Philip O. Ettinger

Arrhythmias and the “Holiday Heart”: Alcoholassociated cardiac rhythm disorders

An association between excessive alcohol use and cardiac rhythm disorders is often difficult to establish in the absence of overt cardiomyopathy. We studied 32 separate hospital admissions for dysrhythmias in 24 patients (20 men, 4 women) with heavy recent alcohol ingestion and prolonged excessive alcohol use. None had evidence of overt heart disease after treatment of arrhythmia. Episodes usually followed heavy weekend or holiday sprees, resulting in hospitalization between Sunday and Tuesday or in proximity to the year-end holidays, a relationship not observed in other alcohol-associated illnesses. Atrial fibrillation was most common, but atrial flutter, atrial tachycardia, junctional tachycardia, multiple APCs, multiple PVCs and ventricular tachycardia were also observed. Transient hypokalemia was present in four of 30. The mean PEPLVET ratio after treatment was 0.412 ± 0.014 (normal 0.299 ± 0.008, P < 0.001). High-speed ECGs showed prolongation of PRc, QRS, and QTc. At cardiac catheterization, intracardiac pressures and volumes, coronary arteriograms and ventricular wall motion were normal at rest and mean cardiac index was slightly low, but the left ventricular response to angiotensin was abnormal. Cardiac arrhythmias presenting during weekend or holiday drinking episodes are associated with conduction delays and depressed cardiac performance indicative of early cardiomyopathy and suggest a “holiday heart” syndrome.

Altered Myocardial Function and Metabolism in Chronic Diabetes Mellitus without Ischemia in Dogs

Myocardial disease in diabetes mellitus is usually attributed to coronary atherosclerosis. To examine the influence of uncomplicated diabetes on the left ventricle, a mild noninsulin-requiring diabetes was produced in male mongrel dogs after three intravenous doses of alloxan were administered at monthly intervals. There was a persistent decline in glucose tolerance and a reduced insulin content in the pancreas of each alloxan-diabetic dog at the termination of the experiment. The dogs were anesthetized for hemodynamic and metabolic studies after approximately 11 months. Left ventricular end-diastolic volume and cardiac output were measured by the indicator-dilution method. An increase in afterload with moderate aortic pressure elevations elicited a significant rise in end-diastolic volume and stroke volume in normal control dogs. In diabetes, despite a similar end-diastolic pressure response, the end-diastolic volume and the stroke volume responses were significantly less than those in control dogs. During acute volume expansion of the ventricle with saline, the end-diastolic pressure increment in diabetic dogs was twice that in control dogs. These responses were attributed to an increased stiffness of the left ventricle that was apparently due to accumulation of glycoprotein (measured by periodic acid-Schiff staining) in the interstitium. Since similar abnormalities were observed in dogs with diabetes occurring spontaneously and were absent when the pancreatic effects of alloxan were inhibited in a separate group of dogs, the pathogenetic role of alloxan via a direct action on myocardium was excluded. Analysis of lipids in the left ventricle revealed elevated triglyceride and cholesterol concentrations despite normal plasma levels. During infusion of 14C-1-oleic acid, cardiac oxidation appeared to be normal, but fatty acid incorporation, which was predominantly into phospholipid in the control dogs, was diverted to triglyceride in the diabetic dogs. Since an aberration of de novo synthesis was not found during studies with 14C-acetate, triglyceride accumulation was attributed to altered intracellular metabolism, perhaps related to glycerol phosphate acyl transferase activity. The basis for cholesterol accumulation was less clear, since neither 14C-acetate nor 14C-oleate incorporation into sterol was enhanced. Myocardial ischemia was excluded on the basis of patency of coronary arteries and normal coronary blood flow, myocardial cation content, and mitochondrial morphology. Thus, it was concluded that chronic diabetes mellitus can alter myocardial composition and function independent of vascular effects.

Systolic Time Intervals as Measures of the Contractile State of the Left Ventricular Myocardium in Man

Previous studies relating systolic time intervals and measures of cardiac performance have suggested that the time intervals may be useful indices of myocardial contractility. To explore this possibility, systolic times and left ventricular (LV) performance and contractility were measured nearly simultaneously in 14 normal subjects and 56 patients with cardiac disease. Preejection period (PEP) and the ratio of PEP to LV ejection time (LVET) changed significantly with acute inotropic influences (exercise and isoproterenol), were normal in patients with right or left ventricular overloads in whom cardiac index and ejection fraction were depressed but contractile element velocity at peak dP/dt and the Frank-Levinson contractility index were normal, and were significantly abnormal in patients with either clinically evident or occult LV decompensation in whom the measures of contractility were reduced. Correlations of PEP and PEP/LVET with measures of both performance and contractility were insignificant for patients with valvular disease, shunts, or cor pulmonale and significant but weak for the entire series. However, in subjects with either normal left ventricles or cardiac disease confined to the left ventricle, PEP and PEP/LVET exhibited good correlations with measures of pump function and excellent correlations with measures of contractility. These results indicate that the systolic times are a valid measure of contractility which should prove useful in comparing patients with cardiac pathology confined to the LV myocardium and in following patients with extramyocardial hemodynamic lesions of constant severity.

Myocardial Function and Lipid Metabolism in the Chronic Alcoholic Animal

In view of the variables that obscure the pathogenesis of cardiomyopathy, a study was undertaken in mongrel dogs fed ethanol as 36% of calories for up to 22 mo. Both the experimental and control groups maintained body weight, hematocrit, plasma vitamin, and protein levels. Left ventricular function was evaluated in the intact anesthetized dog using indicator dilution for end-diastolic and stroke volume determinations. During increased afterload with angiotensin, the ethanol group exhibited a larger rise of end-diastolic pressure (P<0.01), whereas end-diastolic and stroke volume responses were significantly less than in controls. Preload increments with saline elicited a significantly higher end-diastolic pressure rise in the ethanol group (P<0.01). No hypertrophy, inflammation, or fibrosis was present and it was postulated that the enhanced diastolic stiffness was related to accumulation of Alcian Blue-positive material in the ventricular interstitium. To evaluate myocardial lipid metabolism, [1-(14)C]oleic acid was infused systemically. Plasma specific activity and myocardial lipid uptake were similar in both groups. There was a significantly increased incorporation of label into triglyceride, associated with a reduced (14)CO(2) production, considered the basis for a twofold increment of triglyceride content. In addition, diminished incorporation of [(14)C]oleic acid into phospholipid was observed accompanied by morphologic abnormalities of cardiac cell membranes. Potassium loss and sodium gain, like the lipid alteration, was more prominent in the subendocardium. Thus, chronic ethanol ingestion in this animal model is associated with abnormalities of ventricular function without evident malnutrition, analogous to the preclinical malfunction described in the human alcoholic.

Ventricular Conduction Delay and Arrhythmias during Regional Hyperkalemia in the Dog: Electrical and Myocardial Ion Alterations

Coronary arterial potassium (K+) infusion induces RS-T elevation and ventricular arrhythmias. To explore electrical and metabolic events in this model, which simulates ischemia-induced injury currents and rhythm disturbances, isotonic KCl was infused at 0.5–2.0 ml/min into the left anterior descending or the left circumflex coronary artery of 37 intact anesthetized dogs. Conduction was monitored with His and left bundle branch recording catheters and with bipolar electrodes at endocardial and epicardial levels in the perfused zone. RS-T segments rose progressively in the perfused region despite normal coronary blood flow (85Kr) some dogs progressed rapidly to multifocal ventricular extrasystoles and ventricular fibrillation (group 1), but others continued for 20 minutes without developing fibrillation (group 2). His-Q and bundle branch conduction times were not altered by regional perfusion, but the QRS complex was locally prolonged. Ventricular extrasystoles were frequently coupled and left bundle branch depolarizations never preceded them, suggesting origin within the myocardium. Intramural electrodes showed increasing delay in the perfused region; epicardial depression was uniformly more profound and preceded significant arrhythmias, often with 2:1 or higher intramural block and slowing sufficient to allow reentry. Epicardial activation preceding endocardial activation was observed during extrasystoles in the perfused zone, further indicating origin within the myocardium. These phenomena were associated with tissue concentrations of K+ in the outer wall significantly higher than those in the inner wall; and sodium was also significantly elevated. Therefore, it was postulated that reentry from depressed epicardial layers into more normal surrounding tissue constitutes a possible basis for the arrythmias in this model.

The fate of experimentally induced coronary artery thrombosis

Abstract The evolution of a thrombus in a coronary artery is uncertain. To evaluate this process, coronary artery thrombosis was induced in intact anesthetized dogs by passage of current through an electrode catheter in the circumflex or anterior descending branch of the left coronary artery. Twenty-five animals survived the immediate period after arterial occlusion, which was established angiographically. Survivors were placed in one of two groups depending upon the interval between thrombosis and spontaneous death or sacrifice: group I (1 to 17 days; 15 dogs) and group II (30 to 70 days; 10 dogs). Gross examination of the coronary arteries was performed post mortem in all dogs, with additional premortem angiographic studies in group II. At the time of death all dogs but 1 in group I showed continued evidence of significant obstructive thrombosis. In group II angiography and postmortem examination of the involved arteries indicated that within the second month after clot production sufficient changes occur within the thrombus to re-establish varying degrees of blood flow.

Myocardial ischemia and cell acidosis: Modification by alkali and the effects on ventricular function and cation composition

Myocardial cell pH was measured with 5, 5 dimethyl-2, 4-oxazolidinedione (DMO) in intact anesthetized dogs by a transient indicator dilution technique. Bolus injections of labeled DMO, vascular, extracellular and water indicators were made into the left anterior descending coronary artery, and blood samples were collected from the great cardiac vein. The steady state distribution of DMO between cells and plasma was calculated from the mean transit times of the indicator. Normal myocardial cell pH averaged 6.94 and changed by 58% of the concomitant alterations in plasma pH after infusions of acid or alkali. Myocardial ischemia induced by inflation of a balloon tip catheter in the left anterior descending coronary artery resulted in progressive decreases in cell pH to 6.59 by 1 hour. Infusions of sodium carbonate diminished intracellular acidosis. Hemodynamic studies during 4 hours of ischemia with blood pH at 7.55 to 7.60 indicated a significantly reduced left ventricular end-diastolic pressure and increased stroke volume by comparison with findings in animals given infusions of saline solution. Ventriculograms revealed improved wall motion in the ischemic segment after infusion of alkali. Precordial mapping showed a significant reduction in the number of leads with S-T segment elevation as well as in the sum of S-T segment elevations, but R wave amplitudes did not differ from those in control studies. Calculations of extracellular space, tissue water and cation content revealed a reduced gain of cell sodium ion and loss of cell potassium ion during ischemia after alkali treatment. The latter may account for the S-T segment responses, whereas enhanced ventricular performance may be related to reduced competition of hydrogen ion with calcium ion for binding sites on contractile protein.

Hemodynamics at Rest and during Exercise in Combined Aortic Stenosis and Insufficiency

The effects of exercise in combined aortic stenosis and insufficiency were evaluated in 10 patients by pressure measurements and the measurement of forward (QF) and regurgitant (QR) flows by simultaneous upstream and downstream sampling using indocyanine green. While heart rate increased, systolic aortic valve pressure gradient (mean, 37 ± 9 mm Hg) did not change. Increased QF (mean, 4.54 ± 0.34 at rest and 6.89 ± 0.42 liters/min with exercise, P < 0.001) was balanced by decreased QR (means, 4.09 ± 1.02 and 2.33 ± 0.71 liters/min, P < 0.02), and total flow did not change significantly. Although diastolic regurgitant period declined, total diastolic seconds per minute decreased by only 6%, while calculated systemic resistance decreased by 30%. Left ventricular systolic and aortic pressures increased, while left ventricular end-diastolic pressure and diastolic aortic valve gradients were unchanged. Mean systolic aortic valve area, calculated by utilizing total valve flow, was 1.8 ± 0.3 cm2 in both states.Thus, exercise reduces regurgitant fraction in mixed aortic lesions as in pure aortic insufficiency. This observation confirms the necessity of measuring total valve flow in evaluation of mixed valve lesions. Reduced regurgitation primarily reflects an altered relationship of peripheral resistance to backflow resistance. While the stenotic valve is demonstrated to behave as a fixed systolic orifice, the diastolic aortic orifice area cannot be calculated from the Gorlin equation as it does not account for peripheral resistance.

Origin of Acetyl Strophanthidin-Induced Ventricular Arrhythmias

To examine the origin of digitalis-induced ventricular tachycardia (VT), acetyl strophanthidin (AS) (25 mug/min) was perfused into a limited zone of myocardium in intact anesthetized dogs through a catheter placed fluoroscopically in the left anterior descending artery without ischemia. A second catheter in the great cardiac vein sampled venous effluent from this region. His and left bundle branch depolarizations were recorded and bipolar intramural electrograms from endocardial and epicardial sites within the anterior descending region were obtained. No conduction alterations preceded arrhythmia. Cardiac venous K+ rose from 3.3 +/- to 4.4 +/- 0.2 meq/liter (P less than 0.001), indicating egress from the perfused zone. 10 animals (Group 1) were sacrificed 2 min after onset of VT while 11 (Group 2) continued until fibrillation (4-14 min). All showed normal (endocardial leads to epicardial) transmural depolarization during sinus rhythm, but 10/21 demonstrated reversal, usually late during VT, including 8/11 in Group 2. Epicardial activation preceded fascicular activation and QRS. Recordings from the border and circumflex regions in 10 additional dogs (Group 3) demonstrated activation reversal only in the border zone. Myocardial K+ was reduced (mean 63 +/- 1 mueq/g) and Na+ increased (mean 41 +/- 2 mueq/g) in the perfused zone (nonperfused circumflex area K+ 72 +/- 1, Na+ 33 +/- 1 mueq/g, P less than 0.001 for both); changes were similar in inner and outer ventricular wall. In related experiments, subepicardial injections of AS induced activation reversal within the immediate area, similar to recordings during coronary infusion. Reversed transmural activation with early epicardial depolarization suggest VT arises within myocardium; electrolyte gradients between adjacent regions may be causative.

Glucose intolerance in nonischemic cardiac disease. Role of cardiac output and adrenergic function.

To investigate hemodynamic and neurohumoral factors as a basis for glucose intolerance in cardiac patients without ischemic heart disease, intravenous glucose tolerance tests were performed on 70 patients aged 14 to 69 years. The mean glucose fractional disappearance rate (Gk) for 41 patients less than age 50 was 1.19 ± 0.07% fall in glucose/min, while in 19 normal subjects, it was 1.60 ± 0.14, P <0.01. The decreased Gk was independent of age and lipid levels, and was directly related to a reduced plasma concentration of immunoreactive insulin. The glucose response to tolbutamide of 40 patients was similarly reduced and correlated with Gk.Since enhanced adrenergic stimulation may be present in cardiac patients with resultant reduction of insulin secretion, the effects of alpha-and beta-adrenergic blockade were examined in two groups of cardiac patients. Gk was unchanged after administration of phenoxybenzamine and propranolol. Since a reduced cardiac output correlated with diminished Gk, the influence of sustained improvement after corrective cardiac surgery was evaluated in 13 patients. Postsurgical increments of resting cardiac output were usually associated with lower fasting insulin levels, enhanced clearance of administered glucose, and a more rapid rise of plasma insulin to higher peak levels, followed by a relatively rapid decline. Enhanced insulin response to tolbutamide after surgery in seven of these patients was also demonstrated. It is suggested that chronically reduced levels of cardiac output can result in reduced insulin secretion and in glucose intolerance and that these may be reversed with improved cardiac function.