Moderate Alcohol Consumption and Safety of Lovastatin and Warfarin Among Men: The Post-Coronary Artery Bypass Graft Trial
Article Outline
Abstract
Purpose
Although moderate drinking has been associated with lower mortality among patients with coronary heart disease, its safety among patients taking common cardiac medications is unknown.
Subjects and methods
We studied 1244 men enrolled in the Post-Coronary Artery Bypass Graft (CABG) Trial who had undergone previous coronary bypass surgery. Participants were randomly assigned to lovastatin in low (mean 4 mg) or high (mean 76 mg) doses and to low-dose warfarin (mean international normalized ratio [INR] 1.4, goal INR <2.0) or placebo in a factorial design. Participants underwent routine measurement of alanine aminotransferase (ALT) and INR levels every 6 to 12 weeks for 4 to 5 years. We categorized weekly alcohol intake as abstention (<1 drink), light (1-6 drinks), moderate (7-13 drinks), and heavier (≥14 drinks).
Results
During follow-up, 66% of men taking warfarin had an INR of 2.0 or higher, and 7% of men had an ALT of 80 IU/L or higher. Maximum INR (P = .72) and ALT (P = .51) levels did not differ across categories of alcohol intake. The risks of an INR of 2.0 or higher were 67%, 66%, 68%, and 61% among non-, light, moderate, and heavier drinkers (P = .86), respectively. The corresponding risks of an ALT of 80 IU/L or more were 8%, 10%, 9%, and 6% (P = .70), respectively.
Conclusion
Moderate drinking did not adversely influence the safety of low-dose warfarin or even high-dose lovastatin among men in this randomized trial, as measured by INR and ALT levels.
Keywords: Alcohol consumption , Warfarin , Statin , Medication , Safety
Moderate alcohol consumption has been associated with a lower risk of coronary heart disease in a wide array of epidemiologic studies,1 an association supported by the established effects of alcohol intake on cardiovascular risk factors in short-term trials.2 Moderate drinking is also associated with lower morbidity and mortality among patients with established coronary heart disease.3, 4, 5 Current American Heart Association guidelines neither encourage nor prohibit moderate drinking,6 but some physicians nonetheless suggest that patients at risk for coronary heart disease with no contraindications to alcohol use “should be informed that they have, on average, net health benefits from the regular consumption of small-to-moderate amounts of alcohol.”7
Among the important concerns about advising patients at high risk for coronary heart disease to consider moderate drinking is the potential for interactions with commonly used cardiac medications. One panel of experts in alcohol research, geriatrics, internal medicine, and psychiatry defined hazardous drinking among older adults as intake of even 1 drink daily when combined with nitrates or warfarin or 2 drinks daily when used with aspirin or antihypertensives.8 In a review for the National Institute on Alcohol Abuse and Alcoholism, Weathermon and Crabb specifically suggested that patients taking warfarin should avoid alcohol entirely.9 Current labeling for statins consistently advises caution among patients who consume “substantial quantities” of alcohol but does not provide further clarification.
Surprisingly, robust data on the actual risk of adverse drug events associated with moderate drinking are extremely limited. In 1 small controlled study, consumption of 20 g (∼1.5 drinks) daily for 6 weeks affected the pharmacokinetics of fluvastatin but did not influence its hypolipidemic effects or aminotransferase levels.10 Even more strikingly, a case-control study of 289 patients taking warfarin found that moderate alcohol consumption (1 drink every other day to 2 drinks per day) was associated with an 80% lower risk of extreme elevations in the international normalized ratio (INR).11 Thus, the true risk associated with combined use of alcohol in moderation and common medications remains uncertain.
To address these questions, we examined the relationship between alcohol consumption and the risk of elevations in alanine aminotransferase (ALT) or INR levels among men enrolled in the Post-Coronary Artery Bypass Graft (CABG) Trial.12, 13 This trial randomly assigned patients with previous CABG surgery in a factorial design to high-dose or low-dose lovastatin and to low-dose warfarin or placebo, with a standardized, intensive protocol to monitor medication safety.
Methods
Study Population and Design
The Post-CABG Trial was a multicenter, double-blind, randomized, controlled trial. Participants had undergone CABG surgery 1 to 11 years before entry and were 21 to 74 years of age. Other inclusion criteria included: the presence of 2 completely independent and patent saphenous vein grafts, a low-density lipoprotein cholesterol (LDL-C) between 130 and 175 mg/dL, plasma triglycerides less than 300 mg/dL, and a left ventricular ejection fraction of at least 30%. Exclusion criteria included: a likelihood of revascularization or death in 5 years, unstable angina or myocardial infarction within 6 months before the start of the trial, severe angina, heart failure, and absolute contraindications to treatment with any of the study medications. A total of 1351 patients enrolled in the study at the Montreal Heart Institute, Cedars-Sinai Medical Center, Cleveland Clinic, University of Minnesota, and Baylor College of Medicine. All participants provided written informed consent at the onset of the trial.
All prospective participants received active warfarin treatment with 1 mg/d for a month before randomization. Because low-intensity anticoagulation was of primary interest, only participants who demonstrated a minimal reaction to warfarin and consumed more than 90% of the prescribed medication were randomized. All participants also received instruction regarding a National Cholesterol Education Program Step I diet and were offered programs on exercise and smoking cessation.
Patients were randomly assigned in a 2 × 2 factorial design to 1 of 4 treatment groups: aggressive LDL-C lowering with lovastatin 40 to 80 mg/d to achieve an LDL-C of 60 to 85 mg/dL; moderate LDL-C lowering with lovastatin 2.5 to 5 mg/d to achieve an LDL-C of 130 to 140 mg/dL; warfarin 1 to 4 mg/d to achieve an INR of 1.8 to 2.0; and warfarin-placebo. Participants could receive cholestyramine 8 mg/d as necessary to achieve LDL-C goals. Adherence to prescribed treatment with lovastatin was generally excellent; 85% to 90% took the medication as prescribed.12 The mean doses of lovastatin were 4 mg in the low-dose group and 76 mg in the high-dose group.
Alcohol Consumption
In a baseline interview, participants reported their usual weekly consumption of alcohol in drinks per week. A standard drink was specified as a 5-ounce glass of wine, 12 ounces of beer, or a single mixed drink. Because only 17 of the 102 female participants consumed any alcohol, we restricted our analyses to the 1249 male participants. On the basis of the participants’ reported alcohol consumption, we divided men into 4 groups: nondrinkers, those consuming 1 to 6 drinks per week (light drinkers), those consuming 7 to 13 drinks per week (moderate drinkers), and those consuming 14 or more drinks per week (heavier drinkers).
A history of excessive alcohol consumption did not contraindicate enrollment in the Post-CABG Trial. If a prospective participant reported such a history, study personnel were prompted to evaluate the prospective participant’s suitability for a 5-year study.
To validate self-reported alcohol consumption, we assessed the correlation of alcohol intake with levels of high-density lipoprotein cholesterol at baseline. The Spearman correlation coefficient was 0.23 (P<.001), similar to that reported in multiple other studies.14, 15, 16, 17
Outcome Measurements
All participants were followed for 5 years. The primary endpoint of the trial was atherosclerotic progression assessed angiographically. Study staff saw participants every 6 weeks for 15 months and every 3 months thereafter. Participants underwent routine measurement of ALT and INR levels at each visit. Participants could undergo dose reduction of lovastatin in the event of ALT levels more than twice the upper normal range. Participants also underwent dose reduction in the event of creatine kinase levels more than 3 times the normal range, but these were not monitored at regular intervals. No participant developed rhabdomyolysis. With the use of Biotrack machines (Ciba Corning, Mountain View, CA), AQ study staff decreased warfarin doses when INR levels were 2.0 or higher. Four abstainers and one light drinker did not have information on ALT or INR levels during follow-up and were not included in our analyses. Field center staff also reported hemorrhagic events that required hospitalization at any time during follow-up, including the site of bleeding and whether transfusion was required.
Other Covariates
Participants reported their smoking history at baseline, which we categorized as current, former, and never. Participants reported their physical activity level relative to others of their age and sex in 5 categories; we grouped the upper 2 and lower 2 categories to create 3 levels of activity. Study staff measured the height and weight of all participants; we grouped body mass index into 3 categories (<25 kg/m2, 25-29.9 kg/m2, and 30.0+ kg/m2). Macronutrient intake was estimated by a 24-hour dietary recall performed at baseline.18
Statistical Methods
We compared baseline alcohol consumption to 2 measures of medication safety: ALT levels in relation to lovastatin use and INR levels among patients assigned to warfarin. For ALT levels, we examined maximal ALT levels (the highest recorded ALT at any visit) and the risk of having an ALT of at least 80 IU/L at any visit as our primary endpoints. We performed our primary analyses, including all participants, and restricted them to those receiving high-dose lovastatin. For INR levels, we examined maximal INR levels (the highest recorded INR at any visit) and the risk of having an INR of at least 2.0 among those participants assigned to warfarin. As secondary analyses, we explored the risks of having an ALT of at least 60 or 100 IU/L or requiring dose reduction of lovastatin; we also explored the risk of having an INR of 2.5 on warfarin.
We used Fisher exact tests for unadjusted binary variables and analysis of variance for unadjusted continuous variables. We performed logarithmic transformation of maximal ALT and INR levels because their distributions had a rightward skew; we, therefore, present exponentiated results of these variables for comparability with clinical values. In linear regression models, we examined differences in maximal ALT or INR levels according to alcohol use adjusted for potentially confounding factors; we present least-square means from these models. We used logistic regression models for risks of elevated INR or ALT levels. We included: age, race, body mass index (in 3 categories), physical activity (in 3 categories), current smoking, former smoking, history of myocardial infarction, history of hypertension, history of stroke, and intake of calories, fat, and protein (as continuous variables). To confirm the robustness of our models, we adjusted for selected covariates in stepwise selection models (with entry and stay criteria of P = .20, forcing alcohol use into the model). Results from the stepwise models closely approximated results from models with all covariates in all cases and are not shown here.
The SAS System for Windows, Release 8.01 (SAS Institute, Inc., Cary, NC) was used for all analyses. No external funding was used in this study.
Results
Table 1 shows characteristics of the 1244 male Post-CABG participants with information on medication safety, according to usual alcohol consumption. As expected, alcohol consumption was associated with younger age and a greater prevalence of current smoking. Overall, the mean age of participants was 61.5 years, and 94% of men were white.
Table 1. Characteristics of 1244 Men Enrolled in the Post-Coronary Artery Bypass Graft Trial, According to Baseline Alcohol Consumption
| Weekly drinks of alcohol | ||||
|---|---|---|---|---|
| None (n = 674) | 1-6 (n = 346) | 7-13 (n = 138) | 14+ (n = 86) | |
| Median alcohol intake (drinks/wk) | 0 | 3 | 9 | 14 |
| Age (y) | 61.7 ±7.4 | 61.3±7.0 | 61.0±7.6 | 60.8±7.3 |
| White | 624(93) | 335(97) | 134(97) | 82(95) |
| Smoking: | ||||
| Never | 187(28) | 86(25) | 18(13) | 14(16) |
| Quit 2+ y ago | 388(58) | 200(58) | 95(69) | 53(62) |
| Quit <2 y ago | 30(4) | 30(9) | 5(4) | 3(3) |
| Current | 69(10) | 30(9) | 20(14) | 16(19) |
| Body mass index (kg/m2) | 28.0±4.4 | 27.5±4.3 | 27.5±4.4 | 27.5±3.6 |
| Energy intake (kcal/d) | 1835±593 | 1866±555 | 1830±538 | 2011±663 |
| Protein intake (g/d) | 83±28 | 84±29 | 83±28 | 84±30 |
| Fat intake (g/d) | 63±30 | 64±32 | 59±28 | 65±31 |
| LV ejection fraction (%) | 56.7±11.9 | 56.0±11.7 | 55.6±11.7 | 57.9±11.9 |
| History of: | ||||
| Excessive alcohol use | 35(5) | 2(1) | 1(1) | 3(3) |
| Myocardial infarction | 323(48) | 175(51) | 72(52) | 40(47) |
| Stroke | 17(3) | 9(3) | 5(4) | 3(3) |
| Hypertension1 | 244(36) | 114(33) | 48(35) | 33(38) |
| Less active than others2 | 95(14) | 31(9) | 9(7) | 9(10) |
| HDL-C (mg/dL) | 37±8 | 40±9 | 42±10 | 45±11 |
| Treatment assignment | ||||
| High-dose lovastatin/warfarin | 160(24) | 88(25) | 41(30) | 21(24) |
| High-dose lovastatin/placebo | 185(27) | 77(22) | 30(22) | 17(20) |
| Low-dose lovastatin/warfarin | 163(24) | 93(27) | 34(25) | 25(29) |
| Low-dose lovastatin/placebo | 166(25) | 88(26) | 33(24) | 23(27) |
Of the 1244 men, 674 (54%) drank no alcohol or less than 1 drink weekly (nondrinkers), 346 (28%) consumed 1 to 6 drinks per week (light), 138 (11%) consumed 7 to 13 drinks per week (moderate), and 86 (7%) consumed 14 or more drinks per week (heavier). In the heavier drinking group, 42 men consumed more than 14 drinks per week, and 9 men consumed more than 21 drinks per week.
The dose of warfarin required to maintain goal INR levels tended to be slightly higher among men with greater alcohol consumption. For example, age-adjusted mean warfarin doses between clinic visits 13 and 20 (the period during which doses were most stable) were 2.5 mg among abstainers, 2.6 mg among light and moderate drinkers, and 2.8 among heavier drinkers.
Risk of Elevated Alanine Aminotransferase
During follow-up, 45 men (4%) underwent dose reduction because of elevated ALT levels, and only 7 men underwent dose reduction for elevated creatine kinase levels (4 nondrinkers and 3 light drinkers). ALT levels reached or exceeded 80 IU/mL in 106 men (9%). A total of 31 men (2%) had an ALT level of 120 IU/L or greater, too few to be analyzed separately. INR levels reached or exceeded 2.0 in 413 (66%) of the 625 men assigned to warfarin.
The risks of abnormal ALT levels according to alcohol use are shown in Table 2. There was a trend toward greater maximum ALT with greater alcohol consumption, although the absolute magnitude of this difference was small. There was no clear relation of alcohol use to the risk of an ALT level of at least 80 IU/L (our primary endpoint) or to the risk of requiring dose reduction (Table 2). In analyses that explored alternate cutpoints, alcohol use was associated with a higher risk of an ALT of at least 60 IU/L in a stepwise manner, but we did not find a similar graded relationship with a cutpoint of 100 IU/L.
Table 2. Measures of abnormal Alanine Aminotransferase levels according to alcohol use and lovastatin dose
| Weekly drinks of alcohol | P value | ||||
|---|---|---|---|---|---|
| None (n = 674) | 1-6 (n = 346) | 7-13 (n = 138) | 14+ (n = 86) | ||
| All participants | |||||
| Primary analyses | |||||
| Maximum ALT±SE (IU/L) | 39.1±0.8 | 40.0±1.0 | 41.9±2.0 | 40.3±2.1 | .51 |
| Adjusted maximum ALT±SE | 39.0±0.7 | 39.9±1.0 | 41.9±1.7 | 41.7±2.2 | .08 |
| Cases of ALT 80+ (%) | 55(8) | 33(10) | 13(9) | 5(6) | .70 |
| Adjusted odds ratio (95% CI) | Ref | 1.2(0.8-1.9) | 1.2(0.6-2.3) | 0.8(0.3-2.0) | .85 |
| Secondary analyses | |||||
| Cases of ALT 60+ (%) | 109(16) | 58(17) | 28(20) | 19(22) | .38 |
| Adjusted odds ratio (95% CI) | Ref | 1.0(0.7-1.5) | 1.3(0.8-2.1) | 1.7(0.9-3.0) | .05 |
| Cases of ALT 100+ (%) | 21(3) | 11(3) | 7(5) | 3(3) | .22 |
| Adjusted odds ratio (95% CI) | Ref | 1.2(0.5-2.5) | 2.0(0.8-5.2) | 1.4(0.4-5.2) | .22 |
| Cases of dose reduction (%) | 23(3) | 12(3) | 8(6) | 2(2) | .53 |
| Adjusted odds ratio (95% CI) | Ref | 1.0(0.5-2.1) | 1.8(0.8-4.1) | 0.7(0.2-3.3) | .60 |
| High-dose lovastatin | |||||
| Maximum ALT±SE | 40.3±0.8 | 39.9±1.5 | 45.5±2.8 | 41.0±3.0 | .22 |
| Adjusted maximum ALT±SE | 40.4±1.0 | 39.7±1.4 | 45.1±2.5 | 42.3±3.2 | .12 |
| Cases of ALT 80+ (%) | 27(8) | 15(9) | 7(10) | 2(5) | .83 |
| Adjusted odds ratio (95% CI) | Ref | 1.2(0.6-2.3) | 1.1(0.4-2.8) | 0.7(0.1-3.1) | .81 |
Results restricted to men randomized to high-dose lovastatin were similar, with little relation of alcohol use to maximal ALT levels or to the risk of an ALT level of at least 80 (Table 2). The graded positive relation between alcohol use and risk of an ALT of 60 or higher found in exploratory analyses was similar in magnitude (but not statistically significant) among men randomized to high-dose lovastatin (data not shown); the number of cases of dose reduction or an ALT of 100 IU/L or higher were too few for analysis.
Risk of Elevated International Normalized Ratio
Table 3 shows similar results for INR levels among men assigned to warfarin. We found no clear relation between alcohol use and maximal INR levels or the risk of an INR of 2.0 or higher. In alternative analyses that explored a higher INR cutpoint, heavier alcohol intake was associated with a lower risk of an INR of 2.5 or higher in a graded manner.
Table 3. Measures of abnormal International Normalized Ratio levels according to alcohol use among Post-Coronary Artery Bypass Graft Trial participants assigned to warfarin
| Weekly drinks of alcohol | P value | ||||
|---|---|---|---|---|---|
| None (n = 326) | 1-6 (n = 181) | 7-13 (n = 75) | 14+ (n = 46) | ||
| Maximum INR±SE | 2.2±0.05 | 2.2±0.05 | 2.2±0.07 | 2.1±0.08 | .72 |
| Adjusted maximum INR±SE | 2.2±0.04 | 2.1±0.06 | 2.2±0.09 | 2.1±0.11 | .36 |
| Cases of INR 2.0+ | 215(67) | 119(66) | 51(68) | 28(61) | .86 |
| Adjusted odds ratio | Ref | 1.0(0.6-1.4) | 1.0(0.6-1.8) | 0.8(0.4-1.5) | .62 |
| Cases of INR 2.5+ | 113(35) | 61(34) | 21(28) | 9(20) | .16 |
| Adjusted odds ratio | Ref | 0.9(0.6-1.4) | 0.7(0.4-1.2) | 0.5(0.2-1.0) | .03 |
| Hemorrhagic events | 12(4) | 4(2) | 2(3) | 0(0) | .63 |
Only 18 men assigned to warfarin developed hemorrhages that required hospitalization during follow-up (Table 3). Of these, 8 were gastrointestinal and 4 were genitourinary, and 8 required transfusion. Although the number of hemorrhages was too small for formal analysis, no hemorrhages occurred among men consuming 14 or more drinks per week.
Our overall findings were similar when we assessed men consuming more than 14 drinks per week (ie, more than currently recommended levels). Such men had maximal ALT and INR levels just slightly lower than abstainers, and only 1 man developed an ALT of 80 IU/L or higher.
Discussion
In this analysis of men enrolled in the Post-CABG Trial, we found no evidence that moderate drinking was associated with clinically significant elevations in ALT levels among participants taking lovastatin or similar elevations in INR levels among men taking low-dose warfarin. In secondary analyses, alcohol consumption was associated with a greater risk of low-level ALT elevations (≥60 IU/L) but a lower risk of high-level INR elevations (≥2.5 or greater).
The association of moderate alcohol use with medication safety is not well described. With regard to warfarin, some,9 but not all,19 authors recommend alcohol be avoided completely. In a multicenter study of complications of chronic anticoagulation, Fihn and colleagues found that patients with a history of alcohol abuse did not have a higher incidence of bleeding complications.20 Compared with all other patients, binge drinkers older than 65 years showed a nonsignificant trend toward more frequent serious bleeding (hazard ratio 1.7; 95% confidence interval, 0.8-3.9). In a case-control study of an outpatient anticoagulation unit, Hylek and colleagues11 found a U-shaped association between alcohol use and risk of an INR of 6.0 or higher. The odds ratios were 0.7 (95% confidence interval, 0.3-1.6) among consumers of 1 to 2 drinks per week, 0.2 (95% confidence interval, 0.1-0.7) among consumers of 3 drinks per week to 2 drinks per day, and 0.9 (95% confidence interval, 0.2-4.2) among consumers of more than 2 drinks per day or binge drinkers.
Even less is known about the relation of moderate alcohol use and statin toxicity. In a rat model, coadministration of alcohol and simvastatin actually reduced aspartate aminotransferase levels compared with alcohol administration alone.21 In a study of 20 hypercholesterolemic volunteers, Smit and colleagues10 found that coadministration of fluvastatin and alcohol for 6 weeks produced a greater area under the concentration curve and a greater time to maximum concentration for fluvastatin than administration of fluvastatin alone. The authors reported no evidence that coadministration increased aminotransferase levels. Surprisingly, the same authors found that fluvastatin administered with 70 g of alcohol to 10 healthy volunteers reduced the half-life of fluvastatin without a change in the area under the concentration curve.22
ALT elevations of a clinically relevant magnitude were infrequent in the Post-CABG Trial. Current labeling recommends dose reduction only for sustained ALT elevations of 3 times the upper limit of normal, a level rarely reached in this trial. Although this record of safety provides cautious reassurance that concomitant moderate drinking and lovastatin use are generally safe, it also limits our power to detect all clinically meaningful risk estimates. Fortunately, there seems to be an even lower absolute risk of hepatotoxicity of statins in clinical practice settings.23 A similar caution is warranted regarding warfarin. Because hemorrhage occurred rarely in the Post-CABG Trial, we cannot exclude the possibility that the antiplatelet effects of alcohol24 increase the risk of hemorrhage among warfarin users to at least some degree, although the absolute risk seems low.
The potential for interactions between alcohol use and medication use is substantial, particularly among older adults who may be more susceptible to the effects of alcohol. In a survey of 311 retirement community residents, Adams25 found that 38% reported using both alcohol and a medication with which it could interact. Likewise, Fink and colleagues26 found that among 549 older adults, 154 (28%) were identified as hazardous drinkers on the basis of moderate drinking and concurrent use of potentially interacting medications. Our results in a population of men with a mean age of 61.5 years provide some reassurance about 2 common medications; similar studies on other common medications and in older populations are still needed.
Our study has both strengths and limitations. For example, although the Post-CABG Trial was large and based in multiple centers, clinical trial participants may practice healthier behaviors than other patients, potentially limiting the generalizability of our study. Even more important, we could not study women, whose metabolism of alcohol differs from that of men. We also cannot necessarily extrapolate our results to statins other than lovastatin or to full-dose anticoagulation with warfarin. Only studies of higher potency statins and studies including women can provide needed reassurance about the safety of alcohol use in the broader population at risk.
The assessment of ALT and INR levels in the Post-CABG Trial was uniform, regular, and intensive, all important strengths. However, we did not have information on other measures of medication safety, and it is possible that these may have yielded different results.
The run-in period for warfarin use was intended to limit the subsequent variability in INR levels and may have led to a false finding of no association between alcohol use and safety of anticoagulation. Nonetheless, the prevalence of INR levels above goal was quite high, with more than half of participants having an INR over 2.0 at some time, providing this study with substantial power and relatively precise estimates.
We relied on self-reported alcohol use, which mimics standard clinical care and seems to be reliable in other populations.16 We also validated this assessment with measured high-density lipoprotein cholesterol levels. Although the rank order of participants is therefore likely to be substantially correct, we may have over- or underestimated the actual consumption of participants. Moreover, alcohol consumption was ascertained only at baseline, and we could not assess changes in alcohol use. Finally, heavy regular alcohol consumption was uncommon among Post-CABG participants, and participants were not asked about episodic heavy drinking, limiting our ability to determine whether they have particular effects on the safety of lovastatin or warfarin.
In summary, among men who had undergone previous CABG surgery, we found little evidence that moderate alcohol use affects the safety of either lovastatin or low-dose warfarin as measured by ALT and INR levels. Although these results provide no reason for physicians to recommend moderate drinking, they provide reassurance about its safety among the increasing number of patients taking statins or anticoagulants.
Acknowledgments
The Post-CABG Study was conducted and supported by the National Heart, Lung, and Blood Institute (NHLBI) in collaboration with the Post-CABG Study Investigators. This article was prepared using a limited access dataset obtained by the NHLBI and does not necessarily reflect the opinions or views of the Post-CABG Study or the NHLBI. A short abstract of this work was presented at the 2004 American Heart Association Scientific Sessions and published accordingly in Circulation.
References
- . Demonstration of deductive meta-analysis (ethanol intake and risk of myocardial infarction) . Epidemiol Rev . 1993;15:328–351
- . Moderate alcohol intake and lower risk of coronary heart disease (meta-analysis of effects on lipids and haemostatic factors) . BMJ . 1999;319:1523–1528
- . Prior alcohol consumption and mortality following acute myocardial infarction . JAMA . 2001;285:1965–1970
- . Mortality and light to moderate alcohol consumption after myocardial infarction . Lancet . 1998;352:1882–1885
- . Wine drinking and risks of cardiovascular complications after recent acute myocardial infarction . Circulation . 2002;106:1465–1469
- . AHA Science Advisory: Wine and your heart: a science advisory for healthcare professionals from the Nutrition Committee, Council on Epidemiology and Prevention, and Council on Cardiovascular Nursing of the American Heart Association . Circulation . 2001;103:472–475
- . Balancing the risks and benefits of moderate drinking . Ann N Y Acad Sci . 2002;957:1–6
- Using data to enhance the expert panel process. Rating indications of alcohol-related problems in older adults . Int J Technol Assess Health Care . 2001;17:125–136
- . Alcohol and medication interactions . Alcohol Res Health . 1999;23:40–54
- . Effects of alcohol and fluvastatin on lipid metabolism and hepatic function . Ann Intern Med . 1995;122:678–680
- . Acetaminophen and other risk factors for excessive warfarin anticoagulation . JAMA . 1998;279:657–662
- . The effect of aggressive lowering of low-density lipoprotein cholesterol levels and low-dose anticoagulation on obstructive changes in saphenous-vein coronary-artery bypass grafts . N Engl J Med . 1997;336:153–162
- Long-term effects on clinical outcomes of aggressive lowering of low-density lipoprotein cholesterol levels and low-dose anticoagulation in the post coronary artery bypass graft trial . Circulation . 2000;102:157–165
- . High-density lipoprotein cholesterol and alcohol consumption in US white and black adults (data from NHANES II) . Am J Public Health . 1993;83:811–816
- Effects of similarities in lifestyle habits on familial aggregation of high density lipoprotein and low density lipoprotein cholesterol (the NHLBI Family Heart Study) . Am J Epidemiol . 1999;150:910–918
- The assessment of alcohol consumption by a simple self-administered questionnaire . Am J Epidemiol . 1991;133:810–817
- Alcohol consumption and carotid atherosclerosis in older adults (the Cardiovascular Health Study) . Arterioscler Thromb Vasc Biol . 2003;23:2252–2259
- . Intense dietary counseling lowers LDL cholesterol in the recruitment phase of a clinical trial of men who had coronary artery bypass grafts . J Am Diet Assoc . 1992;92:441–445
- . A review of enzyme induction of warfarin metabolism with recommendations for patient management . Pharmacotherapy . 1997;17:917–928
- Risk factors for complications of chronic anticoagulation. A multicenter study . Ann Intern Med . 1993;118:511–520
- The effect of co-administration of simvastatin and alcohol in rats . In Vivo . 2003;17:523–527
- . Effects of alcohol consumption on pharmacokinetics, efficacy, and safety of fluvastatin . Am J Cardiol . 1995;76:89A–96A
- . Screening for statin-related toxicity (the yield of transaminase and creatine kinase measurements in a primary care setting) . Arch Intern Med . 2003;163:688–692
- . Effect of ethanol on platelet function . Alcohol Clin Exp Res . 1999;23:1114–1118
- . Potential for adverse drug-alcohol interactions among retirement community residents . J Am Geriatr Soc . 1995;43:1021–1025
- The alcohol-related problems survey (identifying hazardous and harmful drinking in older primary care patients) . J Am Geriatr Soc . 2002;50:1717–1722
PII: S0002-9343(05)00898-3
doi:10.1016/j.amjmed.2005.09.038
© 2006 Elsevier Inc. All rights reserved.
