Author + information
- Janet M. Catov, PhD, MSa,b,c,d,∗ (, )@PittHealthSci@UPMCnews,
- Malamo Countouris, MDe@malamo512 and
- Alisse Hauspurg, MDa,b
- aDepartment of Obstetrics, Gynecology, and Reproductive Sciences, University of Pittsburgh School of Medicine, Pittsburgh, Pennsylvania
- bMagee-Womens Research Institute, Pittsburgh, Pennsylvania
- cDepartment of Epidemiology, University of Pittsburgh Graduate School of Public Health, Pittsburgh, Pennsylvania
- dClinical and Translational Science Institute, Office of Research, Health Sciences, University of Pittsburgh, Pittsburgh, Pennsylvania
- eDepartment of Cardiology, University of Pittsburgh School of Medicine, Pittsburgh, Pennsylvania
- ↵∗Address for correspondence:
Dr. Janet M. Catov, Department of Obstetrics, Gynecology, and Reproductive Sciences, University of Pittsburgh School of Medicine, 204 Craft Avenue, Suite A208, Pittsburgh, Pennsylvania 15213.
In this issue of the Journal, Stuart et al. (1) provide the first empirical test of whether hypertensive disorders of pregnancy (HDP) can improve the prediction of cardiovascular disease (CVD) events in women. This question is a critical one, and this paper makes an important initial contribution toward an answer. The Nurses’ Health Study II is a cohort of predominantly white women with large numbers, prolonged follow-up, reasonable estimation of traditional cardiovascular risk factors, and maternal report of HDP. The results are mixed. The authors report, as expected and consistent with 2 decades of evidence, that HDP are sex-specific markers associated with CVD events, independent of established risk factors. Inclusion of HDP in CVD risk prediction models improved model fit for women 40 to 49 years of age but did not improve risk discrimination or risk reclassification.
There are a few possible explanations for these findings. First, as Stuart et al. (1) suggest, much of the link between HDP and CVD may be mediated by traditional risk factors. This scenario is perhaps hinted at by the stronger results detected in women 40 to 49 years of age rather than in older women. Current risk prediction approaches, including this report, are hampered by our inability to account for the accrual of risk across the reproductive years. Women <40 years of age were excluded in the present study because of the absence of evidence to guide prediction of long-term CVD events in these age groups. Pregnancy may uniquely reveal risk in young adult women, but this theory has never been directly tested. The importance of this point is amplified by the evidence that the accumulation of exposure to modest elevations in risk factors such as blood pressure across adulthood (e.g., systolic blood pressure >115 or 120 mm Hg) is linked to excess cardiovascular risk (2,3). In addition, blood pressure measured at ∼20 years of age contributes to CVD risk beyond blood pressure measured in middle age (4). Thus, it remains unknown if hypertension in pregnancy may mark women susceptible to a high-risk blood pressure trajectory and therefore act as a novel screen for hypertension treatment. The cardioprotective benefit of risk factor identification at younger ages in women guided by pregnancy history cannot be evaluated by CVD event prediction using our current modalities.
Another possible explanation for the nondiscrimination of CVD risk prediction according to HDP history is that maternal recall of pregnancy history may not distinguish the subtypes and severity that are the primary drivers of long-term CVD risk. This possibility cannot be tested in the current CVD cohorts, which rely on maternal recall, but prospective cohorts of younger women with medical record evidence of HDP (e.g., ALSPAC [Avon Longitudinal Study of Parents and Children]) (5) are approaching ages when CVD events occur. In addition, the Nurses’ Health Study II does not include many black women or other minorities who carry excess burden of pregnancy complications, cardiovascular risk factors such as obesity and hypertension, and CVD events. Established cohorts such as the CARDIA (Coronary Artery Risk Development in Young Adults) (6) study with >30 years of follow-up will soon have accrued events in women and thus will be available to examine race-specific pregnancy associations. Furthermore, contemporary cohorts of young adult women such as the NuMom2b-Heart Health study (7) that recruited women starting in 2010 during a first pregnancy and now monitoring them across the post-pregnancy years will no doubt provide new evidence on the accrual of risk during the reproductive years and how to leverage this history to predict CVD events. Importantly, intervention studies are also needed to directly examine how this risk profile may be leveraged to mitigate CVD risk in women.
Stuart et al. (1) make another important contribution by testing an approach that combines parity and HDP to predict CVD risk. Their inclusion of all women, nulliparous and parous, strengthens their findings. About 20% of women are nulliparous, and strategies to include these women and identify all the reproductive factors that may predict CVD are needed. The authors characterize 7 exposure groups, and it is possible that a simpler approach may be warranted. Although there is a J-shaped association between number of births and CVD, it is possible that this finding is due to socioeconomic confounding rather than biology (8). Thus, 3 groups (nulliparous, parous with normotensive pregnancies, and parous with a history of HDP) may be warranted. As the authors point out, there may also be other pregnancy features that can further refine risk stratification such as history of stillbirth, placental abruption, multiple miscarriages, infertility, gestational diabetes, and preterm delivery.
The major limitation in clinical management after HDP is the lack of evidence to guide risk-reducing interventions. To advance the field and develop successful risk prediction tools and interventions, multidisciplinary collaboration is key. In line with this goal, the American College of Obstetricians and Gynecologists and the American Heart Association recently released a Presidential Advisory emphasizing the critical importance of a partnership between obstetrician/gynecologists, cardiologists, and primary care physicians to optimize early identification and modification of risk factors for heart disease in women (9).
The current American College of Cardiology/American Heart Association (ACC/AHA) guidelines on management of hypertension and hyperlipidemia incorporate CVD risk in their recommendations (10). The updated 2017 ACC/AHA hypertension guidelines identified stage I hypertension as blood pressure >130/80 mm Hg and recommend initiating drug treatment for those with stage I hypertension and elevated 10-year CVD risk ≥10%. In addition, guidelines for statin initiation among women with hyperlipidemia hinge on 10-year CVD risk prediction, with recommendations of moderate- to high-dose statin initiation among women with 10-year CVD risk ≥7.5% (11). To improve cardiovascular health in women with HDP, there is a critical need for risk prediction tools and evidence-based interventions. Stuart et al. (1) have begun to fill this gap.
Although more research is needed, the importance of this first study (1) cannot be overstated. The lesson learned is not that pregnancy complications cannot enhance the prediction of CVD risk in women. Instead, we must follow up this research with larger, more diverse, more contemporary cohorts with more precise characterization of pregnancy health and disease. We should begin testing innovative risk stratification and intervention strategies that leverage this information to mitigate risk in the reproductive years. We must also develop new paradigms that incorporate information across the life course—from birth to old age—to characterize the predictors of risk and, ultimately, the strategies that improve health.
↵∗ Editorials published in the Journal of the American College of Cardiology reflect the views of the authors and do not necessarily represent the views of JACC or the American College of Cardiology.
The authors have reported that they have no relationships relevant to the contents of this paper to disclose.
- 2018 American College of Cardiology Foundation
- Stuart J.J.,
- Tanz L.J.,
- Cook N.R.,
- et al.
- Loria C.M.,
- Liu K.,
- Lewis C.E.,
- et al.
- Cutter G.,
- Burke G.,
- Dyer A.,
- et al.
- Brown H.L.,
- Warner J.J.,
- Gianos E.,
- et al.
- Whelton P.K.,
- Carey R.M.,
- Aronow W.S.,
- et al.
- Stone N.J.,
- Robinson J.G.,
- Lichtenstein A.H.,
- et al.