Amiodarone is a frequently described drug used for the treatment of atrial and ventricular arrhythmias. Yet after >25 years of clinical experience, knowledge of its mode of action and the activity of its metabolite or metabolities in humans is still incomplete. In practice, the use of amiodarone is difficult due to its long elimination half-life (t½: as long as 68 days) (1,2) and the delayed onset of antiarrhythmic action. In humans, amiodarone is eliminated mainly by hepatic metabolism, and its bioavailability varies between 40 and 60% (3,4). Several metabolites are formed; however, only N-monodesethylamiodarone (DEA) has been detected in humans (5,6). The elimination t½ and the blood concentration of DEA have been shown to be comparable to those of amiodarone in long-term therapy (2). In an animal model, Nattel and colleagues (7) demonstrated that DEA has a more potent antiarrhythmic action than amiodarone and may thus have effects additive to those of the parent drug.
Interaction of amiodarone with a variety of other therapeutic agents has been reported, including digoxin, anticoagulants, other antiarrhythmic drugs, and calcium antagonists (8). The primary mechanism of interaction is believed to be the inhibition of hepatic metabolism, which alters the pharmacokinetics and pharmacodynamics of the coadministered drugs. In some clinical conditions, such as local anesthesia or ventricular tachycardia (VT), a patient receiving amiodarone therapy may also receive lidocaine. An interaction between amiodarone and lidocaine has been observed in vivo. Siegmund and associates (9) reported that 65 h after initiation of the amiodarone therapy, lidocaine was cleared from the patient with a t½ of 7 h, which is longer than the 2-3 h measured in healthy individuals (11). The blood concentration of lidocaine (12.6 mg/L) in the report of Siegmund and associates (9) was more than twice the therapeutic level 1.5-5 μg/ml that results from usual lidocaine dose. Keidar and co-workers (10) reported that severe sinus bradycardia and a long sinoatrial arrest occurred in a patient on amiodarone receiving 15 ml 2% lidocaine for local anesthesia. In contrast, Nattel and colleagues (12) reported no interference between lidocaine and amiodarone in cardiac patients. We therefore reinvestigated the problems of amiodarone-lidocaine interaction.
The aim of our study was twofold: to elucidate the mechanism of interaction between amiodarone and lidocaine in vitro by using human liver microsomes and then to evaluate the consequence of the interaction in vivo. Such knowledge could permit better understanding of amiodarone kinetics, side effects, and mechanism of drug interaction on a molecular level.
MATERIALS AND METHODS
In vitro studies
Amiodarone (2-butyl-3-[3,5 diiodo-4-(β- diethylaminoethoxy)-benzoyl] hydrochloride, N-monodesethylamiodarone (2-butyl-3-[3,5 diiodo-4-(β-ethylaminoethoxy)-benzoyl] benzofuran) hydrochloride (DEA), and internal standard L8040 (2-butyl-3-[3,5 dibromo-4-(2-diethylaminopropoxy)-benzoyl] benzothiophene) were supplied by Sanofi (CH-4142 Munchenstein, Switzerland). Lidocaine, monoethylglcinexylidine (MEGEX), and internal standard (N-ethyl,-methyl-glycinexylidine) were obtained from Astra Läkemedel AB (S-151 Södertälje, Sweden). Diisopropylether, methanol, and ammonium sulfate were purchased from Merck (CH-8029 Zurich, Switzerland). All chemicals were of the highest commercially available degree of purity.
Preparation of human liver microsomes. The microsomes were prepared from the liver of kidney transplant donors as described previously (13). The same preparation was also used in another study of lidocaine (14). The protein content was estimated by the method of Bensadoun and co-workers (15).
Incubation conditions for amiodarone metabolism. Microsomal protein (60 μg) was incubated in phosphate buffer 0.1 M containing 1% human albumin in a final volume of 0.8 ml at pH 7.4 at 37°C in the presence of a NADPH-regenerating system (consisting of isocitrate dehydrogenase 1 IU/1 ml, 2 mM NADP, 5 mM trisodium isocitrate, 5 mM magnesium chloride). Amiodarone hydrochloride (8 mg) was dissolved in 0.5 ml ethanol and diluted with water to the desired concentration and added as a twofold concentrated solution. The reaction was incubated at 37°C for 0.5 min. Thereafter, it was started by addition of NADPH. The reaction was stopped by addition of 0.5 ml diisopropylether and then vortexed vigorously for 20 s. After addition of 0.3 μg internal standard L8040 in 0.1 ml methanol, amiodarone and its related compounds were extracted on a reciprocating shaker for 10 min. The reaction was then centrifuged at 2,000 g for 5 min. One volume of 50-150 μl organic phase was injected into the chromotgraph. All experiments were performed in duplicate. Separate assays have demonstrated that the reaction rate was linear with microsomal protein between 20 and 100 μg and time of 2-15 min. Therefore, 60 μg protein and a time of 7 min frequently used in the standard assay.
To investigate the interindividual variations of DEA formation in humans, microsomes from 11 human liver samples were assayed with an amiodarone concentration of 150 μM. We investigated the substrate-dependent formation kinetics of DEA by incubating amiodarone (10-200 μM) with human liver microsomes. The reaction rates were plotted against the substrate concentrations, and the classic equation of Michaelis-Menten kinetics was used to fit the data. The Km and Vmax were estimated directly by using ELSFIT software (16).
Stability of DEA. We tested the stability of DEA in vitro by replacing amiodarone with DEA (0.75, 1.5, 3.2 μM) in the standard biotransformation assay of amiodarone. Incubation was performed between 5 and 60 min. To study the influence of DEA on amiodarone metabolism, the experiment was repeated in the presence of amiodarone (1, 2, 5 μM). The variation in DEA concentrations was compared with that of the control containing no NADPH.
Inhibitory effect of DEA. We studied the ability of DEA to inhibit lidocaine metabolism by coincubating DEA (0.0, 0.32, 0.75, 3.2 μM) with lidocaine (60 μM) in the lidocaine metabolism reaction (described herein). The formation rate of MEGEX was then quantified.
Immunoquantitation of CYP3A4. The enzyme CYP3A4 in the microsome samples was quantified with the monoclonal antibody 13-7-10 (17) as described in an earlier study (14).
In vitro interaction between lidocaine and amiodarone. Protein from human liver microsomes KDL14 (60 μg) was incubated with lidocaine in concentrations of 0.15-1.0 mM and amiodarone in concentrations of 0.010-0.160 mM. After 7-min incubation, 0.4 ml of the incubation medium was removed for the determination of the DEA concentration. The assay for lidocaine metabolism was continued for 30 min as described in the original study (14). The biotransformation of lidocaine was stopped by adding 10 ml perchloric acid 60% and was analyzed by high-performance liquid chromatography (HPLC) (described herein). The apparent inhibition constants (Ki) for lidocaine and amiodarone reactions were determined graphically by the method of Dixon (18).
HPLC assay for amiodarone and its metabolite. The HPLC procedures used for the quantification of DEA and amiodarone were described in detail previously (19). Because the absolute extraction recovery of DEA is quantitative (92%) at neutral pH (19), the extraction of DEA was performed directly at the incubation pH of 7.4. In our hands, the calibration graphs for DEA and amiodarone were linear to 30 μg/ml. The limit of detection was 0.02 μg/ml (29 nM), and the coefficients of variation of the between- and the within-assay were <5%. Lidocaine and MEGEX did not show any interference in this assay.
HPLC assay for lidocaine and its metabolite. For the determination of lidocaine and MEGEX concentrations in the in vitro experiments, the assay reported by Bargetzi and colleagues (14) was used. As much as 0.2 ml of the incubation solution was injected directly onto the HPLC column. Amiodarone and DEA did not interfere in the assay.
In vivo studies
Procedure. The study population consisted of 5 men and 1 woman; in these patients, amiodarone treatment was required because of symptomatic cardiac arrhythmias. None of them had received the drug previously. They were aged 30-74 years (mean 53 years), and all had normal liver and kidney function and no endocronological disorders. Body weight ranged from 56 to 123 kg (mean 79.3 kg). The concomitant diseases and demographic data are summarized in Table 1. The protocol was approved by the Ethics Committee of the University Hospital Zurich.
Because the onset of antiarrhythmic activity of orally administered amiodarone usually occurs after 5-7 days of loading at a daily dose of 600-800 mg, the study was performed in three phases. First, before amiodarone therapy (control phase), 1 mg/kg body weight lidocaine hydrochloride (Xylocaine 1%, Astra, Dietikon, Switzerland) was administered intravenously to each subject in 2 min. Blood samples for the determination of the serum concentrations of lidocaine and its metabolite MEGEX and amiodarone and its metabolite DEA were taken from the contralateral arm through a venous catheter. The blood samples were collected at the following times: before and 5, 10, 15, 30, 60, 120, 180, 240, and 300 min after the lidocaine injection. To avoid changes in liver blood flow, the patients remained supine and were allowed no food during the experiment. Second, the amiodarone therapy was initiated by oral administration of 500 mg amiodarone hydrochloride (Cordarone, Sanofi, Munchenstein, Switzerland) daily. After patients ingested a cumulative dose of 3 g for 6 days (corresponding to the loading period, phase I), lidocaine hydrochloride was injected once more and blood samples were collected as described in the control phase. Third, the therapy was continued by further administration of amiodarone hydrochloride until the total amount of 13 g was reached after 19-21 days (phase II). Lidocaine was then reinjected, and blood samples were collected. The samples were allowed to clot at 4°C; the serum was then separated by centrifugation and stored at -20°C until assayed. ECGs were recorded during all three phases of the investigations.
Analytical assay. The serum concentrations of lidocaine and MEGEX were measured by a modification assay proposed by Oellerich and co-workers (20). In our experiments, the limit of detection (signal to noise ratio <3) for MEGEX and lidocaine was 5 and 10 ng/ml, respectively. The coefficients of variation for the interassay analysis of lidocaine and MEGEX were 5.9-11% at a concentration of 25 ng/ml and 3% for both compounds at a concentration of 200 ng/ml. The accuracy of the assay ranged from 85 to 92% for MEGEX and from 95 to 100% for lidocaine.
Kinetic analysis of lidocaine and its metabolite. All concentrations were expressed with reference to the drug base. The serum concentration-time data of lidocaine of each patient were plotted semilogarithmically with the time and fitted to a two-exponential equation with ELSFIT software (16). Pharmacokinetic parameters were derived by fitting curves by the conventional method (21). Representative parameters were calculated as follows: The t½ of the lidocaine serum concentration was determined by (0.693/β), where β is the slope of the terminal portion of the curve. The systemic clearance (Cl) for lidocaine was calculated by (lidocaine dose/AUCO-∞). The distribution volume at steady-state (Vdss) was calculated as the quotient of systemic clearance and the mean residence time (MRT). The area under the serum concentration-time curve (AUC) of lidocaine and MEGEX between 0 and 300 min was determined directly by the linear trapezoidal rule.
Free concentration of lidocaine. The free concentration of lidocaine in the serum samples was measured 1 h postdose by the ultrafiltration technique of Ha and associates (22).
The results are mean and SD. Within each study phase, the pharmacokinetic data were compared with the corresponding data of the control by a two-tailed, paired Student's t test. Differences were considered statistically significant if the probability of erroneously rejecting the null hypothesis of no difference was <5%. The comparison with the fitted curves was made by an F test (23). The unbound fraction of lidocaine was expressed as the percentage of the total (bound plus unbound) serum lidocaine concentration.
In vitro findings
Under our experimental conditions, the amiodarone biotransformation by human liver microsomes occurred rapidly. The substrate-dependent metabolite formation of amiodarone obeyed Michaelis-Menten kinetics. In the microsomes of three livers (KDL 14, 15, 23), the Vmax and Km were 136 (±67 SD) nmol/h/mg protein and 43.2 (±31.0 SD) μM, respectively. The Eadie-Hoffstee plot suggested that the reaction was monophasic.
To investigate interindividual variations of the formation of DEA, microsomes from 11 different human liver samples were incubated with 150 μM amiodarone, a concentration in which the reaction rate was close to Vmax. In these samples, the DEA formation rates varied as much as eightfold. Amiodarone N-monodesethylase activity correlated well with the amounts of cytochrome P-4503A4 (CYP3A4) quantified by immunoreaction with an appropriate antibody (r = 0.68, p < 0.05, n = 1.1) (Fig. 1A). Moreover, in the same liver samples, the DEA formation rates also correlated well with those of lidocaine N-monodesethylation as determined by Bargetzi and colleagues (14) (r = 0.834, p < 0.001, n = 11) (Fig. 1B).
In vitro lidocaine inhibited DEA formation competitively, with the apparent inhibition constant Ki = 120 μM; vice versa, the lidocaine metabolism was inhibited in the same manner by amiodarone. The apparent inhibition constant of amiodarone (Ki = 47 μM) was 1 order of magnitude lower than the Ki of lidocaine. In comparison to lidocaine, the specific inhibition of CYP3A4 isoform troleandomycin (TAO) inhibited DEA formation with an IC50 = 0.67 μM (data not shown).
Our data show that under the conditions established for amiodarone metabolism, DEA (0.32 and 3.2 μM) was stable for at least 60 min. Moreover, in vitro DEA inhibited the formation of MEGEX in a concentration-dependent manner. As shown in Fig. 2, the MEGEX formation rate from the reaction containing 60 μM lidocaine decreased to 37% of control when DEA (0.32-3.2 μM) was added to this reaction. In the range of the clinically observed concentrations (0.75-3.2 μM or 0.5-2 mg/L), DEA had no effect on amiodarone metabolism.
In vivo interaction between amiodarone and lidocaine
The trough serum concentrations of amiodarone and DEA in 6 patients increased significantly from 1.63 ± 1.28 (mean ± SD) (DEA 0.44 ± 0.15) in phase I to 2.02 ± 1.41 (DEA 1.11 ± 0.42) μg/ml in phase II (amiodarone p = 0.009, DEA p = 0.004). In patient 4, the amiodarone concentration was in the upper limit of the therapeutic range (3-4 μg/ml) (25), however, no side effects were observed. To exclude the possible interference of other substances in analytical assay, we analyzed the serum samples of this patient once more using the reverse-phase HPLC method (26). The results were unchanged. The reason for the relatively high bioavailability in this patient is unknown. The data are summarized in Table 1. The lidocaine serum concentrations of a typical patient (patient 1) measured without amiodarone (control phase) and during the coadministration of amiodarone (phase I and II) are plotted versus time in Fig. 3A. The corresponding variation in lidocaine metabolite MEGEX is shown in Fig. 3B. For the control phase, the average value of lidocaine AUC calculated between 0 and 300 min averaged 111.7 ± 23.2 μg/min/ml (n = 6). After treatment with 3 g amiodarone (phase I), this parameter increased to 135.3 ± 34.6 (p = 0.016), whereas the AUC of MEGEX decreased from 19.2 ± 6.5 to 15.8 ± 8.3 μg/min/ml (p = 0.04). In addition, the systemic clearance of lidocaine decreased from 7.86 ± 1.83 to 6.31 ± 2.21 ml/min/kg body weight (p = 0.003). However, the differences between the elimination t½ (130 ± 19 vs. 155 ± 31 min, p = 0.070), and the Vdss (1.21 ± 0.25 vs. 1.16 ± 0.19 L/kg body weight, p = 0.44) were not significant. The pharmacokinetic parameters of phase II were statistically not different from those of phase I. The data are summarized in Table 2.
Lidocaine serum protein binding
The unbound fraction of lidocaine remained unchanged in the study phases: control 0.198 ± 043 (mean ± SD); phase I, 0.194 ± 0.041; and phase II, 0.203 ± 0.060.
In a previous study using microsomes expressed in the human B-lymphoblastoid cell lines stably transfected with human cDNAs coding specifically for various types of CYP, we showed for the first time the role of CYP3A4 and CYP1A1 in amiodarone metabolism (27). In the present study, the participation of CYP3A4 in amiodarone biotransformation was supported by significant correlations between amiodarone N-monodesethylase activity and (a) the immunochemically determined level of CYP3A4 and (b) the rate of MEGEX formation from lidocaine. Trivier and co-workers (28) and Fabre and associates (29) obtained similar results using other approaches. The Michaelis-Menten constant Km for the amiodarone N-monodesethylation reaction reported by Trivier and co-workers (28) was similar to ours. However, the value reported by Fabre and associates (29) was 1 order of magnitude higher. Because the activity of CYP3A isoform varies considerably between individuals and because the inducible nonhepatic CYP1A1 in another organ or organs may enhance this variation, the amiodarone bioavailability may vary interindividually within a wide range. In addition, other drugs, whose metabolism is mediated by CYP3A4, are candidates for the interference with amiodarone. This phenomenon has been observed in vivo with lidocaine (9,10) quinidine (30), midazolam (31). Our present data demonstrate directly that in vitro amiodarone may inhibit lidocaine metabolism competitively and vice versa. Fabre and associates (29) reported that nifedipine, another substrate of CYP3A4, inhibited the formation of DEA competitively, similar to results in our in vitro study. Its apparent Ki was 38 μM.
In clinical conditions, such as VT, a patient receiving amiodarone therapy may also receive lidocaine (9,10). Due to the drug interaction, the plasma lidocaine concentration may increase to a toxic range, even with a standard dose of intravenous lidocaine infusion or if large doses are used for local anesthesia. Our observations are somewhat in conflict with other data (11). The concentration of a lidocaine metabolite or metabolites was not investigated in any study of the amiodarone-lidocaine interaction (9-11). The pharmacokinetic analysis of lidocaine, including the measurement of lidocaine metabolite MEGEX formation, is important to the understanding of the mechanism of the possible interaction. In the current study, changes in liver blood flow, which may also influence the kinetics of lidocaine, were not apparent. We noted that AUC of lidocaine increased under these conditions, whereas the AUC of its major metabolite MEGEX decreased when amiodarone was administered. We showed that this interaction occurs early during the loading phase, before the onset of the therapeutic effects of amiodarone. After a cumulative amount of 3 g amiodarone hydrochloride (phase I), the mean lidocaine clearance in 6 patients decreased by 19.7%. This change is comparable to that reported by Windle and colleagues (32) and Heimark and co-workers (33) for the procainamide- and warfarin-amiodarone interactions respectively. Moreover, the decrease in lidocaine clearance was accompanied by a decrease in the AUC of its major metabolite MEGEX, suggesting strongly that the metabolism of lidocaine was directly impaired by amiodarone therapy. Earlier studies showed that amiodarone interferes with drugs that have a low hepatic extraction ratio, such as phenytoin (34), procainamide (32), and warfarin (33) by increasing the elimination t½ and AUC of the parent drugs, whereas the formation of the metabolite or metabolites was reduced. Vdss of the affected drugs was unchanged however. In the present study, amiodarone influenced the kinetics of the highextraction drug lidocaine. The changes in lidocaine elimination after a single dose (1 mg/kg i.v. body weight) are not of major clinical significance. However, in other clinical conditions, e.g., when lidocaine is continuously infused, the importance of this interaction might be different. Further studies are therefore needed.
The probable mechanism for the interaction between amiodarone and lidocaine could be the inhibition of the hepatic CYPs (e.g., CYP3A4) by amiodarone, as has been demonstrated in in vitro experiments. Theoretically, the inhibition may be due not only to the direct mutual interaction between the parent drugs (lidocaine/amiodarone), but also to the interaction between their metabolite (or more than one metabolite) and/or parent drugs (e.g., lidocaine-DEA, amiodarone-MEGEX). Therefore, predicting which compound or compounds are mainly involved in the interaction is difficult. However, according to our findings, the role of DEA in the inhibition of the metabolism of coadministered drugs should be taken into consideration.
Although the in vitro inhibition of CYP3A4 was observed with high drug concentration, our data suggest that an interaction between amiodarone and lidocaine may also occur under clinical conditions. The interaction could become especially relevant when lidocaine accumulates in patients with congestive heart failure or hepatic dysfunction.
Acknowledgment: We thank all patients who participated in this study and Schwester Heidy Mattmann, Cardiology Division, for helping in the clinical phase. We also thank Dr. Thomas Buckingham, Cardiology Division, University Hospital Zurich for reviewing the manuscript and H. Boeschenstein-Manner for manuscript preparation.
1. Holt DW, Tucker GT, Jackson PR, Storey GC. Amiodarone pharmacokinetics. Am Heart J
2. Stäubli M, Troendle A, Schmid B, et al. Pharmacokinetics of amiodarone, desethylamiodarone and other iodine-containing amiodarone metabolites. Eur J Clin Pharmacol
3. Nattel S, Talajic M. Recent advances in understanding the pharmacology of amiodarone. Drugs
4. Pourbaix S, Berger Y, Desager JP, Pacco M, Harvengt C. Absolute bioavailability of amiodarone in normal subjects. Clin Pharmacol Ther
5. Flanagan RJ, Storey GA, Holt SW. Identification and measurement of desethylamiodarone in blood plasma specimens from amiodarone treated patients. J Pharm Pharmacol
6. Latini R, Reginato R, Burlingame AL, Kates RE. High performance liquid chromatographic isolation and fast atom bombardment mass spectrometric identification of N
-dedesethylamiodarone, a new metabolite of amiodarone in the dog. Biomed Mass Spectrom
7. Nattel S, Davies M, Quantz M. The antiarrhythmic efficacy of amiodarone and desethylamiodarone, alone and in combination, in dogs with acute myocardial infarction. Circulation
8. Lesko LJ. Pharmacokinetic drug interactions with amiodarone. Clin Pharmacokinet
9. Kaidar S, Grenadier E, Palant A. Sinoatrial arrest due to lidocaine injection in sick sinus syndrome during amiodarone therapy. Am Heart J
10. Siegmund JB, Wilson JH, Imhoff TE. Amiodarone interaction with lidocaine. J Cardiovasc Pharmacol
11. Piper JA, Rodman JH. Lidocaine. In: Evans WE, Schentag JJ, Jusko WJ, Harrison H, eds. Applied pharmacokinetics
, 2nd ed. Spokane, WA: Applied Therapeutics, 1986:639-81.
12. Nattel S, Talajic M, Beaudoin D, Matthews C, Roy D. Absence of pharmacokinetic interaction between amiodarone and lidocaine. Am J Cardiol
13. Meier PJ, Mueller HR, Dick B, Meyer UA. Hepatic monooxygenase activities in subjects with a genetic defect in drug oxidation. Gastroenterology
14. Bargetzi MJ, Aoyama T, Gonzalez FJ, Meyer UA. Lidocaine metabolism in human liver microsomes by cytochrome P450IIIA4. Clin Pharmacol Ther
15. Bensadoun A, Weinstein D. Assay of protein in the presence of interfering materials. Anal Biochem
16. Sheiner LB. ELSFIT: a program for extended least squares fit
. San Francisco: Division of Clinical Pharmacology. University of California. 1983.
17. Beaune P, Kremers P, Letawegoujon F, Gielen JE. Monoclonal antibodies against human liver cytochrome P-450. Biochem Pharmacol
18. Dixon M. The determination of enzyme inhibitor constants. Biochem J
19. Ou CN, Rognerud CL, Duong LT, Frawley VL. Liquid chromatogrpahic determination of amiodarone and N
-desethylamiodarone in serum. Clin Chem
20. Oellerich M, Raude E, Burdelski M, et al. Monoethylglycilidine kinetics: a novel approach to assessment of liver function. J Clin Chem Clin Biochem
21. Wagner JG, ed. Fundamentals of clinical pharmacokinetics
. IL: Drug Intelligence Publications, 1979.
22. Ha HR, Vozeh S, Follath F. Measurement of free lidocaine serum concentration by equilibrium dialysis and ultrafiltration techniques: the influence of pH and heparin. Clin Pharmacokinet
23. Motulsky HJ, Ransnas LA. Fitting curves to data using nonlinear regression: a practical and nonmathematical review. FASEB J
24. Storey GCA, Adams PC, Campbell RWF, Holt DW. Highperformance liquid chromatographic measurement of amiodarone and desethylamiodarone in small tissue samples after enzymatic digestion. J Clin Pathol
25. Harris L, Roncucci R, eds. Amiodarone
. Paris: MEDSI, 1986:147.
26. Stäubli M, Bircher J, Galeazzi RL, Remund H, Studer H. Serum concentrations of amiodarone during long term therapy. Relation to dose, efficacy and toxicity. Eur J Clin Pharmacol
27. Ha HR, Wyss B, Stieger B, Meyer UA, Follath F. Amiodarone metabolism in human liver microsomes [Abstract 5265]. FASEB J
28. Trivier JM, Libersa C, Belloc C, Lhermitte M. Amiodarone N
-deethylation in human liver microsomes: involvement of cytochrome P450 3A enzymes. Life Sci
29. Fabre GF, Julian B, Saint-Aubert B, Joyeux H, Berger Y. Evidence for CYP3A-mediated N
-deethylation of amiodarone in human liver microsomal fractions. Drug Metab Dispos
30. Tartini R, Kappenberger L, Steinbrunn W, Meyer UA. Dangerous interaction between amiodarone and quinidine. Lancet
31. Gascon MP, Dayer P. In vitro forecasting of drugs which may interfere with the biotransformation of midazolam. Eur J Clin Pharmacol
32. Windle J, Prystowsky EN, Miles WM, Heger JJ. Pharmacokinetic and electrophysiologic interactions of amiodarone and procainamide. Clin Pharmacol Ther
33. Heimark LD, Wienkers L, Kunze K, et al. The mechanism of the interaction between amiodarone and warfarin in humans. Clin Pharmacol Ther
34. Nolan PE, Markus F, Hoyer GL, Bliss M, Gear K. Pharmacokinetic interaction between intravenous phenytoin and amiodarone in healthy volunteers. Clin Pharmacol Ther