Parity and Carotid Atherosclerosis in Men and Women
Insights Into the Roles of Childbearing and Child-Rearing
Background and Purpose— Parity appears to be associated with carotid atherosclerosis in women aged 45 years and older. Studying this association among younger women and men may provide insight into whether this association relates predominantly to childbearing or child-rearing.
Methods— The association between parity and carotid atherosclerosis (intima-media thickness and presence of plaques) was assessed in a cohort consisting of 750 women and 1164 men, all with at least one traditional cardiovascular risk factor, aged 18 to 80 years of age. Traditional cardiovascular risk factors were also assessed, and the Framingham Risk Score calculated.
Results— In age-adjusted analyses, the number of children was associated with adiposity, fasting glucose, 2-hour glucose, Framingham risk score, and carotid atherosclerosis in women, but not in men. Multivariate linear regression models indicate that the prevalence of plaques was increased by 15% (95% CI, 2 to 29) per child among women, and 0% (95% CI, −10 to 11) among men, after adjustment for age, socioeconomic and lifestyle factors (including waist circumference). The association between parity and carotid intima-media thickness was similar in younger and older women (PHeterogeneity=0.20).
Conclusions— A higher number of children is associated with increased carotid atherosclerosis in both younger and older women, but not among men. These findings indicate that childbearing, but not child-rearing, may be a risk factor for atherosclerosis, and suggest the potential importance of considering the number of children when assessing the level of cardiovascular risk in women.
There is an emerging body of literature examining the association between parity and cardiovascular disease. The number of pregnancies has been associated with increased risk of cardiovascular disease among women in some studies,1,2 with the absence of an association in another.3 The use of surrogate markers of cardiovascular disease, such as vascular structure or function, may help clarify the relationship between parity and cardiovascular disease due to the long time interval usually present between pregnancy and the onset of clinical cardiovascular events.4 Such noninvasive measures of total atherosclerotic burden include the presence of carotid plaques and the assessment of carotid intima-media thickness (IMT). Previous studies have generally found an association between parity and carotid atherosclerosis among women. In a cohort of 1200 women, Wolff et al found a U-shaped association, where both nulliparity and 4 or more children were associated with increased carotid IMT in comparison with those women with 1 child.5 Humphries et al4 described a weak trend for increasing IMT with increasing parity among over 4000 elderly women and a 10% increase in the risk of having a carotid plaque per live birth. The results of a recent study of 66 women found data suggestive of an association between parity and increasing IMT among women of childbearing age.6 In contrast to these reports, a recent study by Kharazmi et al7 described no association between parity and either IMT or presence of plaques after adjustment for age in a population-based cohort of 746 Finnish women.
With the exception of one smaller study,6 previous studies have been limited to women aged 45 years and older, or 55 years and older, and thus it remains uncertain as to whether these changes in carotid wall thickness are apparent in younger females. In contrast, there are no reports in the literature of the association between parity and carotid atherosclerosis among men. Studying younger women and men, in addition to older women, will provide important information concerning the etiology of any such associations. Indeed a major advantage in studying both males and females is that the males represent a control group who have been exposed only to child-rearing, and not to childbearing.
Therefore, we studied the association between parity and carotid atherosclerosis (both IMT and the presence of plaques) in a cohort consisting of 750 women and 1164 men, all with at least one traditional cardiovascular risk factor, aged 18 to 80 years of age. We sought to describe the nature of this association, with a particular focus on the roles of childbearing and child-rearing.
Materials and Methods
We studied 750 women and 1164 males, aged 18 to 80 years, referred to an outpatient Centre for Prevention and Detection of Atherosclerosis in Lyon, France, between 1995 and 2000. All patients were referred by a primary care physician on the basis of a diagnosis with at least one cardiovascular risk factor (hypercholesterolemia, hypertension, diabetes, obesity, tobacco smoking, or a family history of cardiovascular disease). All patients provided written informed consent, and the project was undertaken in accordance with French legislation (Huriet law) and institutional guidelines.
Carotid IMT and Presence of Plaques
Carotid IMT was investigated as previously described by ourselves and others.8,9 Briefly, high-resolution ultrasound images were obtained of the common carotid arteries in longitudinal section. IMT was assessed in the segment of artery within 1 cm of the bifurcation. The average of both left and right carotid arteries was used for statistical analyses.
The presence or absence of arterial plaques was assessed in the distal portion of the common carotid artery, the carotid bulb, and the proximal portion of the internal carotid artery. Plaques were defined as a focal protrusion into the lumen, or an IMT ≥1.5 mm.
Assessment of Socioeconomic, Lifestyle and Cardiovascular Risk Factors
A structured questionnaire, administered by a study physician, was used to determine parity (assessed as the number of live births for women, and number of children for men), employment and marital status, physical activity, medical history (including hysterectomy and oophorectomy, ever use of oral contraception, and diabetes), menopausal status, current medication use, and smoking status.
Employment status was categorized as unemployed, blue-collar employees, white-collar professionals, and retired. Marital status was categorized as single/never married, married/domestic partnership (outside of marriage), widowed, and separated. Physical activity was categorized into 3 groups according to the number of 20-minute sessions of moderate or intense activity per week: (1) high-level physical activity: 3 or more sessions; (2) moderate-level physical activity: 1 or 2 sessions; (3) sedentary: <1 session. Smoking status was categorized as never smokers, current smokers, ex-smokers who quit within the previous 2 years, ex-smokers who quit 2 to 5 years prior, and ex-smokers who quit more than 5 years prior.
Dietary intake was assessed via a 24-hour dietary recall, assisted by a registered dietitian. An unhealthy diet score was calculated as previous described.10 This score is a composite dietary score designed to provide a snapshot of dietary health, and is derived from dietary intake of cholesterol, alcohol, percentage energy derived from fat and carbohydrates, and total caloric intake. These components and cut-points are based on European and American dietary recommendations.11,12 Blood pressure was measured at rest using an automatic sphygmomanometer. The average of 3 readings was used in these analyses. A venous blood sample was obtained after an overnight fast. Analyses included total cholesterol, HDL cholesterol, triglycerides, glucose and fibrinogen. LDL cholesterol was calculated using the Friedewald equation,13 except when triglycerides were >4.52 mmol/L, in which case LDL subfraction was measured directly. A standard 2-hour oral glucose tolerance test was undertaken in 1575 subjects (82%).
Framingham Risk Score
The Framingham risk score was calculated as described previously.14 This score uses data concerning an individual’s blood pressure, total cholesterol, smoking status, HDL cholesterol and diabetes status to estimate the individual’s 10 years probability of developing coronary heart disease.
Differences in lifestyle, socioeconomic and cardiovascular risk factors between different parity groups were determined by ANOVA and ANCOVA for continuous variables, logistic regression for age-adjusted dichotomous variables, and χ2 (Pearson) for multinomial variables. The association between parity and carotid atherosclerosis was examined using standard logistic and multiple linear regression techniques. Results of logistic regression models are presented as odds ratio (95% CI), and results of multiple linear regression models as unstandardized β-coefficient (SE). The residuals of the multiple linear regression models were normally distributed. Modification of the associations between parity and carotid atherosclerosis by age, sex and obesity were examined formally using a test for heterogeneity.15 All other statistical analyses were performed using SPSS software (version 15.0; SPSS, Chicago, Ill).
The average number of children was similar for women and men (women: 2.07 [SD 1.7], versus men: 2.08 [SD 1.4]). The modal number of children was 2 for both women and men, occurring in 29% and 38%, respectively. A comparison of the age stratified average number of children in this cohort and the population statistics for metropolitan France16 is shown in the supplemental Figure I, available online at http://stroke.ahajournals.org.
Association of Parity With Lifestyle Factors and Cardiovascular Risk Factors
Lifestyle factors and the prevalence and severity of cardiovascular risk factors, stratified by parity, are shown in Table 1 for men and Table 2 for women. Briefly, there was a positive association between number of children and adiposity among women, but not men. The number of children was also negatively related to HDL-cholesterol concentration and positively associated with fasting glucose among women.
Association Between Number of Children and Atherosclerosis in Men
Association Between Number of Children and Atherosclerosis in Women
Our data provide evidence for a linear association between the number of children and the prevalence of carotid plaques among women (risk increased by 15% per child; see Table 3). Women with 4 or more children had a >2-fold increased odds of having a carotid plaque. The association between parity and carotid atherosclerosis remained statistically significant after adjustment for cardiovascular risk factors, suggesting that the association is largely independent of these risk factors. The association between parity and carotid atherosclerosis was not altered by adjustment for menopausal status, hysterectomy/ovariectomy, ever use of oral contraceptive pill, 2-hour glucose, or fibrinogen (data not shown).
A formal test for heterogeneity comparing men and women with regards to the association of parity and carotid atherosclerosis after adjustment for age, marital status, employment status, smoking, physical activity, dietary score and waist circumference gave some evidence for the association being stronger among women than men (PHeterogeneity=0.013 for increase in IMT associated with ≥4 children; PHeterogeneity=0.091 for prevalence of carotid plaques associated with the number of children).
Association Between Parity and Atherosclerosis in Women: Modification by Age and Obesity
We then examined whether the association between parity and carotid atherosclerosis in females was modified by age or obesity. There was a strong association between parity and carotid IMT among women in the lowest quartile of age (≤46.17 years (n=180); β-coefficient [SE]=0.137 [0.048] for ≥4 children versus nulliparous), whereas the association appeared to be less marked among women in the highest quartile of age (>61.01 years (n=178); β-coefficient [SE]= 0.049 [0.050] for 4+ children versus nulliparous, adjusted for age, marital status, employment status, smoking, physical activity, dietary score and waist circumference; PHeterogeneity=0.20).
With regards to obesity, the association between parity and atherosclerosis was weaker among normal weight women (normal weight [n=265]; β-coefficient [SE]=0.035 [0.041] for 4+ children versus nulliparous) than among obese women (obese [n=273]; β-coefficient [SE]=0.123 [0.044] for 4+ children versus nulliparous, adjusted for age, waist circumference, marital status, employment status, and smoking; PHeterogeneity=0.14).
The results of the present study show that increasing parity is associated with carotid atherosclerosis in women but not in men among a population with at least 1 risk factor for cardiovascular disease. This association with carotid atherosclerosis was most marked in women with 4 or more children, whose carotid IMT was 0.068 mm greater and whose risk of having a carotid plaque was >2-fold greater than that of nulliparous women. A difference in common carotid IMT of this magnitude equates to an increase in relative risk for incident stroke or myocardial infarction of approximately 9%.17
Most of the studies on the association between parity and coronary heart disease have included only women. However, comparisons between men and women distinguish whether the mechanisms for the association between parity and atherosclerosis involve biological processes related to pregnancy or socioeconomic or lifestyle factors that are related to family size and child-rearing.18 The present study is the first to assess the relationship between the number of children and subclinical atherosclerosis among men. In contrast to women, we observed that among men there was no significant linear association between the number of children and cardiovascular risk factors. These divergent results between men and women support the view that the impact of parity on cardiovascular disease is due to biological effects of childbearing rather than being related to socioeconomic or lifestyle issues related to child-rearing. Considering that the association appears to be related to the biological effects of pregnancy, it would be of interest to evaluate the role of age at the first pregnancy, and also at the final pregnancy, in order to further define this association and identify at-risk individuals.
Previous large studies examining this association have been limited to women aged 45 years and older. We found some evidence that the association between parity and carotid IMT was stronger among the youngest quartile of women (aged 18 to 46 years) than among the oldest quartile of women (aged 61 to 79 years) suggesting that the increased risk is observed early after childbearing, is not restricted to the elderly, and cannot been explained by a survival bias. The association between increased parity and carotid atherosclerosis observed among the youngest group of women is of interest because the prevalence and severity of cardiovascular risk factors is likely to be similar to that present during childbearing years in these women, in contrast to those women aged over 60 years, and the influence of childbearing may be relatively more pronounced than that of child-rearing among younger women.
Several risk factors are associated with parity such as BMI, smoking, socioeconomic status, and impaired glucose control. These factors are generally associated with carotid atherosclerosis,19,20 and thus may at least partly mediate the association observed among women. In the present study, increased parity was associated with greater adiposity and a lower level of HDL cholesterol, in agreement with previous reports among older women.4 Furthermore, other studies have shown lower HDL cholesterol levels in parous women than nulliparous women, with this effect being observed many years after childbirth.21 These translated into only modest associations between parity and both Framingham risk score and carotid atherosclerosis. However, we controlled for these parameters in the multiple regression analyses, and the association between parity and carotid atherosclerosis appeared to be largely independent of traditional cardiovascular risk factors, including blood pressure, cholesterol, diabetes, obesity, smoking and other lifestyle risk factors. As such, increased parity appears to be an additional risk factor for early atherosclerosis among women, beyond the presence of conventional risk factors.
The biological mechanisms that mediate the association between parity and atherosclerosis remain unclear. Modified exposure to sex steroids during each pregnancy may favor the development of atherosclerosis into later life. However, the role of endogenous sex hormones in the development of atherosclerosis has been debated and remains contentious and poorly defined among women.22,23 Repeated pregnancies may also lead to chronic metabolic alterations, such as increased adiposity, insulin-resistance and dysglycemia, which predispose to the formation of atherosclerotic lesions.4 Adjustment for lipid levels and glycemia did not modify the association between parity and atherosclerotic lesions; however, the association between parity and carotid atherosclerosis appeared to be stronger among obese women than in healthy weight women, implicating weight control in this relationship. Because we did not directly assess insulin-resistance, we cannot exclude that insulin resistance may mediate the observed association in women.
This study cannot directly evaluate the long-term impact of parity on the development of carotid atherosclerotic lesions because of its cross-sectional design. Thus, we cannot draw conclusions on the causality of the link between parity and early atherosclerosis. Furthermore, the cohort studied consists entirely of subjects with at least 1 major cardiovascular risk factor, and as such presents a population at a high-risk of cardiovascular disease. Although the age-specific average number of children in this cohort is similar to that of the population of metropolitan France, whether the observed associations between parity and atherosclerosis are evident in the general population remains to be determined.
The present study shows in a large cohort of subjects at risk of cardiovascular disease that an increased number of children is associated with enhanced carotid atherosclerosis in women but not in men. These findings indicate that childbearing, but not child-rearing, may be a risk factor for atherosclerosis, and suggest the potential importance of considering the number of children when assessing the level of cardiovascular risk in women.
Sources of Funding
M.R.S. was supported by a fellowship from the National Health and Medical Research Council (NHMRC).
- Received August 28, 2008.
- Revision received November 12, 2008.
- Accepted December 15, 2008.
Green A, Beral V, Moser K. Mortality in women in relation to their childbearing history. BMJ. 1988; 297: 391–395.
Colditz GA, Willett WC, Stampfer MJ, Rosner B, Speizer FE, Hennekens CH. A prospective study of age at menarche, parity, age at first birth, and coronary heart disease in women. Am J Epidemiol. 1987; 126: 861–870.
Humphries KH, Westendorp IC, Bots ML, Spinelli JJ, Carere RG, Hofman A, Witteman JC. Parity and carotid artery atherosclerosis in elderly women: The Rotterdam Study. Stroke. 2001; 32: 2259–2264.
Wolff B, Volzke H, Robinson D, Schwahn C, Ludemann J, Kessler C, John U, Felix SB. Relation of parity with common carotid intima-media thickness among women of the Study of Health in Pomerania. Stroke. 2005; 36: 938–943.
Borson-Chazot F, Serusclat A, Kalfallah Y, Ducottet X, Sassolas G, Bernard S, Labrousse F, Pastene J, Sassolas A, Roux Y, Berthezene F. Decrease in carotid intima-media thickness after one year growth hormone (GH) treatment in adults with GH deficiency. J Clin Endocrinol Metab. 1999; 84: 1329–1333.
Krauss RM, Eckel RH, Howard B, Appel LJ, Daniels SR, Deckelbaum RJ, Erdman JW Jr, Kris-Etherton P, Goldberg IJ, Kotchen TA, Lichtenstein AH, Mitch WE, Mullis R, Robinson K, Wylie-Rosett J, St Jeor S, Suttie J, Tribble DL, Bazzarre TL. American Heart Association Dietary Guidelines: revision 2000: A statement for healthcare professionals from the Nutrition Committee of the American Heart Association. Circulation. 2000; 102: 2284–2299.
Friedewald WT, Levy RI, Fredrickson DS. Estimation of the concentration of low-density lipoprotein cholesterol in plasma, without use of the preparative ultracentrifuge. Clin Chem. 1972; 18: 499–502.
Wilson PW, D'Agostino RB, Levy D, Belanger AM, Silbershatz H, Kannel WB. Prediction of coronary heart disease using risk factor categories. Circulation. 1998; 97: 1837–1847.
Altman DG, Bland JM. Interaction revisited: the difference between two estimates. BMJ. 2003; 326: 219.
Institut national de la statistique et des études économiques. Bilan démographique 2007. Available at http://www.insee.fr/fr/themes/detail.asp?ref_id=bilan-demo®_id=99&page=donnees-detaillees/bilan-demo/pop_age3c.htm. Accessed on 08/08/2008.
Kravdal O. Is the relationship between childbearing and cancer incidence due to biology or lifestyle? Examples of the importance of using data on men. Int J Epidemiol. 1995; 24: 477–484.
Skilton MR, Moulin P, Serusclat A, Nony P, Bonnet F. A comparison of the NCEP-ATPIII, IDF and AHA/NHLBI metabolic syndrome definitions with relation to early carotid atherosclerosis in subjects with hypercholesterolemia or at risk of CVD: Evidence for sex-specific differences. Atherosclerosis. 2007; 190: 416–422.
van Stiphout WA, Hofman A, de Bruijn AM. Serum lipids in young women before, during, and after pregnancy. Am J Epidemiol. 1987; 126: 922–928.
Barrett-Connor E. Sex differences in coronary heart disease: why are women so superior? The 1995 Ancel Keys Lecture. Circulation. 1997; 95: 252–264.
Lawlor DA, Ebrahim S, Davey Smith G. Role of endogenous oestrogen in aetiology of coronary heart disease: analysis of age related trends in coronary heart disease and breast cancer in England and Wales and Japan. BMJ. 2002; 325: 311–312.