- Email: [email protected]
Is the apoE4 allele an independent predictor of coronary events?
Is the apoE4 Allele an Independent Predictor of Coronary Events? Angelo Scuteri, MD, PhD, Angelo J. G. Bos, MD, PhD, Alan B. Zonderman, PhD, Larry J. Brant, PhD, Edward G. Lakatta, MD, Jerome L. Fleg, MD PURPOSE: Although the apolipoprotein E genotype ⑀4 (apoE4) has been associated with high cholesterol levels, whether it is an independent predictor of coronary events is not certain. SUBJECTS AND METHODS: We measured apoE genotypes in 730 participants in the Baltimore Longitudinal Study of Aging (421 men and 309 women, mean [⫾ SD] age of 52 ⫾ 17 years) who were free of preexisting coronary heart disease. A proportional hazards regression model was used to study the association between risk factors and the occurrence of coronary events, defined as angina pectoris, documented myocardial infarction by history or major Q waves on the electrocardiogram (Minnesota Code 1:1 or 1:2), or coronary death, adjusted for other risk factors, including total plasma cholesterol level. RESULTS: The apoE4 allele was observed in 200 subjects (27%), including 183 heterozygotes and 17 homozygotes. Cor-
onary risk factor profiles were similar in those with and without apoE4. Coronary events developed in 104 (14%) of the 730 subjects, including 77 (18%) of the 421 men during a mean follow-up of 20 years and 27 (9%) of the 309 women during a mean follow-up of 13 years. Coronary events occurred significantly more frequently in subjects with apoE4 (n ⫽ 40, 20%) than in those without this allele (64, 12%, P ⬍0.05). In a multivariate model, apoE4 was an independent predictor of coronary events in men (risk ratio [RR]⫽ 2.9, 95% confidence interval [CI]: 1.8 to 4.5, P ⬍0.0001) but not in women (RR ⫽ 0.9, 95% CI: 0.4 to 1.9, P ⫽ 0.62). CONCLUSION: The apoE4 genotype is a strong independent risk factor for coronary events in men, but not women. The association does not appear to be mediated by differences in total cholesterol levels. Am J Med. 2001;110:28 –32. 䉷2001 by Excerpta Medica, Inc.
A
lesterol levels is unclear. The aim of the present study was to investigate whether the apoE4 allele is a predictor of coronary events, independent of traditional coronary risk factors, in adults without evidence of preexisting cardiovascular disease.
polipoprotein E (apoE) is a liver polypeptide that serves as a ligand for the low-density lipoprotein (LDL) receptor (1). ApoE is a structural component of both chylomicron and very low density lipoprotein (VLDL) remnants. It is thought to mediate their binding and uptake by both the LDL receptor and the LDL receptor-related protein (1,2). In humans, there are three alleles of apoE (⑀2, ⑀3, and ⑀4) and hence six different genotypes (2/2, 2/3, 2/4, 3/3, 3/4, and 4/4). The frequencies of apoE genotypes vary in different age and racial groups (3–5). The apoE polymorphism explains 4% to 15% of the variation in serum LDL cholesterol concentrations (4,6). The apoE allele ⑀4 (apoE4) is associated with higher LDL cholesterol levels (6), presumably because it downregulates hepatic LDL receptors. The ⑀4 allele has been associated with several pathologic conditions, including gallstone formation (7,8), carcinoma of the proximal colon (9), and Alzheimer’s disease (10 –12). There is some evidence that the apoE genotype is associated with cardiovascular disease (6,13), but whether this is mediated by differences in serum choFrom the Laboratory of Cardiovascular Science (AS, AJGB, EGL, JLF), the Laboratory of Personality and Cognition (ABZ), and the Research Resources Branch (ABZ, LJB), Gerontology Research Center, National Institute on Aging, National Institutes of Health, Baltimore, Maryland. Requests for reprints should be addressed to Angelo Scuteri, MD, PhD, Laboratory of Cardiovascular Sciences, Gerontology Research Center, NIA-NIH, 5600 Nathan Shock Drive, Baltimore, Maryland 21224. Manuscript submitted March 20, 2000, and accepted in revised form August 25, 2000. 28
䉷2001 by Excerpta Medica, Inc. All rights reserved.
METHODS Subjects and Study Design We measured apoE genotypes in a sample of 814 participants in the Baltimore Longitudinal Study of Aging, which is a prospective study of men and women that has been conducted since 1958. Participants are communitydwelling volunteers 21 to 96 years of age not seeking healthcare, who agreed to return to the Gerontology Research Center every 2 years for 2 and a half days of physical, physiologic, and psychologic examinations (14). Subjects were selected for the current investigation if they were free of preexisting coronary artery disease on their initial study visit, as defined by history of angina pectoris, documented myocardial infarction, or major Q waves on a resting electrocardiogram ([ECG] Minnesota Code 1:1 or 1:2) (15). Eighty-four (10%) of the original 814 subjects were excluded because of a prior coronary event, including 54 with angina pectoris, 16 with a history of myocardial infarction, and 14 with pathologic Q waves, leaving 730 subjects for this analysis. The apoE genotype was determined by polymerase chain reaction amplification of leukocyte DNA followed by HhaI digestion and product characterization (16). The 0002-9343/01/$–see front matter PII S0002-9343(00)00639-2
ApoE4 Allele and Coronary Events/Scuteri et al
prevalence of the apoE4 allele in the 84 excluded subjects was 34%.
Follow-Up All participants were evaluated for the development of new coronary events during subsequent biennial visits, supplemented by hospital records, death certificates, and autopsy data. Events were defined by the development of angina pectoris, myocardial infarction, or coronary death (fatal myocardial infarction or sudden death). The development of angina pectoris was determined from the subject’s response to a Rose questionnaire and clinical assessment. Myocardial infarction was diagnosed by conventional clinical criteria during a hospitalization, or by the development of diagnostic Q waves on the resting ECG (Minnesota code 1:1 or 1:2) (15). For participants who died, the cause of death was determined by the consensus of three study physicians after review of the death certificate, hospital records, and autopsy data, and communication with the subject’s family and personal physician, as available. Follow-up time for participants who experienced an event was calculated to the actual event date or approximated to the date of their next visit, as in patients with angina or new Q waves. Only the first coronary event was considered for participants with more than one event. For subjects free of coronary events, follow-up time was calculated to the time of their last biennial visit or their death from a noncoronary cause.
Statistical Analysis All analyses were performed with SAS statistical software (Cary, North Carolina). Continuous variables were expressed as mean ⫾ SD and categorical variables as proportions. An unpaired t test was used for comparison between groups for continuous variables, and the chisquare test was used for categorical variables. HardyWeinberg equilibrium was tested with a chi-square test. A Cox proportional hazards regression model was used to study the association between risk factors and the occurrence of coronary events (17). Predictor variables included the presence of the apoE4 genotype, age at first examination, sex, smoking more than 10 cigarettes per day, systolic blood pressure (per 10 mm Hg increase), fasting total plasma cholesterol level (per 20 mg/dL increase), and body mass index in kg/m2 (per 2 unit increase). Risk ratios (RR) and 95% confidence intervals (CI) were determined. The proportionality assumption of the Cox model was not valid for age (but was valid for the other variables in the model); thus multivariate models were stratified by age (18), with groups defined as younger than 50 years, 50 to 65 years, and older than 65 years). We tested for interactions between apoE4, plasma cholesterol level, and coronary events, and between apoE4, sex, and coronary events, by using interaction terms. Kaplan-Meier survival curves were used to compare survival in women and men with and without the
apoE4 allele, and were tested using the log-rank test. A two-tailed P value ⬍0.05 was considered statistically significant.
RESULTS Distribution of Risk Factors at Enrollment
The mean age of the participants at entry was 52 ⫾ 15 years (range 20 to 91), and 58% were men. The mean follow-up time was 20 ⫾ 10 years (maximum 39) for men and 13 ⫾ 6 years (maximum 20) for women. (The study began enrolling men in 1958 and women in 1978). Coronary events developed in 104 (14%) of the 730 subjects, including 77 (18%) of the 421 men and 27 (9%) of the 309 women (P ⫽ 0.001). The mean time to the first event was 17 ⫾ 7 years (range 3 to 39). These events consisted of 61 cases of angina pectoris, 22 cases of new Q-wave myocardial infarctions on ECG, and 14 cases of new myocardial infarctions by history; in addition, 7 participants had death as their initial coronary event. The distribution of the genotypes conformed with Hardy-Weinberg equilibrium (2/2, 0.3%; 2/3, 12%; 2/4, 1.7%; 3/3, 60.6%; 3/4, 23.1%; and 4/4, 2.3%). The apoE4 allele was present in 200 subjects (183 heterozygotes and 17 homozygotes), for a prevalence of 27%. The prevalence of apoE4 did not differ by sex (25% in men and 30% in women, P ⫽ 0.63). Subjects with apoE4 did not differ in the distribution of traditional cardiovascular risk factors at baseline compared with those without the apoE4 allele (Table 1).
Predictive Role of ApoE4 for New Coronary Events Coronary events were also associated with older age, greater systolic blood pressure, and total serum cholesterol levels but not body mass index or diastolic blood pressure. Coronary events were significantly more frequent in subjects with the apoE4 allele (20% [40 of 200]) than in those without it (12% [64 of 530], P ⬍0.05). After stratification for age and adjustment for other coronary risk factors, apoE4 remained a significant predictor of coronary events (Table 2), as did systolic blood pressure and male sex. We found no evidence of an interaction between plasma cholesterol level and apoE4 (P for interaction ⫽ 0.72). In a separate analysis that included only myocardial infarction and coronary death as coronary events, the presence of apoE4 (RR ⫽ 1.7; 95% CI: 1.1 to 3.3, P ⬍0.01) remained a significant predictor.
Effects of ApoE4 in Subjects with Complete Lipoprotein Profile Data on LDL and high-density lipoprotein (HDL) cholesterol levels were available for 321 subjects (158 men and 163 women), including 94 subjects (29%) with apoE4 (51 [32%] of the men and 43 [27%] of the womJanuary 2001
THE AMERICAN JOURNAL OF MEDICINE威
Volume 110 29
ApoE4 Allele and Coronary Events/Scuteri et al
Table 1. Distribution of Risk Factors at Baseline by ApoE4 Genotype No apoE4 (n ⫽ 530) Characteristic (Unit) Age (years) Male sex Current smoking Body mass index (kg/m2) Systolic blood pressure (mm Hg) Diastolic blood pressure (mm Hg) Total plasma cholesterol (mg/dL)
ApoE4 (n ⫽ 200)
Number (%) or mean ⫾ SD 313 (59) 32 (6) 25 ⫾ 3 125 ⫾ 18 79 ⫾ 10 200 ⫾ 42
en). There were no differences in traditional cardiovascular risk factors between subjects with and without apoE4 in either sex, except that LDL cholesterol levels were higher in men with apoE4 than in the other men (127 ⫾ 35 mg/dL versus 116 ⫾ 31 mg/dL, P ⬍0.05). Seventeen (18%) of the subjects (14 men and 3 women) with apoE4 had a coronary event, compared with 26 (11%) of the remaining subjects. When analyzed by sex, 14 (28%) of the men with apoE4 versus 17 (16%) of the men without the ⑀4 allele had a coronary event (P ⫽ 0.07). The risk of coronary events was the same (7%) in women with and without apoE4. When LDL and HDL cholesterol levels were substituted for total cholesterol levels in proportional hazards model, neither LDL cholesterol (RR ⫽ 1.01 per 20 mg/dL, 95% CI: 0.98 to 1.05) nor HDL cholesterol (RR ⫽ 1.03 per 20 mg/dL, 95% CI: 0.99 to 1.18) levels were associated with coronary events, whereas apoE4 remained a significant predictor (RR ⫽ 3.2, 95% CI: 1.2 to 4.2, P ⬍0.01).
Effects of ApoE4 in Men and Women In a multivariate model, an interaction term (sex and apoE4) was significant (P ⬍0.05), indicating that the effects of apoE4 on coronary events were different in men and in women. Thus, separate analyses were performed to determine those sex-related differences. Of the 108 men with apoE4, 31 (29%) had coronary events, compared with 46 (15%) of the 313 men without apoE4 (P ⬍0.001). In an age-stratified multivariate model, apoE4
108 (54) 10 (5) 25 ⫾ 4 125 ⫾ 18 78 ⫾ 10 204 ⫾ 41
P Value 0.10 0.64 0.50 0.73 0.79 0.67 0.27
was associated with almost a tripling of the rate of coronary events in men (Table 3). For the 158 men for whom LDL and HDL cholesterol levels were available, apoE4 remained an independent predictor of coronary events (RR ⫽ 2.7, a 95% CI: 1.3 to 5.7, P ⬍0.01). By contrast, the incidence of coronary events did not differ between women with (8 [9%] of 94) and those without the apoE4 allele (19 [9%] of 215, P ⫽ 0.96). In an age-stratified multivariate model, neither apoE4 nor any other traditional cardiovascular risk factors were associated with coronary events in women (Table 3). Similar results were found when the analysis was performed in the 163 women for whom HDL and LDL cholesterol levels were available. Men with the apoE4 allele had significantly lower coronary-event–free survival than the other three groups (P ⬍ 0.01). Pairwise comparisons of the survival curves confirmed this difference for men with apoE4 compared with the other three groups (using a Bonferroni multiple comparisons correction, P ⬍0.05). The remaining three curves did not differ from one another.
DISCUSSION We found that the presence of the apoE4 allele was a strong predictor of coronary events among men but not among women. The association was independent of
Table 2. Predictors of Coronary Events: Results of Multivariate Proportional Hazards Model Risk Factor ApoE4 Female sex Current smoking Body mass index (per 2 kg/m2) Total cholesterol level (per 20 mg/dL) Systolic blood pressure (per 10 mm Hg)
Hazard Ratio (91% Confidence Interval) 2.01 0.6 1.2 1.03 1.07 1.23
The model was stratified by age (⬍50; 50 – 65; ⬎65 years; see methods). 30
January 2001
THE AMERICAN JOURNAL OF MEDICINE威 Volume 110
1.4–3.1 0.3–1.0 0.4–3.3 0.90–1.17 0.98–1.17 1.05–1.44
P Value 0.001 0.05 0.84 0.67 0.18 0.05
ApoE4 Allele and Coronary Events/Scuteri et al
Table 3. Sex-Specific Predictors of Coronary Events* Men Risk Factor ApoE4 Current smoking Body mass index (per 2 kg/m2) Total cholesterol level (per 20 mg/dL) Systolic blood pressure (per 10 mmHg)
Women
Hazard Ratio (95% Confidence Interval) 2.8 0.7 1.04 1.09 1.17
1.76–4.48 0.2–2.7 0.90–1.20 0.98–1.24 1.03–1.38
0.0001 0.54 0.61 0.16 0.01
0.81 2.5 1.03 1.12 0.97
P Value 0.35–1.87 0.5–12 0.80–1.32 0.89–1.37 0.78–1.22
0.62 0.25 0.84 0.38 0.80
* As estimated from an age-stratified multivariate proportional hazards model.
plasma cholesterol levels and other conventional coronary risk factors, which did not differ by apoE4 status. Dallongeville et al (19) reported that the frequency of apoE4 is increased in populations at high risk of coronary heart disease and is lower in geographic regions where the disease is less common. Several case-control studies and a few longitudinal studies have linked the presence of apoE4 with a greater risk of coronary heart disease (20 – 25). In one study, for example, the relative frequency of apoE4 predicted 40% of the variation in mortality for coronary events after adjustment for cholesterol levels (20). In a longitudinal study among elderly Finnish men, apoE4 was twice as common in those who died a coronary death as in event-free controls (23). In a nested case-control study in 619 high-risk men from the Multiple Risk Factor Intervention Trial (24), the presence of apoE4 was associated with an increased risk of coronary events, especially those that were fatal. In a meta-analysis, the risk of coronary artery disease for men with the apoE4 allele was estimated to be about 40% greater than for men without apoE4 (13). ApoE4 may also be associated with an increased risk of cerebrovascular ischemic events (26). In this study, we observed that the apoE4 allele increased the risk of coronary events in men but not in
women. A recent meta-analysis of nine studies including 181 women and 1,971 men with coronary heart disease concluded that although the apoE4 allele increased coronary risk overall, the data were not convincing for women (13). Investigators from the Copenhagen City Heart Study reported that although both men and women with the 4/3 or 4/4 genotype had greater levels of cholesterol and apoB than those with the 3/3 genotype, only men with apoE4 had an increased risk of ischemic heart disease (27). The mechanism for a sex-specific effect of apoE4 on the risk for coronary disease is unknown (28). The present study has several limitations. First, the sample included predominantly upper middle class, healthy, community-dwelling volunteers. In addition, given the shorter follow-up time and the lesser number of coronary events in women, it is difficult to ascertain whether the lack of an association between the apoE4 genotype and coronary events in women reflects a biological difference or is a result of low statistical power (the upper limit of the 95% confidence interval for the hazard ratio associated with apoE4 in women was 1.9). Another limitation is that a majority of coronary events consisted of new angina pectoris, although similar results were found when the analysis was restricted to “hard” events. Finally, LDL and HDL cholesterol level measurements were not available for all of the participants, although our results were similar in the subgroup for whom the values were available. We conclude that apoE4 is an independent risk factor for coronary events in men but not in women, and that the association is not mediated by differences in total cholesterol levels. Studies are needed to determine how apoE4 causes coronary events, and why there appear to be sex-related differences. This information might allow the development of preventive strategies that are tailored to the apoE genotype.
REFERENCES Figure 1. Cumulative coronary event-free survival by sex and apoE4 allele status. Event-free survival was significantly lower in men with apoE4 than in the other three subgroups (P ⬍0.01), but did not differ among the other three groups.
1. Mahley RW. Apolipoprotein E: cholesterol transport protein with expanding role in cell biology. Science. 1988;240:622– 640. 2. Beisiegel U, Weber W, Ihrke G, et al. The LDL-receptor-related protein, LRP, is an apolipoprotein E-binding protein. Nature. 1989; 341:162–164. January 2001
THE AMERICAN JOURNAL OF MEDICINE威
Volume 110 31
ApoE4 Allele and Coronary Events/Scuteri et al 3. Eggersten G, Tegelman R, Ericsson S, et al. Apolipoprotein E polymorphism in a healthy Swedish population: variation in allele frequency with age and relation to serum lipid concentrations. Clin Chem. 1993;39:2125–2129. 4. Gomez-Coronado D, Alvarez JJ, Entrala A, et al. Apolipoprotein E polymorphism in men and women from a Spanish population: allele frequencies and influence on plasma lipids and apolipoproteins. Atherosclerosis. 1999;147:167–176. 5. Ferrieres J, Sing CF, Roy M, et al. Apolipoprotein E polymorphism and heterozygous familial hypercholesterolemia: sex-specific effects. Arterioscler Thromb. 1994;14:1554 –1560. 6. Davignon J, Gregg RE, Sing CF. Apolipoprotein E polymorphism and atherosclerosis. Arteriosclerosis. 1988;8:1–21. 7. Juvonen T, Kervinen K, Kairaluoma MI, et al. Gallstone cholesterol content is related to apolipoprotein E polymorphism. Gastroenterology. 1993;104:1806 –1813. 8. Van Erpecum KJ, Carey MC. Apolipoprotein E4: another risk factor for cholesterol gallstone formation? Gastroenterology. 1996;111: 1764 –1767. 9. Kervinen K, Sodervik H, Makela J, et al. Is the development of adenoma and carcinoma in proximal colon related to apolipoprotein E phenotype? Gastroenterology. 1996;110:1785–1790. 10. Roses AD. Apolipoprotein E alleles as risk factors for Alzheimer’s disease. Annu Rev Med. 1996;47:387– 400. 11. Strittmatter WJ, Roses AD. Apolipoprotein E and Alzheimer’s disease. Proc Natl Acad Sci USA. 1995;92:4725– 4727. 12. Poirier J, Minnich A, Davignon J. Apolipoprotein E, synaptic plasticity and Alzheimer’s disease. Ann Med. 1995;27:663– 670. 13. Wilson PWF, Schafer EJ, Larson MG, Ordovas JM. Apolipoprotein E alleles and risk of coronary disease. A meta-analysis. Arterioscler Thromb Vasc Biol. 1996;16:1250 –1255. 14. Shock NW, Greulich RC, Andres RA, et al. Normal Human Aging. The Baltimore Longitudinal Study of Aging. Washington, DC: US Government Printing Office, NIH publication no. 84-2450; 1984:45. 15. Rose GA, Blackburn H. Cardiovascular Survey Methods. Geneva: World Health Organization; 1968. 16. Hixon JE, Vernier DT. Restriction isotyping of human apolipoprotein E by gene amplification and cleavage with HhaI. J Lipid Res. 1990;31:545–548.
32
January 2001
THE AMERICAN JOURNAL OF MEDICINE威 Volume 110
17. Cox DR. Analysis of Binary Data. London: Methuen & Co; 1970. 18. Kalbfleisch JD, Prentice RL. The Statistical Analysis of Failure Time Data. New York: Wiley & Sons; 1980. 19. Dallongeville J. Apolipoproteine E: proprie´te´s physiologiques, polymorphisme et atherosclerose. Sang Thrombose Vaisseaux. 1993; 5:707–717. 20. Stengard JH, Weiss KM, Sing CF. An ecological study of association between coronary heart disease mortality rates in men and the relative frequencies of common allelic variations in the gene coding for apolipoprotein E. Hum Genet. 1998;103:234 –241. 21. Ou T, Yamakawa-Kobayashi K, Arinami T, et al. Methylenetetrahydrofolate reductase and apolipoprotein E polymorphism are independent risk factors for coronary heart disease in Japanese: a case-control strudy. Atherosclerosis. 1998;137:23–28. 22. Wilson PWF, Myers RH, Larson MG, et al. Apolipoprotein E alleles, dyslipidemia, and coronary heart disease: the Framingham Offspring Study. JAMA. 1994;272:1666 –1671. 23. Stengard JH, Zerba KE, Pekkanen J, et al. Apolipoprotein E polymorphism predicts death from coronary artery disease in a longitudinal study of elderly Finnish men. Circulation. 1995;91:265–269. 24. Eichner JE, Kuller LH, Orchard TJ, et al. Relation of apolipoprotein E phenotype to myocardial infarction and mortality from coronary artery disease. Am J Cardiol. 1993;7:160 –165. 25. Katzel L, Fleg JL, Busby-Whitehead MJ, et al. Exercise-induced silent myocardial ischemia in master athletes. Am J Cardiol. 1998;81: 261–265. 26. Margaglione M, Seripa D, Gravina C, et al. Prevalence of apolipoprotein E alleles in healthy subjects and survivors of ischemic stroke: an Italian case-control study. Stroke. 1998;29:399 – 403. 27. Frikke-Schmidt R, Tybjaerg-Hansen A, Steffensen R, et al. Apolipoprotein E genotype: epsilon 32 women are protected while epsilon 43 and epsilon 44 men are susceptible to ischemic heart disease. The Copenhagen City Heart Study. J Am Coll Cardiol. 2000;35: 1192–1199. 28. Davignon J, Cohn JS, Mabile L, Bernier L. Apolipoprotein E and atherosclerosis: insight from animal and human studies. Clin Chim Acta. 1999;286:115–143.
onary risk factor profiles were similar in those with and without apoE4. Coronary events developed in 104 (14%) of the 730 subjects, including 77 (18%) of the 421 men during a mean follow-up of 20 years and 27 (9%) of the 309 women during a mean follow-up of 13 years. Coronary events occurred significantly more frequently in subjects with apoE4 (n ⫽ 40, 20%) than in those without this allele (64, 12%, P ⬍0.05). In a multivariate model, apoE4 was an independent predictor of coronary events in men (risk ratio [RR]⫽ 2.9, 95% confidence interval [CI]: 1.8 to 4.5, P ⬍0.0001) but not in women (RR ⫽ 0.9, 95% CI: 0.4 to 1.9, P ⫽ 0.62). CONCLUSION: The apoE4 genotype is a strong independent risk factor for coronary events in men, but not women. The association does not appear to be mediated by differences in total cholesterol levels. Am J Med. 2001;110:28 –32. 䉷2001 by Excerpta Medica, Inc.
A
lesterol levels is unclear. The aim of the present study was to investigate whether the apoE4 allele is a predictor of coronary events, independent of traditional coronary risk factors, in adults without evidence of preexisting cardiovascular disease.
polipoprotein E (apoE) is a liver polypeptide that serves as a ligand for the low-density lipoprotein (LDL) receptor (1). ApoE is a structural component of both chylomicron and very low density lipoprotein (VLDL) remnants. It is thought to mediate their binding and uptake by both the LDL receptor and the LDL receptor-related protein (1,2). In humans, there are three alleles of apoE (⑀2, ⑀3, and ⑀4) and hence six different genotypes (2/2, 2/3, 2/4, 3/3, 3/4, and 4/4). The frequencies of apoE genotypes vary in different age and racial groups (3–5). The apoE polymorphism explains 4% to 15% of the variation in serum LDL cholesterol concentrations (4,6). The apoE allele ⑀4 (apoE4) is associated with higher LDL cholesterol levels (6), presumably because it downregulates hepatic LDL receptors. The ⑀4 allele has been associated with several pathologic conditions, including gallstone formation (7,8), carcinoma of the proximal colon (9), and Alzheimer’s disease (10 –12). There is some evidence that the apoE genotype is associated with cardiovascular disease (6,13), but whether this is mediated by differences in serum choFrom the Laboratory of Cardiovascular Science (AS, AJGB, EGL, JLF), the Laboratory of Personality and Cognition (ABZ), and the Research Resources Branch (ABZ, LJB), Gerontology Research Center, National Institute on Aging, National Institutes of Health, Baltimore, Maryland. Requests for reprints should be addressed to Angelo Scuteri, MD, PhD, Laboratory of Cardiovascular Sciences, Gerontology Research Center, NIA-NIH, 5600 Nathan Shock Drive, Baltimore, Maryland 21224. Manuscript submitted March 20, 2000, and accepted in revised form August 25, 2000. 28
䉷2001 by Excerpta Medica, Inc. All rights reserved.
METHODS Subjects and Study Design We measured apoE genotypes in a sample of 814 participants in the Baltimore Longitudinal Study of Aging, which is a prospective study of men and women that has been conducted since 1958. Participants are communitydwelling volunteers 21 to 96 years of age not seeking healthcare, who agreed to return to the Gerontology Research Center every 2 years for 2 and a half days of physical, physiologic, and psychologic examinations (14). Subjects were selected for the current investigation if they were free of preexisting coronary artery disease on their initial study visit, as defined by history of angina pectoris, documented myocardial infarction, or major Q waves on a resting electrocardiogram ([ECG] Minnesota Code 1:1 or 1:2) (15). Eighty-four (10%) of the original 814 subjects were excluded because of a prior coronary event, including 54 with angina pectoris, 16 with a history of myocardial infarction, and 14 with pathologic Q waves, leaving 730 subjects for this analysis. The apoE genotype was determined by polymerase chain reaction amplification of leukocyte DNA followed by HhaI digestion and product characterization (16). The 0002-9343/01/$–see front matter PII S0002-9343(00)00639-2
ApoE4 Allele and Coronary Events/Scuteri et al
prevalence of the apoE4 allele in the 84 excluded subjects was 34%.
Follow-Up All participants were evaluated for the development of new coronary events during subsequent biennial visits, supplemented by hospital records, death certificates, and autopsy data. Events were defined by the development of angina pectoris, myocardial infarction, or coronary death (fatal myocardial infarction or sudden death). The development of angina pectoris was determined from the subject’s response to a Rose questionnaire and clinical assessment. Myocardial infarction was diagnosed by conventional clinical criteria during a hospitalization, or by the development of diagnostic Q waves on the resting ECG (Minnesota code 1:1 or 1:2) (15). For participants who died, the cause of death was determined by the consensus of three study physicians after review of the death certificate, hospital records, and autopsy data, and communication with the subject’s family and personal physician, as available. Follow-up time for participants who experienced an event was calculated to the actual event date or approximated to the date of their next visit, as in patients with angina or new Q waves. Only the first coronary event was considered for participants with more than one event. For subjects free of coronary events, follow-up time was calculated to the time of their last biennial visit or their death from a noncoronary cause.
Statistical Analysis All analyses were performed with SAS statistical software (Cary, North Carolina). Continuous variables were expressed as mean ⫾ SD and categorical variables as proportions. An unpaired t test was used for comparison between groups for continuous variables, and the chisquare test was used for categorical variables. HardyWeinberg equilibrium was tested with a chi-square test. A Cox proportional hazards regression model was used to study the association between risk factors and the occurrence of coronary events (17). Predictor variables included the presence of the apoE4 genotype, age at first examination, sex, smoking more than 10 cigarettes per day, systolic blood pressure (per 10 mm Hg increase), fasting total plasma cholesterol level (per 20 mg/dL increase), and body mass index in kg/m2 (per 2 unit increase). Risk ratios (RR) and 95% confidence intervals (CI) were determined. The proportionality assumption of the Cox model was not valid for age (but was valid for the other variables in the model); thus multivariate models were stratified by age (18), with groups defined as younger than 50 years, 50 to 65 years, and older than 65 years). We tested for interactions between apoE4, plasma cholesterol level, and coronary events, and between apoE4, sex, and coronary events, by using interaction terms. Kaplan-Meier survival curves were used to compare survival in women and men with and without the
apoE4 allele, and were tested using the log-rank test. A two-tailed P value ⬍0.05 was considered statistically significant.
RESULTS Distribution of Risk Factors at Enrollment
The mean age of the participants at entry was 52 ⫾ 15 years (range 20 to 91), and 58% were men. The mean follow-up time was 20 ⫾ 10 years (maximum 39) for men and 13 ⫾ 6 years (maximum 20) for women. (The study began enrolling men in 1958 and women in 1978). Coronary events developed in 104 (14%) of the 730 subjects, including 77 (18%) of the 421 men and 27 (9%) of the 309 women (P ⫽ 0.001). The mean time to the first event was 17 ⫾ 7 years (range 3 to 39). These events consisted of 61 cases of angina pectoris, 22 cases of new Q-wave myocardial infarctions on ECG, and 14 cases of new myocardial infarctions by history; in addition, 7 participants had death as their initial coronary event. The distribution of the genotypes conformed with Hardy-Weinberg equilibrium (2/2, 0.3%; 2/3, 12%; 2/4, 1.7%; 3/3, 60.6%; 3/4, 23.1%; and 4/4, 2.3%). The apoE4 allele was present in 200 subjects (183 heterozygotes and 17 homozygotes), for a prevalence of 27%. The prevalence of apoE4 did not differ by sex (25% in men and 30% in women, P ⫽ 0.63). Subjects with apoE4 did not differ in the distribution of traditional cardiovascular risk factors at baseline compared with those without the apoE4 allele (Table 1).
Predictive Role of ApoE4 for New Coronary Events Coronary events were also associated with older age, greater systolic blood pressure, and total serum cholesterol levels but not body mass index or diastolic blood pressure. Coronary events were significantly more frequent in subjects with the apoE4 allele (20% [40 of 200]) than in those without it (12% [64 of 530], P ⬍0.05). After stratification for age and adjustment for other coronary risk factors, apoE4 remained a significant predictor of coronary events (Table 2), as did systolic blood pressure and male sex. We found no evidence of an interaction between plasma cholesterol level and apoE4 (P for interaction ⫽ 0.72). In a separate analysis that included only myocardial infarction and coronary death as coronary events, the presence of apoE4 (RR ⫽ 1.7; 95% CI: 1.1 to 3.3, P ⬍0.01) remained a significant predictor.
Effects of ApoE4 in Subjects with Complete Lipoprotein Profile Data on LDL and high-density lipoprotein (HDL) cholesterol levels were available for 321 subjects (158 men and 163 women), including 94 subjects (29%) with apoE4 (51 [32%] of the men and 43 [27%] of the womJanuary 2001
THE AMERICAN JOURNAL OF MEDICINE威
Volume 110 29
ApoE4 Allele and Coronary Events/Scuteri et al
Table 1. Distribution of Risk Factors at Baseline by ApoE4 Genotype No apoE4 (n ⫽ 530) Characteristic (Unit) Age (years) Male sex Current smoking Body mass index (kg/m2) Systolic blood pressure (mm Hg) Diastolic blood pressure (mm Hg) Total plasma cholesterol (mg/dL)
ApoE4 (n ⫽ 200)
Number (%) or mean ⫾ SD 313 (59) 32 (6) 25 ⫾ 3 125 ⫾ 18 79 ⫾ 10 200 ⫾ 42
en). There were no differences in traditional cardiovascular risk factors between subjects with and without apoE4 in either sex, except that LDL cholesterol levels were higher in men with apoE4 than in the other men (127 ⫾ 35 mg/dL versus 116 ⫾ 31 mg/dL, P ⬍0.05). Seventeen (18%) of the subjects (14 men and 3 women) with apoE4 had a coronary event, compared with 26 (11%) of the remaining subjects. When analyzed by sex, 14 (28%) of the men with apoE4 versus 17 (16%) of the men without the ⑀4 allele had a coronary event (P ⫽ 0.07). The risk of coronary events was the same (7%) in women with and without apoE4. When LDL and HDL cholesterol levels were substituted for total cholesterol levels in proportional hazards model, neither LDL cholesterol (RR ⫽ 1.01 per 20 mg/dL, 95% CI: 0.98 to 1.05) nor HDL cholesterol (RR ⫽ 1.03 per 20 mg/dL, 95% CI: 0.99 to 1.18) levels were associated with coronary events, whereas apoE4 remained a significant predictor (RR ⫽ 3.2, 95% CI: 1.2 to 4.2, P ⬍0.01).
Effects of ApoE4 in Men and Women In a multivariate model, an interaction term (sex and apoE4) was significant (P ⬍0.05), indicating that the effects of apoE4 on coronary events were different in men and in women. Thus, separate analyses were performed to determine those sex-related differences. Of the 108 men with apoE4, 31 (29%) had coronary events, compared with 46 (15%) of the 313 men without apoE4 (P ⬍0.001). In an age-stratified multivariate model, apoE4
108 (54) 10 (5) 25 ⫾ 4 125 ⫾ 18 78 ⫾ 10 204 ⫾ 41
P Value 0.10 0.64 0.50 0.73 0.79 0.67 0.27
was associated with almost a tripling of the rate of coronary events in men (Table 3). For the 158 men for whom LDL and HDL cholesterol levels were available, apoE4 remained an independent predictor of coronary events (RR ⫽ 2.7, a 95% CI: 1.3 to 5.7, P ⬍0.01). By contrast, the incidence of coronary events did not differ between women with (8 [9%] of 94) and those without the apoE4 allele (19 [9%] of 215, P ⫽ 0.96). In an age-stratified multivariate model, neither apoE4 nor any other traditional cardiovascular risk factors were associated with coronary events in women (Table 3). Similar results were found when the analysis was performed in the 163 women for whom HDL and LDL cholesterol levels were available. Men with the apoE4 allele had significantly lower coronary-event–free survival than the other three groups (P ⬍ 0.01). Pairwise comparisons of the survival curves confirmed this difference for men with apoE4 compared with the other three groups (using a Bonferroni multiple comparisons correction, P ⬍0.05). The remaining three curves did not differ from one another.
DISCUSSION We found that the presence of the apoE4 allele was a strong predictor of coronary events among men but not among women. The association was independent of
Table 2. Predictors of Coronary Events: Results of Multivariate Proportional Hazards Model Risk Factor ApoE4 Female sex Current smoking Body mass index (per 2 kg/m2) Total cholesterol level (per 20 mg/dL) Systolic blood pressure (per 10 mm Hg)
Hazard Ratio (91% Confidence Interval) 2.01 0.6 1.2 1.03 1.07 1.23
The model was stratified by age (⬍50; 50 – 65; ⬎65 years; see methods). 30
January 2001
THE AMERICAN JOURNAL OF MEDICINE威 Volume 110
1.4–3.1 0.3–1.0 0.4–3.3 0.90–1.17 0.98–1.17 1.05–1.44
P Value 0.001 0.05 0.84 0.67 0.18 0.05
ApoE4 Allele and Coronary Events/Scuteri et al
Table 3. Sex-Specific Predictors of Coronary Events* Men Risk Factor ApoE4 Current smoking Body mass index (per 2 kg/m2) Total cholesterol level (per 20 mg/dL) Systolic blood pressure (per 10 mmHg)
Women
Hazard Ratio (95% Confidence Interval) 2.8 0.7 1.04 1.09 1.17
1.76–4.48 0.2–2.7 0.90–1.20 0.98–1.24 1.03–1.38
0.0001 0.54 0.61 0.16 0.01
0.81 2.5 1.03 1.12 0.97
P Value 0.35–1.87 0.5–12 0.80–1.32 0.89–1.37 0.78–1.22
0.62 0.25 0.84 0.38 0.80
* As estimated from an age-stratified multivariate proportional hazards model.
plasma cholesterol levels and other conventional coronary risk factors, which did not differ by apoE4 status. Dallongeville et al (19) reported that the frequency of apoE4 is increased in populations at high risk of coronary heart disease and is lower in geographic regions where the disease is less common. Several case-control studies and a few longitudinal studies have linked the presence of apoE4 with a greater risk of coronary heart disease (20 – 25). In one study, for example, the relative frequency of apoE4 predicted 40% of the variation in mortality for coronary events after adjustment for cholesterol levels (20). In a longitudinal study among elderly Finnish men, apoE4 was twice as common in those who died a coronary death as in event-free controls (23). In a nested case-control study in 619 high-risk men from the Multiple Risk Factor Intervention Trial (24), the presence of apoE4 was associated with an increased risk of coronary events, especially those that were fatal. In a meta-analysis, the risk of coronary artery disease for men with the apoE4 allele was estimated to be about 40% greater than for men without apoE4 (13). ApoE4 may also be associated with an increased risk of cerebrovascular ischemic events (26). In this study, we observed that the apoE4 allele increased the risk of coronary events in men but not in
women. A recent meta-analysis of nine studies including 181 women and 1,971 men with coronary heart disease concluded that although the apoE4 allele increased coronary risk overall, the data were not convincing for women (13). Investigators from the Copenhagen City Heart Study reported that although both men and women with the 4/3 or 4/4 genotype had greater levels of cholesterol and apoB than those with the 3/3 genotype, only men with apoE4 had an increased risk of ischemic heart disease (27). The mechanism for a sex-specific effect of apoE4 on the risk for coronary disease is unknown (28). The present study has several limitations. First, the sample included predominantly upper middle class, healthy, community-dwelling volunteers. In addition, given the shorter follow-up time and the lesser number of coronary events in women, it is difficult to ascertain whether the lack of an association between the apoE4 genotype and coronary events in women reflects a biological difference or is a result of low statistical power (the upper limit of the 95% confidence interval for the hazard ratio associated with apoE4 in women was 1.9). Another limitation is that a majority of coronary events consisted of new angina pectoris, although similar results were found when the analysis was restricted to “hard” events. Finally, LDL and HDL cholesterol level measurements were not available for all of the participants, although our results were similar in the subgroup for whom the values were available. We conclude that apoE4 is an independent risk factor for coronary events in men but not in women, and that the association is not mediated by differences in total cholesterol levels. Studies are needed to determine how apoE4 causes coronary events, and why there appear to be sex-related differences. This information might allow the development of preventive strategies that are tailored to the apoE genotype.
REFERENCES Figure 1. Cumulative coronary event-free survival by sex and apoE4 allele status. Event-free survival was significantly lower in men with apoE4 than in the other three subgroups (P ⬍0.01), but did not differ among the other three groups.
1. Mahley RW. Apolipoprotein E: cholesterol transport protein with expanding role in cell biology. Science. 1988;240:622– 640. 2. Beisiegel U, Weber W, Ihrke G, et al. The LDL-receptor-related protein, LRP, is an apolipoprotein E-binding protein. Nature. 1989; 341:162–164. January 2001
THE AMERICAN JOURNAL OF MEDICINE威
Volume 110 31
ApoE4 Allele and Coronary Events/Scuteri et al 3. Eggersten G, Tegelman R, Ericsson S, et al. Apolipoprotein E polymorphism in a healthy Swedish population: variation in allele frequency with age and relation to serum lipid concentrations. Clin Chem. 1993;39:2125–2129. 4. Gomez-Coronado D, Alvarez JJ, Entrala A, et al. Apolipoprotein E polymorphism in men and women from a Spanish population: allele frequencies and influence on plasma lipids and apolipoproteins. Atherosclerosis. 1999;147:167–176. 5. Ferrieres J, Sing CF, Roy M, et al. Apolipoprotein E polymorphism and heterozygous familial hypercholesterolemia: sex-specific effects. Arterioscler Thromb. 1994;14:1554 –1560. 6. Davignon J, Gregg RE, Sing CF. Apolipoprotein E polymorphism and atherosclerosis. Arteriosclerosis. 1988;8:1–21. 7. Juvonen T, Kervinen K, Kairaluoma MI, et al. Gallstone cholesterol content is related to apolipoprotein E polymorphism. Gastroenterology. 1993;104:1806 –1813. 8. Van Erpecum KJ, Carey MC. Apolipoprotein E4: another risk factor for cholesterol gallstone formation? Gastroenterology. 1996;111: 1764 –1767. 9. Kervinen K, Sodervik H, Makela J, et al. Is the development of adenoma and carcinoma in proximal colon related to apolipoprotein E phenotype? Gastroenterology. 1996;110:1785–1790. 10. Roses AD. Apolipoprotein E alleles as risk factors for Alzheimer’s disease. Annu Rev Med. 1996;47:387– 400. 11. Strittmatter WJ, Roses AD. Apolipoprotein E and Alzheimer’s disease. Proc Natl Acad Sci USA. 1995;92:4725– 4727. 12. Poirier J, Minnich A, Davignon J. Apolipoprotein E, synaptic plasticity and Alzheimer’s disease. Ann Med. 1995;27:663– 670. 13. Wilson PWF, Schafer EJ, Larson MG, Ordovas JM. Apolipoprotein E alleles and risk of coronary disease. A meta-analysis. Arterioscler Thromb Vasc Biol. 1996;16:1250 –1255. 14. Shock NW, Greulich RC, Andres RA, et al. Normal Human Aging. The Baltimore Longitudinal Study of Aging. Washington, DC: US Government Printing Office, NIH publication no. 84-2450; 1984:45. 15. Rose GA, Blackburn H. Cardiovascular Survey Methods. Geneva: World Health Organization; 1968. 16. Hixon JE, Vernier DT. Restriction isotyping of human apolipoprotein E by gene amplification and cleavage with HhaI. J Lipid Res. 1990;31:545–548.
32
January 2001
THE AMERICAN JOURNAL OF MEDICINE威 Volume 110
17. Cox DR. Analysis of Binary Data. London: Methuen & Co; 1970. 18. Kalbfleisch JD, Prentice RL. The Statistical Analysis of Failure Time Data. New York: Wiley & Sons; 1980. 19. Dallongeville J. Apolipoproteine E: proprie´te´s physiologiques, polymorphisme et atherosclerose. Sang Thrombose Vaisseaux. 1993; 5:707–717. 20. Stengard JH, Weiss KM, Sing CF. An ecological study of association between coronary heart disease mortality rates in men and the relative frequencies of common allelic variations in the gene coding for apolipoprotein E. Hum Genet. 1998;103:234 –241. 21. Ou T, Yamakawa-Kobayashi K, Arinami T, et al. Methylenetetrahydrofolate reductase and apolipoprotein E polymorphism are independent risk factors for coronary heart disease in Japanese: a case-control strudy. Atherosclerosis. 1998;137:23–28. 22. Wilson PWF, Myers RH, Larson MG, et al. Apolipoprotein E alleles, dyslipidemia, and coronary heart disease: the Framingham Offspring Study. JAMA. 1994;272:1666 –1671. 23. Stengard JH, Zerba KE, Pekkanen J, et al. Apolipoprotein E polymorphism predicts death from coronary artery disease in a longitudinal study of elderly Finnish men. Circulation. 1995;91:265–269. 24. Eichner JE, Kuller LH, Orchard TJ, et al. Relation of apolipoprotein E phenotype to myocardial infarction and mortality from coronary artery disease. Am J Cardiol. 1993;7:160 –165. 25. Katzel L, Fleg JL, Busby-Whitehead MJ, et al. Exercise-induced silent myocardial ischemia in master athletes. Am J Cardiol. 1998;81: 261–265. 26. Margaglione M, Seripa D, Gravina C, et al. Prevalence of apolipoprotein E alleles in healthy subjects and survivors of ischemic stroke: an Italian case-control study. Stroke. 1998;29:399 – 403. 27. Frikke-Schmidt R, Tybjaerg-Hansen A, Steffensen R, et al. Apolipoprotein E genotype: epsilon 32 women are protected while epsilon 43 and epsilon 44 men are susceptible to ischemic heart disease. The Copenhagen City Heart Study. J Am Coll Cardiol. 2000;35: 1192–1199. 28. Davignon J, Cohn JS, Mabile L, Bernier L. Apolipoprotein E and atherosclerosis: insight from animal and human studies. Clin Chim Acta. 1999;286:115–143.