Adverse pregnancy outcomes and future maternal cardiovascular disease

1. INTRODUCTION

Cardiovascular disease (CVD), including coronary artery disease (CAD) and stroke, is the leading cause of death in women.1 Adverse pregnancy outcomes, including preeclampsia, preterm birth, gestational diabetes mellitus (GDM), and small‐for‐gestational‐age (SGA) infants have been associated with future maternal CVD risk. Normal pregnancy requires significant maternal cardiometabolic adaptation, with a 30% to 50% increase in cardiac output, alterations in lipid profile including an approximate 50% increase in total cholesterol, and a significant increase in insulin resistance during the second half of gestation to facilitate transfer of glucose to the fetus.2 Given these drastic cardiometabolic changes, it has been hypothesized that pregnancy may act as a “stress test” for future CVD. Women with adverse pregnancy outcomes have maladaptive responses to pregnancy. An increased mean arterial pressure and systemic vascular resistance, typically accompanied by hyperlipidemia and exaggerated insulin resistance, is seen in preeclampsia (Table 1), whereas a more extreme insulin resistance in the absence of an overt hypertensive syndrome occurs in GDM.2 Whether adverse pregnancy outcomes merely unmask a woman's preexisting CVD risk, or are actually mediators of future CVD pathogenesis, is unclear (Figure 1).

Cardiovascular disease is often under‐recognized in women compared to men secondary to differences in presentation between the sexes, which can result in delayed implementation of preventive strategies in women. Primary prevention, if applied to high‐risk populations early enough to avert cumulative damage of chronic disease, could reduce CVD incidence. In this context, a history of adverse pregnancy outcomes could identify women who would benefit from more targeted primary prevention efforts, particularly more intensive lifestyle interventions. Approximately 80% of women have at least 1 pregnancy during their lifetime, and approximately 30% of singleton pregnancies are complicated by an adverse pregnancy outcome. Therefore, these factors could be particularly informative of future CVD risk.2

A full description of all the types of cardiovascular disorders that may be related to pregnancy (e.g., including arrhythmias, spontaneous coronary artery dissection, peripartum cardiomyopathy) and the cardiovascular conditions that impact a women's health if she becomes pregnant (including congenital heart disease, aortopathies, or valvular disease), are beyond the scope of this review. We focus this review on the implications of adverse pregnancy outcomes of preeclampsia, GDM, birth of SGA infants, and preterm delivery (PTD) on future CVD risk in women.

2. PREECLAMPSIA

Preeclampsia, a heterogeneous syndrome characterized by new‐onset hypertension and proteinuria after 20 weeks, is thought to most often result from uteroplacental insufficiency (the failure of the placenta to deliver sufficient nutrients to the fetus, often the result of insufficient blood flow to the placenta), eventually leading to systemic endothelial dysfunction and impaired maternal hemodynamics.3 In the most severe forms, end‐organ dysfunction (such as eclampsia or acute renal or hepatic failure) develops, often necessitating immediate delivery, and leads to significant maternal and neonatal morbidity.4 The association of preeclampsia with later‐life CVD is well established and has been replicated in diverse populations across multiple studies.5, 6, 7 The largest meta‐analysis, including more than 6.4 million women with 39 years of follow‐up, demonstrated that women with a history of preeclampsia have a 71% increased risk of CVD mortality, a 2.5‐fold increase in risk of CAD, and a 4‐fold increase in heart failure risk when compared to women without a history of preeclampsia (Table 2).7

The biologic plausibility of this relationship is reinforced by the observation that the risk is amplified in women with recurrent preeclampsia, more severe, or early‐onset disease.6, 8 A Canadian study including over 1 million women with 25 years of follow‐up found that recurrent preeclampsia was associated with a significantly greater risk of CVD.8 Relative to women without preeclampsia, women with recurrent preeclampsia had 4 times the risk of atherosclerosis when compared to women with more than 2 deliveries, and twice the risk of atherosclerosis when compared to women with nonrecurrent preeclampsia. This relationship held across all cardiovascular outcomes.8

Recent data have also demonstrated further increased cardiovascular risk among women with preeclampsia and an additional complication, such as PTD or an SGA infant. A Norwegian study with a mean follow‐up of 17.2 years found that women with preeclampsia alone had a 2‐fold increased risk of a major coronary event, whereas women with preeclampsia plus PTD or SGA were 4 times as likely to have a major coronary event as women with uncomplicated pregnanices.9 Overall, women with a history of preeclampsia before 34 weeks or preeclampsia combined with PTD have a higher risk of death from CVD than women with preeclampsia alone, and 4 to 8 times the risk of women with an uncomplicated pregnancy.5, 10

Although many hypotheses have been generated to explain this association, our understanding of the underlying mechanisms is quite limited. Proposed etiologies for this increased risk include the presence of shared risk factors in preeclampsia and CVD and the possibility of common physiologic processes.11 A study of cardiovascular risk factors present both before and after pregnancy suggests that about half of the elevated risk for future hypertension after preeclampsia can be attributed to pre‐pregnancy risk factors.12 Furthermore, the pathologic vascular lesion of the placenta found in preeclampsia, termed acute atherosis, is similar to that observed in atherosclerosis, including lipid‐laden foam cells.13 This is supported by the clinical observation that preeclampsia is associated with the presence of coronary artery calcium.14 Finally, the distinct possibility that preeclampsia generates lasting endothelial dysfunction and causes the increased risk of CVD cannot be overlooked.15 Prior studies have shown that the endothelial dysfunction in women with preeclampsia persists into the postpartum period, suggesting a possible mechanism for the development of future CVD.16, 17 These mechanistic pathways are not mutually exclusive.

3. GESTATIONAL DIABETES

GDM affects between 6% and 7% of pregnancies in the United States, with the prevalence increasing secondary to rising obesity, maternal age, and maternal comorbidities. Normal pregnancy is accompanied by significant insulin resistance to ensure adequate transfer of glucose to the developing fetus.2 Women with GDM have metabolic dysregulation with exaggerated insulin resistance and an inability of their pancreatic function to match the physiologic insulin resistance of pregnancy.

GDM is well‐known to be associated with increased future maternal risk of type 2 diabetes mellitus (T2DM). A meta‐analysis of 675 455 women with up to 28 years of follow‐up showed that women with GDM have a more than 7‐fold increased risk of developing T2DM.18 Approximately one‐third of women in that study developed T2DM in the 3 to 5 years following delivery, and nearly 70% of women developed T2DM more than 10 years postpartum.

The association between GDM and CVD has been demonstrated in numerous studies (Table 2).19, 20, 21 Shah et al. showed that among 8191 women with and 81 262 women without GDM, GDM was associated with a 71% greater risk of future CVD events.19 This association was attenuated after adjusting for the subsequent development of T2DM. However, other studies have shown that GDM is associated with future CVD independent of future T2DM. A Canadian study with 10 years of follow‐up showed that women with GDM who did not subsequently develop T2DM still had 30% greater risk of CVD and 41% greater risk of CAD compared to women without GDM.20 Additionally, a prospective analysis of 2787 women in the CARDIA (Coronary Artery Risk Development in Young Adults) study demonstrated that a history of GDM was associated with increased carotid intima media thickness, a marker of early atherosclerosis, prior to the development of subsequent diabetes and metabolic syndrome, suggesting that a history of GDM alone, even without progression to T2DM, may influence atherosclerotic risk.22

GDM is also associated with a number of other cardiometabolic risk factors, including hypertension, hyperlipidemia, metabolic syndrome, and elevated markers of inflammation, including C‐reactive protein and interleukin‐6.23 Moreover, women with an abnormal glucose challenge test, but without GDM, have been shown to carry a higher risk for CVD compared to women without glucose intolerance, suggesting that even mild hyperglycemia portends increased CVD risk.24

The underlying mechanisms explaining the association between GDM and future CVD risk are thought to be similar to that of T2DM.23 Endothelial and vascular smooth muscle reactivity, as assessed by flow‐mediated dilation and arterial pulse wave velocity, has been found to be reduced in women with GDM at various points postpartum.23 Additionally, biomarkers of endothelial dysfunction, including intracellular adhesion molecule‐1, vascular cell adhesion molecule‐1, and E‐selectin, are elevated in the early postpartum period in women with GDM.25 These studies indicate a possible mechanism for the development of future cardiometabolic disorders in women with GDM; however, interpretation of these results is limited by small sample sizes and presence of confounding factors, and warrants further study.

4. SMALL FOR GESTATIONAL AGE

SGA newborns are those who are smaller in size than normal for the gestational age and fetal sex, most commonly defined as a weight below the 10th percentile for the gestational age at delivery. Not all newborns that are SGA fail to reach their optimal growth potential and may simply be constitutionally small. There are multiple etiologies for SGA associated with intrauterine growth restriction (IUGR), but the most frequent is related to underlying placental insufficiency and placental malperfusion. Women with impaired placental implantation are at increased risk of fetal growth abnormalities.2

Across various cohort studies, women with SGA infants have been found to be roughly twice as likely to experience future CVD (Table 2).26, 27 The strength of this association increases with increasing severity of SGA.26 Mortality is also increased in women with SGA infants. A study using data from the National Patient Registry in Denmark found a 2.5‐fold increase in death from cardiovascular causes and a 1.9‐fold increase in all‐cause mortality associated with birth of an SGA infant.28 Notably, a meta‐analysis of 6 studies demonstrated a 33% increase in maternal cardiovascular mortality associated with every approximately 500‐g decrease in infant birth weight.29

Although a history of SGA is known to be associated with increased CVD risk, mothers with a history of large infants may also be at risk. Bonamy et al. showed that infant birth weight 2 standard deviations above the mean is associated with 26% higher risk of maternal CVD.26 This observation may be related to the strong association between macrosomia and diabetes, as the risk is attenuated after adjusting for cardiometabolic risk factors.

The pathophysiology of this relationship between SGA and maternal CVD is poorly understood, but may relate to the possibility that women with CVD risk factors may be unable to compensate for the significant hemodynamic changes in pregnancy, placing them at greater risk for placental insufficiency and malperfusion.26 This is supported by the observation that dysregulation of maternal arterial distensibility is associated with low infant birth weight.30 It is also likely that environmental and nutritional factors may play a role.29 Additional research is necessary to further elucidate these mechanisms.

5. PRETERM DELIVERY

PTD, or delivery prior to 37 weeks gestational age, impacts approximately 10% of pregnancies. Spontaneous PTD occurs following preterm labor and can be caused by infection, or more commonly, is idiopathic. Elective or iatrogenic PTD occurs secondary to medical or obstetric indication, most commonly preeclampsia or SGA.2

PTD has been found across numerous studies to be associated with CVD (Table 2).26, 27, 28, 31, 32, 33, 34 A meta‐analysis of 10 studies from various European countries showed that PTD was associated with a 38% increased risk of ischemic heart disease, 71% increased risk of stroke, and 2‐fold increased risk of overall CVD.32 Because preeclampsia and IUGR are common reasons for iatrogenic PTD and are each on their own associated with CVD risk, this meta‐analysis only included studies that excluded cases of preeclampsia or IUGR, or adjusted for hypertension/preeclampsia and birthweight. An analysis of the NHSII (Nurses' Health Study II) found that PTD was independently associated with a 42% greater risk of CVD after adjusting for prepregnancy lifestyle and CVD risk factors.33 This effect remained significant after excluding pregnancies complicated by preeclampsia.

Similar to preeclampsia, the magnitude of the risk of CVD associated with prior PTD is further increased with the number of preterm deliveries and the gestational age at delivery. In the previously mentioned NHSII study, PTD at <32 weeks gestational age was associated with a 2‐fold greater risk of CVD, compared to a 1.2‐fold greater risk of CVD with PTD at ≥32 to <37 weeks gestational age, after adjusting for lifestyle and CVD risk factors.33 Additionally, a study of women in Denmark found that among multiparous women, recurrent PTD was associated with higher risk for CVD compared to 1 PTD (hazard ratio [HR] = 1.44 vs 1.24).31

The mechanisms underlying the association between spontaneous PTD and CVD are poorly understood; however, the relationship between PTD and the development of traditional CVD risk factors after delivery may play a role. In a longitudinal study of 1205 women in CARDIA, those with a history of PTD had an increased risk of future metabolic syndrome, independent of GDM, preeclampsia, and metabolic syndrome prior to pregnancy (HR = 1.47; 95% confidence interval: 1.19‐1.84).35 PTD has also been found to be associated with a greater change in systolic and diastolic blood pressure over 20 years.36 As with the other adverse pregnancy outcomes, whether PTD is the cause of future CVD or merely unmasks preexisting CVD risk that predispose to PTD is unclear.

6. GUIDELINES AND RECOMMENDATIONS

There is compelling evidence that adverse pregnancy outcomes serve as warning signs for future cardiovascular risk and can be useful in identifying high‐risk women for early intervention. As such, in 2011, the American Heart Association (AHA) updated its guidelines for the prevention of CVD in women to include preeclampsia, GDM, and pregnancy‐induced hypertension as major risk factors for CVD.19 These guidelines also advocate taking detailed histories of pregnancy complications, including GDM, preeclampsia, PTD, and birth of an SGA infant. There is little consensus on the most effective screening methods; however, questionnaires as outlined in Table 3 have been proposed.37

Other obstetric and cardiovascular organizations have also issued recommendations for risk reduction in women following an adverse pregnancy outcome (Table 4). Specifically, the American College of Obstetricians and Gynecologists recommends patient education for women with recurrent or preterm preeclampsia outlining the increased risk of future CVD, as well as implementation of a healthy lifestyle with weight loss, physical activity, and smoking cessation in addition to aspirin prophylaxis in future pregnancies.4 The AHA similarly recommends smoking cessation, Dietary Approaches to Stop Hypertension–like diet, regular physical activity, as well as weight management to maintain body mass index <25 kg/m2.38 In women with a history of GDM, the American Diabetes Association recommends screening for persistent diabetes 6 to 12 weeks postpartum and every 1 to 3 years, as well as encouraging healthy eating patterns and lifestyle interventions to support weight loss.39

These guidelines serve as a step forward toward increasing recognition and prevention of CVD risk in women with adverse pregnancy outcomes; however, many gaps still remain. As discussed in this review, a history of PTD and SGA infants predicts CVD; however, these complications were not included in the 2011 AHA guidelines. Additionally, these guidelines are not consistent across obstetrics and cardiology, with each of the groups focused on different patient populations and no recommendation on when screening should begin. A shared consensus statement across organizations would be helpful. Furthermore, prior studies have shown a low awareness of the benefits of screening by obstetricians and primary care providers, and additional efforts are needed to increase both patient and provider awareness.40

7. CONCLUSIONS AND FUTURE DIRECTIONS

Adverse pregnancy outcomes, including preeclampsia, GDM, PTD, and birth of an SGA infant, have been consistently shown to be associated with long‐term maternal CVD risk across multiple studies and diverse patient populations. These pregnancy complications likely share common pathophysiologic pathways and may be related to similar predisposing factors in women. Because pregnancy occurs early in a woman's life, and often before the onset of clinically evident CVD, it serves as a unique opportunity to assess a woman's later life CVD risk and initiate meaningful risk‐reduction strategies, particularly more intensive lifestyle changes. Recent updates to guidelines highlight the importance of including adverse pregnancy outcomes in CVD risk assessment and improving postpartum screening. Current CVD risk‐estimation tools, such as the American College of Cardiology/AHA Pooled Cohort Equations, may underestimate risk among these women with adverse pregnancy outcomes, and thus a careful pregnancy history assessment might be an important tool to guide the clinician‐patient risk discussion.41

Despite the progress of current research, many unanswered questions still remain. Whether pregnancy complications are markers that expose underlying subclinical CVD or borderline risk factors, or whether they cause vascular and metabolic damage that lead to future CVD, is unknown. The pathophysiologic mechanisms linking adverse pregnancy outcomes to CVD risk are not well understood. Even among women with clinically “uncomplicated” pregnancies, aberrant metabolic or cardiovascular responses to the pregnancy might identify additional women at elevated CVD risk.3 Identifying women with a history of adverse pregnancy outcomes who may be at elevated CVD risk is important; however, there is no consensus on the optimal timing of postpartum screening. Initiation of screening decades after pregnancy may be hampered by poor patient recall.42 Moreover, there is currently a lack of evidence and protocols for implementation of risk stratification strategies beyond simply screening for traditional risk factors (for example, sex‐specific risk score calculations, coronary artery calcium scoring, high‐sensitivity C‐reactive protein measurements). The development of CVD risk prediction tools that include pregnancy history alongside traditional CVD risk factors may improve screening for high‐risk patients. Lastly, there are limited data on the long‐term benefits of screening and risk‐reduction strategies. Further clinical studies on the efficacy of preventive and therapeutic interventions are needed to develop evidence‐based guidelines for women with adverse pregnancy outcomes.

Hauspurg A, Ying W, Hubel CA, Michos ED, Ouyang P. Adverse pregnancy outcomes and future maternal cardiovascular disease. Clin Cardiol. 2018;41:240–247. 10.1002/clc.22887