Journal of Analytical Toxicology, Vol. 28, January/February 2004
Technical Note I
A RapidProcedurefor the Monitoringof Amiodarone and N-Desethylamiodaroneby HPLC-UV Detection loEtta M. luenke 1,*, Paul I. Brown 1, Gwendolyn A. McMillin 1,2, and Francis M. Urry 1,2
tARUP Institute for Clinical and Experimental Pathology, ARUP Laboratories Inc., Salt Lake City, Utah 84108 and 2Department of Pathology, University of Utah Health Sciences Center, Salt Lake City, Utah 84 I32
t Abstract [ This article describesa rapid isocratic high-performance liquid chromatographic (HPLC) method for the simultaneous measurementof the antiarrhythmic drug amiodarone and its potentially active metabolite N.desethylamiodarone (DEA). Following a simple liquid-liquid extraction, amiodarone and its metabolite are quantitated (0.3-6.0 rag/L) by analysison an HPLC-UV system. The analytical time was reduced by 50%, without compromisingthe assayperformance, when RocketTM column technologywas employed. The assay'slimit of quantitation, linearity, imprecision,and accuracy adequately covered the therapeutic range for appropriate patient monitoring. Amiodarone and DEA can be simultaneouslyand accurately quantitated in serum or plasma by HPLC-UV detection with imprecision < 6% at therapeutic concentrations and a quantitation range from 0.3 to 6.0 mg/L. Monitoring of this drug can allow for effective use, while minimizing seriousside effects.
Introduction
Amiodarone (Cordarone| Pacerone| is a predominately Vaughan Williams'class III antiarrhythmic drug introduced in the early 1970s and later approved (1984) in the United States for the treatment of disturbances in the normal heartbeat rhythm generally diagnosed as hemodynamicallyunstable ventricular tachycardiaand ventricular fibrillation.An intravenous formulation was introduced in 1995 for therapy in an urgent care setting. Amiodarone blocks sodium channels and prolongs the duration of the action potential and refractory period of all cardiac fibers. Followingoral administration, amiodarone is slowlyand variably absorbed. Studies have shown bioavailabilityto be between 35 and 65%. Maximum plasma concentrations are attained 3--7 h after single dose with onset of action commonly taking 1-3 weeks. Its large, variablevolume of distribution (approximately 60 IJkg) and its high protein binding (96%) pre* Author to whom correspondenceshould be addressed:loEttaM, luenke, ARUP Institute |or Clinical and ExperimentalPathology,ARUP LaboratoriesInc., 500 ChipetaWay, Salt Lake City, Utah 84108. E-mail:
[email protected].
clude it from being dialyzable.Amiodaroneand, to a greater extent, its metaboliteN-desethylamiodarone(DEA)accumulate in adipose tissue and highly perfused organs such as liver, lung, and spleen (1,2). The pharmalogical activity of DEA,the major metabolite in humans, is debated in the literature. However,it has been noted that a high concentration of DEAcan be correlated with toxicity (3). In chronic therapy, the plasma ratio of metabolite to parent is approximately one. Recent publications have noted altered metabolite concentrations with amiodarone generic substitutions (4) and the dramatic inhibition of amiodarone metabolism by grapefruit juice (5) and other modulators of the CYP3Ametabolic pathway (6). The route of elimination of amiodarone is via hepatic excretion into bile, and some enterohepatic recirculation may occur. Arniodarone has been shown to have a biphasic elimination pattern with an initial half reduction of plasma levels in 2-10 days, followed by a much slower elimination half-life of 26-107 days. This biphasic pattern is likely due to the large volume of distribution, with the drug being rapidly eliminated from highly perfused organs and slowly from less perfused organs, such as adipose tissue, where it has accumulated. DEA has an elimination half-life of approximately 61 days. In absence of a loading dose, steady-state plasma concentrations can take up to 535 days or 180 days with a loading dose. Other potential indications include maintaining the stability of sinus rhythm, secondaryprevention to an additional myocardial infarction in primary survivors, and increased survival time in certain therapy-resistant heart failure patients (7). Adverseside effects reported include hypersensitivityand interstitial alveolarpneumonitis, in which 10% of cases end in fatality. Other adverse effects include impairedvision; hyper or hypothyroidism, possibly related to the drug containing 37.3% iodine by weight; liver injury; and photosensitivity. Plasma concentrations below I mg/L are usually therapeutically ineffective,where as levels greater than 2.5 mg/L are not therapeutically necessary.Adverseeffects may be more related to duration of treatment than to dose (8-13). Because of its narrow therapeutic range (1-3 mg/L) for parent with no established ranges for metabolite and the possibility of serious adverse effects, monitoring of drug concentrations in plasma or serum is useful.
Reproduction (photocopying) of editorial content of this journal is prohibited without publisher's permission.
63
Journal of Analytical Toxicology, Vol. 28, January/Februarv2004
Amiodarone and DEA are generally measured by liquid chromatographic (LC) assays, although an enzyme-linked immunosorbent (ELISA)assay method has been described (14). In some high-performance liquid chromatography (HPLC) assays, a multistep extraction technique using several solvents is used (15,16). Others use protein precipitation methods (17,18), whereas still others use solid-phase extraction techniques (19,20). Some methods have interference from collection tube peaks (21) or, in the case of the ELISAassay, a cross reactivity with metabolites. The need to measure amiodarone and DEA in plasma or serum samples in a simplified manner prompted the development of an LC assay utilizing solvent extraction. The need for a more rapid method lead to the utilization of Rocket column technology, commonly used in the pharmaceutical industry, but not yet commonly applied to therapeutic management.
Methods The assay was performed using an HP 1100 LC (Agilent Technologies, Palo Alto, CA) with autosampler, variable wave detector, and a Perkin Elmer Nelson 1022 integrator (Wellesley, MA) for result recording. The detector wavelength was set at 242 nm. The anayltes initially were separated on a 50 mm x 4.6-ram Phenosphere Cyano (Phenomenex, Torrance, CA) 3-p particle column at a flow rate 0.9 mL/min with a backpressure of approximately 95 bar. Comparatively, the analytes were separated using a Plantinum Cyano 33 mm x 7.0-mm Alltech Rocket column (Alltech, Deerfield, IL) with 3 p 100 A packing
A
at a flow rate of 3.0 mL/min, with a back pressure of approximately 51 bar. The mobile phase consisted of acetonitrile/ methanol/0.05M ammonium acetate at 40:56:3, respectively, and was recycled for up to seven days without chromatogram deterioration. HPLC-grade acetonitrile and methanol were purchased from Fisher Scientific (Hampton, NH). The mobile phase was filtered and degassed prior to use through a 0.45-p nylon membrane under vacuum. The internal standard (IS), L8040 [2 ethyl-3-(3,5-dibromo-4di-n-propylaminopropoxybenzoyl)benzothiophene], amiodarone, and DEA drug stocks were obtained from Sanofi Recherche (Paris, France). The primary stock solutions (100 mg/mL) of amiodarone, DEA,and IS were prepared in methanol. Drug-free plasma was obtained by filtering blood bank plasma though C-18 filters. Filtered plasma calibrators containing both amiodarone and DEA in equal concentrations were prepared over the concentration range of 0.3-6.0 mg/L. The working IS was diluted to a 50-ng/pL concentration in methanol/saline, mixed in a 1:1 ratio. Samples were prepared by transferring 0.25 mL of patient plasma or serum, plasma positive control, plasma negative control (blank),and calibrators to respective microcentrifuge tubes. Then, 20 pL of the working IS was added to each tube, followed by 0.2 mL of 1M sodium phosphate monobasic (pH 4.5). The tubes were mixed by vortex, then 0.5 mL of HPLC-grade methylt-butyl-ether (Fisher Scientific) was added to each tube. The tubes were then capped, vortex mixed for 30 s, and centrifuged for 5 min at approximately 13,000 rotations per minute (RPM). For analysis, 0.2 mL of the organic phase was transferred to an autosampler vial. Injection volumes utilized were 6 pL for the Phenosphere column and 15 pL for the Rocket column. The integrator then uses a ratio of the peak height of the analyte divided by the peak height of the IS to quantitate the analytes of interest from the four-point calibration curve.
B Table I. Summary of Intra- and Interrun Variance for Amiodarone and DEA Using Phenosphere and Rocket Columns Target Value (mg/L)
Amiodarone
DEA
0.6 0.6 2.0 2.0 6.0 6.0
1.00 2.57 0.82 5.46 0.46 2.51
0.55 2.57 0.25 3.7 0.43 1.72
0.6 0.6 2.0 2.0 6.0 6.0
1.10 2.67 0.9 5.54 0.92 2.23
0.59 2.59 0.35 3.78 0.46 1.72
E Phenosphere column (n = 9) Intrarun %CV Interrun %CV Intrarun %CV Interrun %CV Intrarun %CV Interrun %CV
\ Time (min)
Time (min)
Figure 1. (A) Chromatogram obtained with a representative patient sample analyzed with the Phenospherecolumn. Amiodarone elutes at 2.067 rain, followed by DEA at 3.203 min, and IS at 3.917 min. (B) The same sample analyzed with the Rocket column, with amiodarone eluting at 0.727, DEA at 1.037, and IS at 1.520 min, respectively.
64
Rocket column (n = 9) Intrarun %CV Interrun %CV Intrarun %CV Interrun %CV Intrarun %CV Interrun %CV
Journal of Analytical Toxicology, Vol. 28, January/February 2004
Results Figure 1 illustrates an example of a chromatogram from a representative patient sample obtained with both the Phenosphere (1A) and Rocket (1B) columns. The ratios of amiodarone/LS040 and DEA/L8040 were linear between the lower and the upper limits of detection of 0.3 and 6.0 rag/L, respectively, using the criteria of 85-115% of target for accuracy and a coefficientof variation (CV)less than 10% for imprecision.The Table II. Table of Drugs Tested in Interference Study Timolol Nifedipine Itraconazole Flecainide Digoxin Lidocaine and encainide Fluphenazine Oxazepam Carisprodol and metabolite Zonisamicle Phenytoin and metabolites Lamotrigine Phenobarbital Acetaminophen Ibuprofen Nortriptyline Doxepin and metabolite Valproic acid Mephenytoin and metabolite Methsuximide and metabolite Clozapine Pentobarbital Citalopram
Propranolol Cimetidine Mexi[etine Digitoxin Disopyramide Carbamazepine and metabolites Perphenazine Alprazolam Oxcarbazapine Levetiracetam Felbamate Clonazepam Primidone Salicylate Amitriptyline Desipramine Imipramine Topiramate Mycophenolicacid Ethotoin Barbital Quetiapine Risperidoneand serlraline
Table Ilk Direct Comparison of Method with Previously Published Rapid Method Weiret al. (18) Column
Phenomenex Rocket Column Column
C-18/ precolumn Wavelength 244 Mobile phase fresh daily Temperature ambient Flow 1.5 Injection volume 200 pL Run time 7 rain Extraction acetonitrile technique precipitation Inferences many Analytical 0.1-10 measurablerange pg/mL
Phenosphere Cyano 242 recycle 30 days ambient 0.9 6 pL 4.5 min MTBE extraction none 0.3-6.0 pg/mL
Platinum Cyano 242 recycle 30 days ambient 3 15 pL 2.2 min MTBE extraction none 0.3-6.0 pgJmL
Column lifetime
4-6000 injections
4-6000 injections
unknown
assay was tested at three concentrations in triplicate on three separate days during one month to study inter- and intrarun variation. This data is summarized in Table I. Recovery was also determined by comparing the average peak height for six extracted plasma samples prepared to contain 1 mg/L concentration with that for unextracted samples of identical concentration. The mean (• SD) percent recoverieson the Phenospere column were 94 • 4.5%, 93 • 5.2%, and 95 • 2.6%, respectively, for amiodarone, DEA,and L8040. An identical study, performed with the Rocket column, produced mean (• SD) percent recoveries of 95 • 3.5%, 94 • 4.2%, and 96 • 1.6%, respectively, for amiodarone, DEA,and L8040. The linear regression of the data obtained with 30 unidentified patient samples using the Rocket column was 1.02x + 0.01 for amiodarone and 0.99x + 0.03 for DEA,as compared to the Phenosphere column. An alternative IS, promazine (Sigma, St. Louis, MO) at 35 mg/mL,was found to be equivalent to the current IS in both chromatographic properties and retention time (data not shown). No interference or coelution with amiodarone or its metabolite were found with the assay on either column for therapeutic concentrations of drugs found in Table II. No difference was noted in recoveries from plasma and serum (data not shown). A liquid-liquid extraction was chosen to expand column lifetime for the assay. The assay runs for a total time of 4.5 rain from injection to injection on the Phenosphere column. A substantial decrease to 2.2 rain was achieved when Rocket technologywas applied. This technology utilizes a larger column diameter and a shorter length to produce a faster separation. Rocket columns' manufacturing controls the coverage of the bonded phase to the silica backbone. By doing so, a dual-mode separation medium is developed,allowing both polar and nonpolar sites, extending its polar selectivitybeyond other reversephase columns. Being built for speed, Rocket columns run more efficientlyat higher-than-standardsflowrates, but provide excellentpeak shape at lower flowrates. Analytespresent at low concentrations or numerous peaks closely eluting in the chromatogram may limit this technology (22). When compared with a previously published assay (18), our assay shows its merit and cost effectiveness,as shown in Table III. Potential cost savings lie in the lack of a precolumn, mobile phase stability, and column longevity.
Conclusions In conclusion, this paper describes a robust assay for the measurement of amiodarone and its metabolite DEA in serum or plasma. The method analysistime was reduced twofoldby the application of Rocket column technology,without compromise to clinical utility.
References 1. Physicians Desk Reference, 56th ed. Medical Economics Company, Montvale, NJ, 2002, pp 3487-3494. 65
Journal of Analytical Toxicology,Vol. 28, January/February2004 2. T.J.Campbell and K.M. Williams. Therapeutic drug monitoring: antiarrhythmic drugs. Br. J. Clin. Pharmacol. 52:21 S-34S (2001). 3. RT. Pollak. Altered metabolite concentrations with amiodarone generic substitutions cannot be observed without monitoring. Can. J. Cardiol. 17:1159-1163 (2001). 4. S.C. Sauro, D.D. De Carilos, G.L. Pierpont, and C.C. Gornick. Comparision of plasma concentrations for two amiodarone products. Ann. Pharmacother. 36:1682-1685 (2002). 5. C.C. Libersa, S.A. Brique, K.B. Motte, J.F. Caron, L.M. GuedonMoreau, L. Humbert, A. Vincent, P. Devos, and M.A. Lhermitte. Dramatic inhibition of amiodarone metabolism induced by grapefruit juice. Br. J. Clin. PharmacoL 49:373-378 (2000). 6. S. Micuda, M. Hodac, P. Parizek, M. Pleskot, L. Sispera,J. Cerman, J. Malakova, and V. Pidrman. Influence of CYP3A metabolizer status on the pharmacokinetics and pharmacodynamics of amiodarone. Acta Medica 43" 95-101 (2000). 7. B.N. Singh. Amiodarone: the expanding antiarrhythmic role and how to fo]low a patient on chronic therapy. Clin. CardioL 20: 608-618 (1997). 8. J.A. Oates, A.J.]. Wood, and J.W. Mason. Medical intelligence drug therapy amioclarone. N. Engl. J. Med. 31@ 455-466 (1987). 9. R. Falik, B.T. Flores, L. Shaw, G.A. Gibson, M.E. Josephson,and EE. Marchlinski. Relationship of steady-state serum concentrations of amiodarone and desethylamiodarone to therapeutic efficacy and adverse effects. Am. J. Med. 82:1102-1108 (1987). 10. J.J. Heger, E.N. Prystowsky, W.M. Jackman, G.V. Naccarelli, K.A. Warfel, R.L. Rinkinberger, and D.P. Zipes. Amiedarone: clinical efficacy and electrophysiology during long term therapy for recurrent ventricular tachycardia or ventricular fibrillation. N. Engl. ]. Med. 305:539-545 (1981 ). 11. M. Staubli, J. Bircher, H. Galeazzi, and H. Studer. Serum concentrations of amiodarone during long term therapy. Relation to dose, efficacy and toxicity. Eur. J. Pharmacol. 24:485-494 (1983). 12. B.N. Singh, J.T. Collett, and C.Y.C. Chew. New perspectives in the pharmacologic therapy of cardiac arrhythmias. Prog. Cardiovasc. Dis. 22:243-301 (1980). 13. RJ. Dean, K.D. Groshart, J.G. Porteffield, D.H. lansmith, and E.B.
66
14. 15.
16. 17.
18. 19. 20.
21. 22.
Golden. Amiodarone~associated pulmonary toxicity. Am. J. Clin. Path. 87:7-13 (1987). T. Saita, H. Fujito, and M. Mori. Development of an enzymelinked immunosorbent assayfor the quantification of amiodarone. Biol. Pharm. Bull. 25:954-958 (2002). N.D. Mostow, D.L. Noon, C.M. Myers, L. Rakita, and J.L. Blumer. Determination of amiodarone and its N-deethylated metabolite in serum by high-performance liquid chromatography. J. Chromatogr. 227:229-237 (1983). L.J.Lesko, A. Marion, A.T. Canada, and C. Haffajee. High-pressure liquid chromatography of amiodarone in biological fluids. J. Pharm. Sci. 70:1366-1368 (1981). T.A. Plomp, M. Engels, E.O. Robles de Medina, and R.A.A. Maes. Simultaneous determination of amiodarone and its major metabolite desethylamiodarone in plasma, urine and tissuesby high performance liquid chromatography. J. Chromatogr. 273:379-392 (1983). S.J.Weir and C.T. Ueda. Rapid liquid chromatographic determination of amiodarone and its N-deethyl metabolite in plasma, urine and bile. J. Pharm. $ci. 74:460-465 (I 985). P.T.Pollak, S.G. Carruthers, and D.J. Freeman. Simplified liquidchromatographic assay of amiodarone and desethylamiodarone after solid phase extraction. Clin. Chem. 32:890-893 (I 986). G.L. Lensmeyer,D.A. Wiebe, and T. Doran. Application of the Empore solid-phase extraction membrane to the isolation of drugs from blood. I. Amiodarone and desethylamiodarone. Ther. Drug. Monit. 13:244-250 (1991 ). V.V. Murthy. Unusual interference from primary collection tube in a high-performance liquid chromatographic assay of amiodarone. J. Clin. Lab. Anal 11:232-234 (1997). Overcome limitations of time and space with AIItech Rocket columns. AIItech Bulletin #422-B, Alltech, Deeffield, IL, 2001, pp I-8. TM
Manuscript received March 24, 2003; revision received June 25, 2003.
