Women who have given birth may have a significantly higher risk for cardiovascular disease (CVD) and stroke compared with women who have never had children, and the risk appears to increase with each live birth, new research suggests.
Investigators conducted a meta-analysis of 10 studies involving more than 150,000 cases of CVD among more than 3 million participants and found a 14% higher risk with parity vs nulliparity.
Moreover, when the association was analyzed with respect to the number of births, the investigators found a J-shaped relationship, in which each birth was associated with a 4% higher likelihood of developing CVD, regardless of factors such as body mass index (BMI), diabetes, hypertension, smoking, and income levels.
Each live birth was associated with a 5% higher risk for coronary heart disease (CHD) and a 4% higher risk for stroke.
“We think doctors could play a role in the dissemination [of these findings] for women patients,” study author Dongming Wang, MD, of the Department of Occupational and Environment Health, School of Public Health, Tongii Medical College, Huazhong University of Science and Technology, Hubei, China, told Medscape Medical News.
“The doctors could tell them that ever-parity is related with CVD risk and there is an association between number of pregnancies and risk for CVD,” he said.
The study was published online December 19 in the European Journal of Preventive Cardiology.
“Many studies have shown that reproductive factors may affect women’s health in later life, and parity as an important reproductive factor has also been shown to play an important role in the development of CVD,” Wang said.
The association, however, “remains controversial, given the inconsistency of previous findings. Thus, we aimed to conduct a meta-analysis of cohort studies to quantitatively assess the association between parity and CVD risk,” he reported.
To investigate the question, the researchers searched PubMed and Web of Science to identify studies on parity and CVD risk through June 7, 2018. They used the search terms “cardiovascular disease,” “coronary heart disease,” “ischemic heart disease,” and “stroke.”
Studies were required to have a cohort design, with parity number as the exposure of interest. CVD risk, rather than CVD mortality, was the required outcome.
Additionally, studies had to report relative risks (RRs) with 95% confidence intervals (CIs) for ≥3 quantitative categories of parity numbers, or to provide risk estimates per live birth.
To analyze dose and response, the researchers assigned the median or mean parity number in each category to the corresponding RR for each study.
Increased Cardiometabolic Changes
From a total of 4746 citations identified in the initial literature search, the researchers included 10 cohort studies (all published between 1987 and 2018) in the meta-analysis. Of these, nine studies (13 records) analyzed parous vs nulliparous women, and eight studies (11 records) included the dose-response analysis of parity number.
Studies were conducted in Sweden (n = 1), the United States (n = 4), China (n = 2), the United Kingdom (n = 2), and in multiple European countries (n = 1).
Study samples ranged from 867 to 1,332,062 women; the number of CVD cases ranged from 45 to 65,204; and the average duration of follow-up ranged from 6 years to 52 years.
All studies were considered to be of high quality.
For the nine studies that compared parous to nulliparous women (n = 3,014,987 participants and 148,169 CVD cases), the RR of CVD for ever-parity vs nulliparity was 1.14 (95% CI, 1.09 – 1.18), with moderate heterogeneity (I 2 = 62.0%; P = .002).
A sensitivity analysis that excluded one study at a time from the pooled estimate had “little impact” on the overall effect size.
For the eight studies that focused on the relationship between parity number and CVD risk (n = 3,051,107 participants and 144,163 CVD cases), the summary risk estimates for an increase of one live birth was 1.04 (95% CI, 1.02 – 1.05).
The researchers note that “significant heterogeneity” (I 2 = 89.6%) was found in these studies.
Additionally, they observed a J-shaped association in the nonlinear dose-response meta-analysis of parity number and CVD risk when using a cubic spline model.
Similar associations were found between parity number and CHD, ischemic heart disease, and stroke risk.
Egger’s and Begg’s tests showed no publication bias in any of the studies.
The researchers conducted subgroup analyses to “explore the potential source of statistical heterogeneity among the studies and assess the stability of the results.”
These analyses evaluated category of CVD, geographical location, whether there was adjustment for BMI, diabetes mellitus, hypertension, cigarette smoking, income, physical activity, and cholesterol.
They found that the associations of parity number with risk for CVD were similar in subgroup analyses.
“We found a significant association between parity and CVD risk when comparing parity with nulliparous,” Wang summarized.
“In the dose-response analysis, we observed a potential nonlinear J-shaped dose-response relationship between the number of births and CVD risk,” he added.
Wang described the biological mechanisms underlying these associations as “complex.”
He suggested that pregnancy “could lead to the accumulation of abdominal fat, endothelial dysfunction, atherosclerosis, and increased pro-atherogenic lipid levels and systemic inflammation.”
These cardiometabolic changes “may permanently impact the cardiovascular system, leading to a higher risk of CVD in parous women later in life,” he said.
Not Necessarily Applicable to Stroke
Commenting on the study for Medscape Medical News, Cheryl Bushnell, MD, MHS, professor of neurology, vice chair of research, and chief of the stroke division, Wake Forest Baptist Health, Winston Salem, North Carolina, who was not involved with the study, said the analysis contributes to the field through “the assessment of quality of design and outcome capture for each study included in the meta-analysis.”
Moreover, “since population studies on this topic are challenging, it is always useful to combine the results from studies from a variety of countries and geographic regions into one source and perform a meta-analysis,” she continued.
“The primary take-home message is that the results of studies assessing the relationship between parity and CVD is relatively consistent, showing a small but significant increase of CVD with increasing parity,” she said.
However, despite these findings, “there are several issues not addressed in the paper,” she cautioned.
“There is no mention of whether investigators studied the impact of the number of children on fathers, since some studies have suggested that this phenomenon may not be an isolated maternal problem and the same risk for CVD may apply to fathers as the number of children increases.”
She noted that some of the studies “do not include adjustment for socioeconomic status, which could be a significant bias.”
Finally, she added, “the premise was to evaluate CVD, although the majority of studies focused on CHD, and only one included stroke — therefore, the results may not be ideally extrapolated to include stroke.”
The authors state, “Further prospective large-scale studies are warranted to confirm our findings, and to establish causality and to elucidate the underlying mechanism.”