emitted energy at 10 and 10.6 ev. The use of a dopant chemical for enhanced sensitivity was explored by post-column infusion of either acetone or toluene at 0.06 ml/min from a KDS100 syringe pump (KD Scientific, Holliston, MA) into the HPLC mobile phase stream. Neither dopant resulted in significant enhancement of the response across the range of the standard curve and, therefore, none was added for the analyses. Quantitation was based on multi-point, internal standard linear regression. The peak-area ratio of amlodipine (rn/z 238) to that of the internal standard (m/z 480) was used to calculate the response factor. Identification was based on retention time matching within _+ 2% and ion ratios matching within + 20% to the mean values of all calibrators. 347
Results The urine specimen from case 1 was analyzed for drugs of abuse by enzyme-linked immunoassay and for therapeutic and abused drugs by GC-MS, following an alkaline drug extrac- il... ~i Nicarcb'plne(ISTD) ~.,~, ~ Nic,~dpine (ISTD) 0.5 1 1.5 2 Time (min) ~ 1,481- Nicardil~ne (ISTD) tion. The heart blood and vitreous were tested for volatile chemicals, including ethanol by headspace GC. No volatile compounds were detected. Venlafaxine and norvenlafaxine were identified in the alkaline urine extract and were confirmed in an alkaline heart blood extract at concentrations of 0.3 mg/l for venlafaxine and 1.0 mg/l for norvenlafaxine. The heart blood for case 2 was tested for volatile substances, and the heart blood and bile specimens were tested for therapeutic and abused drugs. This included volatile testing for methanol, ethanol, acetone, and isopropanol by headspace GC, acid/neutral drug testing by GC-nitrogen-phosphorus detection (NPD), alkaline drug testing by GC-NPD, morphine by immunoassay, and acetaminophen, ethchlorvynol, and salicylate by color test. No volatile substances were detected in the heart blood. The alkaline drug screen identified doxylamine and dextromethorphan in the bile; the presence of each drug was confirmed by 2.5 GC-MS. Neither drug was detected in the heart blood at a limit of quantitation of 0.05 mg/l. Because of the history surrounding these cases, a special assay was developed for amlodipine. The assay was able to accurately quantify amlodipine with a limit - W of quantitation of 0.1 mg/l and limit of detection of 0.025 mg/l in whole blood. With an increase in both sample size and injection volume, the method could be adapted to measure therapeutic concentrations as well. All specimens were found to contain amlodipine upon confirmation by LC-MS (Figure 2), and the concentrations appear in Table I. 0.5 1 1.5 2 2.5 Time (rain) Figure 2. Extracted ion chromatograms of the liver extract from case 1 (A) and a negative blood sample (g). Discussion Physical symptoms of amlodipine overdose are nonspedfic and may go unreported. These include headache, nausea, Table I. Tissue Distributions for Amlodipine Overdoses Head Peripheral Stomach Blood Blood Bile Brain Kidney Liver Lung Spleen Contents Urine (mg/l) (mg/l) (mg/l) (mg/kg) (mg/kg) (mg/kg) (mg/kg) (mg/kg) (mg/kg) (mg/l) Case 1 2.4 NA* 77 5.4 39.9 68.8 136.4 64.4 583 1.1 Case 2 0.95 0.87 69 NA NA 91.1 NA NA NA NA * NA = not available. 348
Journal of Analytical Toxicology, VoL 30, lune 2006 fatigue, abdominal pain, and edema. Because of the long halflife of amlodipine, compared with other calcium channel blockers, the onset of action is slower and acute side effects (hypotension, pulmonary edema, and bradycardia) may be delayed. Reports of amlodipine overdose have typically been associated with suicidal intent and have involved doses of between 50 and 500 rag. Koch et al. (18) reported a maximal serum concentration of 0.185 mg/l in a 63-year-old woman at 11 h after ingesting 70 mg of amlodipine and an unknown amount of oxazepam (serum = 5.25 rag/l). Treatment included gastric lavage, activated charcoal, calcium gluconate, and vasopressors; however, death occurred 26 h after the ingestion. Cosbey and Carson (19) related a postmortem blood concentration of 2.7 mg/l in a 15-year-old female who had intentionally ingested 140 mg of amlodipine and an unspecified dose of mefenamic acid. Death occurred 6 h postingestion, and autopsy found the lungs congested and edematous with the gastric contents containing a little over two 10-rag doses of residual amlodipine. Stanek et al. (20) measured a serum level of 0.088 mg/l at 2.5 h after the intentional ingestion of 50-100 mg of amlodipine by a 42-year-old woman, and a similar value of 0.067 mg/l was measured in the serum of a 76-year-old man by Adams and Browne (21) following the accidental ingestion of 100 mg. Though both of these ingestions were nonfatal, the latter patient died of complications weeks later. A failed suicide attempt was reported by Yuan et al. (22) for a 37-year-old male who ingested amlodipine (6.7 mg/kg, 470 mg estimated dose), atenolol, and alprazolam. A peak serum level of 0.13 mg/l was measured 12 h postingestion and was probably affected by resuscitative efforts, including gastric lavage and activated charcoal. Recently, Johansen and Genner (23) related a fatal amlodipine overdose by a 50-year-old male. The deceased was discovered with moderate decomposition, and his postmortem blood and liver amlodipine levels were 2.2 mg/l and 8.7 mg/kg, respectively. The authors could not definitively assign an acute overdose based on the presence of only 0.1 mg of amlodipine in the stomach contents. In case 1, the significance of two empty bottles of the antihypertensive angiotensin receptor antagonist, olmesartan medoxomil, was considered. Olmesartan medoxomil is the inactive prodrug of the active, de-esterified olmesartan. No reference standard could be obtained for measurement of olmesartan; however, full scan LC-MS analysis of the urine specimen gave tentative identification that matched well with a published mass spectrum (25). The two cases presented herein correlate well with reported postmortem blood concentrations for amlodipine overdoses. Comparisons with fatal and nonfatal serum levels are complicated because of an unknown serum to whole blood ratio; however, the blood values are far in excess of 0.001-0.025 mg/l therapeutic plasma concentrations (24). The residual gastric load of amlodipine calculated from case I corresponds to 47 mg or approximately 5 doses of amlodipine remaining after the removal (at autopsy) of a bolus of undigested medication. The large liver concentrations seen for both cases suggest significant tissue sequestration and, when viewed along with the additional tissue values in case 1, is consistent with the expected large volume of distribution. Evidence for postmortem redis- tribution was unavailable because of the absence of peripheral blood collection in case I and the similarity of central and peripheral blood concentrations in case 2. However, the potential exists for redistribution to have occurred in case I because of the large undigested quantity of drug remaining at autopsy. The medical examiner ruled the cause of death in the first case as mixed drug intoxication with atherosclerotic coronary vascular disease as a possible contributing factor. The manner of death was ruled suicide. The second case was determined to be a suicide, with the cause of death being amlodipine intoxication. Acknowledgments This work was funded in part by the American Registry of Pathology, Washington, D.C. 20306-6000. References 1. M. Haria and A.J. Wagstaff. Amlodipine: A reappraisal of its pharmacological properties and therapeutic use in cardiovascular disease. Drugs 50:560-586 (1995). 2. A.P. Beresford, D. McGibney, MJ. Humphrey, P.V. Macrae, and D.A. Stopher. Metabolism and kinetics of amlodipine in man. Xenobiotica 18" 245-254 (1988). 3. A.P. Beresford, P.V. Macrae, D. Alker, and R.J. Kobylecki. Biotransformation of amlodipine: Identification and synthesis of metabolites found in rat, dog and human urine--confirmation of structures by gas chromatography-mass spectrometry and liquid chromatography-mass spectrometry. Arzneimittelforschung 39: 201-209 (1989). 4. K. Watanabe, Y. Ochiai, T. Washizuka, T. Inomata, Y. Miyakita, M. Shiba, T. Izumi, A. Shibata, Y.L. Qu, and T. Nagatomo. Clinical evaluation of serum amlodipine level in patients with angina pectoris. Gen. Pharmacol. 27" 205-209 (1996). 5. J. Luksa, D. Josic, M. Kremser, Z. Kopitar, and S. Milutinovic. Pharmacokinetic behaviour of R-(+)- and S-(-)-amlodipine after single enantiomer administration. ]. Chromatogr. B 703:185-193 (1997). 6. M. Carvalho, C.H. Oliveira, G.D. Mendes, M. Sucupira, M.E.A. Moraes, and G. De Nucci. Amlodipine bioequivalence study: Quantification by liquid chromatography coupled to tandem mass spectrometry. Biopharm. Drug Dispos. 22:383-390 (2001). 7. J.Y. Park, K.A. Kim, G.S. Lee, P.W. Park, S.L. Kim, Y.S. Lee, Y.W. Lee, and E.K. Shin. Randomized, open-label, two-period crossover comparison of the pharmacokinetic and pharmacodynamic properties of two amlodipine formulations in healthy adult male Korean subjects. Clin. Ther. 26" 715-723 (2004). 8. K.K. Pan@a, M. Satia, T.P. Gandhi, I.A. Modi, R.I. Modi, and B.K. Chakravarthy. Detection and determination of total am- Iodipine by high-performance thin-layer chromatography: a useful technique for pharmacokinetic studies. J. Chromatogr. B 667: 315-320 (1995). 9. G. Bahrami and S. Mirzaeei. Simple and rapid HPLC method for determination of amlodipine in human serum with fluorescence detection and its use in pharmacokinetic studies. J. Pharm. Biomed. Anal. 36:163-168 (2004). 10. F. Abad-Santos, J. Novalbos, M.A. Galvez-Mugica, S. Gallego- Sandin, S. Almedia, F. Vallee, and A.G. Garcia. Assessment of sex differences in pharmacokinetics and pharmacodynamics of am- Iodipine in a bioequivalence study. PharmacoL Res. 51:445-452 (2005). 11. S.C. Monkman, J.S. Ellis, S. Cholerton, J.M. Thomason, R.A. Sey- 349
mour, and J.R. Idle. Automated gas chromatographic assay for amlodipine in plasma and gingival crevicular fluid. J. Chromatogr. B 678:360-364 (1996). 12. K. Matalka, T. EI-Thaher, M. Saleem, T. Arafat, A. Jehanli, and A. Badwan. Enzyme linked immunosorbent assay for determination of amlodipine in plasma. ]. Clin. Lab. Anal. 15:47-53 (2001). 13. S. Tatar and S. Atmaca. Determination of amlodipine in human plasma by high-performance liquid chromatography with fluorescence detection. J. Chromatogr. B 758:305-310 (2001). 14. K. Shimooka, Y. Sawada, and H. Takematsu. Analysis of am- Iodipine in serum by a sensitive high-performance liquid chromatographic method with amperometric detection. J. Pharm. Biomed. Anal. 7:1267-1272 (1989). 15. A. Marzo, L. Dal Bo, P. Mazzucchelli, N.C. Monti, F. Crivelli, S. Ismaili, M.R. Uhr, and P. La Commare. Amlodipine bioequivalence achieved with a very sensitive liquid chromatography tandem mass spectrometric bioassay. Arzneimittelforschung 50' 688-694 (2000). 16. B. Streel, C. Laine, C. Zimmer, R. Sibenaler, and A. Ceccato. Enantiomeric determination of amlodipine in human plasma by liquid chromatography coupled to tandem mass spectrometry. J. Biochem. Biophys. Methods 54:357-368 (2002). 17. A.B. Baranda, C.A. Mueller, R.M. Alonso, R.M. Jimenez, and W. Weinmann. Quantitative determination of the calcium channel antagonists amlodipine, lercanidipine, nitrendipine, felodipine, and lacidipine in human plasma using liquid chromatography- tandem mass spectrometry. Ther. DrugMoniL 27:44-52 (2004). 18. A.R. Koch, D.P. Vogelaers, J.M. Decruyenaere, B. Callens, A. Verstraete, and W.A. Buylaert. Fatal intoxication with am- Iodipine. J. ToxicoL Clin. ToxicoL 33:253-256 (1995). 19. S.H. Crosbey and D.J.L. Carson. A fatal case of amlodipine poisoning. J. Anal. Toxicol. 21:221-222 (1997). 20. E.J. Stanek, C.E. Nelson, and D. DeNofrio. Amlodipine overdose. Ann. Pharmacother. 31:853-856 (1997). 21. B.D. Adams and W.T. Browne. Amlodipine overdose causes prolonged calcium channel blocker toxicity. Am. ]. Emerg. Med. 16: 527-528 (1998). 22. T.H. Yuan, W.P. Kems, C.A. Tomaszewski, M.D. Ford, and J.A. Kline. Insulin-glucose as adjunctive therapy for severe calcium channel antagonist poisoning. Clin. ToxicoL 37:463-474 (1999). 23. S.S. Johansen and J. Genner. A fatal case of amlodipine poisoning. ]. Clin. Forensic Med. 10:169-172 (2003). 24. R.C. Baselt. Disposition of Toxic Drugs and Chemicals in Man, 7th ed. Biomedical Publications, Foster City, CA, 2004, pp 59-60. 25. N. Kobayashi, I. Fujimori, M. Watanabe, and T. Ikeda. Real-time monitoring of metabolic reactions by microdialysis in combination with tandem mass spectrometry: hydrolysis of CS-866 in vitro in human and rat plasma, livers, and small intestines. Anal Biochem. 287:272-278 (2000). Manuscript received January 6, 2006; revision received February 14, 2006, 350