time of onset of supraventricular tachyarrhythmia in relation to alcohol consumption

Time of Onset of Supraventricular Tachyarrhythmia in Relation to

Alcohol Consumption Markku Kupari, MD, and Pekka Koskinen, MD

It is widely believed but has never been proved that idiopathic supraventricular tachyarrhythmias beginning during or after weekends or winter holidays are frequently alcohol-related (“holiday heart” syndrome). The time of arrhythmia onset was therefore studied in relation to self-reported ethanol consumption and results of a screening test for alcoholism (CAGE questionnaire) in 289 patients aged <65 years admitted for supraventricular tachyanhythmias. There were 102 patients having an etiologically idiopathic arrhythmia with a known time of onset. Among them, but not among those with disease-related arrhythmias, patients with arrhythmic episodes beginning on Saturdays or on Sundays were more often chronic akohol abusers (9 of 19,47%) than either patients with episodes beginning from Mondays through Fridays (18 of 62,22%; p = 0.040) or control subjects from the out-of-hospital population (8 of 66, 12%; p = 0.002). In multivariate analysis, the time of arrhythmia onset was related to the CAGE response (G* = 6.0, p = 0.014) but not to the most recent ethanol use. However, the increased frequency of problem drinkers among patients with weekend-onset idiopathic arrhythmias was only relative, and resulted from a decreased number of abstainers and non-problem drinkers. No conspicuous clustering of alcohol-related arrhythmias was seen after New Year’s or May Day. Thus, although the present study confinns an association between heavy drinking and idiopathic arrhythmias beginning during weekends, it shows that the question may be of a relative rather than an absolute overrepresentation. The term holiday heart may also be somewhat misleading since no postholiday accumulation of akohol-related arrhythmias was found.

(Am J Cardiol 1991;67:716-722)

From the Division of Cardiology, First Department of Medicine, Hel- sinki University Central Hospital, Helsinki, Finland. This study was supported in part by YrjS Jahnsson Foundation, Helsinki, Finland; and by the Foundation for Alcohol Research, Helsinki, Finland. Manuscript received September 10,199O; revised manuscript received and accepted November 27,199O.

Address for reprints: Markku Kupari, MD, Cardiovascular Labo- ratory, Helsinki University Central Hospital, 00290 Helsinki, Finland.

H eavy alcohol consumption can contribute to the onset of supraventricular tachyarrhythmia.1-4 The term “holiday heart” syndrome has been

coined to indicate that alcohol-related arrhythmias cluster around weekends and after festivities such as the Christmas-New Year period.5 A captivating designation, it has become popular both colloquially and in sci- entific communications.1,2~6-8 Yet, the underlying idea was not based on any formal prospective or retrospective investigation but on uncontrolled observations made in a small group of alcoholics.5>g Unable to track down more tenable evidence for the holiday heart phenomenon, we studied the relation of the onset of supraventricular tachyarrhythmias to alcohol use in a larger and less selected patient population.

METHODS Study population: We studied 289 patients (203

men and 86 women) admitted to the emergency ward of our hospital for symptomatic supraventricular tachyarrhythmias and aged <65 years (mean age 50). Between January and October 1985 we studied each patient with new-onset atria1 fibrillation (n = 98),t” between Janu- ary and September 1986 we studied each patient with recurrent atria1 fibrillation (n = 98),” and between De- cember 1986 and September 1987 we studied each patient with supraventricular tachyarrhythmia other than atria1 fibrillation (n = 99: 50 patients with reentry supraventricular tachycardia, 30 with atria1 flutter and 19 with paroxysmal atria1 tachycardia). The classification of the arrhythmias was based on a 1Zlead electrocardiogram taken on admission.12 The total number of admissions (n = 295) was higher than the number of patients because 6 subjects were admitted twice during these studies; only the first admission was included in the present investigation.

Patient evaluation: Each patient was studied by us in the emergency ward, usually within 12 hours of admission. The date and the day of the week the symp- toms of the current arrhythmia had begun were ques- tioned and recorded. In subsequent analyses, patients with weekend arrhythmias (beginning on Saturdays or on Sundays) were compared with those having weekday arrhythmias (beginning from Mondays through Fri- days). Each arrhythmia was classified as either disease related or idiopathic on the basis of clinical examination, 1Zlead electrocardiogram, chest x-rays, laboratory tests,lO and results of all previous cardiac studies. In ad- dition, 205 patients underwent a complete echocardio-

716 THE AMERICAN JOURNAL OF CARDIOLOGY VOLUME 67

graphic examination l3 (Irex System III or Toshiba SSH 60A). The diagnostic criteria for the underlying cardiovascular conditions have been specified earlier.lO

The alcoholic drinks consumed each day during the week preceding the arrhythmia were recorded as grams of absolute ethanol and added up to give the total consumption per week. Subsequently, the patients were cat- egorized into 3 groups by their recent ethanol use: (1) 0 g/week, (2) 1 to 210 g/week, and (3) >210 g/week. The last 2 days’ (day of arrhythmia onset and the preceding day) consumption was calculated separately and classified as either small to moderate (I 150 g) or large (> 150 g). To screen for alcohol abuse, the patients were given the 4 CAGE questions14-16: Have you ever felt you should Cut down on your drinking? Have people Annoyed you by criticizing your drinking? Have you ever felt bad or Guilty about your drinking? Have you ever had a drink first thing in the morning to steady your nerves and to get rid of a hangover (Eye-opener)? The patients giving L2 affirmative responses were classified as chronic heavy drinkerW6; the rest (with 0 to 1 positive response) were classified as abstainers or non- problem drinkers.

Controlling: For a control group, we studied 66 subjects (44 men and 22 women aged <65 years [mean 491) from the local out-of-hospital population, picked up for us by the Central Office of Statistics. They underwent the same studies as our patients, including the screening for alcoholism and the assessment of the last week’s (preceding our contact) ethanol consumption. Eight had hypertension, 3 had coronary artery disease and 2 had chronic pulmonary disease.

Statistics: Group differences in rates and propor- tions were analyzed by the Fisher exact test and the Pearson chi-square test. Multivariate analyses were made, separately for idiopathic and disease-related arrhythmias, using log-linear models17 in 4-way tables of the following variables: time of onset of arrhythmia, amount of recent ethanol consumption, CAGE response and type of arrhythmia. The fit of a model was evaluat- ed by the likelihood ratio chi-square test (G2). The first model to try consisted of only the main effects of the variables. If the fit was poor, 2-way interactions relevant to the purpose of the present study were added to the model one at a time. The improvement of the fit between any 2 nested models Ml and M2 was tested by comparing the difference G2~1 - G2 ~2 to the chi- square distribution with dfMl - dfMz degrees of freedom. The improvement was considered statistically significant at p <0.05.

RESULTS The arrhythmia was disease-related in 185 patients

and idiopathic in 104. The most common underlying conditions in the former group were coronary artery disease (n = 62), systemic hypertension (n = 38), valvular heart disease (n = 17), idiopathic dilated cardiomyopathy (n = 11) and hypertrophic cardiomyopathy (n = 7). The onset time was uncertain in 25 patients, of whom 23 had a disease-related and 2 an idiopathic arrhythmia. All subsequent analyses pertaining to the time of

TABLE I Recent Ethanol Consumption and Response to the CAGE Questionnaire in 66 Subjects Aged <65 Years and Selected Randomly from Out-of-Hospital Population

Ethanol consumption category (g/week) 0 I-210

>210

Number of positive CAGE responses o-1 2-4

* Percentage of the total sample.

No. of Pts. % *

19 29 38 57

9 14

58 88 8 12

onset of arrhythmia are based on data of the remaining 264 patients.

Relation of the time of onset of arrhythmia to recent ethanol use: IDIOPATHIC ARRHYTHMIAS: Figure 1 shows the occurrence of idiopathic arrhythmias over the days of the week and the distribution of the patients into the 3 categories of recent ethanol consumption. Ta- ble I gives the ethanol consumption data of the out-of- hospital control subjects. A higher proportion of patients with weekend arrhythmias (18 of 19 [95%]) had drunk alcohol during the last 7 days than of either patients with weekday arrhythmias (54 of 83 [65%], Fish- er exact test, p = 0.011) or of the control subjects (47 of 66 [72%], Fisher exact test, p = 0.035). The frequency of recent drinking in the control group remained the same even if the 13 subjects with cardiorespiratory dis- eases were excluded (38 of 53 [72%]). The last 2 days’ ethanol intake totaled >150 g in 5 of 19 patients (26%) with weekend arrhythmias and in 8 of 83 (10%) with weekday arrhythmias (Fisher exact test, p = 0.060).

DISEASE-RELATED ARRHYTHMIAS: Figure 2 shows that among patients with disease-related arrhythmias there was no apparent difference in the frequency of recent ethanol consumption between subjects with weekend- onset episodes and those with weekday episodes (29 of 46 [63%] vs 67 of 116 [57%]; Fisher exact test, p = 0.597). The last 2 days’ consumption totaled >150 g in

1

N 10

5

0 Mon Tue Wed Thu Fri Sat Sun

FIGURE 1. The relation between recent alcohol drinking and the day of onset of idiopathic supraventricuiar tachyawhyth- mias. The columns show the number of patients (N) each day of the week and the distributii of the patients into the 3 categories of total ethanol consumption over the preceding 7 days (Og, l-12Og,and >2lOg).

THE AMERICAN JOURNAL OF CARDIOLOGY APRIL 1, 1991 719

4 patients from the former group (9%) and in 3 from the latter (3%) (Fisher exact test, p = 0.101).

Relation of the time of onset of arrhythmia to theCAGEresponse: IDIOPATHICARRHYTHMIAS: Figure 3 shows that a larger proportion of patients with weekend-onset arrhythmias than of those with weekday episodes responded affirmatively to 12 CAGE questions (9 of 19 [47%] vs 18 of 83 [22%]; Fisher exact test, p = 0.040). However, when the daily arrhythmia rates of problem drinkers were examined, it was found that the number of weekend-onset episodes (9 of the total 27) was not different from the 2 of 7 expected (chi-square = 0.31, p = 0.578). By contrast, in non-problem drinkers, only 10 of the 75 arrhythmias had begun during weekends, which was much less than the 2 of 7 expected (chi-square = 9.17, p = 0.003). This created a relative overrepresentation of problem drinking among patients with weekend-onset idiopathic arrhythmias.

The CAGE responses of the population controls are summarized in Table I. The frequency of 12 positive answers was clearly lower than among patients with idiopathic weekend arrhythmias (12 vs 47%; Fisher exact test, p = 0.002), but not significantly different from that among patients with weekday arrhythmias (12 vs 22%; Fisher exact test, p = 0.136). The frequency of >-2 positive responses among controls free of cardiorespira-

30

25

20

-

N 15

10

5

0 Mon Tue Wed Thu Fri Sat Sun

FIGURE 2. The relation between rece-nt alcohol drinking and the day of onset of disease-related supraventricular tachyarrhythmias. For detailed explanation, see the legend to Figure 1.

20 T

15

N 10

Mon Tue Wed Thu Fri Sat Sun

FIGURE 3. The relation between the CAGE questionnaire response and the day of onset of idiopathic supraventricular tachyanhythmias. The cdumnr show the number (N) of patients each day of the week and their disbfbution into the 2 CAGE categories (5 1 positive respowx, 22 pesitive re- Opon~).

tory disease was 13% (7 of 53 subjects; Fisher exact test, p = 0.004 compared with the group of idiopathic weekend arrhythmias).

DISEASE-RELATED ARRHYTHMIAS: Figure 4 shows the frequency of 12 positive CAGE responses in patients with disease-related arrhythmias. No statistically significant difference was found between subjects with weekend-onset and those with weekday-onset episodes (Fish- er exact test, p = 0.123).

Relation of recent ethanol use and the CAGE response to the type of arrhythmia (Table II): In assessing whether the type of arrhythmia was alcohol-related, patients with atria1 fibrillation (n = 192) were compared with those having other supraventricular tachyarrhythmias (n = 97). Table II shows that idiopathic atria1 fibrillation was more often associated with problem drinking by the CAGE survey than were the other idiopathic arrhythmias. The type of arrhythmia was unrelated to recent ethanol use.

Multivariate analysis: Table III shows the results of log linear modeling for the idiopathic arrhythmias; the factors included and their categories are specified in a footnote to the table. As expected, there was a highly significant relation between recent ethanol consumption and the CAGE response. The time of arrhythmia onset was independently associated with the CAGE response (i.e., adding their interaction to the model improved its fit to the data) but not with the preceding week’s ethanol consumption. An association between the type of arrhythmia and the CAGE response was also confirmed. The results were essentially the same even if the last 2 days’ ethanol intake (categories: < 150 g, >150 g) was substituted for the whole preceding week’s consumption.

In patients with disease-related arrhythmias, the time of arrhythmia onset was associated with neither the CAGE response nor the recent ethanol consumption, and there was no association between the type of arrhythmia and the CAGE response (p >0.20 for each interaction).

Holiday arrhythmias: New Year’s and May Day are festivities characterized by alcoholic binges in our coun- try. Over the years 1985 to 1987, only 1 patient was admitted for a supraventricular tachyarrhythmia that had begun during the 3-day period from December 31

30 -

25 --

20 --

N 15.- 1 r

Mon Tue Wed Thu Fri Sat Sun

FIGURE 4. The relation between the CAGE questionnaire re- ~ponse and the day of onset of disease-related wpraventricu- lar tachyarrtrythmias. For detailed explanation, see the legend to Figure 3.


TABLE II Recent Ethanol Consumption and CAGE Questionnaire Response in Relation to Type of Supraventricular Tachyarrhythmia in 289 Patients Aged <65 Years

Idiopathic Arrhythmias Disease-Related Arrhythmias

AF (n = 59) non-AF (n = 45) AF(n = 133) non-AF (n = 52)

n (%)* n (%) n (%) n (%) Recent ethanol consumption (g/week)

0 16 (27) 14(31) 52 (39) 23 w l-210 23 (39) 24 (53) 62 (47) 21(40) >210 20 (34) 7 (16) 19 (14) 8(16) P 0.10 0.74

Number of positive CAGE responses O-1 39 (66) 38 VW 106 (80) 46 (88) 2-4 20 (34) 7 (16) 27 (20) 6 (12) P 0.04 0.16

* The data are gwen as numbers (percentages) of patients in each category. AF = atrlal fibrillation; p = probability values from comparing the recent ethanol consumption (Pearson’s chi-square test) and the CAGE responses (Rsher exact test) between

patients with AF and non-AF arrhythmias.

TABLE Ill Log-Linear Models Showing Independent Relations Among Recent Ethanol Consumption, CAGE Questionnaire Response, Type of Arrhythmia, and Time of Onset of Arrhythmia in 102 Patients Admitted for Idiopathic Supraventricular Tachyarrhythmias

Fit of the Model Improvement of the Fit

Model

Ethanol + CAGE + time + arrhythmia? Ethanol X* CAGE t time t arrhythmia Ethanol X CAGE t time X CAGE t arrhythmia Ethanol X CAGE t time X CAGE + arrhythmia X CAGE Ethanol X CAGE t time X CAGE + arrhythmia X CAGE t time X ethanol

* Likelihood ratio chi-square. 7 The model includes only the main effects of the factors.

G2” P df G2Change p df

83.9 0.000 11 14.9 0.137 10 69.0 0.000 1 8.9 0.446 9 6.0 0.014 1 4.0 0.858 8 4.9 0.027 1 3.4 0.845 7 0.6 0.437 1

*The multiplication sign indicates that both the main effects and the interaction of the Zfactors are effective in the model. Arrhythmia = the type of arrhythmia in 2 categories (atrial ftbrillabon. other supraventricular tachyarrhythmla); CAGE = the CAGE questionnaire response in 2 categories (O-1,2-

4 positive answers); df = degrees of freedom: ethanol = the preceding week’s ethanol consumption in 2 categories (3210 g, >210 g); p = probability value; time = the time of ar-- rhythmia onset in 2 categories (weekend onset, weekday onset).

through January 2, and 5 patients were admitted for worthy that the time of arrhythmia onset remained un- similar arrhythmias beginning from April 30 through known since only the day of admission was recorded. May 2. These rates were not higher than the average 3- For all that, the idea of holiday heart syndrome has day frequencies of arrhythmias in January (3.7) or in survived and become widely accepted.1~2~6-8 However, April and May (3.5). Of the 6 holiday arrhythmias, 5 only 1 later study has produced additional data di- were idiopathic, but only 1 patient was a problem drink- rectly pertinent to this phenomenon. In a retrospective er according to the CAGE response. Five patients had case record analysis, Rich et all8 could not find either drunk either no ethanol or <150 g over the last 2 days postweekend or postholiday accumulation of admissions preceding the arrhythmia. for alcohol-related atria1 fibrillation.

DISCUSSION The “holiday heart” syndrome was originally de-

fined as “a weekend or holiday presentation of an acute rhythm or conduction disturbance associated with heavy ethanol consumption in a person without clinically evi- dent heart disease and disappearing with abstinence.” 5 The description was based on observations in 24 alcoholics who had a total of 32 hospital admissions with either premature beats or various tachyarrhythmias, mostly atria1 fibrillation. Nineteen of the 32 admissions occurred between Sunday and Tuesday, and of the remaining 13 episodes, 6 were seen close to the New Year, The patients did not constitute either a retrospective or a prospective consecutive series, however, and no nonalcoholics with arrhythmias were studied. The fact that several patients had >l admission included in the study may also have biased the findings, and it is note-

Methodologic considerations: The present study combined 3 separate series of patients admitted for acute symptomatic and sustained supraventricular tachyarrhythmias. During each study period, all patients aged <65 years with similar arrhythmias were included. Any selection bias in the emergency ward, such as was possible in the study by Ettinger et a1,5,9 was thus avoided. Our hospital is the largest referral center for the city of Helsinki and its vicinities (population 0.9 million), and patients contacting private doc- tors or community health centers for acute unremitting tachyarrhythmias are as a rule referred to a hospital in our area. However, as we only studied arrhythmias leading to hospital admission, our data may be poor- ly representative of asymptomatic or nonsustained arrhythmic episodes. Since we, in contradistinction to Et- tinger5 and Rich18 and their co-workers, recorded the day of arrhythmia onset instead of the day of admission,

THE AMERICAN JOURNAL OF CARDIOLOGY APRIL 1. 1991 721

we contrasted Saturdays and Sundays (instead of Sun- days, Mondays and Tuesdays) with the remaining days of the week.

In uncovering chronic heavy drinking, we relied on the CAGE questionnaire14 because it is easy to adminis- ter and far superior to conventional laboratory measure- ments in the detection of alcoholism.15J6J9 The score of L2 positive responses has recently been validated as an index of problem drinking in 2 separate general hospital populations.‘6J9 The limitation of the CAGE survey is that it may not detect the occasional nonalcoholic binge drinker.

Clustering of problem drinkers’ idiopathic arrhythmias during weekends- relative rather than absolute: We found that patients who had weekend-onset supraventricular tachyarrhythmias without detectable heart disease were chronic heavy drinkers at least twice as often as patients with similar but weekday-onset arrhythmias. Surprisingly, this did not reflect an increased number of problem drinkers presenting with arrhythmias during weekends but resulted, instead, from a decreased rate of rhythm disturbances in abstainers and non-problem drinkers (Figure 3). The weekend-clustering of alcohol-related arrhythmias was thus apparent rather than real. The possiblity that a reduction of alcohol-unrelated arrhythmias during weekends could contribute to or even be the main cause of the holiday heart phenomenon has not been entertained previously. The mechanism of the variation in the occurrence of alcohol- unrelated idiopathic arrhythmias is unknown but could be related, for instance, to a heavier physical or mental (occupational) stress in the middle of the week than on the weekends.

Role of recent ethanol consumption: Although there was a difference between idiopathic weekend and weekday arrhythmias also with respect to the patients’ recent use of ethanol, this was neither striking nor independently related to the time of onset of arrhythmia. Re- cent electrophysiologic studies6*7 have shown that alcohol-related arrhythmias result from an interaction of the acute effect of ethanol with the underlying subclini- cal cardiomyopathy,20 the arrhythmogenic substrate, and that relatively modest acute amounts will do for this. Although the role of excessive binge drinking has thus probably been overemphasized in the past, it cer- tainly is crucial in those rare cases where healthy, nonalcoholic persons have transient hangover-related tachyarrhythmias.21

Holiday arrhythmias: During the period of our study, the average New Year sales of alcohol in Finland were 1.9 times higher than the mean sales before ordi- nary winter weekends, and the respective ratio was 1.6 for the May Day sales compared with spring weekends (unpublished statistics of the Finnish State Alcohol Company). Nevertheless, we did not see any conspicuous accumulation of idiopathic arrhythmias in CAGE- positive persons close to either holiday. Excessive drinking immediately preceding the holiday arrhythmias was not very common either. Although these data speak against postholiday clustering of alcohol-related ar-


rhythmias, they are probably too few to entitle defini- tive conclusions thereof.

The designation: On the one hand, our data do sup- port the idea of the holiday heart syndrome. The frequency of alcohol abuse among patients with weekend- onset idiopathic arrhythmias was more than double the frequency of abuse in patients with similar but weekday-onset arrhythmias and up to 4 times the prevalence of problem drinking in the out-of-hospital population. On the other hand, the weekend-clustering of problem drinkers’ arrhythmias was not absolute, as the original idea indicated,j but resulted from a decreased rate of arrhythmias in abstainers and non-problem drinkers. Neither did we find any particular accumulation of alcohol-related arrhythmias close to New Year’s or May Day. Captivating as it is, the designation “holiday heart syndrome” may thus be beside the point, and we think that it would be more appropriate to speak simply of alcohol-related arrhythmias without any epithet. After all, ethanol is the culprit instead of Saturdays, Sundays or holidays.

Acknowledgment: We are grateful to Kimmo Luo- manmaki, MD, and Hannu Leinonen, MD, for cooper- ation during this investigation.

REFERENCES 1. Anonymous. Alcohol and atria1 fibrillation (editorial). Lancet 1985;1:1374. 2. Luck JR, Engel TR. Arrhythmias and social drinking. Ann Intern Med 1983;98:253. 3. Lowenstein SR, Gabow PA, Cramer J, Oliva PB, Ratner K. The role of alcohol in new-onset atria1 fibrillation. Arch Intern Med 1983;143:1882-1885. 4. Cohen EJ, Klatsky AL, Armstrong MA. Alcohol use and supraventricular arrhythmia. Am J Cardiol 1988;62:971-973. 5. Ettinger PO, Wu CF, De La Crw C, Weisse AB, Ahmed SS, Regan TJ. Arrhythmias and the “holiday heart”: alcohol-related cardiac rhythm disorders. Am Heart J 1978;95:555-562. 6. Engel TR, Luck JC. Effect of whisky on atria1 vulnerability and “holiday heart.” J Am Call Cardiol 1983;1:816-$18. 7. Greenspon AJ, Schaal SF. The “holiday heart”: electrophysiologic studies of alcohol effects in alcoholics. Ann Intern Med 1983;98:135-139. 8. Engel TR. The holiday heart: effect of ethanol on atria1 vulnerability. Primary Cardiol 198493%101. 9. Ettinger PO. Holiday heart arrhythmias. Int J Cardiol 1984;5:540-542. 10. Koskinen P, Kupari M, Leinonen H, LuomanmLki K. Alcohol and new-onset atrial fibrillation: a case-control study of a current series. Br Heart J 1987;57: 468-473. 11. Koskinen P, Kupari M, Leinonen H. Role of alcohol in recurrences of atria1 fibrillation in persons aged <65 years. Am J Cardioi 1990;66:954-958. 12. Keefe DL, Miura D, Somberg JC. Supraventricular tachyarrhythmias: their evaluation and therapy. Am Heart / 1986;111:1150-1161. 13. Kupari M, Leinonen H, Koskinen P. Value of routine echocardiography in new-onset atria1 fibrillation. Int J Cardiol 1987;16:106-108. 14. Ewing JA. Detecting alcoholism. The CAGE questionnaire. JAMA 1984; 252:1905-1907. 15. Bernadt MW, Mumford J, Taylor C, Smith B. Comparison of questionnaire and laboratory teats in the detection of excessive drinking and alcoholism. Lancet 1982;1:325-328. 16. Bush B, Shaw S, Cleary P, Delbanco TL, Aronson MD. Screening for alcohol abuse using the CAGE questionnaire. Am J Med 1987;82:231-235. 17. Wilkinson L. Systat: the system for statistics. Evanston, IL: Systat Inc., 1988:821-X41. 18. Rich EC. Siebold CM. Camoion B. Alcohol-related acute atrial fibrillation. A case-control dtudy and review of h0 patients. Arch Intern Med 1985;145:830-833. 19. Beresford TP, Blow FC, Hill E, Singer K, Lucey MR. Comparison of CAGE questionnaire and computer-assisted laboratory profiles in screening for covert alcoholism. Lancet 1990;2:482-485. 20. Kupari M, Suokas A. Effects of ethanol and its metabolites on the heart. In: Crown K, Batt RD, eds. Human Metabolism of Alcohol. vol III. Metabolic and Physiologic Effects of Alcohol. Boca Raton, FL: CRC Press, 1989:49-60. 21. Thornton JR. Atria1 fibrillation in healthy non-alcoholic people after an alcoholic binge. Lancet 1984;2:1013-1014.

time of onset of supraventricular tachyarrhythmia in relation to alcohol consumption

Documents