Volume 120, Issue 8 p. 924-931
Original Article
Free Access

Risk of cardiovascular disease after pre-eclampsia and the effect of lifestyle interventions: a literature-based study

D Berks

Corresponding Author

D Berks

Division of Obstetrics and Prenatal Medicine, Department of Obstetrics and Gynaecology, Erasmus MC, Rotterdam, the Netherlands

Correspondence: Dr D Berks, Erasmus MC, Department of Obstetrics and Gynaecology, Division of Obstetric and Prenatal Medicine, Doctor Molewaterplein 50, 3015 GE, Rotterdam, the Netherlands. Email [email protected]Search for more papers by this author
M Hoedjes

M Hoedjes

Department of Public Health, Erasmus MC, Rotterdam, the Netherlands

Search for more papers by this author
H Raat

H Raat

Department of Public Health, Erasmus MC, Rotterdam, the Netherlands

Search for more papers by this author
JJ Duvekot

JJ Duvekot

Division of Obstetrics and Prenatal Medicine, Department of Obstetrics and Gynaecology, Erasmus MC, Rotterdam, the Netherlands

Search for more papers by this author
EAP Steegers

EAP Steegers

Division of Obstetrics and Prenatal Medicine, Department of Obstetrics and Gynaecology, Erasmus MC, Rotterdam, the Netherlands

Search for more papers by this author
JDF Habbema

JDF Habbema

Department of Public Health, Erasmus MC, Rotterdam, the Netherlands

Search for more papers by this author
First published: 26 March 2013
Citations: 96

Abstract

Objective

This study addresses the following questions. Do cardiovascular risk factors fully explain the odds ratio of cardiovascular risk after pre-eclampsia? What is the effect of lifestyle interventions (exercise, diet, and smoking cessation) after pre-eclampsia on the risk of cardiovascular disease?

Design

Literature-based study.

Setting

N/A.

Population or Sample

N/A.

Methods

Data for the calculations were taken from studies identified by PubMed searches. First, the differences in cardiovascular risk factors after pre-eclampsia compared with an uncomplicated pregnancy were estimated. Second, the effects of lifestyle interventions on cardiovascular risk were estimated. Validated risk prediction models were used to translate these results into cardiovascular risk.

Results

After correction for known cardiovascular risk factors, the odds ratios of pre-eclampsia for ischaemic heart disease and for stroke are 1.89 (IQR 1.76–1.98) and 1.55 (IQR 1.40–1.71), respectively. After pre-eclampsia, lifestyle interventions on exercise, dietary habits, and smoking cessation decrease cardiovascular risk, with an odds ratio of 0.91 (IQR 0.87–0.96).

Conclusions

Cardiovascular risk factors do not fully explain the risk of cardiovascular disease after pre-eclampsia. The gap between estimated and observed odds ratios may be explained by an additive risk of cardiovascular disease by pre-eclampsia. Furthermore, lifestyle interventions after pre-eclampsia seem to be effective in decreasing cardiovascular risk. Future research is needed to overcome the numerous assumptions we had to make in our calculations.

Introduction

Pre-eclampsia occurs as a complication in 2–8% of pregnancies.1 After pre-eclampsia, women have an increased risk of cardiovascular disease, including ischaemic heart disease and stroke.2 Moreover, as pre-eclampsia and cardiovascular disease share most risk factors, such as hypertension, obesity, diabetes, and hypercholesterolemia,3, 4 pre-eclampsia functions as a marker for cardiovascular risk.5 However, it is not yet known whether pre-eclampsia itself independently adds to cardiovascular risk.6, 7 If this is true, then pre-eclampsia would be an independent risk factor, and not just a marker for cardiovascular disease.

Furthermore, the burden of cardiovascular disease in women is high. In 2004 cardiovascular disease accounted for 57.0% of global female mortality.8 This emphasises the need for preventive lifestyle (or medical) interventions. The early detection of high-risk individuals maximises the effect of such interventions. After pre-eclampsia, women are thought to be good subjects for such interventions, as they are young and probably well motivated. Other authors have indeed suggested postpartum lifestyle interventions after pre-eclampsia to lower cardiovascular risk.1, 6 However, although previous research has shown that lifestyle interventions are effective if cardiovascular risk factors are already present,9 the effect remains uncertain in the absence of cardiovascular risk factors.10 Thus, as the majority of women after pre-eclampsia have no cardiovascular risk factors,11 and studies of lifestyle interventions after pre-eclampsia are lacking, the possible effects remain unknown; however, effects of lifestyle interventions after pre-eclampsia could be estimated.

This study addressed the following questions. Do cardiovascular risk factors fully explain the increased odds ratios of cardiovascular risk after pre-eclampsia? What is the effect of lifestyle interventions after pre-eclampsia on the risk of cardiovascular disease?

Methods

Estimating both the contribution of cardiovascular risk factors and pre-eclampsia itself on cardiovascular risk, and the effect of lifestyle interventions on cardiovascular risk, involves a number of steps. The estimation process is visualised in Figure 1.

Details are in the caption following the image
Steps in estimating the contribution of cardiovascular risk factors, additive cardiovascular risk of pre-eclampsia, and the effect of lifestyle interventions on future cardiovascular disease.

Cardiovascular risk after pre-eclampsia (steps 1a–1f)

In step 1a we searched for studies that reported on differences in cardiovascular risk factors between women with a history of pre-eclampsia and women with a history of an uncomplicated pregnancy. We focused on cardiovascular risk factors used in the prediction models in step 1c: systolic and diastolic blood pressure, cholesterol levels, weight, smoking, diabetes, familial cardiovascular disease, and the level of C–reactive protein. A PubMed search with the term ‘pre-eclampsia AND risk factors AND cardiovascular disease’ was used to identify original case–control studies that reported on these cardiovascular risk factors after pre-eclampsia (Figure 2). Pre-eclampsia was defined according to the International Society for the Study of Hypertension in Pregnancy (ISSHP) criteria: blood pressure of 140/90 mmHg or higher and proteinuria of at least 300 mg/day at a gestational age of at least 20 weeks. Studies with women who had chronic hypertension were allowed, but we excluded studies that included women with pre-existing renal or cardiovascular disease. As for this and all other searches (steps 1c and 2a), publications until January 2010 were included, and non-English and non-human studies were excluded.

Details are in the caption following the image
PubMed searches (all searches included studies until January 2010).

In step 1b we calculated the difference in cardiovascular risk factors between women with a history of pre-eclampsia and women with a history of a healthy pregnancy by subtraction. For each cardiovascular risk factor, the median value and interquartile range of the reported differences was calculated. This resulted in a collection of median differences of the cardiovascular risk factors, as shown in the first column of Table 1.

Table 1. Differences in cardiovascular risk factors in women 1–30 years after pre-eclampsia, compared with controls without pre-eclampsia, and estimated effects of lifestyle intervention programmes
Risk factor Differences of risk factors after pre-eclampsia compared with controlsa Median (IQR) Effect of lifestyle interventionsb Median (IQR)
Potentially modifiable risk factors
Systolic blood pressure (mmHg) 7.5 (6.3 to 10.5) −2.4 (−5.7 to −1.3)
Diastolic blood pressure (mmHg) 7.0 (5.0 to 9.0) −1.8 (−4.5 to −0.9)
Total cholesterol (mg/dl) 11.5 (5.0 to 18.0) −2 (−16 to −2)
HDL–cholesterol (mg/dl) 0.0 (−1.5 to 0.0) 0 (−3 to −1)
Body mass index (kg/m2) 2.8 (1.8 to 3.0) −1.0% (−3.6 to 0.0%)
Smoking (%) −10.9% (−18.0 to 5.0%) −2.2% (−7 to −0.7%)
Diabetes mellitus (%) 1.7% (0.0 to 3.0%)
Treatment for hypertension (%) 15.4% (12.3 to 22.6%)
Non-modifiable risk factors
Familial cardiovascular disease (<60 years; %) 22.2% (21.8 to 26.3%)
(hs-)C–reactive protein (mg/l) 0.5 (0.42 to 0.59)
  • a Based on the reported studies listed in Appendix S1.
  • b Paired analysis before versus after interventions, based on the reported studies listed in Appendix S2.
  • hs-, high sensitive.

In step 1c we searched for cardiovascular risk prediction models. A recent review by Cui et al.12 was used to identify validated cardiovascular risk prediction models. An additional search by hand provided the Framingham 30–years prediction model.13 We used the original articles that reported on the coefficients that were used in the prediction models. The following models were used: MONICA,14 PROCAM,15 SCORE,16 Eurostroke,17 CUORE,18 Framingham 30-years prediction model,13 and the Reynolds risk score.19

In step 1d we calculated the odds ratios of cardiovascular risk between women with a history of pre-eclampsia and women with a history of a healthy pregnancy, based on the differences in cardiovascular risk factors provided in step 1b. Here, we used the cardiovascular risk prediction models as supplied by step 1c. The interquartile range in step 1b was used to calculate a confidence interval of the odds ratios.

In step 1e the meta-analysis of Bellamy et al.2 was used to provide the observed risk of cardiovascular disease after pre-eclampsia. This study reported separately on ischaemic heart disease and stroke.

In step 1f we calculated the additive effect of pre-eclampsia on cardiovascular risk, corrected for the cardiovascular risk factors mentioned in step 1a. To obtain this effect the observed odds ratios (step 1e) were divided by the calculated odds ratios (step 1d), both for ischaemic heart disease and stroke.

Estimated effect of lifestyle interventions on cardiovascular risk after pre-eclampsia (steps 2a–2d)

In step 2a we searched PubMed for studies that reported on the effect of lifestyle interventions on cardiovascular risk factors. The recent review by our research group provided us with studies reporting on weight reduction and smoking cessation after pregnancy.20 For improving the lipid profile and lowering blood pressure, we used the identical PubMed search strategy as in the review: ‘(puerperium OR postpartum period OR postpartum OR post-partum) AND (lifestyle OR life style OR risk reduction behaviour OR risk reduction behavior OR health promotion OR <specific search term>)’. As a specific search term we used ‘lowering lipids OR dyslipidaemia OR hypercholesterolemia’ and ‘lowering blood pressure OR hypertension’, respectively. Non-human studies and non-English articles were excluded; however, neither search provided any usable studies (Figure 2). Therefore, we used the Cochrane Library to select studies reporting on the effects of primary interventions on lipid profile and blood pressure.21 We selected studies reporting effects in women aged 30–60 years. The studies were categorised according to cardiovascular risk factor (Appendix S2). Next, for each cardiovascular risk factor the median value of the reported effects was calculated. This resulted in a collection of median effects of lifestyle intervention on the cardiovascular risk factors, as shown in the second column of Table 1. For each risk factor, a first- and third-quartile value was calculated.

In step 2b we calculated the difference in cardiovascular risk factors between women with a history of pre-eclampsia, with or without lifestyle interventions. The same studies found in step 1a were used. The reported values of women with a history of pre-eclampsia were considered to be without the effect of lifestyle interventions. To compile a new fictive data sample with values from women who had a history of pre-eclampsia, and should have had lifestyle interventions, the collection of median effects of lifestyle interventions (step 2a) was used to recalculate the values of the cardiovascular risk factors, to give new values, as if these women had implemented the lifestyle interventions after pre-eclampsia.

In step 2c we calculated the difference in cardiovascular risk factors between women with a history of pre-eclampsia, with or without lifestyle interventions. For each cardiovascular risk factor the median value and interquartile range of the difference between the original and the fictive, recalculated values, as provided in step 2b, was calculated by subtraction. This resulted in a collection of median differences of the cardiovascular risk factors (values not shown).

In step 2d we used the risk prediction models found in step 1c to calculate the odds ratios of cardiovascular risk between women with a history of pre-eclampsia, with or without lifestyle interventions, based on the differences of cardiovascular risk factors, as provided in step 2c. The interquartile range in step 2c was used to calculate a confidence interval of the odds ratios.

Results

Cardiovascular risk after pre-eclampsia

The search in step 1a resulted in 878 articles. As shown in Figure 1, 13 articles were then included in the analysis. Screening the references of these articles resulted in an additional three studies. Appendix S1 lists all 16 studies included in the analysis, categorised by cardiovascular risk factor. For each cardiovascular risk factor, the median and interquartile range of the difference in risk between women who had and women who had not developed pre-eclampsia previously is shown in Table 1.

The calculated odds ratios based on these values are shown in Table 2. For ischaemic heart disease, a median odds ratio of 1.14 was calculated based on the differences of cardiovascular risk factors after a history of pre-eclampsia, compared with a healthy pregnancy. Dividing the observed odds ratio of 2.16 with this calculated odds ratio of 1.14, an odds ratio of 1.89 (IQR 1.76–1.98) remained as the additive effect of pre-eclampsia for risk of ischaemic heart disease, corrected for cardiovascular risk factors. For stroke, an odds ratio of 1.17 was calculated based on the difference of cardiovascular risk factors after a history of pre-eclampsia compared with a healthy pregnancy. Dividing the observed odds ratio of 1.81 with this calculated odds ratio of 1.17, an odds ratio of 1.55 (IQR 1.40–1.71) remained as the additive effect of pre-eclampsia for risk of stroke, corrected for cardiovascular risk factors.

Table 2. Calculated odds ratios (ORs) for cardiovascular disease (CVD)
Risk prediction model OR for CVD after pre-eclampsia (pre-eclampsia versus control) OR (Q1–Q3)a OR for CVD of hypothetical lifestyle interventions after pre-eclampsia (before versus after) OR (Q1–Q3)a
Ischaemic heart disease
MONICA14 1.23 (1.19–1.36) 0.92 (0.84–0.96)
PROCAM15 1.09 (0.98–1.23) 0.94 (0.88–0.97)
SCORE16 1.14 (1.02–1.30) 0.87 (0.81–0.96)
Stroke
EuroStroke17 1.17 (1.06–1.29) 0.96 (0.91–0.98)
Ischaemic heart disease and stroke
CUORE18 1.31 (1.16–1.57) 0.88 (0.85–0.99)
Reynolds risk score19 1.42 (1.25–1.71) 0.87 (0.80–0.98)
Framingham 30–year BMI13 1.27 (1.10–1.46) 0.95 (0.87–0.98)
Framingham 30–year lipids13 1.25 (1.10–1.49) 0.87 (0.83–0.98)
  • a Interquartile 50% confidence interval for the OR.

Estimated effect of lifestyle interventions on cardiovascular risk after pre-eclampsia

The included studies are listed in Appendix S2, categorised by cardiovascular risk factor. Table 1 shows the median and interquartile range of the effects for each cardiovascular risk factor.

Based on these values, we estimated the odds ratios for cardiovascular disease after lifestyle interventions in women who had developed pre-eclampsia to be between 0.87 and 0.96 (Table 2).

Discussion

Main findings

Our estimates showed that the cardiovascular risk factors used in the prediction models did not fully explain the risk of cardiovascular disease after pre-eclampsia. A major part of the observed odds ratios of cardiovascular disease after pre-eclampsia remained after adjustment for these cardiovascular risk factors.

According to our estimates, lifestyle interventions after pre-eclampsia will decrease cardiovascular risk by 4–13%. This might be an underestimation, as we had to use studies that did not account for the possible motivational effects of having had pre-eclampsia.21, 25, 26 On the other hand, the duration of follow-up of the studies was well below the time span of the prediction models. Thus, a possible rebound effect after cessation of the interventions would lower the effect on cardiovascular risk. This might have led to an over-estimation of the possible effect.

Interpretation

We hypothesise that pre-eclampsia itself is a true risk factor, rather than a marker, for cardiovascular disease. As transient endothelial dysfunction can last up to 2 years after pre-eclampsia,11 it is likely that pre-eclampsia has a permanent effect on cardiovascular health. This may partly be reflected by permanent effects on cardiovascular risk factors. Indeed, Romundstad et al. showed that the difference in cardiovascular risk factors for women after pre-eclampsia compared with women after a healthy pregnancy is 28–60% higher than the pre-pregnancy difference.22 However, pre-eclampsia may also directly increase cardiovascular risk through other, yet unknown, pathways.

Another explanation for this independent additive effect of pre-eclampsia on cardiovascular risk might be a higher prevalence of thrombophilic factors in women who have developed pre-eclampsia. Indeed, thrombophilic factors like protein–C and -S deficiencies increase the risk of early-life cardiovascular disease.23, 24 However, whether these thrombophilic factors are more prevalent in women who have previously developed pre-eclampsia remains unclear, but the effect is modest at best.1

The studies we included to calculate the differences in cardiovascular risk factors between women with a history of pre-eclampsia and women with a healthy pregnancy (Appendix S1) were very heterogeneous. Pre-eclampsia has different phenotypes. For instance, pre-eclampsia can be early or late, severe or mild, with or without HELLP (a syndrome characterized by haemolysis, elevated liver enzyme levels and a low platelet count) or intrauterine growth restriction (IUGR). The observed odds ratios used were also based on very heterogeneous studies.2 It is to be expected that each phenotype is differently associated with cardiovascular disease.

Limitations and strengths

To calculate our results we had to make several assumptions. The first assumption was that cardiovascular risk factors have the same effect on cardiovascular risk in women as they do in men, as the cardiovascular risk prediction models used were mainly based on male cohorts; however, it is suggested that cardiovascular disease in women might have a different pathogenesis than cardiovascular disease in men.27 The second assumption was that the effects of lifestyle interventions after pre-eclampsia are the same as the effects of lifestyle interventions in general, or after any pregnancy, because specific literature was lacking. The third assumption was that the effects of lifestyle interventions will last for the time span of the cardiovascular risk prediction models (10–30 years), as the studies we could use lasted for at most 6 years.

Our study has several limitations. First, the aforementioned assumptions weaken the results of our study.

Second, we only included studies that were published in PubMed-registered journals and that were written in English. This could have resulted in an overestimation in our calculations of the contribution of cardiovascular risk factors and effects of lifestyle interventions, as a result of publication bias.

Third, we were limited to the cardiovascular risk factors that had been used in the prediction models. Other cardiovascular risk factors, like ethnicity, apolipoproteins, and others, could not be included in our model. This could have led to an over-estimation of the independent effect of pre-eclampsia on cardiovascular risk.

Fourth, because of the heterogeneity of the studies included in Appendix S1 and S2,28-68 we could not perform meta-analyses for each risk factor. We therefore used less accurate medians and interquartile ranges in our calculations.

Conclusion

Cardiovascular risk factors do not fully explain the risk of cardiovascular disease after pre-eclampsia. The gap between estimated and observed odds ratios may be explained by an additive risk of cardiovascular disease after pre-eclampsia. Furthermore, lifestyle interventions after pre-eclampsia seem to be effective in decreasing cardiovascular risk.

Further studies should focus on the effects of lifestyle interventions after pre-eclampsia on cardiovascular risk and cardiovascular risk factors. Preferably, the follow-up would last for 10–30 years. This would eliminate the necessity to make the aforementioned assumptions in our model.

As we found an additive effect of pre-eclampsia on future cardiovascular disease, possible prevention of pre-eclampsia by preconception or early-pregnancy lifestyle modification is of even greater importance.

Disclosure of interests

All authors declare they had no support from any organisation for the submitted work, no financial relationships with any organisations that might have an interest in the submitted work in the previous 3 years, and no other relationships or activities that could appear to have influenced the submitted work.

Contribution to authorship

All authors participated in the planning of the study. DB and MH performed the PubMed searches and wrote the article. DB and JH contributed to the outline of the analysis. JD, HR and ES critically reviewed the final version.

Details of ethics approval

Not applicable.

Funding

None.

Acknowledgements

The authors want to thank Ed Hull and Charles Frink for their critical appraisal of the article.