High-performance liquid chromatographic measurement of amiodarone and desethylamiodarone in small tissue samples after enzymatic digestion

N-nitrosylation potential of mono-N-desethylamiodarone at physiological pH

Amiodarone (AMI) is frequently used for the treatment of supraventricular arrhythmias. The parent drug is rapidly dealkylated to mono-N-desethylamiodarone (MDEA) and the plasma concentrations of AMI and MDEA are comparable. MDEA is a secondary amine and may thus undergo formation to the corresponding N-nitrosamine in combination with coadministered nitrovasodilators. Previous studies have shown that nitrovasodilators release the vasoactive NO? which may nitrosylate thiol or secondary amine groups in aqueous solutions. Therefore, the nitrosylation potential of MDEA at physiological pH was investigated. N-Nitroso-monodesethylamiodarone (NO-MDEA) was synthesized, characterized and used as a reference product for the detection of the corresponding N-nitrosamine. HPLC and NMR results have shown that the NO-MDEA product is an equilibrium of two configurational isomers (syn and anti). NO-release was generated by sodium nitroprusside (SNP) which was exposed to light. The formation to NO-MDEA was assayed by HPLC-UV. It has been found that MDEA is nitrosylated in the higher nanomolar range and that varying oxygenation of the reaction mixture did not significantly affect the reaction yields. The addition of thiols such as serum albumin (0.6mM), l-cysteine (2.5mM) or N-acetylcysteine (2.5mM) inhibited the NO-MDEA formation indicating that they may prevent N-nitrosamine formation in vivo. However, as S-nitrosothiols may also release NO?, in long term exposure to elevated levels of nitric oxide the nitrosylation of secondary amines may be taken into account.

Interaction between amiodarone and lidocaine

We investigated the in vitro and in vivo interaction between amiodarone and lidocaine. The interaction on a molecular level was first studied in microsomes from 11 human livers. Close correlations between amiodarone N-monodesethylase activities and (a) the amounts of cytochrome P-4503A4 (CYP3A4), and (b) the rates of lidocaine N-monodesethylation were observed. Lidocaine inhibited amiodarone N-monodesethylation (Ki = 120 microM) competitively; inversely, amiodarone suppressed lidocaine N-monodesethylase activity in the same manner (Ki = 47 microM). Moreover, the metabolite N-monodesethylamiodarone (DEA) was stable and inhibited lidocaine metabolism in a concentration-dependent manner. The in vivo interaction was investigated in 6 cardiac patients. Each of them received a dose of 1 mg/kg lidocaine hydrochloride intravenously (i.v.) on three different occasions: before amiodarone treatment (control), and after cumulative doses of 3 g (phase I) and 13 g (phase II), respectively, amiodarone hydrochloride. The analysis of lidocaine pharmacokinetics showed an increase in lidocaine area under the curve (AUC) when amiodarone was administered, whereas that of N-monodesethylated lidocaine decreased. Moreover, the systemic clearance of lidocaine decreased, while the elimination half-life (t1/2) and the distribution volume at steady state of lidocaine remained unchanged. The pharmacokinetic parameters during phase II were the same as those during phase 1, indicating that the interaction had already occurred early in the loading phase of amiodarone administration. The interaction between amiodarone and lidocaine may be explained by the inhibition of CYP3A4 by amiodarone and/or by its main metabolite DEA.

A systematic review and critical comparison of internal standards for the routine liquid chromatographic assay of amiodarone and desethylamiodarone

Despite potential adverse effects, clinical use of amiodarone is increasing because of its efficacy in treating arrhythmias. Thus there is a continued need for a rapid, practical amiodarone assay to better study the relationship between serum concentrations and clinical effects and to guide safer dosing. Because the most widely used internal standard, L8040, is no longer available, a systematic comparison of potential alternatives was undertaken based on physicochemical and chromatographic characteristics. All amiodarone assays indexed on Medline were reviewed to produce a list of alternatives and five other potential substances considered based on previous experiences. An isocratic high-performance liquid chromatographic method was modified to allow simultaneous resolution of multiple compounds. The internal standard was expected to perform well in the solid-phase extraction of small sample volumes. No commercially available substances were able to duplicate all the advantages of L8040. Tamoxifen, the most acceptable alternative, was used to develop an assay to measure amiodarone and desethylamiodarone at concentrations as low as 0.25 mg/L in 100 microliters of serum (5 ng detected in a 20 microliters injection). Standard curves were linear over the range of concentrations found in our patients (0.25 to 8 mg/L), within-run coefficients of variation (CVs) averaged 5.3% for amiodarone and 2.9% for desethylamiodarone, and between-run CVs were 4.5% for amiodarone and 1.6% for desethylamiodarone.

Differences in amiodarone, digoxin, flecainide and sotalol concentrations between antemortem serum and femoral postmortem blood

1. The concentrations of amiodarone/desethylamiodarone, digoxin, flecainide and sotalol were measured in serum collected immediately prior to death and in postmortem blood collected from the femoral vein and artery of an 18-year-old male with congenital heart disease who developed a fatal arrhythmia. 2. The concentrations of all four drugs in the sample collected during life were consistent with the dosage given and in the range accepted for normal therapy. 3. There were no differences in amiodarone/desethylamiodarone, flecainide and sotalol concentrations in arterial or venous postmortem blood. 4. The concentrations of desethylamiodarone, digoxin, flecainide and sotalol but not amiodarone, were higher in postmortem blood than in antemortem serum. The flecainide concentration was significantly greater than the upper limit associated with toxicity in life. Without knowledge of the true concentration measured in life, this apparently high, toxic concentration would have suggested that death could have resulted from arrhythmogenic/proarrhythmic effects of the drug in excess. 5. These results further demonstrate the hazards in interpreting postmortem blood concentrations following suspected drug intoxication.

Pharmacokinetics of amiodarone: implications for drug therapy

Amiodarone is a complex molecule with multiple pharmacologic properties and a complex electrophysiologic profile. Its disposition kinetics and relation between plasma drug concentration and efficacy can be analyzed using principles identical to those applicable to other antiarrhythmic drugs. However, the drug's affinity for lipophilic tissues, its extremely slow elimination rate, and the likelihood that some of its effects may not be mediated by the usual antiarrhythmic mechanisms confounds traditional pharmacokinetic analysis. Further data that deal with the fundamental mechanisms of action of the drug, in addition to the nature of the relation between dose and uptake into cellular and subcellular fractions and its pharmacologic effects, will be of value in understanding how the drug exerts salutary actions in cardiac arrhythmias.

Amiodarone hepatotoxicity: prevalence and clinicopathologic correlations among 104 patients

The prevalence of apparent amiodarone-related hepatic injury in 104 patients followed prospectively is compared to that reported in the literature. Asymptomatic elevation of serum aminotransferase levels was detected in approximately one-fourth of the patients, a figure similar to the average of reported cases. The frequency of extrahepatic organ toxicity was increased in patients with elevated levels. Symptomatic "hepatitis" developed in 3% of this series and in less than 1% of cases in the literature. Evidence of hepatic phospholipidosis and the development of pseudoalcoholic liver injury is most likely due to the biochemical effects of the drug and to possible metabolic idiosyncrasy, respectively. Serial blood enzyme measurements, as recommended by the manufacturer, may offer some protection against the development of more serious liver injury. However, levels of amiodarone may persist in various tissues for weeks to months following withdrawal, and stopping the drug does not guarantee the prompt reversal of any organ toxicity. Accordingly, the risks posed and benefits offered by amiodarone should be carefully weighed prior to discontinuing the drug, as the risk of sudden cardiac death may outweigh the hazards of ongoing hepatic, pulmonary or other toxicity.

High performance liquid chromatographic measurement of lignocaine in tissue samples following transabdominal placental biopsy

A method is described for the measurement of lignocaine in small samples of fetal and placental tissue. Tissue samples (ca 100 mg) are digested using a preteolytic enzyme. Lignocaine and an internal standard are extracted into methyl tert-butyl ether and analysed by high-performance liquid chromatography with electrochemical detection (+1.0 V vs Ag/AgCl). The limit of accurate measurement is better than 0.1 mg/kg wet weight for a 100 mg sample. This method has been used to assess fetal exposure to the drug when used as a local anaesthetic during transabdominal placental biopsy (chorionic villus sampling). The range of lignocaine concentrations found in the tissue samples was large (from less than 0.04 mg/kg wet weight to 15.4 mg/kg wet weight) although most samples contained less than 1.0 mg lignocaine/kg wet weight.

Determination of amiodarone hydrochloride in pharmaceutical formulations by derivative UV spectrophotometry and high-performance liquid chromatography (HPLC)

Assay procedures based on derivative ultraviolet spectrophotometry and high-performance liquid chromatography (HPLC) have been developed for the specific determination of amiodarone hydrochloride in pharmaceutical dosage forms. The use of first- and second-order derivative spectrophotometry was found to have suppressed the background absorption from the excipients with comparable accuracy and precision to the reversed-phased HPLC reference method. A conventional UV absorption method (lambda = 242 nm) is subject to possible interference by formulation excipients.

Electrophysiologic effects of desethylamiodarone, an active metabolite of amiodarone: comparison with amiodarone during chronic administration in rabbits

During acute superfusion studies by means of the standard microelectrode technique, we previously showed that both amiodarone and its major metabolite, desethylamiodarone, had a modest effect on the lengthening of the action potential duration (APD) at high drug concentrations and produced a rate-dependent block of the sodium channel in cardiac muscle. In this study the comparative electrophysiologic effects of the two compounds in rabbits treated chronically with these compounds were determined with particular reference to repolarization and sinus node automaticity. The changes were correlated with those in serum and tissue drug levels and in thyroid hormone indices. After 1 week neither compound had a significant effect on atrial or sinus nodal potentials; after 3 weeks, amiodarone increased the atrial APD at 90% repolarization time by 10.5% (p less than 0.05) and the effective refractory period (ERP) by 6.7% (p less than 0.05). The corresponding figures for desethylamiodarone were 13% (NS) and 18% (NS). The sinus cycle length was increased 12% (NS) by amiodarone and 27.9% (p less than 0.05) after the metabolite. In animals treated for 6 weeks, amiodarone increased the ventricular APD at 90% repolarization by 58.8% (p less than 0.01) and desethylamiodarone by 42.0% the corresponding figures for the ERP were 63.4% (p less than 0.01 and 47.4% (p less than 0.01), respectively. At the stimulation frequency used, neither compound exerted a significant effect on Vmax. Both amiodarone and desethylamiodarone significantly decreased serum triiodothyronine and increased reverse triiodothyronine levels but had no effect on thyroxine.(ABSTRACT TRUNCATED AT 250 WORDS)

Myopathy during amiodarone treatment: a case report

The case of a 69-year-old woman who developed muscle weakness on chronic amiodarone treatment is reported. Muscle biopsy showed vacuolar alterations and a marked accumulation of amiodarone and of its metabolite desethylamiodarone was found in the muscle sample despite normal blood levels. Amiodarone-induced morphological alterations of the muscle may be related to the actual tissue concentration rather than to the blood level.

Resting, and rate-dependent depression of Vmax of guinea-pig ventricular action potentials by amiodarone and desethylamiodarone

1 The cellular electrophysiological effects of amiodarone and its metabolite desethylamiodarone (DEA) were studied in guinea-pig ventricular myocardium by use of standard microelectrode techniques. 2 Both compounds produced significant increases in action potential duration (Class III antiarrhythmic effect) and decreases in maximum rate of depolarization (Class I effect), at clinically relevant concentrations. 3 The Class I effects were rate-dependent, with small (0-16%) falls in maximum depolarization rate in the absence of stimulation ('resting block') and progressively larger effects at decreasing interstimulus intervals (range 1200-300 ms). 4 The kinetics of onset and offset of the Class I effect in response to a step change in driving rate were quite fast for both drugs (comparable to those reported for Class Ib agents). 5 It is concluded that this unique combination of Class III action plus Class I effects with fast onset and offset kinetics may help explain the great efficacy of amiodarone in antiarrhythmic therapy.

Sensitive method for the measurement of amiodarone and desethylamiodarone in serum and tissue and its application to disposition studies

A high-performance liquid chromatographic (HPLC) method for the measurement of amiodarone (AM) and its metabolite(s) in serum and tissues was developed. The method uses a 5-micron silica column, methanol containing 0.02% perchloric acid at pH 4 as the mobile phase, and ultraviolet detection at 240 nm. The standard curves for AM and desethylamiodarone (DAM) were linear for serum (range 0.025-6.0 microgram/ml) and tissues (range 0.1-0.5 micrograms for 10-25 mg wet weight). There was a significant decrease as a function of time in AM and DAM concentrations in patients' sera left at ambient temperature in the presence of light. This HPLC method was applied to studies on serum AM elimination kinetics in patients and on tissue uptake during chronic AM administration to rabbits. The elimination half-life (5.8 h) of AM after a 5 mg/kg intravenous dose to a patient was similar to that after acute oral doses. AM, being lipophilic, accumulated maximally in the fat tissue (56 micrograms/g wet weight), followed by lung and liver in rabbits injected with AM for six weeks. The latter two tissues also contained nearly equal quantities of DAM. The high concentrations of AM and DAM in the liver and lungs may be related to the hepatic toxicity and pulmonary fibrosis associated with chronic AM therapy. Two new metabolites were found in the lung and bile of AM-treated rabbits, but these have not yet been identified.

[Amiodarone therapy--behavior of serum and fatty tissue concentrations]

Thirty-eight patients with refractory supraventricular and ventricular tachyarrhythmias were administered a mean oral dosage of 400 mg amiodarone daily (200-600 mg). A high-pressure liquid chromatography method was used to measure serum concentrations of amiodarone and its metabolite desethylamiodarone after one week, one month, three months, and then at 6-month intervals. In 24 patients subcutaneous fatty tissue concentrations were also measured. The mean follow-up was 9 months (4 days to 29 months). A linear correlation was found between amiodarone and its metabolite in serum (r = 0.56, p less than 0.001) as well as in subcutaneous fatty tissue (r = 0.67, p less than 0.001). While serum concentrations were dose dependent, tissue concentrations accumulated during chronic therapy (p less than 0.01, both). Clinical efficacy was achieved in 84% of the patients. No statistically significant difference was found between responders and non-responders as regards serum and subcutaneous fatty tissue concentrations. Side effects of amiodarone occurred in 63%. The incidence of adverse effects was related to significantly higher serum and subcutaneous fatty tissue concentrations of amiodarone and its metabolite (p less than 0.001, both). Thus, although the determination of serum and subcutaneous fatty tissue concentrations does not seem to be helpful for assessing clinical efficacy of this antiarrhythmic drug, these values may predict the occurrence of adverse effects.

Effect of renal failure or biliary stasis on the pharmacokinetics of amiodarone in the rat

The single dose intravenous pharmacokinetics of amiodarone (50 mg/kg) were examined in rats with 72 h of biliary stasis secondary to bile duct ligation compared with paired control animals; and in rats with uranyl nitrate induced acute renal failure compared with paired control animals. Plasma and tissue levels (liver, kidney, heart, and lung) of amiodarone (1) and its N-deethyl metabolite 2 were obtained at 4 and 24 h following drug administration. Pharmacokinetic parameters were derived from plasma samples obtained over a 24-h period. Compared with controls, biliary stasis caused a decrease in the total clearance of 1 (1.74 versus 0.35 L/h/kg) and in the volume of distribution at steady state (21.1 versus 5.0 L/kg); renal failure caused a decrease in total clearance (1.67 versus 0.9 L/h/kg) and an increase in apparent elimination half-life (13.7 versus 10.1 h). Both disease processes produced significantly higher plasma levels of 1 when compared with control animals at 4 and 24 h. However, only the cholestatic animals had consistently higher tissue levels of 1 in the face of elevated plasma levels. In normal rats, no 1 or 2 was detected in the urine after a 50 mg/kg intravenous dose of 1, and less than 0.5% of the total dose of amiodarone (1) was excreted into bile by 12 h.

Amiodarone and its desethyl metabolite: tissue distribution and morphologic changes during long-term therapy

The pharmacokinetic characteristics of amiodarone suggest extensive tissue deposition. We confirmed this by measuring tissue concentrations of the drug and of its major metabolite, desethylamiodarone, in human tissues. These were obtained at autopsy (n = 9), surgery (n = 7), or biopsy (n = 2) from 18 patients who had been treated with amiodarone for varying periods of time. High concentrations of amiodarone were found in fat (316 mg/kg wet weight in autopsy specimens, 344 mg/kg wet weight in biopsy specimens). Amiodarone and desethylamiodarone concentrations (mg/kg wet weight, autopsy samples) were also high in liver (391 and 2354), lung (198 and 952), adrenal gland (137 and 437), testis (89 and 470), and lymph node (83 and 316). We also found high concentrations of amiodarone (306 mg/kg wet weight) and desethylamiodarone (943 mg/kg wet weight) in abnormally pigmented ("blue") skin from patients with amiodarone-induced skin pigmentation. These values were 10-fold higher than those in unpigmented skin from the same patients. These high concentrations were associated with lysosomal inclusion bodies in dermal macrophages in the pigmented skin. The inclusion bodies were intrinsically electron dense and were shown to contain iodine by energy dispersive x-ray microanalysis. Lysosomal inclusion bodies shown by electron microscopy to be multilamellar were seen in other tissues. These tissues included terminal nerve fibers in pigmented skin, pulmonary macrophages, blood neutrophils, and hepatocytes and Kupffer cells. These characteristic ultrastructural findings occur in both genetic lipidoses and lipidoses induced by other drugs, e.g., perhexiline. We conclude that during therapy with amiodarone, widespread deposition of amiodarone and desethylamiodarone occurs. This leads to ultrastructural changes typical of a lipidosis. These changes are seen clearly in tissues associated with the unwanted effects of amiodarone, e.g., skin, liver and lung.

Value of hepatic computerized tomographic scanning during amiodarone therapy

Amiodarone therapy is difficult to monitor because of the poor correlation between plasma amiodarone levels and clinical efficacy or toxicity. Monitoring tissue levels may give a better measure of effectiveness, but tissue levels cannot be easily measured. The iodine-containing amiodarone and its major metabolite, desethylamiodarone (DA), are highly tissue-bound, and it has been shown that computerized tomographic (CT) scanning of the abdomen will detect drug deposition in the liver. Ten patients receiving chronic amiodarone therapy were studied by abdominal CT scanning. Liver CT density was increased in 6; 68 to 94 CT units (normal 50 to 65) and liver/spleen relative CT density increased in 5; 1.4 to 2.0 (normal 1.0 to 1.3). Estimates of liver drug levels (based on a calibration curve with inorganic iodide) gave values of up to 3 g of amiodarone and DA per kilogram weight of liver. Absolute and relative liver CT density correlated significantly with plasma levels of DA (r = 0.65, p less than 0.05), but not with amiodarone (r = 0.55, p less than 0.1). No significant correlation was found with QTc intervals. This indirect estimate of liver deposition of amiodarone and DA may prove useful in guiding antiarrhythmic therapy.

Peripheral neuropathy during longterm high-dose amiodarone therapy

Three patients developed peripheral neuropathy after taking amiodarone for more than 18 months. All had high serum concentrations of amiodarone and its desethyl metabolite; in one patient concentrations in a sural nerve biopsy were 80 times higher than in serum. Peripheral neuropathy is a complication of large doses of amiodarone taken over long periods.

Use of amiodarone during pregnancy

Two cases are reported in which amiodarone was administered during pregnancy for longer periods than has been reported previously. Limited placental transfer of amiodarone and its desethyl metabolite was observed in both cases. A normal child resulted from each pregnancy despite, in one case, amiodarone therapy throughout the entire pregnancy. However, caution is urged in the use of amiodarone during pregnancy in view of the limited data available.

Amiodarone for long-term management of patients with hypertrophic cardiomyopathy

Fifty-three patients with hypertrophic cardiomyopathy who had serious arrhythmias (45 patients), refractory chest pain (5 patients) or a high risk of sudden death (3 patients) received amiodarone for 6 to 96 months (median 18) after completion of a loading and an initial maintenance period. The dose of amiodarone was altered by 50 to 200 mg/day at 3- to 6-month intervals, guided by electrocardiographic monitoring, plasma drug level measurements and side-effect questionnaires. Ventricular tachycardia was suppressed in 24 patients (92%) with doses of 100 to 400 mg/day (median 300); none died suddenly during a mean follow-up of 27 months. Although symptomatic episodes of frequent or prolonged supraventricular tachycardia or paroxysmal atrial fibrillation/flutter were abolished in 8 of 9 patients on 100 to 600 mg/day (median 300), in 1 patient incessant atrial flutter developed that was relatively refractory to direct-current cardioversion. In 11 patients with atrial fibrillation, sinus rhythm was restored in 7 (after direct-current cardioversion in 3) with doses of 100 to 600 mg/day (median 300) and has been maintained in 5 with associated improvement in symptoms. Despite discontinuation of beta-blocker therapy, chest pain was unchanged in 17 patients, was impaired in 11 and was worse in only 2. Amiodarone was discontinued in 3 patients; in 1 because of hair loss, in 1 because of neurologic symptoms and in 1 because of facial discoloration; in the latter 2 patients, amiodarone was restarted after 1 and 14 months, and was tolerated and effective at the lower dosage.(ABSTRACT TRUNCATED AT 250 WORDS)

Clinical pharmacokinetics of the newer antiarrhythmic agents

This article reviews clinical pharmacokinetic data on 8 new antiarrhythmic agents. Some of these drugs have been studied extensively while others are relatively new, with incomplete data due to limited evaluation. Amiodarone is a class III antiarrhythmic drug which is effective in treating many atrial and ventricular arrhythmias that are refractory to other drugs. Amiodarone accumulates extensively in tissues and its disposition characteristics are best described by models with 3 and 4 compartments. Its apparent volume of distribution is very large (1300 to 11,000L) and its elimination half-life very long (53 days). A delay of up to 28 days from of treatment to onset of antiarrhythmic effect may be observed, and the antiarrhythmic effect may persist for weeks to months following cessation of therapy. Clinically significant drug interactions have been observed with amiodarone and warfarin, digoxin, quinidine and procainamide. Encainide is a class Ic antiarrhythmic drug. Although it has a short elimination half-life (1 to 3h), 2 major metabolites with antiarrhythmic effects accumulate in the plasma of patients during long term therapy. Plasma concentrations of O-demethyl encainide appear to correlate with the antiarrhythmic effect. Flecainide, another class Ic antiarrhythmic agent, has an elimination half-life of 14 hours which makes it suitable for twice daily dosing. Flecainide elimination is prolonged in patients with low output heart failure. Significant drug interactions with digoxin and cimetidine have been reported. Lorcainide is also a class Ic antiarrhythmic drug, the bioavailability of which is nonlinear. Clearance of the drug is reduced during long term therapy. A major active metabolite, norlorcainide, accumulates in the plasma of patients during long term therapy and its concentration exceeds that of lorcainide by a factor of 2. The elimination half-lives of lorcainide (9h) and norlorcainide (28h) allow for once or twice daily dosing. Mexiletine, a class Ib antiarrhythmic drug, is structurally similar to lignocaine (lidocaine). A sustained release formulation provides effective plasma concentrations when administered twice daily. The apparent volume of distribution of mexiletine is 5.0 to 6.6 L/kg, and the elimination half-life varies from 6 to 12 hours in normal subjects and from 11 to 17 hours in cardiac patients. Mexilitine is extensively metabolised but the metabolites are not pharmacologically active. Renal elimination of mexiletine is pH dependent. Drugs which induce hepatic metabolism significantly alter the pharmacokinetics of mexiletine.(ABSTRACT TRUNCATED AT 400 WORDS)

Clinical use and pharmacology of amiodarone

Amiodarone, an investigational drug in the United States, has had considerable use in this country and worldwide in the treatment of cardiac arrhythmias. This article reviews the clinical pharmacology of this potentially useful antiarrhythmic agent.

The pathogenesis of amiodarone-induced pigmentation and photosensitivity

A new method has been used to measure tissue levels of amiodarone and its major metabolite desethylamiodarone. Amiodarone-pigmented skin has a drug and metabolite concentration ten times that of non-pigmented skin. Iodine-rich amiodarone and its metabolite have been detected in secondary lysosomes by energy dispersive analysis of X-rays. Amiodarone induces a phototoxic reaction with an action spectrum in both the UV-B and UV-A wavelengths.

Amiodarone pharmacokinetics

The single-dose pharmacokinetics of amiodarone have been studied in volunteer subjects given 400 mg doses by the intravenous and oral routes. The data show the compound to have a very large volume of distribution, a low total clearance, and a long and variable terminal elimination half-life. In patients the terminal elimination half-life was on the order of 40 days, with a more rapid phase of elimination in the first few days following the withdrawal of therapy. The terminal elimination half-life of desethylamiodarone was longer than that of the parent compound. High concentrations of amiodarone and its desethyl metabolite were found in tissue samples, with fat forming a potentially large tissue reservoir of the drug.