By Michael H. Crawford, MD, Editor

SYNOPSIS: An analysis of the Women’s Health Study based on a recent questionnaire about adverse pregnancy outcomes showed hypertensive disorders of pregnancy and low birth weight are independent predictors of subsequent atherosclerotic cardiovascular disease.

SOURCE: Søndergaard MM, Hlatky MA, Stefanick ML, et al. Association of adverse pregnancy outcomes with risk of atherosclerotic cardiovascular (CV) disease in postmenopausal women. JAMA Cardiol 2020;5:1390-1398.

Adverse pregnancy events have been associated with the development of atherosclerosis risk factors and atherosclerotic cardiovascular (CV) disease later in life. However, it is unclear whether they are risk factors themselves or are just associated with risk factors for CV disease. Søndergaard et al sought to clarify this issue by analyzing data from the Women’s Health Initiative (WHI), a large, multiethnic cohort study with prospective follow-up and adjudicated CV disease outcomes.

The WHI authors enrolled postmenopausal women between 1994 and 1998, but no data on pregnancy outcomes were obtained at that time. In 2017, researchers sent a questionnaire on pregnancy outcomes to all surviving enrollees who were still participating in WHI. The questionnaire asked about five conditions requiring a yes or no answer: gestational diabetes, hypertensive disorders of pregnancy, low birth weight (< 2.5 kg), high birth weight (> 4.1 kg), and preterm delivery by > 3 weeks. The primary outcome was atherosclerotic CV disease, defined as myocardial infarction, stroke, peripheral artery disease, or coronary revascularization, all of which were collected during the study period up to 2017 and were adjudicated. The outcomes data were adjusted for other known risk factors, including hyperlipidemia, hypertension, diabetes not pregnancy-related, and smoking. All participants who were pregnant longer than six months were included. Subjects were excluded if they reported any CV disease at baseline.

At baseline, there were 161,808 participants (72,063 were excluded, mainly for death, and 42,940 did not respond to the survey). This left 46,805 responders (65% of those eligible) with a median age of 60 years at enrollment. Adverse pregnancy outcomes (APO) were reported in 13,482. CV disease was more common in these women than those without APO (7.6% vs. 5.8%). Each of the five APOs analyzed separately was significantly associated with CV disease in the univariate model. However, when adjusted for other CV disease risk factors, only gestational diabetes, hypertension, low birth weight, and preterm delivery remained significant.

When all five APOs were analyzed together in an adjusted model, only hypertensive disorders of pregnancy (OR = 1.34; 95% CI, 1.15-1.54) and low birth weight (OR = 1.18; 95% CI, 1.03-1.35) remained independently associated with CV disease. Adjustments for race/ethnicity, income, education, body mass index, breastfeeding, and parity did not materially change the results. The authors concluded hypertensive disorders of pregnancy and low birth weight were independently associated with the development of CV disease after adjustment for atherosclerosis risk factors and other APOs in this multiethnic cohort of postmenopausal women.


Current guidelines recommend considering APO in the CV disease risk estimation in women and for decision-making about prescribing cholesterol-lowering agents for primary prevention.1 However, it is unclear whether APOs are risk factors in themselves or if they are associated with other risk factors. For example, gestational diabetes is known to be associated with the development of type 2 diabetes later in life, and gestational hypertension often is a precursor to later hypertension. In fact, some have suggested that pregnancy is a stress test that can bring out the predisposition to cardiometabolic abnormalities. Thus, this analysis of the WHI is of interest because Søndergaard et al parsed whether APOs are independent of other risk factors for CV disease.

The authors showed that all APOs, except high birth weight, are independently associated with CV disease when adjusted for other risk factors. Some APOs often occur together in the same women. When those APOs were analyzed together with traditional risk factors, the results showed that only hypertensive disorders of pregnancy and low birth weight remained independently associated with CV disease. Low birth weight, prematurity, and gestational hypertension may share a common mechanism of placental dysfunction. Prematurity and low birth weight often occurred together and were not independent of one another.

There were limitations to this study. Gestational hypertension and pre-eclampsia were lumped together as hypertensive disorders of pregnancy. This may not be appropriate, but since they often occurred together, this seemed expedient to the investigators. Although touted as multiethnic, 90% of the subjects were white, 5% Black, 2.5% Hispanic, and 2.5% other. This disparity could be explained by the fact minorities in WHI died more often and were less likely to respond to the survey. Also, there was a survival bias in that the highest-risk subjects died before the APO survey was conducted. There may be a recall bias, too, in that it is easier to recall low birth weight than gestational diabetes. However, these biases would only serve to reduce the odds ratios.

APOs are risk predictors for CV disease that are unique to women, but since about 85% of women experience at least one pregnancy, of which about 20% are complicated by an APO, these are important risk factors. This is especially true for hypertensive disorders of pregnancy, low birth weight, and gestational diabetes. It would be useful if at least these three APOs were included in electronic medical record templates.


  1. Piepoli MF, Hoes AW, Agewall S, et al. 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. Eur Heart J 2016;37:2315-2381.