Association of Adverse Pregnancy Outcomes With Risk of Atherosclerotic Cardiovascular Disease in Postmenopausal Women | Cardiology | JAMA Cardiology | JAMA Network
[Skip to Navigation]
Sign In
Figure 1.  Flow Diagram for the Selection of the Study Participants From the Women’s Health Initiative (WHI)
Flow Diagram for the Selection of the Study Participants From the Women’s Health Initiative (WHI)
Figure 2.  Association of Individual Adverse Pregnancy Outcomes (APOs) With Atherosclerotic Cardiovascular Disease (ASCVD)
Association of Individual Adverse Pregnancy Outcomes (APOs) With Atherosclerotic Cardiovascular Disease (ASCVD)

Each line displays the odds ratio and its 95% CI from the comparison of yes and no responses based on a multinomial logistic model. For each APO, the top line shows the odds ratio for the APO from an unadjusted model and the bottom line shows the odds ratio for the APO from a model that adjusted for all traditional ASCVD risk factors, including age, hyperlipidemia, hypertension, diabetes, and smoking.

Figure 3.  Association of Multiple Adverse Pregnancy Outcomes (APOs) With Atherosclerotic Cardiovascular Disease (ASCVD)
Association of Multiple Adverse Pregnancy Outcomes (APOs) With Atherosclerotic Cardiovascular Disease (ASCVD)

Each line displays the odds ratio and it 95% CI from the comparison of yes and no responses based on a multinomial logistic regression model. For each APO, the top line shows the odds ratio adjusted for traditional ASCVD risk factors, including age, hyperlipidemia, hypertension, diabetes, and smoking, but for no other APOs. The next 4 lines show the odds ratio adjusted for the traditional ASCVD risk factors plus an additional APO. The bottom line shows the odds ratio for the APO adjusted for the traditional ASCVD risk factors plus all other 4 APOs.

Table 1.  Baseline Characteristics of Survey Responders and Nonresponders
Baseline Characteristics of Survey Responders and Nonresponders
Table 2.  Baseline Characteristics According to Adverse Pregnancy Outcome (APO) Statusa
Baseline Characteristics According to Adverse Pregnancy Outcome (APO) Statusa
1.
Xu  J, Murphy  SL, Kochanek  KD, Bastian  B, Arias  E.  Deaths: final data for 2016 (National Vital Statistics Reports volume 67, number 5).   Natl Vital Stat Reports. 2018;67(6):1-76.Google Scholar
2.
Rich-Edwards  JW, Fraser  A, Lawlor  DA, Catov  JM.  Pregnancy characteristics and women’s future cardiovascular health: an underused opportunity to improve women’s health?   Epidemiol Rev. 2014;36(1):57-70. doi:10.1093/epirev/mxt006PubMedGoogle ScholarCrossref
3.
Martinez  G, Daniels  K, Chandra  A.  Fertility of men and women aged 15-44 years in the United States: National Survey of Family Growth, 2006-2010.   Natl Health Stat Report. 2012;51(51):1-28.PubMedGoogle Scholar
4.
Hutcheon  JA, Lisonkova  S, Joseph  KS.  Epidemiology of pre-eclampsia and the other hypertensive disorders of pregnancy.   Best Pract Res Clin Obstet Gynaecol. 2011;25(4):391-403. doi:10.1016/j.bpobgyn.2011.01.006PubMedGoogle ScholarCrossref
5.
Bertoni  AG, Burke  GL, Owusu  JA,  et al.  Inflammation and the incidence of type 2 diabetes: the Multi-Ethnic Study of Atherosclerosis (MESA).   Diabetes Care. 2010;33(4):804-810. doi:10.2337/dc09-1679PubMedGoogle ScholarCrossref
6.
Bellamy  L, Casas  JP, Hingorani  AD, Williams  DJ.  Pre-eclampsia and risk of cardiovascular disease and cancer in later life: systematic review and meta-analysis.   BMJ. 2007;335(7627):974-977. doi:10.1136/bmj.39335.385301.BEPubMedGoogle ScholarCrossref
7.
Fraser  A, Nelson  SM, Macdonald-Wallis  C,  et al.  Associations of pregnancy complications with calculated cardiovascular disease risk and cardiovascular risk factors in middle age: the Avon Longitudinal Study of Parents and Children.   Circulation. 2012;125(11):1367-1380. doi:10.1161/CIRCULATIONAHA.111.044784PubMedGoogle ScholarCrossref
8.
Grandi  SM, Filion  KB, Yoon  S,  et al.  Cardiovascular disease-related morbidity and mortality in women with a history of pregnancy complications.   Circulation. 2019;139(8):1069-1079. doi:10.1161/CIRCULATIONAHA.118.036748PubMedGoogle ScholarCrossref
9.
Wu  P, Haththotuwa  R, Kwok  CS,  et al.  Preeclampsia and future cardiovascular health: a systematic review and meta-analysis.   Circ Cardiovasc Qual Outcomes. 2017;10(2):1-9. doi:10.1161/CIRCOUTCOMES.116.003497PubMedGoogle ScholarCrossref
10.
Wu  P, Gulati  M, Kwok  CS,  et al.  Preterm delivery and future risk of maternal cardiovascular disease: a systematic review and meta-analysis.   J Am Heart Assoc. 2018;7(2):e007809. doi:10.1161/JAHA.117.007809PubMedGoogle Scholar
11.
Haas  DM, Parker  CB, Marsh  DJ,  et al; NHLBI nuMoM2b Heart Health Study.  Association of adverse pregnancy outcomes with hypertension 2 to 7 years postpartum.   J Am Heart Assoc. 2019;8(19):e013092. doi:10.1161/JAHA.119.013092PubMedGoogle Scholar
12.
Haas  DM, Ehrenthal  DB, Koch  MA,  et al; National Heart, Lung, and Blood Institute nuMoM2b Heart Health Study Network.  Pregnancy as a window to future cardiovascular health: design and implementation of the nuMoM2b heart health study.   Am J Epidemiol. 2016;183(6):519-530. doi:10.1093/aje/kwv309PubMedGoogle ScholarCrossref
13.
Mosca  L, Benjamin  EJ, Berra  K,  et al; American Heart Association.  Effectiveness-based guidelines for the prevention of cardiovascular disease in women, 2011 update: a guideline from the American Heart Association.   J Am Coll Cardiol. 2011;57(12):1404-1423. doi:10.1016/j.jacc.2011.02.005PubMedGoogle ScholarCrossref
14.
Grundy  SM, Stone  NJ, Bailey  AL,  et al.  2018 AHA/ACC/aacvpr/aapa/abc/acpm/ada/ags/apha/aspc/nla/pcna guideline on the management of blood cholesterol: a report of the American College of Cardiology/American Heart Association Task Force on Clinical Practice Guidelines.   J Am Coll Cardiol. 2019;73(24):e285-e350. doi:10.1016/j.jacc.2018.11.003PubMedGoogle ScholarCrossref
15.
Piepoli  MF, Hoes  AW, Agewall  S,  et al; ESC Scientific Document Group.  2016 European guidelines on cardiovascular disease prevention in clinical practice: the Sixth Joint Task Force of the European Society of Cardiology and Other Societies on Cardiovascular Disease Prevention in Clinical Practice (constituted by representatives of 10 societies and by invited experts) developed with the special contribution of the European Association for Cardiovascular Prevention & Rehabilitation (EACPR).   Eur Heart J. 2016;37(29):2315-2381. doi:10.1093/eurheartj/ehw106PubMedGoogle ScholarCrossref
16.
The Women’s Health Initiative Study Group.  Design of the Women’s Health Initiative clinical trial and observational study.   Control Clin Trials. 1998;19(1):61-109. doi:10.1016/S0197-2456(97)00078-0PubMedGoogle ScholarCrossref
17.
Hays  J, Hunt  JR, Hubbell  FA,  et al.  The Women’s Health Initiative recruitment methods and results.   Ann Epidemiol. 2003;13(9)(suppl):S18-S77. doi:10.1016/S1047-2797(03)00042-5PubMedGoogle ScholarCrossref
18.
Curb  JD, McTiernan  A, Heckbert  SR,  et al; WHI Morbidity and Mortality Committee.  Outcomes ascertainment and adjudication methods in the Women’s Health Initiative.   Ann Epidemiol. 2003;13(9)(suppl):S122-S128. doi:10.1016/S1047-2797(03)00048-6PubMedGoogle ScholarCrossref
19.
Brown  HL, Warner  JJ, Gianos  E,  et al; American Heart Association and the American College of Obstetricians and Gynecologists.  Promoting risk identification and reduction of cardiovascular disease in women through collaboration with obstetricians and gynecologists: a Presidential Advisory From the American Heart Association and the American College of Obstetricians and Gynecologists.   Circulation. 2018;137(24):e843-e852. doi:10.1161/CIR.0000000000000582PubMedGoogle ScholarCrossref
20.
Arnett  DK, Blumenthal  RS, Albert  MA,  et al.  2019 ACC/AHA guideline on the primary prevention of cardiovascular disease: a report of the American College of Cardiology/American Heart Association Task Force on Clinical Practice Guidelines.   Circulation. 2019;140(11):e596-e646. doi:10.1161/CIR.0000000000000678PubMedGoogle Scholar
21.
Markovitz  AR, Stuart  JJ, Horn  J,  et al.  Does pregnancy complication history improve cardiovascular disease risk prediction? findings from the HUNT study in Norway.   Eur Heart J. 2019;40(14):1113-1120. doi:10.1093/eurheartj/ehy863PubMedGoogle ScholarCrossref
22.
Siddiqui  N, Hladunewich  M.  Understanding the link between the placenta and future cardiovascular disease.   Trends Cardiovasc Med. 2011;21(7):188-193. doi:10.1016/j.tcm.2012.05.008PubMedGoogle ScholarCrossref
23.
Kim  C, Newton  KM, Knopp  RH.  Gestational diabetes and the incidence of type 2 diabetes: a systematic review.   Diabetes Care. 2002;25(10):1862-1868. doi:10.2337/diacare.25.10.1862PubMedGoogle ScholarCrossref
24.
Ahmed  R, Dunford  J, Mehran  R, Robson  S, Kunadian  V.  Pre-eclampsia and future cardiovascular risk among women: a review.   J Am Coll Cardiol. 2014;63(18):1815-1822. doi:10.1016/j.jacc.2014.02.529PubMedGoogle ScholarCrossref
25.
Li  C-Y, Chen  H-F, Sung  F-C,  et al.  Offspring birth weight and parental cardiovascular mortality.   Int J Epidemiol. 2010;39(4):1082-1090. doi:10.1093/ije/dyq045PubMedGoogle ScholarCrossref
26.
Bonamy  AK, Parikh  NI, Cnattingius  S, Ludvigsson  JF, Ingelsson  E.  Birth characteristics and subsequent risks of maternal cardiovascular disease: effects of gestational age and fetal growth.   Circulation. 2011;124(25):2839-2846. doi:10.1161/CIRCULATIONAHA.111.034884PubMedGoogle ScholarCrossref
27.
Romero  R, Kusanovic  JP, Chaiworapongsa  T, Hassan  SS.  Placental bed disorders in preterm labor, preterm PROM, spontaneous abortion and abruptio placentae.   Best Pract Res Clin Obstet Gynaecol. 2011;25(3):313-327. doi:10.1016/j.bpobgyn.2011.02.006PubMedGoogle ScholarCrossref
28.
Valdiviezo  C, Garovic  VD, Ouyang  P.  Preeclampsia and hypertensive disease in pregnancy: their contributions to cardiovascular risk.   Clin Cardiol. 2012;35(3):160-165. doi:10.1002/clc.21965PubMedGoogle ScholarCrossref
29.
Carpenter  MW, Coustan  DR.  Criteria for screening tests for gestational diabetes.   Am J Obstet Gynecol. 1982;144(7):768-773. doi:10.1016/0002-9378(82)90349-0PubMedGoogle ScholarCrossref
30.
Carter  EB, Stuart  JJ, Farland  LV,  et al.  Pregnancy complications as markers for subsequent maternal cardiovascular disease: validation of a maternal recall questionnaire.   J Womens Health (Larchmt). 2015;24(9):702-712. doi:10.1089/jwh.2014.4953PubMedGoogle ScholarCrossref
Original Investigation
September 16, 2020

Association of Adverse Pregnancy Outcomes With Risk of Atherosclerotic Cardiovascular Disease in Postmenopausal Women

Author Affiliations
  • 1Aalborg University School of Medicine and Health, Aalborg, Denmark
  • 2Stanford University School of Medicine, Stanford, California
  • 3UCSF School of Medicine, University of California, San Francisco
  • 4UC San Diego School of Medicine, University of California, San Diego
  • 5Johns Hopkins Bloomberg School of Public Health, Baltimore, Maryland
  • 6Northwestern University Feinberg School of Medicine, Chicago, Illinois
  • 7University of Florida School of Medicine, Gainesville
  • 8Brown University Alpert School of Medicine, Providence, Rhode Island
  • 9Ohio State University College of Public Health, Columbus
  • 10MedStar Health, Washington, DC
  • 11Brigham and Women’s Hospital, Harvard Medical School, Boston, Massachusetts
JAMA Cardiol. 2020;5(12):1390-1398. doi:10.1001/jamacardio.2020.4097
Key Points

Question  Do adverse pregnancy outcomes represent independent risk factors for atherosclerotic cardiovascular disease in women?

Findings  In this cohort study including 46 805 Women’s Health Initiative participants, hypertensive disorders of pregnancy and low birth weight were independently associated with atherosclerotic cardiovascular disease after adjustment for standard risk factors and other adverse pregnancy outcomes.

Meaning  Adverse pregnancy outcomes are sex-specific risk factors for atherosclerotic cardiovascular disease in women.

Abstract

Importance  Atherosclerotic cardiovascular disease (ASCVD) may have unique risk factors in women. Most women have a history of pregnancy; common adverse pregnancy outcomes (APOs) appear to be associated with ASCVD, but prior studies have limitations.

Objective  To assess whether APOs are associated with increased ASCVD risk independently of traditional risk factors.

Design, Setting, and Participants  The APO history among participants in the Women’s Health Initiative, a large multiethnic cohort of postmenopausal women, was assessed. The associations of 5 self-reported APOs (gestational diabetes, hypertensive disorders of pregnancy, low birth weight [ie, birth weight less than 2.49 kg], high birth weight [ie, birth weight greater than 4.08 kg], and preterm delivery by 3 weeks or more) with ASCVD were analyzed, adjusting for traditional ASCVD risk factors. Data were collected and analyzed in 2017.

Exposures  APOs (gestational diabetes, hypertensive disorders of pregnancy, low birth weight, high birth weight, and preterm delivery).

Main Outcomes and Measures  Adjudicated ASCVD.

Results  A total of 48 113 Women’s Health Initiative participants responded to the survey; the median (interquartile range) age at time of enrollment was 60.0 (55.0-64.0) years. A total of 13 482 participants (28.8%) reported 1 or more APOs. Atherosclerotic cardiovascular disease was more frequent in women who reported an APO compared with those without APOs (1028 of 13 482 [7.6%] vs 1758 of 30 522 [5.8%]). Each APO, analyzed separately, was significantly associated with ASCVD, and gestational diabetes, hypertensive disorders of pregnancy, low birth weight, and preterm delivery remained significant after adjustment for traditional ASCVD risk factors. When all APOs were analyzed together, hypertensive disorders of pregnancy (odds ratio, 1.27; 95% CI, 1.15-1.40) and low birth weight (odds ratio, 1.12; 95% CI, 1.00-1.26) remained independently associated with ASCVD. All findings were materially unchanged by additional adjustment for parity, body mass index, and socioeconomic factors.

Conclusions and Relevance  In this large multiethnic cohort of women, hypertensive disorders of pregnancy and low birth weight were independently associated with ASCVD after adjustment for risk factors and other APOs.

Introduction

Cardiovascular disease is the leading cause of death among women in the US,1 and identification of sex-specific risk factors could enhance cardiovascular risk assessment and prevention. Pregnancy is an exposure unique to women: 85% of women give birth at least once in their lives, and up to 30% may experience an adverse pregnancy outcome (APO).2,3 The associations of APOs, such as gestational diabetes, hypertensive disorders of pregnancy, preterm delivery, and low birth weight, with atherosclerotic cardiovascular disease (ASCVD) later in life have only recently been examined.

Gestational diabetes, hypertensive disorders of pregnancy, preterm delivery, and low birth weight have each been associated with subsequent development of ASCVD risk factors as well as with an increased incidence of clinically apparent ASCVD.2,4-12 Consequently, the American Heart Association recommends that taking a history of APOs be part of ASCVD risk evaluation,13 and the 2018 cholesterol treatment guidelines suggest that hypertensive disorders of pregnancy, preeclampsia, gestational diabetes, small-for-gestational-age infants, and preterm deliveries be considered risk enhancers when discussing initiation of statins for primary prevention of ASCVD in women.14 The European Society of Cardiology 2016 guidelines on cardiovascular disease prevention state that it is not clear whether APOs represent risk factors themselves or if they predispose women by their association with other ASCVD risk factors.15

The associations between APOs (including hypertensive disorders of pregnancy, gestational diabetes, small-for-gestational-age infants, and preterm deliveries) and ASCVD risk are supported by several lines of evidence, yet there are limitations and gaps in this evidence. For example, most prior studies of these APOs have not adjusted for all established ASCVD risk factors, many were not prospective or lacked long-term follow-up, and few have been conducted in multiethnic cohorts. Furthermore, most studies have only examined the association of a single APO with ASCVD risk, even though several APOs may occur in the same woman. We sought to overcome several of these limitations by examining data from the Women’s Health Initiative (WHI), a large multiethnic cohort study with rigorous, prospective follow-up and adjudicated ASCVD outcomes.

Methods
Study Population

The WHI is a longitudinal cohort study, consisting of 3 randomized clinical trials (of hormone therapy, dietary changes, and calcium/vitamin D supplementation) and an observational study.16 Women were enrolled between 1994 and 1998 and have been followed up prospectively.17 At entry, WHI participants were aged 50 to 79 years, postmenopausal, and represented ethnically diverse populations.17 Race and ethnicity were self-reported by the participants. Extensive baseline data were collected at an in-person visit, including the participant’s reproductive history, but no data on APOs were collected at study entry.17 This study was approved by the University of California, San Francisco, institutional review board, and all participants gave written informed consent to participate in WHI and its extension study.

APO Questionnaire

In 2017, a questionnaire surveyed participants’ history of APOs (in 1 or more pregnancies) and asked about their dental health, memory, and changes in mental functioning (eAppendix in the Supplement). The survey was sent to all surviving WHI enrollees who were still participating in the study and included 6 questions on APOs during any pregnancy, with possible responses of no, yes, and don’t know. The APOs surveyed were gestational diabetes, gestational hypertension, preeclampsia, low birth weight (ie, birth weight less than 2.49 kg), high birth weight (ie, birth weight greater than 4.08 kg), and preterm delivery by 3 weeks or more (preterm delivery). More than 1 APO may have occurred in the same woman but not necessarily during the same pregnancy.

Outcomes

The primary study outcome was ASCVD, defined as either myocardial infarction, stroke, peripheral artery disease, or coronary revascularization. These outcomes were prospectively ascertained between study entry and the time of the 2017 survey and were adjudicated using standard criteria.18

Covariates

We analyzed baseline characteristics from the data collected at enrollment into the WHI. The ASCVD risk factors included hyperlipidemia (defined as use of cholesterol-lowering drugs), hypertension (including use of antihypertensive drugs), diabetes (not pregnancy related and self-reported), and smoking status. Data on income, highest education level attained, and self-reported ethnicity were also recorded at study entry.

Statistical Analyses

Women were eligible for this study if they had a history of pregnancy that lasted for more than 6 months and were alive and still participating in the WHI in 2017 when the survey was performed. Women were excluded if, at baseline, they reported previous stroke, myocardial infarction, or coronary revascularization procedure. Nonresponders were defined as eligible WHI participants who did not answer the survey.

We assessed differences between responders and nonresponders using counts and percentages for categorical variables and medians and interquartile ranges for continuous variables. We also compared the characteristics of women who reported an APO with those who did not, by counts and percentages for categorical variables and medians and interquartile ranges for continuous variables. In the primary analysis, we used multinomial logistic regression to assess the association between each APO and ASCVD; in this model, each APO was coded using 3 categories (yes, no, or don’t know), and the responses yes and don’t know were compared with the reference category (no) for their association with ASCVD. In secondary analyses, we tested whether results were changed by using 2 alternative model specifications, in which the don’t know responses were (1) treated as missing (ie, omitted from the model) and (2) combined with the yes responses. For simplicity, we present the results of the primary analysis in the main article and the results of the 2 sensitivity analyses in the eAppendix in the Supplement because the findings were similar.

For each APO, we first used a univariate model and then adjusted for the traditional ASCVD risk factors (age, hyperlipidemia, hypertension, diabetes, and smoking). Finally, we assessed the joint associations of each APO with ASCVD in a multivariable model that simultaneously adjusted for all APOs and risk factors. We tested each model for interaction by race/ethnicity. A 2-sided P value less than .05 was considered statistically significant.

All statistical analyses were conducted using RStudio 1.1.456 by RStudio (R version 3.5.1; The R Foundation). Data retrieval was done using SAS Enterprise version 9.4 (SAS Institute).

Results
Survey Responders

Of the 161 808 women enrolled in the WHI, 71 661 were still being followed up in 2017 and were mailed a survey about APOs. A total of 46 805 eligible study participants (65.3%) responded and answered questions on APOs (Figure 1). The median (interquartile range) age at the time of enrollment was 60 (55.0-64.0) years. Responders were younger and had less hypertension, diabetes, and hyperlipidemia than eligible nonresponders (Table 1). Responders were also more likely than nonresponders to be White, college educated, and have higher incomes (Table 1). Atherosclerotic cardiovascular disease was more frequent in women who reported an APO compared with those without APOs (1028 of 13, 482 [7.6%] vs 1758 of 30, 522 [5.8%]).

APOs

Adverse pregnancy outcomes were reported by 13 482 women (28.8%). A single APO occurred in 8726 women, and 2 or more APOs were reported by 4756 women (eFigure 1 in the Supplement). The single most common APO was preterm delivery, which was reported by 6790 women (14.5%), followed by low birth weight, which was reported by 5773 women (12.3%) (eTable 1 in the Supplement). The most common combination of APOs was preterm delivery and low birth weight, with 3719 women (7.9%) reporting having had both conditions. Because many women reported both gestational hypertension (n = 1506) and/or preeclampsia (n = 2065), in subsequent analyses, we combined these responses into the single category of hypertensive disorders of pregnancy.

Baseline characteristics of women at entry into the WHI differed by the presence and type of APO (Table 2). Hypertension at entry in the WHI was significantly more prevalent among women who reported a prior hypertensive disorder of pregnancy (1457 of 2936 [49.6%]) than women reporting other APOs (low: 1782 of 6790 [26.2%]; high: 332 of 959 [34.6%]) or no APOs (6551 of 30 522 [21.5%]). Diabetes at entry into the WHI was significantly more prevalent among women who reported prior gestational diabetes (145 of 959 [15.1%]) compared with women reporting other APOs (low: 202 of 6790 [3.0%]; high: 158 of 2691 [5.9%]) or no APOs (556 of 30 522 [1.8%]). Other ASCVD risk factors, including age, hyperlipidemia, and smoking at study entry, were generally similar in women with different APOs. Baseline characteristics according to the survey responses yes, don’t know, and no for each APO are detailed in the eAppendix and eTable 2 in the Supplement.

A total of 1028 of 13 482 women with an APO (7.6%) had developed ASCVD between entry into WHI and the time of the survey compared with 1758 of 30 522 women without an APO (5.8%) (Table 2; eFigure 2 in the Supplement). The mean (SD) age at onset of ASCVD was lower among women with a history of an APO than those without (70.8 years; IQR, 66.3-75.7 years vs 71.9 years; IQR, 66.9-76.9 years).

APOs and ASCVD

The associations of the responses yes and don’t know for each APO with ASCVD were similar in the multinomial logistic model, which separately compared the yes and don’t know responses with the reference category (no); for simplicity, we present the odds ratios (ORs) for the yes responses in the article and present the ORs for the don’t know responses in eFigure 3 in the Supplement. The ORs for the yes responses in the multinomial model were also similar to the ORs for yes responses in 2 alternative models of the association of APOs with ASCVD, including a binary logistic model in which don’t know responses were treated as missing data and a binary logistic model in which don’t know and yes responses were combined into a single category and compared with no responses (eTable 3 in the Supplement).

Each of the 5 APOs was significantly associated with ASCVD in a univariate model (Figure 2). After adjustment for traditional ASCVD risk factors, all APOs remained statistically significantly associated with ASCVD, with the exception of high birth weight (Figure 2). The associations of low birth weight and preterm delivery with ASCVD were largely unchanged by adjustment for traditional risk factors, while the associations of other APOs with ASCVD were attenuated by roughly half after adjustment for traditional risk factors, although they remained significant (Figure 2). The association of hypertensive disorders of pregnancy with ASCVD was primarily attenuated by adjustment for subsequent development of hypertension, while the association of gestational diabetes with ASCVD was primarily attenuated by adjustment for subsequent development of diabetes (eFigure 4 in the Supplement).

Multiple APOs

When all 5 APOs were tested jointly for their association with ASCVD in a model that adjusted for traditional risk factors, only hypertensive disorders of pregnancy and low birth weight were independently associated with ASCVD. Hypertensive disorders of pregnancy had an adjusted OR of 1.34 (95% CI, 1.15 to 1.54), and low birth weight had an adjusted OR of 1.18 (95% CI, 1.03 to 1.35) (Figure 3).

Additional Adjustments and Effect Modification

All results were not materially changed after additional adjustment for race/ethnicity, income, education, body mass index, breastfeeding, and parity (eFigure 5 in the Supplement). There was no statistically significant effect modification by race/ethnicity of any of the associations between APOs and ASCVD (eFigure 5 in the Supplement). Results from all analyses were virtually unchanged when multiple imputation of missing baseline data was used.

Discussion

In this large, multiethnic cohort of women, gestational diabetes, hypertensive disorders of pregnancy, preterm delivery, and low birth weight were significantly associated with ASCVD, even after adjustment for traditional ASCVD risk factors. When all 5 APOs were assessed together in a single model (which also adjusted for ASCVD risk factors), hypertensive disorders of pregnancy and low birth weight were each independently associated with ASCVD. All of the associations between individual APOs and ASCVD remained significant after further adjustment for race/ethnicity and socioeconomic status, without evidence of effect modification by race/ethnicity. Our study supports guideline recommendations that clinicians should consider a history of pregnancy-associated disorders when assessing ASCVD risk in older women.19,20 Indeed, our results suggest that hypertensive disorders of pregnancy and low birth weight are the most important APOs to consider as potential ASCVD risk enhancers.

The association between pregnancy outcomes and women’s cardiovascular health has become a focus of research interest and might identify unique mechanisms of ASCVD development in women. Previous studies have been inconsistent in demonstrating whether specific individual APOs add to risk prediction. A 2019 study21 reported that adding preeclampsia, gestational hypertension, preterm delivery, and small for gestational age to risk prediction models modestly increased risk discrimination for ASCVD events and added to information provided by traditional risk factors. Adverse pregnancy outcomes, which have been termed a positive stress test, are associated with late-onset ASCVD, but it is not clear whether these associations are independent of established ASCVD risk factors because hypertensive disorders of pregnancy and gestational diabetes are antecedents of diabetes and hypertension.22-24

Prior studies have not found significant associations between high birth weight (or large for gestational age) and later maternal cardiovascular disease.25,26 Low birth weight and small for gestational age are common metrics of the health of a pregnancy, and each has been associated with cardiovascular disease in prior studies.2 Infants may be small for gestational age because of poor implantation of the placenta into the uterus or from vascular insufficiency of the maternal uterine artery, leading to inadequate levels of oxygen and nutrients to the fetus. Low birth weight is defined as the weight of an infant at birth, regardless of gestational age, and may be owing to premature birth. Genetic factors and maternal nutrition may also affect fetal size.

Our study confirms that traditional ASCVD risk factors are more common in women who have had the APOs we studied but also that APOs remain significantly associated with ASCVD even after adjusting for these risk factors. Our study extends prior work by evaluating the joint associations of several APOs with ASCVD, while most prior studies have assessed only single APOs. Because a woman can have more than 1 APO (eFigure 1 in the Supplement), it has been uncertain whether some of the specific APOs that we studied are more closely associated with ASCVD risk, particularly because hypertensive disorders of pregnancy, low birth weight, and preterm delivery may share the same underlying pathophysiology.2,22,27 Our finding that hypertensive disorders of pregnancy and low birth weight were independently associated with ASCVD suggests that these APOs merit particular scrutiny in future epidemiologic and mechanistic studies.

Hypertensive disorders of pregnancy have been found to be associated with late ASCVD in meta-analyses.6,8,9 Low birth weight was not significantly associated with ASCVD in a 2019 meta-analysis of cohort studies,8 but we found a strong independent association between low birth weight and ASCVD, which is plausible because low birth weight, preterm birth, and hypertensive disorders of pregnancy may all arise from placental dysfunction.2,22,27-30 Preterm delivery has also been found to be associated with ASCVD in a 2018 meta-analysis,10 and while our study confirms this association, it was no longer significant after adjustment for low birth weight because these APOs commonly occur in the same pregnancy and share pathophysiologic mechanisms. Hypertensive disorders of pregnancy and low birth weight remained independently associated with ASCVD when adjusting for a full APO history, confirming the results of previous studies.6-9

Prior meta-analyses8 have demonstrated an association between gestational diabetes and ASCVD, which we confirmed in this study. However, this association was attenuated by adjustment for subsequent development of diabetes and was not independent of other APOs. It is important to note that the definition of gestational diabetes has evolved over the past 50 years, and with the advent of oral glucose tolerance testing in the 1970s and more aggressive care for gestational diabetes, the accuracy of the diagnosis of gestational diabetes has improved, and more recent studies will be less subject to exposure misclassification.29

It is unclear whether APOs are associated with ASCVD because they are involved in novel pathophysiologic processes or whether their association with ASCVD serves as an indicator of longer-term and more severe underlying cardiometabolic risk, either by heralding earlier development of diabetes, hypertension, and metabolic syndrome or by serving as a stress test for preexisting metabolic risk.22 Prior studies suggest that APOs remain associated with ASCVD even after adjustment for traditional risk factors, but they could still be serving as markers for more severe metabolic abnormalities. Novel pathophysiologic pathways might connect APOs with late ASCVD, and this possibility will be important to assess in future studies.

Limitations

This study has limitations. This study was based on 46 805 WHI participants who had survived to 2017 and were able to respond to our survey about APOs, which may have introduced survival bias into the estimated strengths of association. However, it is most likely that the women at highest risk of ASCVD would have died prior to the survey, which would have biased our findings toward the null and might account for the lower ORs in our study than reported by prior studies.8

We relied on self-reported APOs, and while women’s ability to recall APOs has been validated in short-term studies,30 it has not been studied over the longer intervals in this study. Study participants’ recall of birth weight and early delivery seemed to be more confident than their recall of gestational diabetes or hypertensive disorders of pregnancy. It is possible that recall bias may have affected our results. It would have been ideal for us to have obtained objective information from medical records about APOs in the study participants; however, the fragmented health care system in the US and the lack of longitudinal, searchable records going back 4 to 5 decades makes retrieval of objective records on APOs infeasible. To the extent that self-reported APOs are uncertain, the likely bias in our results would be toward the null.

This study did not include as many minority women as the entire WHI because of their higher mortality and lower response rates to the survey. Nevertheless, we had a sufficient sample of nonwhite women to document the consistency of the associations between APOs and ASCVD by race/ethnicity. We had data on body mass index only at the time of enrollment, not prior to pregnancy, and we lacked data on gestational weight gain and hence could not assess their effects on ASCVD risk.

Additionally, we assessed the birth weight and preterm delivery in categories and not as continuous variables. It would have been ideal to test associations of measured birth weight and gestational age at delivery as well as continuous measurements of glucose tolerance and blood pressure with the development of ASCVD.

Conclusions

One or more APOs occurred in roughly one-third of women in this large multiethnic cohort study and were independently associated with late ASCVD. Hypertensive disorders of pregnancy and low birth weight were noteworthy because their associations with ASCVD were independent of both traditional cardiac risk factors and preterm delivery and gestational diabetes; consequently, hypertensive disorders of pregnancy and low birth weight may be particularly important risk enhancers to consider when performing ASCVD risk stratification in older women. Further research is needed to understand the potential mechanisms that might link hypertensive disorders of pregnancy and low birth weight with late ASCVD.

Back to top
Article Information

Accepted for Publication: May 22, 2020.

Corresponding Author: Nisha I. Parikh, MD, MPH, UCSF School of Medicine, University of California, San Francisco, 555 Mission Bay Blvd S, San Francisco, CA 94158 (nisha.parikh@ucsf.edu).

Published Online: September 16, 2020. doi:10.1001/jamacardio.2020.4097

Author Contributions: Drs Hlatky and Parikh had full access to all of the data in the study and take responsibility for the integrity of the data and the accuracy of the data analysis.

Study concept and design: Hlatky, Van Horn, Manson, Parikh.

Acquisition, analysis, or interpretation of data: All authors.

Drafting of the manuscript: Søndergaard, Hlatky, Nah.

Critical revision of the manuscript for important intellectual content: Søndergaard, Hlatky, Stefanick, Vittinghoff, Allison, Gemmill, Van Horn, Park, Salmoirago-Blotcher, Sattari, Sealy-Jefferson, Shadyab, Valdiviezo, Manson, Parikh.

Statistical analysis: Søndergaard, Vittinghoff, Nah, Parikh.

Obtained funding: Parikh.

Administrative, technical, or material support: Hlatky, Van Horn, Manson, Parikh.

Study supervision: Hlatky, Stefanick, Allison, Van Horn, Park, Parikh.

Conflict of Interest Disclosures: Dr Hlatky has received grants from the National Heart, Lung, and Blood Institute during the conduct of the study; has served as scientific advisor and on the medical advisory panel of Blue Cross Blue Shield Association; Associate Editor or global advisor of the Journal of the American College of Cardiology; independent outcome event adjudicator for the CANVAS, VEST PREDICTS, and VALOR CKD studies; is a member of the data and safety monitoring board for the TAILOR PCI trial; and is a consultant to the American Heart Association and the Medicines Company. Dr Vittinghoff has received salary support from the National Institutes of Health during the conduct of the study. Dr Salmoirago-Blotcher has received grants and personal fees from National Heart, Lung, and Blood Institute and personal fees from National Institute of Arthritis and Musculoskeletal and Skin Diseases outside the submitted work. No other disclosures were reported.

Funding/Support: The Women’s Health Initiative program was funded by the National Heart, Lung, and Blood Institute, National Institutes of Health, and US Department of Health and Human Services through contracts HHSN268201100046C, HHSN268201100001C, HHSN268201100002C, HHSN268201100003C, and HHSN268201100004C. This study was also supported, in part, by grant 1R01HL139844 from the National Heart, Lung, and Blood Institute. Mr Søndergaard was supported by a Danish American Research Exchange (DARE) award from the Lundbeck Foundation.

Role of the Funder/Sponsor: The funders had no role in the design and conduct of the study; collection, management, analysis, and interpretation of the data; preparation, review, or approval of the manuscript; and decision to submit the manuscript for publication.

References
1.
Xu  J, Murphy  SL, Kochanek  KD, Bastian  B, Arias  E.  Deaths: final data for 2016 (National Vital Statistics Reports volume 67, number 5).   Natl Vital Stat Reports. 2018;67(6):1-76.Google Scholar
2.
Rich-Edwards  JW, Fraser  A, Lawlor  DA, Catov  JM.  Pregnancy characteristics and women’s future cardiovascular health: an underused opportunity to improve women’s health?   Epidemiol Rev. 2014;36(1):57-70. doi:10.1093/epirev/mxt006PubMedGoogle ScholarCrossref
3.
Martinez  G, Daniels  K, Chandra  A.  Fertility of men and women aged 15-44 years in the United States: National Survey of Family Growth, 2006-2010.   Natl Health Stat Report. 2012;51(51):1-28.PubMedGoogle Scholar
4.
Hutcheon  JA, Lisonkova  S, Joseph  KS.  Epidemiology of pre-eclampsia and the other hypertensive disorders of pregnancy.   Best Pract Res Clin Obstet Gynaecol. 2011;25(4):391-403. doi:10.1016/j.bpobgyn.2011.01.006PubMedGoogle ScholarCrossref
5.
Bertoni  AG, Burke  GL, Owusu  JA,  et al.  Inflammation and the incidence of type 2 diabetes: the Multi-Ethnic Study of Atherosclerosis (MESA).   Diabetes Care. 2010;33(4):804-810. doi:10.2337/dc09-1679PubMedGoogle ScholarCrossref
6.
Bellamy  L, Casas  JP, Hingorani  AD, Williams  DJ.  Pre-eclampsia and risk of cardiovascular disease and cancer in later life: systematic review and meta-analysis.   BMJ. 2007;335(7627):974-977. doi:10.1136/bmj.39335.385301.BEPubMedGoogle ScholarCrossref
7.
Fraser  A, Nelson  SM, Macdonald-Wallis  C,  et al.  Associations of pregnancy complications with calculated cardiovascular disease risk and cardiovascular risk factors in middle age: the Avon Longitudinal Study of Parents and Children.   Circulation. 2012;125(11):1367-1380. doi:10.1161/CIRCULATIONAHA.111.044784PubMedGoogle ScholarCrossref
8.
Grandi  SM, Filion  KB, Yoon  S,  et al.  Cardiovascular disease-related morbidity and mortality in women with a history of pregnancy complications.   Circulation. 2019;139(8):1069-1079. doi:10.1161/CIRCULATIONAHA.118.036748PubMedGoogle ScholarCrossref
9.
Wu  P, Haththotuwa  R, Kwok  CS,  et al.  Preeclampsia and future cardiovascular health: a systematic review and meta-analysis.   Circ Cardiovasc Qual Outcomes. 2017;10(2):1-9. doi:10.1161/CIRCOUTCOMES.116.003497PubMedGoogle ScholarCrossref
10.
Wu  P, Gulati  M, Kwok  CS,  et al.  Preterm delivery and future risk of maternal cardiovascular disease: a systematic review and meta-analysis.   J Am Heart Assoc. 2018;7(2):e007809. doi:10.1161/JAHA.117.007809PubMedGoogle Scholar
11.
Haas  DM, Parker  CB, Marsh  DJ,  et al; NHLBI nuMoM2b Heart Health Study.  Association of adverse pregnancy outcomes with hypertension 2 to 7 years postpartum.   J Am Heart Assoc. 2019;8(19):e013092. doi:10.1161/JAHA.119.013092PubMedGoogle Scholar
12.
Haas  DM, Ehrenthal  DB, Koch  MA,  et al; National Heart, Lung, and Blood Institute nuMoM2b Heart Health Study Network.  Pregnancy as a window to future cardiovascular health: design and implementation of the nuMoM2b heart health study.   Am J Epidemiol. 2016;183(6):519-530. doi:10.1093/aje/kwv309PubMedGoogle ScholarCrossref
13.
Mosca  L, Benjamin  EJ, Berra  K,  et al; American Heart Association.  Effectiveness-based guidelines for the prevention of cardiovascular disease in women, 2011 update: a guideline from the American Heart Association.   J Am Coll Cardiol. 2011;57(12):1404-1423. doi:10.1016/j.jacc.2011.02.005PubMedGoogle ScholarCrossref
14.
Grundy  SM, Stone  NJ, Bailey  AL,  et al.  2018 AHA/ACC/aacvpr/aapa/abc/acpm/ada/ags/apha/aspc/nla/pcna guideline on the management of blood cholesterol: a report of the American College of Cardiology/American Heart Association Task Force on Clinical Practice Guidelines.   J Am Coll Cardiol. 2019;73(24):e285-e350. doi:10.1016/j.jacc.2018.11.003PubMedGoogle ScholarCrossref
15.
Piepoli  MF, Hoes  AW, Agewall  S,  et al; ESC Scientific Document Group.  2016 European guidelines on cardiovascular disease prevention in clinical practice: the Sixth Joint Task Force of the European Society of Cardiology and Other Societies on Cardiovascular Disease Prevention in Clinical Practice (constituted by representatives of 10 societies and by invited experts) developed with the special contribution of the European Association for Cardiovascular Prevention & Rehabilitation (EACPR).   Eur Heart J. 2016;37(29):2315-2381. doi:10.1093/eurheartj/ehw106PubMedGoogle ScholarCrossref
16.
The Women’s Health Initiative Study Group.  Design of the Women’s Health Initiative clinical trial and observational study.   Control Clin Trials. 1998;19(1):61-109. doi:10.1016/S0197-2456(97)00078-0PubMedGoogle ScholarCrossref
17.
Hays  J, Hunt  JR, Hubbell  FA,  et al.  The Women’s Health Initiative recruitment methods and results.   Ann Epidemiol. 2003;13(9)(suppl):S18-S77. doi:10.1016/S1047-2797(03)00042-5PubMedGoogle ScholarCrossref
18.
Curb  JD, McTiernan  A, Heckbert  SR,  et al; WHI Morbidity and Mortality Committee.  Outcomes ascertainment and adjudication methods in the Women’s Health Initiative.   Ann Epidemiol. 2003;13(9)(suppl):S122-S128. doi:10.1016/S1047-2797(03)00048-6PubMedGoogle ScholarCrossref
19.
Brown  HL, Warner  JJ, Gianos  E,  et al; American Heart Association and the American College of Obstetricians and Gynecologists.  Promoting risk identification and reduction of cardiovascular disease in women through collaboration with obstetricians and gynecologists: a Presidential Advisory From the American Heart Association and the American College of Obstetricians and Gynecologists.   Circulation. 2018;137(24):e843-e852. doi:10.1161/CIR.0000000000000582PubMedGoogle ScholarCrossref
20.
Arnett  DK, Blumenthal  RS, Albert  MA,  et al.  2019 ACC/AHA guideline on the primary prevention of cardiovascular disease: a report of the American College of Cardiology/American Heart Association Task Force on Clinical Practice Guidelines.   Circulation. 2019;140(11):e596-e646. doi:10.1161/CIR.0000000000000678PubMedGoogle Scholar
21.
Markovitz  AR, Stuart  JJ, Horn  J,  et al.  Does pregnancy complication history improve cardiovascular disease risk prediction? findings from the HUNT study in Norway.   Eur Heart J. 2019;40(14):1113-1120. doi:10.1093/eurheartj/ehy863PubMedGoogle ScholarCrossref
22.
Siddiqui  N, Hladunewich  M.  Understanding the link between the placenta and future cardiovascular disease.   Trends Cardiovasc Med. 2011;21(7):188-193. doi:10.1016/j.tcm.2012.05.008PubMedGoogle ScholarCrossref
23.
Kim  C, Newton  KM, Knopp  RH.  Gestational diabetes and the incidence of type 2 diabetes: a systematic review.   Diabetes Care. 2002;25(10):1862-1868. doi:10.2337/diacare.25.10.1862PubMedGoogle ScholarCrossref
24.
Ahmed  R, Dunford  J, Mehran  R, Robson  S, Kunadian  V.  Pre-eclampsia and future cardiovascular risk among women: a review.   J Am Coll Cardiol. 2014;63(18):1815-1822. doi:10.1016/j.jacc.2014.02.529PubMedGoogle ScholarCrossref
25.
Li  C-Y, Chen  H-F, Sung  F-C,  et al.  Offspring birth weight and parental cardiovascular mortality.   Int J Epidemiol. 2010;39(4):1082-1090. doi:10.1093/ije/dyq045PubMedGoogle ScholarCrossref
26.
Bonamy  AK, Parikh  NI, Cnattingius  S, Ludvigsson  JF, Ingelsson  E.  Birth characteristics and subsequent risks of maternal cardiovascular disease: effects of gestational age and fetal growth.   Circulation. 2011;124(25):2839-2846. doi:10.1161/CIRCULATIONAHA.111.034884PubMedGoogle ScholarCrossref
27.
Romero  R, Kusanovic  JP, Chaiworapongsa  T, Hassan  SS.  Placental bed disorders in preterm labor, preterm PROM, spontaneous abortion and abruptio placentae.   Best Pract Res Clin Obstet Gynaecol. 2011;25(3):313-327. doi:10.1016/j.bpobgyn.2011.02.006PubMedGoogle ScholarCrossref
28.
Valdiviezo  C, Garovic  VD, Ouyang  P.  Preeclampsia and hypertensive disease in pregnancy: their contributions to cardiovascular risk.   Clin Cardiol. 2012;35(3):160-165. doi:10.1002/clc.21965PubMedGoogle ScholarCrossref
29.
Carpenter  MW, Coustan  DR.  Criteria for screening tests for gestational diabetes.   Am J Obstet Gynecol. 1982;144(7):768-773. doi:10.1016/0002-9378(82)90349-0PubMedGoogle ScholarCrossref
30.
Carter  EB, Stuart  JJ, Farland  LV,  et al.  Pregnancy complications as markers for subsequent maternal cardiovascular disease: validation of a maternal recall questionnaire.   J Womens Health (Larchmt). 2015;24(9):702-712. doi:10.1089/jwh.2014.4953PubMedGoogle ScholarCrossref
×