V. Ranade

Bioavailability and pharmacokinetics of magnesium after administration of magnesium salts to humans.

Therapeutically, magnesium salts represent an important class of compounds and exhibit various pharmacologic actions. Examples of magnesium salts are ionic magnesium and magnesium citrate in nephrolithiasis, magnesium salicylate in rheumatoid arthritis, magnesium hydroxide as an antacid as well as a cathartic, and magnesium mandelate as urinary antiseptic. Various anions attached to the cation magnesium, such as oxide, chloride, gluconate, and lactate, affect the delivery of the amounts of elemental magnesium to the target site and thereby produce different pharmacodynamic effects. This review examines the bioavailability and pharmacokinetics of various magnesium salts and correlates pharmacodynamic action with the structure-activity relationship.

Evaluation of a 12-Lead Digital Holter System for 24-Hour QT Interval Assessment

Background: Drug induced QT prolongation may precipitate life threatening cardiac arrhythmias. Evaluation of the QT prolonging effect of new pharmaceutical agents in a ‘thorough QT/QTc study’ is being mandated by FDA. The purpose of this study was to evaluate an automated 12-lead digital Holter system for a thorough QT/QTc study. Methods: Five healthy volunteers underwent 24-hour digital Holter monitoring. Each recording underwent a fully automated QT analysis (AQA) followed by an onscreen complete manual over read (MOR). Each recording was analyzed twice at least 2 weeks apart. The effect of data sampling (5-min segment/hour), the system sensitivity to detect 5-ms increase in QT, and the ability to assess circadian variation were evaluated. Results: The AQA resulted in identical QT for the first and second analyses, but with obvious errors in QT measurements. Compared to the complete onscreen MOR, the mean QT was longer with AQA (416 ± 41 vs. 387 ± 30 ms, p < 0.001), correlation; r = 0.3. The reproducibility of AQA with complete MOR was very good (QT: 387 ± 30 vs. 387 ± 30 ms, coefficient of variation: 0.2%, r = 0.986. The 5-min mean QT intervals correlated well with the hourly mean QT intervals (r = 0.994, p < 0.001, coefficient of variation = 1 ms) and both showed a similar circadian variation. The system was sensitive to detect a 5-ms change in QT intervals (5 ± 2 ms, coefficient of variation = 0.6%, r = 0.998, p < 0.001). Conclusions: The AQA is not an acceptable method, while the automatic analysis with complete MOR is a highly sensitive and reproducible method. Data sampling by analyzing 5-min segments per hour is sensitive and reproducible.

83 IS THE PH-INDUCED CHANGE IN IKR INHIBITION BY IBUTILIDE ARRHYTHMOGENIC?

Ibutilide (I), a class III antiarrhythmic agent, is employed in conversion of atrial fibrillation and atrial flutter. Ibutilide inhibits the cardiac IKr channel and prolongs the Q-T interval and can give rise to ventricular tachycardia. In previous studies, we have shown that extracellular acidosis significantly attenuates the IKr inhibitory effect of proarrhythmic drugs (quinidine) and that extracellular acidosis has little impact on the inhibitory effect of less proarrhythmic drugs such as amiodarone. We hypothesized that I would behave more like quinidine than amiodarone with extracellular acidosis. Cardiac IKr was studied using human-ether-a-go-go-related gene (HERG) expressed on Xenopus oocytes and two-electrode voltage clamp technique. The recording solution contained 96 mM NaCl, 5.0 mM KCl, 2.0 mM CaCl2, and 5 mM HEPES. The pH of the solution was adjusted to 6.2 or 7.4 to represent the acidic or normal conditions. At pH 7.4, I 0.3, 1, 3, and 10 μM inhibited current by 22 ± 5, 54 ± 5, 80 ± 3, and 93 ± 1%, respectively. When I was applied at pH 6.2, I 0.3, 1, 3, and 10 μM, I decreased HERG current by 10 ± 4, 29 ± 4, 32 ± 8, and 36 ± 5%, respectively. I 30 μM produced only a 48 ± 5% current block at pH 6.2. There were significant differences in the percentage current inhibition by 1, 3, and 10 μM I at normal pH versus pH 6.2 (p values < .01). The IC50 of I was 0.9 ± 0.1 μM at pH 7.4 and the IC50 was increased to 31 ± 6 μM at pH 6.2. Our results indicated that I is a potent IKr inhibitor and that extracellular acidosis markedly attenuated IKr inhibition. Diminished IKr inhibition in the ischemic region with low extracellular pH and potent IKr inhibition in the normal pH regions could result in heterogeneity in action potential duration, which may trigger and sustain arrhythmias and contribute to the proarrhythmic toxicity of ibutilide.

82 INHIBITION OF THE HERG CHANNEL BY ASPIRIN: PH OR DIRECT EFFECT?

Aspirin (ASA) has been widely used for many years for relieving pain and fever and in preventing heart attack and stroke. ASA overdose can result in an anion gap, metabolic acidosis, tinnitus, and, in severe cases, encephalopathy and cardiovascular collapse. There are studies showing the inhibitory effect of ASA on heat-evoked currents in rat dorsal root ganglion neurons and the augmented effect of ASA on the NMDA type of glutamate responses in spiral ganglion neurons. The effect of ASA on cardiac ion channels has not been studied. We evaluated the effect of ASA on cardiac IKr channel using HERG expressed on Xenopus oocytes. A two-microelectrode voltage clamp technique was used for recording, and the recording solution contained 96 mM NaCl, 5.0 mM KCl, 2.0 mM CaCl2, 1.0 mM MgCl2, and 5 mM HEPES, and the pH of the solution was adjusted with NaOH to 7.4. ASA was dissolved in the recording solution. At a concentration less than 1 mM, ASA has little effect on HERG current. ASA 1 mM and 2 mM inhibited current by 12 ± 2 and 22 ± 4%, respectively. ASA 3 mM inhibited current by 80 ± 6%. Considering the acidifying influence of ASA, the pH values of 1, 2, and 3 mM ASA solutions were determined as 7.1, 6.5, and 4.9. The pH of the recording solutions was adjusted to the corresponding pH, and the results showed that pH 7.1, 6.5, and 4.9 reduced HERG current by 6 ± 2, 10 ± 2, and 49 ± 5%. ASA 1, 2, and 3 mM caused greater inhibition of current than the recording solutions with the corresponding pH adjustment. There were significant differences in current inhibition between 2 mM ASA and the pH 6.5 recording solution (p < .05) and between 3 mM ASA and pH 4.9 recording solution (p < .01). ASA inhibits HERG current not only through acidification but also by a direct effect. The potent inhibition at 3 mM (54 mg/dL) further suggests that ASA at therapeutic doses and at doses seen with acute and chronic salicylism (40-120 mg/dL) may be arrhythmogenic owing to potent inhibition of the cardiac IKr channel.

68 THE EFFECT OF ACIDOSIS AND HYPERKALEMIA ON THE IKr BLOCKING ACTION OF ANTIARRHYTHMIC DRUGS.

Regional myocardial acidosis resulting from the impaired coronary blood flow has been observed in both animal models and in man. The ischemic myocardium also releases potassium into the extracellular space, which can cause regional hyperkalemia. Antiarrhythmic agents are frequently prescribed for patients with ischemic heart disease and regional changes in pH and potassium may alter the effect of these agents. In this study, we evaluated the effect of extracellular acidosis and hyperkalemia on the action of IKr blocking antiarrhythmic drug-quinidine (Q). The IKr channel was studied at room temperature by employing human-ether-a-go-go-related gene (HERG) expressed in Xenopus oocytes and two-electrode voltage clamp technique was employed for recording. The pH of the recording bath solution was adjusted with NaOH to 6.8 or 7.4 and the recording bath solutions contained either 5 or 7.5 mmol/L KCl (5 or 7.5 K). The recording solution with 5 K, pH 7.4 represented the normal condition and 7.5 K, pH 6.8 represented acidic and hyperkalemic conditions. Q 3, 10, and 30 μM when applied at 5 K, pH 7.4 inhibited current by 17 ± 3, 39 ± 3, and 63 ± 4%. The percentage current block by Q at 7.5 K, pH 7.4 was similar to current block at 5 K, pH 7.4. Q at 7.5 K, pH 7.4 decreased HERG current by 18 ± 1, 42 ± 3, and 65 x3%. But if Q was applied at 5 K, pH 6.8, the HERG inhibitory effect of Q was decreased, and 3, 10, and 30 μM Q produced 8 ± 2, 24 ± 3, and 50 ± 4% current block. Q 3, 10, 30 μM administered in acidic and hyperkalemic condition (7.5 K, pH 6.8) caused 13 ± 2, 25 ± 2, and 47 ± 2% current inhibition, which was similar to the inhibitory effect of Q at 5 K, pH 6.8. There was a significant difference in current block by Q between 5 K, pH 7.4 and 5 K, pH 6.8, and there was also a significant difference in current inhibition by Q between 5 K, pH 7.4 and 7.5 K, pH 6.8 (p < .05). Our data suggest that extracellular acidosis (pH 6.8) attenuates the HERG inhibitory effect of Q, and when extracellular acidosis is combined with hyperkalemia (7.5 K), the effect on IKr inhibition is similar to acidosis alone. The attenuated effect of Q at low pH may cause heterogeneity of repolarization between ischemic and normal regions, and this may set the stage for reentrant arrhythmias, contributing to Qs proarrhythmic toxicity.

69 THE EFFECT OF NARINGENIN (GRAPEFRUIT JUICE) AND ANTIARRHYTHMIC DRUGS ON IKr INHIBITION.

Grapefruit juice has been reported to cause significant QT prolongation in healthy volunteers. Naringenin (N), the principal flavonoid in grapefruit juice, has been identified as the most potent HERG channel blocker among several dietary flavonoids. In light of these reports, we thought that combining naringenin with Ikr-inhibiting antiarrhythmic drugs would increase IKr inhibition and possibly pose an increased health risk by increasing repolarization delay and ensuing arrhythmias. In this study, we investigated the effect of N combined with quinidine (Q) on IKr inhibition. The study was performed in an oocyte system with heterogeneously expressed human-ether-a-go-go-related gene (HERG) employing two electrodes voltage clamp technique for recording. The experiments were performed at room temperature. Doses of 10 μM and 100 μM N were found to inhibit HERG channel by 15 ± 4 and 40 ± 7%. Q at 1 and 10 μM caused 9 ± 1 and 39 ± 3% inhibition of HERG current. When 10 μM N was combined with 1 μM Q, 9 ± 3% current was blocked. HERG current was blocked by 29 ± 2% when 10 μM N was combined with 10 μM Q. N 100 μM combined with 1 μM Q and N 100 μM combined with 10 μM Q caused 21 ± 2 and 36 ± 2% inhibition in HERG current, respectively. Combining naringenin and quinidine does not show an additive effect but rather a diminution in IKr inhibition. Further studies on the interaction of N with other known IKr channel blockers are indicated.

58 THE EFFECTS OF QUINIDINE AND ITS CHIRAL ISOLATES ON ERG-1SM POTASSIUM CURRENTS AND CORRELATION WITH GASTROINTESTINAL AUGMENTATION.

Erg-1sm potassium channel has been recently reported to participate in modulation of gastrointestinal contractility. Since quinidine inhibits cardiac potassium channel and augments gastrointestinal contractility, it was thought that quinidine (Q) may affect erg-1sm. Studies were undertaken to evaluate the effects of Q and its chiral isolates on gastrointestinal erg-1sm potassium current and correlate these effects with colon contractility. Chiral separation (HPLC technique), mass spectrometry, and optical rotation determination were performed. The erg-1sm potassium channel was expressed in Xenopus oocytes and the two-electrode patch clamp technique was employed for recording. An isolated rat colon preparation was employed to measure changes in contractility. As a result of chiral separation, two peaks were obtained with elution times of 8.31 and 8.66 minutes all with a MW of 324; the optical rotations of racemate, isolates X and Y were + 258°, ± 0°; + 217°, respectively. The percentage changes in amplitudes of colon contraction (from baseline) were determined at different concentrations of Q and the two isolates in five experiments in each group. Quinidine 0.1, 1, and 10 μM increased contractility by 79 ± 34, 125 ± 42, 217 ± 51 (p ≥ .05) for isolate X and 70 ± 20, 115 ± 32, 272 ± 32 (p ≥ .05), and 22 ± 12, 46 ± 17, 59 ± 22 for isolate Y. The inhibition of erg-1sm currents by Q was 19 ± 4, 21 ± 5, and 48 ± 6 (p ≥ .05), respectively, for isolate X, 20 ± 4, 23 ± 5, 39 ± 7 (p ≥ .05), and for isolate Y, 22 ± 4, 21 [x 4, 31 ± 6. One chiral isolate and Q markedly augment contractility, while Q and the two chiral isolates inhibit the erg-1sm potassium currents to a similar extent. These results suggest that erg-1sm inhibition does not explain GI contractile augmentation caused by the Q racemate and its chiral isolates.

67 PHARMACOLOGY AND TOXICOLOGY OF A NEW AQUEOUS FORMULATION OF INTRAVENOUS AMIODARONE (AMIO-AQUEOUS) IN COMPARISON TO CORDARONE IV

Background Hypotension is the most frequent adverse event reported with intravenous amiodarone (Cordarone IV). The hypotension has been attributed to the vasoactive solvents of the formulation, polysorbate 80 and benzyl alcohol, both known to exhibit negative inotropy and hypotensive effect. A new aqueous formulation of intravenous amiodarone (Amio-Aqueous) does not contain vasoactive excipients and may be less toxic and causes less hypotension than Cordarone IV. This hypothesis was tested in a series of animal studies with direct comparison of Amio-Aqueous and Cordarone IV. Methods Four studies were performed in anesthetized Sprague Dawley rats weighing between 450 and 550 grams: 1) the lethal dose 50% (LD50) and lethal dose 100% (LD100) were determined in 6 rats for each drug; 2) the effects of the two drugs on myocardial contractility were compared at 5, 10 and 20 mg/kg doses using a Walton-Brody strain gauge (n = 30); 3) the effects of the drugs on arterial blood pressure were compared at doses of 3, 5, 10, and 20 mg/kg (n = 10); 4) the antiarrhythmic effects were compared at doses between 0.5 and 20 mg/kg following left anterior descending coronary artery ligation (n = 12). The studies were conducted in accordance to the NIH Guide for the Care and Use of Laboratory Animals. Results The acute toxicology study showed that both LD50 and LD100 were 30% greater for Amio-Aqueous than for Cordarone. At the dose where all animals expired on Cordarone, 50% of animals were still alive on Amio-Aqueous. The study on myocardial contractility showed that Amio-Aqueous was far less negative inotropic than Cordarone IV (p<0.001). Amio-Aqueous did not have an effect on contractility at 5 and 10 mg/kg dose levels while Cordarone resulted in a 25% (p<0.01) and 29% (p<0.002) decrease, respectively. The study on arterial blood pressure showed that Cordarone caused a significant decrease in blood pressure at each of the 3, 5, 10, and 20 mg/kg dose (p<0.05 to p<0.001) while Amio-Aqueous did not. The study on the antiarrhythmic effects showed comparable efficacy for both formulations. Conclusions Cordarone IV was more toxic, caused significant hypotension and negative inotropy, while Amio-Aqueous lacked the hypotensive and cardiotoxic properties of Cordarone IV. Therefore Amio-Aqueous is safer than Cordarone IV and the generic formulations.

68 CHIRAL SEPARATION OF THE INOTROPIC AND CHRONOTROPIC ACTIONS OF DIGOXIN

Background Digoxin (D) prolongs conduction at the AV node and augments cardiac contractility by inhibition of Na+,K+-ATPase. The digoxin molecule is chiral having asymmetries at the C3 and C17 carbon centers that could give rise to stereoscopic isomers. The actions of D on cardiac conduction and contractility could be mediated through the different isoforms of Na+,K+-ATPase that exist with different isomers having different degrees of inhibition of the different isoforms. Methods Using cyclobond chiral column we separated digoxin into two distinct chromatographic peaks each with different retention times. Optical rotation was +17° and +3° respectively for the two peaks but both isolates showed the same mass/change ratio of 780, identical to that of racemate digoxin. The effects of the isolates on cardiac contractility and AV conduction were evaluated in anesthetized guinea pigs (GP): 15 GPs were randomly given D, isolate 1 or 2. AV conduction was assessed by measuring PR interval and contractility by a Walton Brody strain gauge arch sutured to the left ventricular free wall. D or the isolates were infused continuously at 6 μg/kg/min. Results D and isolate 1 caused a progressive increase in the PR interval while isolate 2 did not progressively increase PR. D and isolate 2 caused a progressive increase in contractility (% change from the baseline) while isolate 1 caused little change in contractility. Conclusions We concluded that D can be chirally separated with one isolate causing progressive PR prolongation and the other contractile augmentation. (Figure)

2 VENOUS IRRITATION WITH CORDARONE IV AND TWO DIFFERENT FORMULATIONS OF AQUEOUS AMIODARONE WHEN ADMINISTERED VIA PERIPHERAL VEINS

Background Intravenous amiodarone is irritable to the venous system and should be administered through a central venous catheter whenever possible. However, amiodarone is often administered via a peripheral vein. The controlled trials for Cordarone IV primarily employed central venous lines and did not provide sufficient information about the risk of peripheral drug administration. This study evaluated the venous irritation with Cordarone IV (Wyeth) and 2 aqueous formulations of intravenous amiodarone: Amio-Aqueous, the original aqueous formulation (Academic Pharmaceuticals), and Amiodarone Aqueous IV, produced by a different manufacturing process (Wyeth). Each intravenous formulation suspends amiodarone as a micelle. Methods In a series of controlled trials, patients were randomized to either Amio-Aqueous or lidocaine, or to either Cordarone IV or Amiodarone Aqueous IV. Therapy with Amio-Aqueous and Amiodarone Aqueous IV were initiated with boluses of 150 mg amiodarone (10 mL, 15 mg/mL), Cordarone IV was initiated with 10-minute infusion (75 mL, 2 mg/mL), all were followed with a 24-hour infusion delivering 900-1150 mg amiodarone over 24 hours. Therapy with lidocaine was initiated with 100 mg boluses followed by a 24-hour infusion. Results A total of 177 patients received Cordarone IV, 168 patients received Amiodarone Aqueous IV, 185 received Amio-Aqueous, and 127 received lidocaine. The most common adverse events for Amiodarone Aqueous IV and Cordarone IV were injection site inflammation (60% and 32%), injection site pain (55% and 22%), injection site edema (50% and 17%), and phlebitis (35% and 6%), p<0.001. In contrast, the incidence of venous irritation was 2% for Amio-Aqueous with phlebitis in 1.6% (p<0.001). There was no venous irritation with lidocaine. X-ray diffraction studies showed that the physical nature of the two aqueous formulations (Amiodarone Aqueous IV and Amio-Aqueous) were different. Conclusions Both Cordarone IV and Amiodarone Aqueous IV are irritative to the venous system, while Amio-Aqueous is not, a difference that could relate to the different micelle structure of amiodarone among the formulations. The original aqueous formulation (Amio-Aqueous) appears to lack significant venous irritation and may be administered via a peripheral vein.