Comparison of Outcomes Among Moderate Alcohol Drinkers Before Acute Myocardial Infarction to Effect of Continued Versus Discontinuing Alcohol Intake After the Infarct
Article Outline
Light-to-moderate alcohol consumption has been previously associated with a lower risk of acute myocardial infarction (AMI) and mortality. The association of changes in drinking behavior after an AMI with health status and long-term outcomes is unknown. Using a prospective cohort of patients with AMI evaluated with the World Health Organization's Alcohol Use Disorders Identification Test, we investigated changes in drinking patterns in 325 patients who reported moderate drinking at the time of their AMI. One-year alcohol consumption, disease-specific (angina pectoris and quality of life) and general (mental and physical) health status and rehospitalization outcomes, and 3-year mortality were assessed. Seattle Angina Questionnaire Angina Frequency and Quality of Life, Short Form-12 Mental and Physical Component Summary Scales were modeled using multivariable hierarchical linear models within site. Of the initial 325 moderate drinkers at baseline, 273 (84%) remained drinking and 52 (16%) quit. In fully adjusted models, Physical Component Scale scores (beta 6.47, 95% confidence interval 3.73 to 9.21, p <0.01) were significantly higher during follow-up in those who remained drinking. Persistent moderate drinkers had a trend toward less angina (relative risk 0.65, 95% confidence interval 0.39 to 1.10, p = 0.11), fewer rehospitalizations (hazard ratio 0.79, 95% confidence interval 0.44 to 1.41, p = 0.42), lower 3-year mortality (relative risk 0.75, 95% confidence interval 0.23 to 2.51, p = 0.64), and better disease-specific quality of life (Seattle Angina Questionnaire Quality of Life, beta 3.88, 95% confidence interval −0.79 to 8.55, p = 0.10) and mental health (Mental Component Scale, beta 0.83, 95% confidence interval −1.62 to 3.27, p = 0.51) than quitters. In conclusion, these data suggest that there are no adverse effects for moderate drinkers to continue consuming alcohol and that they may have better physical functioning compared to those who quit drinking after an AMI.
After an acute myocardial infarction (AMI), moderate alcohol drinkers may alter their alcohol consumption in response to such a life-changing event. To date, there have been no studies describing the drinking patterns and associated outcomes of moderate drinkers after an AMI. Given the potential benefits of moderate alcohol consumption on minimizing ischemic events, we sought to describe the proportion of patients who drank moderate amounts of alcohol before their AMI and then quit compared to those who continued consuming alcohol after their AMI. We then examined the association of continued alcohol consumption with cessation on 1-year disease-specific (angina pectoris and quality of life) and general (mental and physical) health status outcomes, 1-year rehospitalizations, and 3-year mortality rates.
Methods
From January 1, 2003, to June 28, 2004, 2,498 patients with AMI were recruited into the Prospective Registry Evaluating Myocardial Infarction: Event and Recovery (PREMIER) study from 19 US hospitals. Details of the study have been published elsewhere.1 Eligible patients were ≥18 years old with enzymatic evidence supporting the diagnosis of AMI and had prolonged (>20 minutes) ischemic signs/symptoms or electrocardiographic ST changes. Institutional research board approval was obtained at each participating institution and patients signed informed consent for baseline and follow-up interviews. Trained data collectors interviewed patients during hospitalization and abstracted charts that included data regarding patients' presentation, clinical history, and treatment during hospitalization. A centralized follow-up center performed follow-up interviews by telephone at 1 month and 6 and 12 months after patients' MI. Nondrinkers and binge drinkers were removed from the analysis because they were not the focus of this study.
Alcohol consumption was assessed by patient self-report at the time of a confirmed AMI and during follow-up interviews. Three brief screening questions of the World Health Organization's Alcohol Use Disorders Identification Test2 were used at baseline and follow-up to quantify the frequency of drinking, how many drinks were consumed in 1 day, and binge drinking (≥6 drinks at a single time). Moderate drinking was defined as drinking a minimum of 2 to 4 times/month with no more than 1 drink to 4 drinks per occasion, thus averaging <1 drink/day. Binge drinkers were excluded because binge drinking has been associated with adverse cardiovascular and all-cause mortality3 and is not supported by any medical evidence. Thus, of the initial 2,498 patients enrolled in PREMIER, only the 325 moderate drinkers were retained to address our study question of the changes in drinking patterns and outcomes.
The analyzed outcomes were (1) 1-year disease-specific and general health status, (2) 1-year rehospitalization, and (3) 3-year mortality assessed after the 1-year interview. To quantify patients' health status, the disease-specific Seattle Angina Questionnaire (SAQ)4 and the generic Short Form-12 (SF-12)5 were administered at the time of enrollment and at the 1-year follow-up interview. The SAQ is a 19-item questionnaire measuring patients' perspectives of how their coronary disease affects health status domains. The SAQ is valid, reliable, sensitive to clinical changes, and prognostic of subsequent mortality and admissions for acute coronary syndromes.6, 7 For the purposes of this study, scores on the Angina Frequency and Quality of Life subscales were used, which range from 0 to 100, with higher scores indicating less angina and better quality of life. Angina was dichotomized into any angina versus no angina due to the skewed distribution of angina because most patients were angina free 1 year after their AMI and in recognition of the clinical goal to have patients attain complete angina relief. The SF-12 is a valid and reliable measurement of patients' general functional status and generates mental and physical component scores (Mental Component Scale and Physical Component Scale).5 Rehospitalization data were collected by self-report during each patient telephone interview throughout the 1-year follow-up. Mortality data were collected by querying the Social Security Death Master File in addition to information obtained from family members through follow-up interviews. Baseline for mortality was the 1-year interview date because a patient had to survive to the 1-year interview to assess changes in alcohol use.
Descriptives between categories of change in alcohol use were compared to chi-square or Fisher's exact tests for categorical variables or to t tests for continuous variables, as appropriate. Kaplan-Meier estimates were used for 1-year rehospitalization, 3-year mortality, and creation of Kaplan-Meier survival curves to compare time to outcomes with the log-rank test. Multivariable models adjusted for baseline health status (in health status models only), age, gender, Caucasian race, smoking, diabetes, renal failure, and congestive heart failure. SAQ Quality of Life subscale, SF-12 Mental Component Scale, and SF-12 Physical Component Scale were modeled using multivariable hierarchical linear models within a site and estimates are reported as beta regression weights and 95% confidence intervals. The hierarchical model structure accounted for clustering of patients within a site.8 Angina, rehospitalization, and mortality were modeled using multivariable hierarchical modified Poisson regression within a site.9 Typical analyses often use logistic regression to estimate adjusted odds ratios, which are then interpreted as relative risks. However, in this study the events being modeled (presence of angina and adverse outcomes) were not rare, in which case odds ratios are poor estimates of relative risks. To address this issue, we estimated adjusted relative risks directly using a modified Poisson regression model.9 These estimates are reported as relative risks and 95% confidence intervals. Three-year mortality and 1-year rehospitalization were modeled using time-to-event multivariable proportional hazards regression stratified by site. Results are reported as hazard ratios and 95% confidence intervals. All tests for statistical significance were 2-tailed with an alpha level of 0.05. All analyses were conducted using SAS 9.1.3 (SAS Institute, Cary, North Carolina) and R 2.6.0.10
Results
Of the initial 325 patients who were included in the analysis, 273 (84%) continued to drink, and 52 (16%) quit (Figure 1). Most patients were Caucasian, men, and >50 years of age. A larger percentage of the quitters had a history of smoking, diabetes, congestive heart failure, and chronic renal failure compared to those who continued to drink alcohol (Table 1).
Figure 1.
Study population.
Table 1. Unadjusted patient baseline and follow-up data
| Persistent Moderate Drinkers | Quitters | p Value | |
|---|---|---|---|
| (n = 273) | (n = 52) | ||
| Unadjusted baseline data | |||
 Age (years) | 61±11.7 | 62±13.2 | 0.550 |
| Caucasian | 256(94%) | 36(69%) | <0.001 |
| Men/women | 227(83%)/46(17%) | 40(77%)/12(23%) | 0.282 |
| Smoked within previous 30 days | 63(23%) | 17(33%) | 0.144 |
| Diabetes mellitus | 28(10%) | 7(14%) | 0.494 |
| Congestive heart failure | 7(3%) | 6(12%) | 0.009 |
| Chronic renal failure | 6(2%) | 2(4%) | 0.619 |
| SAQ-assessed angina (any vs none) at baseline | 137(50%) | 29(56%) | 0.460 |
| SAQ-assessed quality of life (baseline) | 67±21.0 | 61±23.5 | 0.055 |
| SF-12 Physical Component Scale score (baseline) | 48±10.8 | 46±11.0 | 0.282 |
| SF-12 Mental Component Scale score (baseline) | 53±9.0 | 50±11.9 | 0.022 |
| Unadjusted outcome data | |||
| SAQ-assessed presence of angina (any vs none) at 1 year | 35(13%) | 12(23%) | 0.055 |
| SAQ-assessed quality of life at 1 year | 87±14.5 | 81±23.7 | 0.015 |
| SF-12 Physical Component Scale score (1 year) | 50±8.9 | 41±12.0 | <0.001 |
| SF-12 Mental Component Scale score (1 year) | 54±7.3 | 52±10.0 | 0.061 |
| Rehospitalizations (1 year)⁎ | 75(28%) | 18(35%) | 0.26 |
| Mortality (3 years)⁎ | 16(6.0%) | 5(10%) | 0.3 |
⁎Kaplan-Meier estimates. |
Unadjusted Kaplan-Meier estimates of 3-year mortality rates were lower for persistent moderate drinkers (6.0%) compared to quitters (10%, p = 0.3; Figure 2, Table 1). Unadjusted Kaplan-Meier estimates of 1-year rehospitalization rates were also lower for persistent moderate drinkers (28%) compared to quitters (35%, p = 0.26; Table 1). Neither of these associations persisted after multivariable adjustment for mortality and rehospitalization (Figure 3).
Figure 2.
Kaplan-Meier curve for alcohol pattern and 3-year mortality.
Figure 3.
(A) Adjusted alcohol drinking pattern and adverse outcomes, angina, and rehospitalization at 1-year follow-up and 3-year mortality. (B) Alcohol drinking pattern and SAQ Quality of Life (QoL) subscale, SF-12 Mental Component Scale (MCS), and Physical Component Scale (PCS) summary outcomes at 1-year follow-up. The 95% confidence intervals (error bars) are presented.
In unadjusted analyses, persistent drinkers had significantly better physical function and disease-specific quality of life at 1 year and trended toward having better mental health and less angina (Table 1).
In fully adjusted analyses, including adjustments for patients' baseline health status, persistent drinkers had significantly better physical function and a trend toward better disease-specific quality of life and mental health and less angina (Figure 3). Although not statistically significant, in part due to small samples and large confidence intervals, all point estimates suggested better outcomes in those who continued drinking compared to those who quit.
Discussion
We found that ≥16% than 1 of 7 patients who were engaged in the healthiest type of alcohol consumption at the time of their AMI stopped after the event. These patients had worse physical function 1 year later and trended toward having more angina pectoris and worse quality of life. This study is the first to describe the patterns and health status consequences of moderate drinkers after an AMI.
These observations are consistent with mechanistic data that suggest numerous potential pathways by which alcohol consumption may be beneficial. Such mechanisms include increased insulin sensitivity, increased high-density lipoprotein cholesterol,11 and anti-inflammatory properties of alcohol with decreases in C-reactive protein,12, 13 tumor necrosis factor-α, interleukin-6, and sympathetic tone.11, 14 Enhancement in insulin sensitivity is thought to be due to suppression of fatty acid release from adipose tissue resulting in decreased substrate competition, which in turn increases mitochondrial glucose uptake and use.15 Improvements in inflammation, abdominal obesity, and autonomic tone may also be playing lesser roles in the cardioprotective effects of light-to-moderate regular alcohol consumption.16 These characteristics of alcohol may contribute to the altering of disease-specific and general health statuses.
Our findings are also consistent with an emerging literature on the association of changes in alcohol consumption and cardiovascular disease. A recently published study by King et al17 illustrated the effects of changes in drinking behavior and long-term cardiovascular outcomes. In that study, the investigators followed middle-aged nondrinkers and compared those who began to drink to those who continued to abstain. They found that the cohort that began to drink had a 38% lower risk of developing cardiovascular disease than the cohort who continued to abstain from alcohol. The findings from this study are also consistent with studies in the general population that examined changes in drinking behavior with cardiovascular outcomes.
An alternative explanation for why persistent moderate drinkers have better physical functioning than the quitters could be related to reasons for alcohol cessation after AMI. In this study, the reasons for alcohol cessation are unknown; however, they could be secondary to a significant lifestyle change or being too sick to drink alcohol. Supporting the possibility that patients who quit were sicker, we found that a larger percentage of quitters had a history of smoking, diabetes, congestive heart failure, and chronic renal failure compared to those who continued to drink. Importantly, after adjusting for these co-morbidities and other potential sources of confounding, the associations persisted with no clear evidence for harm by quitting alcohol after an AMI.
Our observations should be interpreted in the context of the following potential limitations. White men comprised most of the present study population, and thus we cannot extrapolate our findings to women or other ethnicities. In addition, our study did not address the health effects of other drinking patterns such as binge drinking, new drinking, very light sporadic drinking, or specific types of alcohol beverages. It is also possible that our definition of light-to-moderate drinking may have influenced our results. Most studies on the health effects of alcohol have defined moderate consumption as 1 drink to 2 drinks/day, which has consistently been associated with decreased risks of cardiovascular mortality and other adverse cardiovascular events,18 whereas we included drinking a minimum of 2 to 4 times/month with no more than 1 drink to 4 drinks per occasion in our definition. Thus, our categorization of light-to-moderate drinking averages to less than 1 drink/day and is not substantially different from previous efforts to categorize drinking behaviors. Future, larger studies may be needed to further refine the levels of alcohol consumption most associated with outcomes.
Our observational data suggest the need for further clarification of alcohol-consumption recommendations for patients who have been light-to-moderate drinkers before an AMI. Explicitly counseling patients about this is important because it is a common clinical experience that patients associate alcohol consumption, like smoking, as an “unhealthy” lifestyle and may mistakenly believe that they can improve their prognosis by stopping. Our observational study suggests that patients after an AMI should engage in a discussion with their personal physicians regarding the continuation of alcohol consumption.
References
- . The Prospective Registry Evaluating Myocardial Infarction: Events and Recovery (PREMIER)—evaluating the impact of myocardial infarction on patient outcomes. Am Heart J. 2006;151:589–597
- . The AUDIT alcohol consumption questions (AUDIT-C): an effective brief screening test for problem drinking. Ambulatory Care Quality Improvement Project (ACQUIP). Alcohol Use Disorders Identification Test. Arch Intern Med. 1998;158:1789–1795
- . Binge drinking and mortality after acute myocardial infarction. Circulation. 2005;112:3839–3845
- . Development and evaluation of the Seattle Angina Questionnaire: a new functional status measure for coronary artery disease. J Am Coll Cardiol. 1995;25:333–341
- . A 12-item short-form health survey: construction of scales and preliminary tests of reliability and validity. Med Care. 1996;34:220–233
- . Monitoring the quality of life in patients with coronary artery disease. Am J Cardiol. 1994;74:1240–1244
- . Health status predicts long-term outcome in outpatients with coronary disease. Circulation. 2002;106:43–49
- . Hierarchical modeling: its time has come. Am Heart J. 2003;145:16–18
- . A modified poisson regression approach to prospective studies with binary data. Am J Epidemiol. 2004;159:702–706
- . Language and Environment for Statistical Computing. 2005;Vienna: Austria
- . Alcohol and cardiovascular health: the razor-sharp double-edged sword. J Am Coll Cardiol. 2007;50:1009–1014
- . Alcohol consumption and plasma concentration of C-reactive protein. Circulation. 2003;107:443–447
- . Moderate alcohol consumption reduces plasma C-reactive protein and fibrinogen levels; a randomized, diet-controlled intervention study. Eur J Clin Nutr. 2002;56:1130–1136
- . C reactive protein, moderate alcohol consumption, and long term prognosis after successful coronary stenting: four year results from the GENERATION study. Heart. 2004;90:419–424
- . Moderate alcohol consumption, estrogen replacement therapy, and physical activity are associated with increased insulin sensitivity: is abdominal adiposity the mediator?. Diabetes Care. 2003;26:2734–2740
- . Alcohol and coronary heart disease: the answer awaits a randomized controlled trial. Circulation. 2005;112:1379–1381
- . Adopting moderate alcohol consumption in middle age: subsequent cardiovascular events. Am J Med. 2008;121:201–206
- . Alcohol dosing and total mortality in men and women: an updated meta-analysis of 34 prospective studies. Arch Intern Med. 2006;166:2437–2445
This study was funded in part by CV Therapeutics, Inc., Palo Alto, California, and CV Outcomes, Inc., Kansas City, Missouri.
Dr. Spertus discloses that he owns the copyright to the Seattle Angina Questionnaire.
PII: S0002-9149(10)00469-8
doi:10.1016/j.amjcard.2010.01.339
© 2010 Elsevier Inc. All rights reserved.
