| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
(Circulation. 1996;94:1263-1268.)
© 1996 American Heart Association, Inc.
Articles |
55 Years of Age
the Department of Cardiology, Juntendo University, Tokyo, Japan.
| Abstract |
|---|
|
|
|---|
Methods and Results Serum Lp(a) concentrations were determined in 354 women and 706 men with or without angiographically defined CAD (one or more coronary arteries with narrowing of
75%). The age-specific impact of elevated Lp(a) (
30 mg/dL) on CAD was examined in each sex. In the younger age group (<55 years old), elevated Lp(a) was independently associated with CAD in both sexes (adjusted odds ratio [OR]: women, 6.90, P<.01; men, 2.63, P<.05). The age-specific ORs declined with age, and elevated Lp(a) no longer conferred an increased CAD risk in either elderly men or women
65 years old. In the age group of 55 to 64 years, elevated Lp(a) was positively associated with CAD for men (adjusted OR: 2.45, P<.05) but not for women (adjusted OR: 0.56, P=NS).
Conclusions For both sexes, elevated Lp(a) appears to be an independent risk factor for premature CAD and the importance of Lp(a) appears to decrease with age. However, for women, the risk estimate of Lp(a) began to decline at an age
10 years younger than for men. These data suggest that not only age- but also sex-specific factors such as menstrual status may interact with the association between Lp(a) and CAD.
Key Words: lipoprotein(a) women aging risk factors coronary disease
| Introduction |
|---|
|
|
|---|
Recently, the sex and age differences in the nature of the association between CAD and various risk factors have been suggested. For example, data from a recent prospective study suggested that women did not have an increased risk of CAD if they had elevated LDL cholesterol,20 and low HDL cholesterol and elevated triglyceride levels have been reported to be stronger predictors of CAD in women than in men.20 21 The clinical importance of dyslipidemia in older persons also remains controversial. In particular, conflicting views exist regarding the atherogenic role of Lp(a) in the elderly.9 16 22 23 The importance of elevated Lp(a) as a risk factor for CAD may be different between sexes or ages. However, there have been few published studies that addressed this issue. The purpose of the present study was to investigate the association between Lp(a) and CAD, taking into account the differences in sex and age.
| Methods |
|---|
|
|
|---|
Coronary Angiography
Selective coronary angiography was performed in patients after a 12-hour fast. The coronary arteries were injected with 4 to 8 mL Iopamidol and recorded on 35-mm cineangiographic film in multiple views. Films were evaluated for obstructive coronary lesions by experienced angiographers who were blinded to the patients' risk profiles. Stenoses of coronary arteries were categorized according to the reporting system proposed by the American Heart Association,27 and stenosis of
75% was considered to be hemodynamically significant. CAD cases were defined as those who had one or more coronary arteries with significant stenosis, and control subjects in this study were those who did not. The cases were also divided into three subgroups on the basis of the number of diseased coronary arteries: the 1VD group had significant stenosis in one coronary vessel, the 2VD group had significant stenosis in two coronary vessels, and the 3VD group had significant stenosis in three coronary vessels.
Lipid and Lipoprotein Analysis
Blood samples were collected at the time of arteriotomy before systemic heparinization for the cardiac catheterization and were allowed to clot at room temperature for 1 to 2 hours. All sera were obtained under conditions in which proteolysis was minimized with EDTA, NaN3, and protease inhibitor and were maintained at 4°C until the assay. Serum cholesterol, triglyceride, and HDL cholesterol levels were measured by hospital personnel in the clinical chemistry department using standard enzymatic methods. Serum Lp(a) concentrations were measured as the total lipoprotein mass according to the ELISA method28 or the latex immunoassay method (Sanwa Co).29 These two assays correlated well with each other, with the correlation coefficient r=.996 and the linear regression of y=1.04x+0.05, where x indicates the latex immunoassay method and y indicates the ELISA method).29 The frequency distribution of serum Lp(a) concentrations in healthy Japanese subjects was highly skewed, with a median of 10 to 15 mg/dL,29 30 which is not significantly different from the results for white populations.30 Because the 90th percentile of Lp(a) concentrations in healthy Japanese populations was
30 mg/dL30 and many previous reports have suggested that Lp(a) confers its pathological effect at
30 mg/dL,7 8 we defined subjects with Lp(a) of
30 mg/dL as those with elevated Lp(a). The crude values of LDL cholesterol were calculated with the use of Friedewald's formula,31 and correction was made for the contribution of Lp(a) cholesterol. Because Lp(a) is reported to contain 30% cholesterol by weight,2 Lp(a) value was multiplied by 0.3, and the resulting value was subtracted from the crude LDL cholesterol estimate.
Determination of Other Variables
Patient data were obtained from medical records regarding current smoking habits, presence of diabetes mellitus or hypertension, and use of the lipid-lowering drugs. Information about menstrual status of each woman was not completed; however, the majority of the women in this study were presumed to be postmenopausal. We did not have complete information about postmenopausal estrogen use in the women. Although postmenopausal estrogen use is prevalent in western countries, especially in the United States,32 33 only few Japanese women are presumed to be receiving hormone replacement therapy. To confirm these speculations, we also carried out another review of the medical records and telephone interviews with the female subjects aged 55 to 64 years (n=127) regarding their menstrual status and postmenopausal estrogen use. As expected, none of the women were premenopausal at the time of the catheterization or had been treated with the estrogen.
Statistical Analysis
We used Pearson's
2 statistic to compare the risk factor prevalence and Student's t test to compare continuous variables, except the concentrations of Lp(a). Because the distribution of Lp(a) is highly skewed, data were summarized as median and 25th and 75th percentiles, and differences of Lp(a) concentrations were evaluated through nonparametric statistical procedures (Mann-Whitney and Kruskal-Wallis tests). All probability values for statistical significance were two-tailed. As a measure of association between the elevated Lp(a) concentrations (
30 mg/dL) and the CAD, age-specific and multivariate-adjusted odds ratios (ORs) were estimated for each sex, and results are given as the OR and 95% CIs. The adjusted ORs were estimated with the stepwise logistic regression model, in which smoking habit, presence of hypertension or diabetes mellitus, LDL cholesterol level (
3.34 mmol/L or taking the lipid-lowering drugs versus <3.34 mmol/L), HDL cholesterol level (
1.03 mmol/L versus >1.03 mmol/L), and triglyceride level (
1.69 mmol/L versus <1.69 mmol/L) were included as confounding variables. All analyses were performed with SPSS version 4.0 for Macintosh.
| Results |
|---|
|
|
|---|
|
Age- and Sex-Specific ORs of Elevated Lp(a)
Table 2
lists age-specific crude ORs of the elevated Lp(a) (
30 mg/dL) for the CAD within each sex. For both men and women <55 years old, although the ORs in some of the age groups did not reach statistical significance, the estimates were >1.0, indicating a positive association between the elevated Lp(a) and CAD. The crude risk estimates appeared to decline with age, and the significant association was no longer seen in subjects
65 years old in either sex. In age groups of 55 to 64 years, the risk estimates were >1.0 for men; however, they were <1.0 for women.
|
Elevated Lp(a) as a Risk Factor for Women
The association between Lp(a) and CAD abruptly disappeared at
55 years of age in women. Thus, the women were divided into two age groups (<55 years and
55 years), and the associations between CAD and various risk factors were determined through univariate comparisons (Table 3
). In both age groups, the cases were more likely to be diabetic and to be treated with lipid-lowering drugs than were the controls. The prevalences of smoking and hypertension were higher in the cases than in the controls in the younger but not in the older women. The cases had higher mean LDL cholesterol concentrations than the controls, but only in the younger women. Higher median Lp(a) concentration was associated with CAD in the younger women (P<.05) but not in the older women. In addition, Lp(a) concentrations increased with the number of vessels affected only in the younger (P<.01 for trend, Figure).
|
|
Age- and Sex-Specific Contributions of Elevated Lp(a) and Other Risk Factors
Table 4
shows age- and sex-specific prevalences and ORs of elevated Lp(a). The prevalence of elevated Lp(a) tended to be higher in women than in men (37% versus 32%, P>.1); however, no significant difference was seen between the age groups within each sex. In the younger age group (<55 years), the elevated Lp(a) concentration conferred the crude ORs of 5.94 (P<.001) and 2.59 (P<.05) for women and men, respectively. These risk estimates remained significant after the multivariate adjustment (adjusted ORs of 6.90; P<.01 and 2.63 for women; P<.05 for men). In the younger group, other significant variables were LDL cholesterol (adjusted OR of 18.20; P<.001), smoking (adjusted OR of 9.93; P<.01), and diabetes mellitus (adjusted OR of 7.17; P<.05) for women and LDL cholesterol (adjusted OR of 2.78; P<.01), HDL cholesterol (adjusted OR of 3.59; P<.01), and hypertension (adjusted OR of 2.22; P<.05) for men. For both men and women
65 years old, the crude estimates of the elevated Lp(a) were
1.0, indicating a lack of the association between Lp(a) and CAD in the elderly. For the older patients, the significant variables were HDL cholesterol (adjusted OR of 9.40; P<.05) for women and LDL cholesterol (adjusted OR of 2.34; P<.05) for men. In the age group of 55 to 64 years, the crude ORs of Lp(a) were 2.46 (P<.01) and 0.64 (P=NS) for men and women, respectively. In multivariate stepwise analysis, Lp(a) (adjusted OR of 2.45; P<.01), LDL cholesterol (adjusted OR of 1.87; P<.05), HDL cholesterol (adjusted OR of 2.34; P<.05), and smoking (adjusted OR of 2.04; P<.05) were selected as statistically significant variables for the middle-aged men, and diabetes mellitus (adjusted OR of 3.53; P<.05) was selected as a statistically significant variable for middle-aged women.
|
| Discussion |
|---|
|
|
|---|
65 years old). However, in patients aged 55 to 64 years, elevated Lp(a) was positively associated with CAD in men only. The present data confirm previous studies showing the association between elevated Lp(a) and premature CAD in men.9 10 11 12 13 14 15 16 17 18 19 This association has also been examined in women9 10 11 12 13 14 15 34 35 36 and the elderly9 16 22 23 ; however, conflicting views still exist regarding the association in these populations. We found the age- and sex-specific interactions with this association, and conflicting views on this issue may be in part explained by the results of the present study.
Effect of Age
Because advancing age is one of the striking and nonmodifiable risk factors for CAD, the impact of other various risk factors may decrease with age.37 Some previous reports have suggested a lack of an association between elevated Lp(a) and CAD in the elderly.9 16 22 We also found a lack of an association in both elderly men and women
65 years old. The impact of elevated Lp(a) on CAD appeared to decrease with age for both sexes; however, for women, the risk estimates began to decline at
55 years of age, which was 10 years sooner than for men. In the age group of 55 to 64 years, elevated Lp(a) was associated with CAD for men but not for women. This implies that other sex-specific factors may interact with the association in this age group.
Estrogen Deficiency as a Risk Factor for CAD
It is well known that risk of CAD increases after the menopause in women, suggesting a cardioprotective effect of estrogen. Recently, the protective effect of estrogen was confirmed through observational studies of estrogen replacement therapy, suggesting a 50% reduction in CAD risk in postmenopausal women using estrogen.32 Favorable changes in serum lipid levels in the postmenopausal women taking estrogen (eg, increase in HDL cholesterol, decrease in LDL cholesterol,38 39 and decrease in Lp(a)38 40 ) have been proposed as the most promising explanation of the beneficial effect of estrogen.
On the other hand, the effects of lipid levels on CAD in women remain controversial. For example, recent prospective data from the Lipid Research Clinic Follow-up Study, in which 1405 women 50 to 69 years old had been prospectively followed for 14 years, indicated that LDL cholesterol concentration was of little value in predicting CAD in women.20 Thus, if estrogen lowers LDL cholesterol, it is uncertain whether the reduced LDL cholesterol level lowers CAD risk in the postmenopausal women. This appears to be true for the other lipid risk factors.
Recently, other mechanisms of the action, including antioxidant effect, beneficial alterations of hemostatic factors, attenuation of arterial vascular tone, and inhibitory effects on proliferation and biosynthesis of the vascular smooth muscle cells, have been suggested.32 41 Moreover, a sex-related difference in the nature of the coronary arteries has also been suggested. Collins and colleagues42 studied postmenopausal women and middle-aged men to investigate the direct effect of estrogen on the coronary circulation and demonstrated that estrogen attenuated acetylcholine-induced coronary vasoconstriction in the women but not in the men. This evidence appears to suggest the sex-specific cardioprotective effect of estrogen. Moreover, if the effect of estrogen deficiency itself was more important than expected in the women, the impact of the lipid risk factors such as that of LDL cholesterol and Lp(a) may decrease after menopause. In the present study, we found an abrupt decline in ORs of elevated Lp(a) at
55 years old in women but not in men. At 55 years of age, almost all women have experienced menopause,43 44 45 and this age coincides with the age at which sex hormones reach the postmenopausal levels in most of the women.45 Therefore, we believe that the sex-specific interaction with the association between Lp(a) and CAD shown in the present study is related to the menopause.
Menstrual Status and Lp(a)
Recent epidemiological studies have suggested that concentrations of Lp(a) are higher in postmenopausal women than in premenopausal women,46 47 and that postmenopausal estrogen use is associated with a reduced concentration of Lp(a).38 39 40 However, in the present study, we failed to show an association between CAD and elevated Lp(a) in women
55 years old. Because almost all of the women
55 years old are postmenopausal,43 44 45 this evidence suggests that even though Lp(a) increases after the menopause, it may lose its significance as a risk factor for CAD.
Study Limitations
In the present study, actual menopausal information for each woman was not taken into account because we did not have complete information. We have no direct evidence of whether menstrual status interacts with the association between Lp(a) and CAD. However, the effects of age and menopause on CAD are difficult to separate, and we found a sex-specific interaction with the association that may be related to menopause.
Conclusions
For both women and men, elevated Lp(a) appears to be an independent risk factor for premature CAD and the importance of Lp(a) appears to decrease with age. However, for women, the risk estimate of Lp(a) began to decline
10 years sooner than that for men. These data suggest that not only age but also menstrual status may interact with the association between elevated Lp(a) and CAD. Furthermore, we speculate that risk factors other than Lp(a) may play a more important role in postmenopausal women. Careful prospective studies will be necessary to clarify this issue.
| Acknowledgments |
|---|
| Footnotes |
|---|
Received January 16, 1996; revision received March 13, 1996; accepted March 26, 1996.
| References |
|---|
|
|
|---|
This article has been cited by other articles:
![]() |
J. Suk Danik, N. Rifai, J. E. Buring, and P. M Ridker Lipoprotein(a), measured with an assay independent of apolipoprotein(a) isoform size, and risk of future cardiovascular events among initially healthy women. JAMA, September 20, 2006; 296(11): 1363 - 1370. [Abstract] [Full Text] [PDF] |
||||
![]() |
S. Tsimikas, E. S. Brilakis, E. R. Miller, J. P. McConnell, R. J. Lennon, K. S. Kornman, J. L. Witztum, and P. B. Berger Oxidized Phospholipids, Lp(a) Lipoprotein, and Coronary Artery Disease N. Engl. J. Med., July 7, 2005; 353(1): 46 - 57. [Abstract] [Full Text] [PDF] |
||||
![]() |
E. Zambrelli, E. Emanuele, S. Marcheselli, L. Montagna, D. Geroldi, and G. Micieli Apo(a) size in ischemic stroke: Relation with subtype and severity on hospital admission Neurology, April 26, 2005; 64(8): 1366 - 1370. [Abstract] [Full Text] [PDF] |
||||
![]() |
C. Hernandez, G. Francisco, P. Chacon, and R. Simo Lipoprotein(a) as a Risk Factor for Cardiovascular Mortality in Type 2 Diabetic Patients: A 10-year follow-up study Diabetes Care, April 1, 2005; 28(4): 931 - 933. [Full Text] [PDF] |
||||
![]() |
A. A. Ariyo, C. Thach, R. Tracy, and the Cardiovascular Health Study Investigators Lp(a) Lipoprotein, Vascular Disease, and Mortality in the Elderly N. Engl. J. Med., November 27, 2003; 349(22): 2108 - 2115. [Abstract] [Full Text] [PDF] |
||||
![]() |
C. Sarti, L. Pantoni, G. Pracucci, A. Di Carlo, P. Vanni, and D. Inzitari Lipoprotein(a) and Cognitive Performances in an Elderly White Population : Cross-Sectional and Follow-Up Data Stroke, July 1, 2001; 32(7): 1678 - 1683. [Abstract] [Full Text] [PDF] |
||||
![]() |
F. WAHN, V. DANIEL, F. KRONENBERG, G. OPELZ, D. V. MICHALK, and U. QUERFELD Impact of Apolipoprotein(a) Phenotypes on Long-Term Renal Transplant Survival J. Am. Soc. Nephrol., May 1, 2001; 12(5): 1052 - 1058. [Abstract] [Full Text] |
||||
![]() |
G. Ricci, G. Tamaro, R. Simeone, E. Giolo, G. Nucera, F. De Seta, and S. Guaschino Lipoprotein(a) changes during natural menstrual cycle and ovarian stimulation with recombinant and highly purified urinary FSH Hum. Reprod., March 1, 2001; 16(3): 449 - 456. [Abstract] [Full Text] [PDF] |
||||
![]() |
J. M. Foody, J. A. Milberg, K. Robinson, G. L. Pearce, D. W. Jacobsen, and D. L. Sprecher Homocysteine and Lipoprotein(a) Interact to Increase CAD Risk in Young Men and Women Arterioscler. Thromb. Vasc. Biol., February 1, 2000; 20(2): 493 - 499. [Abstract] [Full Text] [PDF] |
||||
![]() |
K. J. Harjai Potential New Cardiovascular Risk Factors: Left Ventricular Hypertrophy, Homocysteine, Lipoprotein(a), Triglycerides, Oxidative Stress, and Fibrinogen Ann Intern Med, September 7, 1999; 131(5): 376 - 386. [Abstract] [Full Text] [PDF] |
||||
![]() |
C. Gazzaruso, A. Garzaniti, P. Buscaglia, G. Bonetti, C. Falcone, P. Fratino, G. Finardi, and D. Geroldi Association between apolipoprotein(a) phenotypes and coronary heart disease at a young age J. Am. Coll. Cardiol., January 1, 1999; 33(1): 157 - 163. [Abstract] [Full Text] [PDF] |
||||
![]() |
E. Sbarouni, Z. S. Kyriakides, and D. T. Kremastinos The effect of hormone replacement therapy alone and in combination with simvastatin on plasma lipids of hypercholesterolemic postmenopausal women with coronary artery disease J. Am. Coll. Cardiol., November 1, 1998; 32(5): 1244 - 1250. [Abstract] [Full Text] [PDF] |
||||
![]() |
W. Su, H. Campos, H. Judge, B. W. Walsh, and F. M. Sacks Metabolism of Apo(a) and ApoB100 of Lipoprotein(a) in Women: Effect of Postmenopausal Estrogen Replacement J. Clin. Endocrinol. Metab., September 1, 1998; 83(9): 3267 - 3276. [Abstract] [Full Text] |
||||
![]() |
P. N. Hopkins, L. L. Wu, S. C. Hunt, B. C. James, G. M. Vincent, and R. R. Williams Lipoprotein(a) Interactions With Lipid and Nonlipid Risk Factors in Early Familial Coronary Artery Disease Arterioscler. Thromb. Vasc. Biol., November 1, 1997; 17(11): 2783 - 2792. [Abstract] [Full Text] |
||||
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
|
Circulation Home | Subscriptions | Archives | Feedback | Authors | Help | AHA Journals Home | Search Copyright © 1996 American Heart Association, Inc. All rights reserved. Unauthorized use prohibited. |