Cardiac function in alcohol-associated systemic hypertension

CardiacFunctionin Alcohol-Associated SystemicHypertension HOWARD S. FRIEDMAN, MD, BALENDU C. VASAVADA, MD, ABDEL M. MALEC, MD, KHALID K. HASSAN, MD, ASHOK SHAH, MD, and SALMA SIDDIQUI

and normotensive alcoholics (67 f 4 dynes/cm* X 103) LV stress was elevated (normal 46 f 3 dynes/cm* X 103, both p <0.02). However, LV mass was not increased (hypertenstve 96 f 4 g/m* vs normotensive 100 f 4 g/m*; (normal 92 f 5 g/m*), resutting in a markedly increased stress to mass ratio (hypertensive 0.6 f 0.06; Normal 0.5 f 0.05, p <0.02). Hypertensive alcohoks atso had LV “hyperfunctii,” with an increased stress/LV end-systolic volume ratio (1.7 f 0.1 vs 1.3 f 0.1 dynes/cm* X 103/ml, p <0.02). Thus, hypertensive alcoholics, even those with transitory hypertension, have more abnormal cardiac function than normotensive alcohoks. Presence of hypertension wfth hyperdynamic LV features may be a prelude to heart failure. (Am J Cardtol 1966;57:227-231)

The pathogenesis of alcohol cardiomyopathy is obscure. Because systemic hypertension is observed in one-third of alcoholics, the relation of this finding to left ventricular (LV) function was analyzed in 66 alcoholics (26 with a blood pressure of 160195 mm Hg or htgher) 4 to 5 days after alcohol withdrawal. Hypertensive alcoholics had a more abnormal ratio of preejection period/LV ejection time (PEP/ET) (0.396 f 0.01 vs 0.35 f 0.01, p cO.02) than normotenstve alcoholics (matched normal 0.290 f 0.01). Hypertensive alcoholics (transitory hypertension) with blood pressures of 120/60 mm Hg or less at time of study also had more abnormal PEP/LVET than matched normotensive alcoholics (0.415 f 0.03 vs 0.331 f 0.01, p <0.05). In both hypertensive (77 f 6 dynes/cm* X 103)

A

mained hospitalized for 4 to 5 days in a detoxification program. None of the patients had delirium tremens, a history of hypertension or heart disease. Sixteen healthy male physicians of a comparable age served as normal control subjects. The 66 patients were separated into 2 groups: those with hypertension, defined as a systolic blood pressure (BP) of at least 160 mm Hg or a diastolic BP of at least 95 mm Hg, or both, some time during the hospitalization, and those whose BP was always lower than these values. The clinical features of the alcoholics from which this group was derived were the subject of another report.ll In that study, patients with hypertension were older than the normotensive group; however, in the present study, hypertensive and normotensive alcoholic patients were of a comparable age (Table I). Most of the alcoholics were black or Hispanic. Hypertensive and normotensive alcoholics did not differ with respect to estimated average recent use of alcohol, duration of alcohol usage or alcohol blood levels on admission. Except for comparable elevations of serum glutamic oxaloacetic transaminase in both groups of alcoholics on admission, average values of other blood chemical determinations were within the normal range. Although the age, weight, body surface area or obesity indexI of the alcoholic patients was not different from the normal control subjects, the hypertensive alcoholics had significantly greater body weight and surface

lcoholics are at risk for development of heart failure.’ The heart of patients with this complication of alcohol abuse is generally dilated and hypocontractile.2J However, even before alcoholics show such severe cardiac dysfunction, more subtle evidence of impaired left ventricular (LV) performance may be found.4-g Although an association of alcoholism with the development of cardiomyopathy is generally accepted, the mechanism whereby ethanol produces this disorder is unknown. Recently, a relation between problem drinking and systemic hypertension was identified.lO-l2 Because hypertension is the leading risk factor for the development of heart failure,13 we report the relation of hypertension in alcoholic patients to cardiac function.

Methods Subjects: The study population was derived from male alcoholics admitted for withdrawal who reFrom the Section of Cardiology and the Department of Medicine, The Brooklyn Hospital and Downstate Medical Center, State University of New York, Brooklyn, New York. Manuscript received May 6, 1985; revised manuscript received July 3, 1985, accepted July 5,1985. Address for reprints: Howard S. Friedman, MD, Chief of Cardiology, The Brooklyn Hospital, 121 Dekalb Avenue, Brooklyn, New York 11201.

227

ALCOHOL-ASSOCIATED

220

TABLE

I

Clinical

HYPERTENSION

Features

TABLE

Normal (n = 16)

Age W Weight (lb) Body surface area (m*) Obesity index (lb X 100)

Hypertensive (n = 26) 36f 169 f 1.94 f

1 4 0.03

35 f 1 152 f 6’ 1.83 f 0.02’

3.41 f

3.45 f

0.05

3.25 f 0.10

l p <0.02 vs hypertensive patients. Values are mean f standard error of the mean.

II

Mass

and Dimensional

Normal (n = 16) PWT EDV EDV ESV ESV

(cm) (ml) (ml/m*) (ml) (ml/m*)

0.9 f llOf5 60 f 37 f 20 f 171 f 92 i 3.1 f

Mass (9) Mass (g/m*) Left atrium (cm)

in*

TABLE

Volume,

Normotensive (n = 40)

33 f 2 15866 1.84 f 0.04 0.07

Ill

Values are mean f standard EDV = enddiastollc volume: wall thickness.

0.04 2 3 1 12 5 0.3

Relations

Hypertensive (n = 26) 0.9 f 117f5 62 f 48 f 25 f 182 f 96 f 3.2 f

Normotensive (n = 40)

0.03

0.9 f 120f4 66 f2 48 f 26 f 187 f 100 f 3.2 f

3 3 2 9 4 0.1

error of the mean. ESV = end-systolic

volume;

0.02

3 2 7 4 0.1

PWT = posterior

Hemodynamlcs TABLE Normal (n = 16)

Heart rate (beatslmin) Systolic BP (mm Hg) Diastolic BP (mm W Cardiac index (liters/minlm2)

62 f 6 117f3

Hypertensive (n = 26) 77 f 2’ 126f6

73 f

2

116 f

I+

73 f 2

85 f

5’

76*

I+

2.7 f 0.1

2.7 f

0.1

2.9 f

0.1

l p <0.02 vs normal subjects; + p <0.02 vs hypertensive Values are mean f standard error of the mean. BP = blood pressure.

IV

Left

Ventricular

Normotensive (n = 40)

Systolic Normal (n = 16)

PW In systole FS(%) EF (%)

(cm)

1.7 i 37 i 66f

0.07 1 1

Function Hypertensive (n = 26) 1.5 f 0.05’ 32f I’ 81f 1’

* p <0.02 vs normal subjects. Values are mean f standard error of the mean. EF = ejection fraction: FS = fractional shortening;

Normotensive (n = 40) 1.4 f 0.03’ 33f 1’ 61 f 1’

PW = posterior

wall.

patients.

area than the normotensive alcoholics (Table I]. None of the alcoholic patients had atria1 fibrillation, left bundle branch block or mitral regurgitation, and none was receiving drugs with cardiotonic actions. Methods: Cardiac function was assessed by systolic time intervals and echocardiography on the fourth or fifth day after alcohol withdrawal. All subjects underwent %dimensional echocardiography to ensure that no regional wall motion abnormalities were present: hemodynamic measurements were derived from Mmode images. Echocardiography was performed using a system with a transducer using a piezocrystal that had a resonant frequency of 2.5 MHz. The echocardiograms and the simultaneous electrocardiographic leads were recorded on a strip-chart recorder at paper speeds of 50 and 100 mm/s. Patients were examined in the supine position; no patient was excluded for technical reasons. The transducer was placed at the left border in the fourth or fifth intercostal space. Echocardiograms were recorded continuously as transducer was tilted through an arc from the apex to the base of the heart. The echocardiographic LV dimensions were measured at a site where some parts of both mitral leaflets or chordae were seen, as described by Corya et al-l5 For each patient a phonocardiogram, an electrocardiogram and a carotid pulse tracing were recorded simultaneously, at a paper speed of 100 mm/s, at the time of each echocardiographic study. The carotid pulse tracing was monitored through a funnel held manually over the carotid artery and connected by a plastic air-

filled tube, 5 mm in diameter, to a pulse transducer with a time constant longer than 4 seconds. Total electromechanical systole (Q&l was measured from the onset of ventricular depolarization to the first high-frequency vibration of the aortic component of the second heart sound. LV ejection time (ET) was measured from the onset of the rapid upstroke to the trough of the incisura of the carotid pulse tracing. The preejection period (PEP) was determined by subtracting LVET from the QS,. LVET was corrected for heart rate using the regression formula of Weissler et a1.16 The PEP/LVET ratio was also calculated. Heart rate was determined by dividing the RR interval into 60. Wall thickness and cavity dimensions were measured following the recommendations of the American Society of Echocardiography for determining ventricular septal thickness (VS], LV internal dimensions (LVID), posterior wall thickness (PWT) and left atria1 size.” Measurements at end-diastole were made at the beginning of the QRS complex; measurements at endsystole were made at the point of the smallest distance separating septum from the posterior wall. All calculations were made using the mean values of at least 3 cardiac cycles. Fractional shortening (%) was taken as the difference of the LVID in end-diastole and in endsystole divided by LVID in end-diastole X 100. LV end-diastolic and end-systolic volumes and the difference, stroke volume, were calculated using the formula of Teichholz et alI8 for LV volume: (7.0/2.4 + LVID)LVID3. All LV volumes were also normalized for body surface area. LV ejection fraction was determined by dividing stroke volume by LV end-diastolic volume. Cardiac

February

TABLE

V

Systolic

Time

Intervals

Normal (n = 16) PEP (ms) LVET (ms) LVETI (ms) PEP/LVET

86 299 418 0.290

f f f f

3 5 3 0.01

1, 1986

THE

TABLE Hypertensive (n = 26) 101 f 257 f 388 f 0.398 f

JOURNAL

Left Ventricular

Normotensive (Il = 40)

3’ 1’ 4’ 0.01

VI

AMERICAN

l

95 f 2’ 273f4” 397 f 3’ 0.350 f 0.01’

OF CARDIOLOGY

Stress-Volume Normal (n = 16)

46f

Stress

(5x

Volume

Mass

229

Relations

Hypertensive (n = 26)

2

57

Normotensive (n = 40)

77f6'

67f4'

0.4

3.3 f 0.3

2.6 f 0.2

1.3 f 0.1

1.7 f 0.1’

1.5 i

0.5 f 0.05

0.8 f 0.06"

0.7 f 0.05

103)

+

’ p <0.02 vs normal subjects; 7 p <0.02 vs hypertensive patients. Values are mean f standard error of the mean. LVET = left ventricular ejection time: LVETI = LVET index; PEP = preejection period.

SBP/ESV (mm Hg/ml) StresslESV J!F!Z ( cm2

3.1 f

X 103/ml >

Stress/mass

output was calculated from the derived stroke volume and heart rate and expressed as the cardiac index by normalizing for the body surface area. LV mass [g) was calculated using the formula of Devereux and Reicheklg: 1.04 ([LVID + PWT + IVS13 - LVID3) - 14. End-systolic LV meridional wall stress (dynes/cm2 X 103) an index of myocardial afterload, was determined using the formulazO: [(0.344 X SPXLVIDS)/PWTS X (1 X PWTS/LVIDS)], where LVIDS = LVID in systole, PWTS = PWT in systole and SP = cuff systolic pressure. Systolic pressure or end-systolic wall stress/endsystolic volume, indexes of myocardial contractility,21 and the ratio of end-systolic stress to LV mass were also calculated. Twelve-lead electrocardiograms were obtained on the same day as were the echocardiograms. Criteria of Romhilt and Estesz2 were used for diagnosing LV hypertrophy and those by Morris et alz3 for diagnosing left atria1 enlargement. Statistical analysis: Electrocardiographic findings were compared by chi-square test. All other differences were analyzed by the Student t test for unpaired data. Linear regressions and correlation coefficients were made using the least-squares method. To correct for simultaneous multiple comparisons, the Bonferroni method24 was used (P
Results The electrocardiograms of normotensive and hypertensive alcoholic subjects were not different with respect to the presence of LV hypertrophy (normotensive 14%; hypertensive LZYO), left atria1 enlargement (normotensive 23%; hypertensive 12%), or abnormalities of the ST segment and T wave [normotensive 23%; hypertensive 47%). However, the likelihood of having 1 or more of these electrocardiographic findings was greater in hypertensive alcoholics (65% vs 3470, x2 = 4.48, p <0.05). Hemodynamic values are listed in Table II. Hypertensive alcoholics had a greater average heart rate than the normal group; cardiac output was not different. Hypertensive alcoholics at this time still had a higher systolic BP than normotensive alcoholics and a higher diastolic BP than both normotensive alcoholics and normal control subjects.

dynes __ ( cm2

0.1

X 103/g/m2 >

* p <0.02 vs normal subjects. Values are mean f standard error of the mean. ESV = end-systolic volume; SBP = systolic blood pressure.

TABLE VII Hypertensive

Comparlson Alcoholics

of Matched

Normotensive Age W Body surface area (m2) Heart rate (beats/min) Blood pressure (mm Hg) PEPILVET

Normotenslve

(n = 7)

34f 1 1.9 f 0.07 73 f 4 117*3/77*3 0.331 f 0.014

and Transltory

Hypertensive

(n = 7)

35i2 1.9 f

0.07

735 5 117f3/76f 0.415

* p <0.05. Values are mean f standard error of the mean. PEP/LVET = ratio of preejection period to left ventricular

1

f 0.033’

ejection

time.

Alcoholic patients did not have different average LV wall thickness, volume or mass or left atria1 size than the normal control subjects (Table III). However, LV systolic function, as indicated by posterior wall thickness in systole, fractional shortening, and ejection fraction, was reduced in alcoholic patients [Table IV). Systolic time intervals were markedly abnormal in alcoholics (Table V). Moreover, average PEP/LVET in hypertensive alcoholics was significantly worse than that in normotensive alcoholics. Both hypertensive and normotensive alcoholics had significantly increased LV wall stress (Table VI). When LV wall stress was divided by LV end-systolic volume, an index of LV contractility, values in hypertensive alcoholic patients significantly exceeded those of normal control subjects (Table VI). Also, the ratio of LV stress to mass was increased in hypertensive alcoholics. Because BP influences systolic time intervals, patients with transitory hypertension who had BPS of 120/80 mm Hg or less at the time of the study were contrasted with a closely matched group of normotensive alcoholics. The 2 groups were comparable with respect to age, body surface area, heart rate and BP, yet hypertensive alcoholics had a significantly worse PEP/LVET [Table VII). The relation of PEP/LVET is dependent on loading conditions as well as contractility; therefore, this

230

ALCOHOL-ASSOCIATED

HYPERTENSION

variable was analyzed to determine the factors that may have caused the abnormality. The increased PEP/LVET could not be explained in this fashion, because PEP/LVET did not correlate with LV wall stress (afterload) and an increased value would not be expected from the directional changes of the LV enddiastolic volume or the index of contractility observed in these patients. Only in hypertensive alcoholics did PEP/LVET correlate with heart rate (r = 0.68, p
Discussion In alcoholics, a form of systemic hypertension may develop that generally abates with abstinence.lOJ1 On resuming use of alcohol, hypertension may recur.lOJ2 Alcoholics may also have hypertension during withdrawal from alcohol, with the highest BP sometimes not occurring for 2 to 3 days after stopping its useel Because the elevations of BP are generally transitory,l”J1 hypertension may go unrecognized or may be dismissed as innocuous. BP elevations, however, may be marked, with values of 160/95 mm Hg or higher occurring in one-third of alcoholics undergoing detoxification.ll Moreover, the occurrence of an exaggerated BP response to cold pressor test after 4 to 5 days of abstinence in alcoholics with transitory hypertension suggests that such BP elevations reflect a diathesis to hypertension rather than being merely an acute reactive response.ll Alcoholics also show cardiac functional abnormalities even after several days of abstinence.4-Q Since alcoholic-associated hypertension may be transitory, a possible relation of such hypertension to cardiac dysfunction may therefore have gone unrecognized. In the present study, which was undertaken to examine this possible association, hypertensive alcoholics were indeed found to have worse systolic time intervals than normotensives, even when their BP at the time of the hemodynamic study was comparable. Also, both hypertensive and normotensive alcoholics had lower LV ejection phase indexes and higher LV wall stresses than matched normal subjects. Hypertensive alcoholics, however, showed an increased ratio of LV wall stress to end-systolic volume, suggesting an increased myocardial contractility, but had normal LV mass and therefore had an increased stress to mass ratio. Relation to previous studies: Abnormal cardiac function has been found in alcoholics both by noninvasive methods4-Q and by hemodynamic measurements made at catheterization.25 Pathologic abnormalities found at autopsy have also been described.26 Although in various experimental animals prolonged administration of alcohol has produced abnormalities of cardiac function,27J8 alcoholic cardiomyopathy has still not been replicated, and therefore, an indirect action of alcohol through its effects on BP must be considered at least as a contributory influence.

In our study alcoholics were, on average, 5 to 10 years younger than those of previous similar clinical and pathologic studies. 7-Q~26Unlike the findings in most earlier investigations,7-Q we did not find an increase in LV mass in alcoholics. However, the marked increase in LV wall stress in our patients could explain the increased LV mass observed in older alcoholics. The acute response to this increased wall stress appears to be an augmentation of contractility, to maintain cardiac function, whereas the chronic response, to reduce these high wall tensions and thereby lessen myocardial oxygen requirements, is the development of LV hypertrophy. 2QMyocardial injury, with the ensuing development of myocardialz6 and perivascular fibrosis,30 and of cardiac dilatation and hypocontractility,3J may therefore be the consequence of a failure to reduce sufficiently these LV wall stresses. The chronic metabolic effects of alcohol and its metabolites may be especially important in this setting. Alcohol and acetaldehyde depress mitochondrial function.3* These substances also retard protein synthesis,32,33which could interfere with the development of compensatory hypertrophy. Moreover, the chronic elevations of catecholamines produced by these substances may also contribute to myocardial injury.34 Implications: The presence of hypertension in alcoholics, dismissed in the past as merely an acute reactive response to alcohol withdrawal, has been shown to have an adverse effect on cardiac function. The increased LV wall stress, reflecting, in part, these elevations of BP, could explain the increased heart weight observed in older, chronic alcoholics. Moreover, the presence of compensatory “hyperfunction” of the left ventricle in hypertensive alcoholics may denote a prodromal phase before the development of cardiomyopathy; such a stage has been postulated to exist for hypertensives at risk for heart failure.35 Follow-up studies of hypertensive alcoholics are required to determine whether these alcoholics are the ones in whom cardiomyopathy develops. In any case, given the injurious effects of hypertension on the cardiovascular system, this study serves to identify a subset of alcoholics at risk for cardiac complications. Acknowledgment: The assistance of Drs. Luther Clark and Nathan Kraut in the recruitment of patients for this study and of Daisy Frankson for her secretarial assistance is acknowledged.

References 1. Friedman HS. Geller SA. Lieber CS. The effect of alcohol on the heart, skeletal muscle and smooth muscle. In: Lieber CS. ed. Major Problems in Internal Medicine. Aspects of Alcoholism. Philadelphia: W.B. Saunders, 1982;436-479. 2. Alexander CS. Idiopathic heart disease. I. Analysis of 100 cases with special reference to chronic alcoholism. Am J Med 1966;41:213-226. 3. Tobin JR, Driscoll JF. Lim MT, Sutton GC, Szanto PB, Gunnar RM. Primary myocardial disease and alcoholism: the clinical manifestations and course of the disease in a selected population of patients observed for three or more years. Circulation 1967;35:754-764. 4.SpodickDH,PigottVM,ChirifeR.Preclinicalcardiacmalfunctioninchronic alcoholism. Comparison with matched normal controls and with alcoholic cardiomyopathy. N Engl J Med 1972;267:677-680. 5. Levi GF, Quadri A, Ratti S, Basagni M. Preclinical abnormality of left ventricular function in chronic alcoholics. Br Heart J 1971;39:35-37. 6. Askanas A, Udoshi M, Sadjadi S. The heart in chronic olcohohsm: a noninvasive study. Am Heart J 1980:99:9-X

February

7. Kino M, Imamitchi H, Morigutchi M, Kawamura K. Takatsu T. Cardiovascular status in asymptomatic olchoholics. with reference to the level of ethanol consumption. Br Heart 1 1981;46:545-551. 8. Mathews EC, Gardin JM, Henry WL, Del Negro AA, Fletcher RD. Snow jA, Epstein SE. Echocardiographic abnormalities in chronic alcoholics with and without overt congestive heart failure. Am r Cardiol 1981;47:570-578, 9. Ballas M, Zoneraich S, Yunis M, Zoneraich 0, Rosner F. Noninvasive cardiac evaluation in chronic alcoholic patients with alcohol withdrawal syndrome. Chest 1982;82:148-153. 10. Saunders JB, Beevers DG. Paton A. Alcohol-induced hypertension. Loncet 1981;2:652-656. 11. Clark LT. Friedman HS. Hypertension associated with alcohol withdrawal: assessment of mechanisms and complications. Alcoholism Clin Exp Res 19a5;9:125-130. 12. Potter JF. Beevers DG. Pressor effect of alcohol in hypertension. Lancet 1984;1:119-122. 13. Kannel WB, Castelli WP. McNamara PM, McKee PA, Feinlico M. Role of blood pressure in the development of congestive heart failure. N Engl [ Med 1972;287:781-787. 14. Khosla T, Lowe CR. Indicies of obesity derived from body weight and height. Br \ Prev Sot Med 1967;21:122-128. 15. Corya BC, Feigenbaum H, Rasmussen S, Black MJ. Echocardiographic features of congestive cardiomyopathy compared with normal subjects and patients with coronary artery disease. Circulation 1974;49:1153-1159, 16. Weissler AM, Harris WS, Schoenfeld CD. Systolic time intervals in heart failure in man. Circulation 1968;37:149-159. 17. Sahn D1, DeMaria A, Kisslo 1, Weyman A. The Committee on M-Mode Standardization of the American Society of Echocardiography. Recommendations regarding quantification in M-mode echocardiography: results of a survey of echocardiographic measurements. Circulation 1978;58:1072-1083. 18. Teichholz LE. Kreulen T, Herman MV, Gorlin R. Problem in echocardiographic volume determinations: echocardiographic-angiographic correlations in the presence or absence of asynergy. Am 1 Cardiol 1976;37:7-Il. 19. Devereux RB. Reichek N. Echocardiographic determination of left ventricular mass in man. CircuJation 1977;55:613-618. 20. Reichek N, Wilson J, Martin JS. Plappert TA, Goldberg S, Hirshfeld JW. Noninvasive determination of left ventricular end-diastolic stress: validation

1, 1935

-‘WE

AMEQ!CAN

,‘C)L’RNA:

OF CARDIOLOGY

Volume

57

231

of the method anti inftial applrcrltion Circuiotlon 1982;65:99-108. 21. Nivatpurmin ‘T. Katz S. Scheuer 1. Pf~ok l?ft ventricular systolic pressure/end- :ystohc volume ratio. a sensitive detector of left ventricular disease. (4rn / Cardiol l~~!l:43:r~~ii9-97-1. 22. Romhilt DLV, I:stos EX A point-score system Jar the ECG diagnosisof left ventricular hypertrophy. Am Ileart f 1968;75:752-758. 23. Morris JJ. Estes EH, Whalen RE. Thompson HK. McIntosh H. P-wave analysis in valvular heart disease. Circulation 1964;29:242-252. 24. Wallcnstein S. Zucker CI.. Fleiss JL. Some statistical methods useful in circulation research. Circ Res 1980;47:1-8. 25. Gould L, Shariff M, Zahir M, DiLleto M. Cardiac hemodynamics in alcoholic patients with chronic liver disease cd a presystolic gallop. r Clin Invest 1969:48:&O-868. 26. Steinberg ID. Hayden MT. Prevalence of chnicaJJy occult cardjomyopathy in chronic alcoholism. Am Heart J 1981;101:461-464. 27. Friedman HS, Liebcr CS. Alcohol and the heart. In: Feldman EB, ed. Nutrition and the Heart. New York: Churchill Livingstone, 1983:149-150. 28. Noren GR, Stalcy NA, Einzig S. Mike11 FL, Asinger RW. Alcohol-induced congestive cardiomyopathy: an animal model. Cardiovasc Res 1983:17:8187. 29. Grossman W, Jones D, McLaurin L. Wall stress and patterns of hypertrophy in the human left ventricle. [ Clin Invest 1975;56:56-64. 30. Factor SM. Intramyocardial small-vessel disease in chronic alcoholism. Am Ifeart 1 1976;92:561-575. 31. Segal LD. Mason DT. Acute effects of acetaldehyde and ethanol on rat heart mitochondria. Res Commun Chem Pathol PharmacoJ1979;25:461-474. 32. Schreiber SS, Evans CD, Reff F, Rothschild MA, Oratz M. Prolonged feedings of ethanol to the young growing guinea pig. 1. The effect on protein synthesis in the afterloaded right ventricle measured in vitro. Alcoholism Clin Exp

Res 1982:6:384-390.

33. Schreiber SS, Briden K, Oratz M, Rothschild MA. Ethanol, acetaldehyde and myocardial protein synthesis. J Clin Invest 1972;51:2820-2826. 34. Rossi MA, Oliveira JSM, Zucoloto S, Becker PFL. Norepinephrine levels and morphologic alteratjons of myocardium in chronic alcoholic rats. Bietraege zur Pathologic 1976;159:51-60. 35. Meerson FZ. Mechanism of hypertrophy of the heart and experimental prevention of acute cardiac insufficiency. Br Heart J 1971;33:100-108.