Women who have had preeclampsia or eclampsia are at increased risk for cardiovascular (cardiovascular) disease compared with pregnant women without preeclampsia or eclampsia.1 – 3 Studies have linked preeclampsia or eclampsia with later life myocardial infarction (MI),1,4 – 7 stroke,5,8 – 12 death,6,11,13 and coronary heart disease.14,15 Data are lacking on the long-term cardiovascular outcomes of women with preeclampsia or eclampsia who experienced a first MI or stroke during or after hospitalization for preeclampsia or eclampsia. The purpose of this study was to assess the subsequent occurrence of MI, stroke, or cardiovascular death after the first admission of MI and stroke in these women.
MATERIALS AND METHODS
Using a matched case-control design, women with preeclampsia or eclampsia diagnosed were identified using the Myocardial Infarction Data Acquisition System and were followed-up for cardiovascular outcomes and survival.
The objective was to assess the relationship between the medical conditions of preeclampsia or eclampsia, stroke, and MI by assessing cardiovascular outcomes. We compared four groups (Fig. 1). In analysis 1, we estimated the effect of preeclampsia or eclampsia by comparing women with preeclampsia or eclampsia (case 1) with a first MI and stroke diagnosis when the MI and stroke event occurred either on admission for preeclampsia or eclampsia or afterward and women without documented preeclampsia or eclampsia and a first MI and stroke (control 1). To estimate the effect of a first MI and stroke event accompanying preeclampsia or eclampsia, we conducted analysis 2, which further evaluated all women with preeclampsia or eclampsia by comparing women with preeclampsia or eclampsia with a first MI and stroke on admission for preeclampsia or eclampsia only (pregnancy-related), which represented the most severe case (case 2) and women with preeclampsia or eclampsia and no MI or stroke on preeclampsia or eclampsia admission (control 2). We compared the occurrence of the subsequent combined cardiovascular outcomes in these four groups defined as the first occurrence of hospital admission for MI or stroke or death attributable to a cardiovascular cause. The two case groups were matched to the respective control groups by age and year of discharge from the hospital. In each analysis, separate assessments were performed for MI and stroke. Women in all groups were followed-up for up to16 years in the Myocardial Infarction Data Acquisition System database.
Our data source was the Myocardial Infarction Data Acquisition System, which contains hospital records of all patients admitted to nonfederal hospitals in New Jersey with a cardiovascular disease diagnosis or procedure. The information from the Myocardial Infarction Data Acquisition System database including the diagnoses has been previously validated using a random sample of 726 medical charts.16,17 In addition, a random chart review performed for patients with stroke showed 89.5% specificity for diagnosis of stroke. The New Jersey State Institutional Review Board and the University of Medicine and Dentistry of New Jersey Institutional Review Board approved the study with waiver of consent.
The selection of case and control group participants included a total of 6,628 women (13–54 years, mean±standard deviation 31.2±6.4) who were admitted for the first time with a primary diagnosis (reason for admission) or secondary diagnosis of preeclampsia or eclampsia (International Classification of Diseases, 9th Revision, Clinical Modification [ICD-9-CM] codes 642.3–642.7) from years 1994 to 2009. A description of the case and control group participants used in both analyses is provided in Figure 1. An ICD-9-CM code of 410 was used to identify a diagnosis of MI, whereas ICD-9-CM codes 430, 431, 433.X1, 434.X1, 437.4, and 437.6 were used to identify stroke. Up to three control group participants were selected from the eligible pool by matching to the case group participant's age and year of discharge. If a control group participant of the same age or discharge year as the case group participant was not available, then a control group participant ±2 years in age or discharge year was selected. In MI and stroke analyses, case group participants with MI and stroke diagnoses and history of malignant neoplasm before the preeclampsia or eclampsia admission were excluded.
Patient characteristics included age, race, primary insurance, and source of admission, which were available in the Myocardial Infarction Data Acquisition System database. We examined the presence of the comorbid conditions recorded either at or before the admission for the first or index event using ICD-9-CM codes for diabetes mellitus, renal disease, cardiac dysrhythmia, anemia, obesity, hyperlipidemia, systemic lupus erythematosus, and hypercoagulable state. We also examined the presence of chronic hypertension recorded before the preeclampsia or eclampsia admission using ICD-9-CM codes.
The ascertainment of death was obtained from death information (date and cause of death) by matching the Myocardial Infarction Data Acquisition System records to New Jersey death registration files using a public automated and previously validated record linkage and consolidation software (The Link King).18 Deaths were classified into two categories: cardiovascular death and noncardiovascular death.
In both analyses, time to the combined cardiovascular outcome (MI, stroke, cardiovascular death) was calculated as the number of days from the index event date to the admission date of the cardiovascular outcome (MI or stroke) or the date of death, whichever occurred first. If a case group participant or control group participant did not experience any of the cardiovascular outcomes or death, then she was followed-up from the index event to December 31, 2009. In addition to the combined cardiovascular outcome, time to the individual components of the combined cardiovascular outcome (admission for MI, admission for stroke, cardiovascular death) was separately calculated as the number of days from the index event to the occurrence of the cardiovascular outcome. When the outcome of interest was MI or stroke, women who died of any cause before experiencing that outcome were censored on the date of their death. When the outcome of interest was cardiovascular death, patients who died of noncardiovascular-related causes were censored on the date of their death.
Patient characteristics and the frequencies of each outcome were summarized by case or control status in each analysis group. Event rates were calculated using Cox proportional hazards regression models. Models were adjusted for race, primary insurance, obesity, and history of diabetes mellitus in analysis 1, and for race, primary insurance, and obesity in analysis 2. The reason for adjustments was to avoid bias in the comparison of case and control group participants because their baseline characteristics differed between the groups. The Cox models also were adjusted for the matching by stratifying by age and discharge year. In each Cox model, event rates were compared between case and control group participants through a hazard ratio (HR) with 95% confidence intervals (CIs), and the Kaplan-Meier survival product limit estimates for the “combined cardiovascular outcome” were plotted over time separately for case and control group participants. In both analysis 1 and analysis 2, a sensitivity analysis was performed by replacing cardiovascular deaths with all-cause deaths to explore whether this change had an effect on the outcome. We conducted two additional analyses: a subgroup analysis to explore the potential difference in event rates between younger and older women, using the median age as the cut-off, and an analysis of the risk of admission for heart failure separately and as part of the combined cardiovascular outcome. The results of the latter sensitivity analyses are presented in the Appendix (available online at http://links.lww.com/AOG/A318). All statistical analyses were performed using SAS/STAT software 9.3.
Analysis 1 included comparison of case group participants with preeclampsia or eclampsia and anytime occurrence of a first MI and stroke with control group participants with a first MI and stroke and without a documented history of preeclampsia or eclampsia. Patient characteristics are presented in Table 1. A higher percentage of control group participants in both the MI and stroke study groups were admitted from an emergency department compared with case group participants. In the MI study group, as well as in the stroke group, a higher percentage of control group participants had a history of diabetes mellitus, renal disease, obesity, and anemia compared with the case group participants. The majority of women had commercial insurance.
Results of the frequencies and adjusted risk of the cardiovascular outcomes among case and control group participants for the MI and stroke study groups are presented in Table 2. In the MI study group, none of the outcomes were significantly different between case and control group participants after adjusting for covariates. Figure 2A shows that there were no significant differences in the combined cardiovascular outcome-free survival among case and control group participants (P=.23). In the stroke study group, control group participants were at significantly higher risk for the combined cardiovascular outcome with all-cause and cardiovascular deaths compared with case group participants (for all-cause deaths: absolute risk in case group participants=0.167, absolute risk in control group participants=0.338, absolute risk difference=0.171, number needed to treat=5.8; for cardiovascular deaths: absolute risk in case group participants=0.106, absolute risk in control group participants=0.235, absolute risk difference=0.129, number needed to treat=7.7). Figure 2B shows that control group participants had a significantly lower combined cardiovascular outcome-free survival rate compared with case group participants (P=.04).
Results of the analysis by age groups indicated that when the analyses for the MI study group were performed separately for younger (39 years or younger) and older (older than 39 years) women, MI control group participants from the younger age group were at significantly higher risk for all-cause deaths (HR 5.91, 95% CI 1.31–26.67) and the combined cardiovascular outcome with all-cause deaths (HR 4.26, 95% CI 1.23–14.8) than case group participants. No significant differences in the outcomes were observed between case group participants and control group participants older than age 39 years. In the stroke study group, control group participants were at significantly higher risk for all-cause death (HR 2.36, 95% CI 1.01–5.52) and for the combined cardiovascular outcome with all-cause deaths (HR 2.67, 95% CI 1.23–5.82) than case group participants in the younger age group (33 years or younger), whereas older control group participants (older than 33 years) were at significantly higher risk for cardiovascular deaths (HR 8.02, 95% CI 1.02–63.01) and the combined cardiovascular outcome with cardiovascular deaths (HR 2.78, 95% CI 1.15–6.74) compared with case group participants.
In analysis 2, comparison of case group participants with preeclampsia or eclampsia and a first (pregnancy-related) MI and stroke on the same admission with control group participants with preeclampsia or eclampsia but without a pregnancy-related MI and stroke on the same admission was performed. Results of patient characteristics indicated a higher percentage of case group participants in the MI and stroke study groups were admitted from an emergency department compared with control group participants (Table 1). A higher percentage of case group participants had a history of anemia in the MI study group compared with control group participants, whereas a higher percentage of control group participants had a history of cardiac dysrhythmia compared with case group participants in the stroke study group.
Results of cardiovascular outcomes indicated that for the MI group, there were no stroke admissions or deaths in both case group participants and control group participants. None of the outcomes was significantly different between the case group participants and the control group participants after adjusting for covariates (Table 2). Figure 2C shows that there were no significant differences in the combined cardiovascular outcome-free survival among case group participants and control group participants. In the stroke study group, case group participants were at significantly higher risk for all-cause death and the combined cardiovascular outcome with all-cause deaths compared with control group participants (for all-cause deaths: absolute risk in case group participants=0.111, absolute risk in control group participants=0.019, absolute risk difference=0.092, number needed to treat=10.9; for the combined cardiovascular outcome with all cause deaths: absolute risk in case group participants=0.111, absolute risk in control group participants=0.027, absolute risk difference=0.084, number needed to treat=11.9). However, case group participants were not significantly different from control group participants with respect to the combined cardiovascular outcome with cardiovascular deaths (Fig. 2D).
Results of the analysis by age groups showed that none of the outcomes was significantly different between the case group participants and the control group participants for either the younger (34 years or younger) or the older (older than 34 years) MI study group. In the younger stroke study group (31 years or younger), case group participants were at significantly higher risk for all-cause death (HR 10.5, 95% CI 2.1–53.0), the combined cardiovascular outcome with all-cause deaths (HR 6.82, 95% CI 1.67–27.83), and the combined cardiovascular outcome with cardiovascular deaths (HR 14.04, 95% CI 1.53–129.34) compared with control group participants. No significant differences in the outcomes were observed between case group participants and control group participants in the older age group (older than 31 years).
A history of preeclampsia is a risk factor for coronary artery disease, MI, and cardiovascular death.6,19 The link between a higher risk of cardiovascular disease in later life in women with preeclampsia (PE) was suggested in a previous study20 that reported an increased prevalence of previous preeclampsia or eclampsia in women who had later experienced an MI.20 Our study indicates that preeclampsia or eclampsia not complicated by MI or stroke during that pregnancy may not confer a very high risk for subsequent MI and stroke in up to 16 years of follow-up. Regarding stroke risk, our data suggest that other known risk factors for stroke put women at greater risk for stroke than preeclampsia or eclampsia complicated by a stroke.
Studies examining long-term effects of preeclampsia or eclampsia suggest more adverse cardiovascular outcomes5,8 – 10,12,14 that our study did not address. Taiwanese cohort studies of women using an index date 90 days before delivery8 resulted in a significantly higher risk of stroke during pregnancy and in the first postpartum year,8 as well as a significantly higher risk of major adverse cardiovascular events (MI and stroke) during pregnancy with continuing significant risk to more than 36 months postpartum.5 However, these studies did not compare women without preeclampsia or eclampsia who had stroke.
The risks of cardiovascular sequelae in women with preeclampsia or eclampsia cannot be dismissed. The relationship and etiology of PE and cardiovascular disease are complex and characterized by several markers, including endothelial dysfunction,12,21 – 24 metabolic abnormalities, oxidative stress, enhanced inflammatory response, hypercoagulability,25 arterial stiffness,23 and common molecular biomarkers.26 The Stroke Prevention in Young Women study showed that women who had a history of PE were 60% more likely to have a remote nonpregnancy-related ischemic stroke than those without a history of PE.9 Pregnancy-related acute MI may convey significant maternal morbidity and mortality in women.7 Our study does not include women without preeclampsia or eclampsia and without MI for comparison. In our study, women with preeclampsia or eclampsia with an index stroke were at significantly higher risk for all-cause death and the combined cardiovascular outcome (with all-cause death) compared with women with preeclampsia or eclampsia without index stroke.
Limitations in our study include a lack of patient-level data on typical cardiac-related risk factors (eg, high cholesterol, medications, smoking, blood pressure, and blood transfusions),27 and there was no information pertaining to the trimester in which the women in our study with preeclampsia or eclampsia had diagnoses and delivered. Early PE (less than 34 weeks) appears to be more related to the development of a markedly altered cardiovascular response (probably initiated by a placental disorder factor), which has implications for cardiovascular disease sequelae.25,28 – 31
Our study design cannot generate incidence data, and data were collected retrospectively.32 Case-control studies may prove an association, but they do not demonstrate causation.32 We did not have data on the control group participants (control 1) relating to their pregnancy history.2 Another methodologic limitation with this study is that we used only ICD-9 codes. Because some of our preeclampsia or eclampsia case group participants may have had preexisting hypertension, we included only the diagnosis of preeclampsia or eclampsia in the study.
The strengths of the study include the size of the population studied and the 16-year time interval on data from patients admitted to all nonfederal hospitals in New Jersey. Our results confirm the supposition that women with preeclampsia or eclampsia but without index stroke have a lower risk of all-cause deaths than women with preeclampsia or eclampsia with an index stroke. The rates of subsequent MI and stroke events were much lower in our case group participants and control group participants from analysis 1 than those observed for the general population in the state (for MI: rate in case group participants 5.3% and rate in control group participants 9.7%, compared with general population 17.7%; for stroke: rate in case group participants 3.8% and rate in control group participants 11.6%, compared with rate in general population=11.9%). This may be attributable to the older age of patients in the general population experiencing these conditions (median age for MI 69 years, median age for stroke 75 years).
A follow-up study is needed to compare control 2 group women with matched groups of normotensive pregnant women in the general population whose information is not presently in the Myocardial Infarction Data Acquisition System database. To better understand the relationship between cardiovascular events during pregnancy and later life cardiovascular sequelae, clinicians should consider including a more detailed preeclampsia or eclampsia history in all previously pregnant women.29
In conclusion, one of the most important findings of our study is that women with preeclampsia or eclampsia but without an index stroke have lower risk of cardiovascular outcomes than women with preeclampsia or eclampsia with an index stroke. Women with preeclampsia or eclampsia and an index MI and stroke fare better than those with an MI and stroke without preeclampsia or eclampsia. This finding may be attributable to increased comorbidities in the control group. These data lend support to the designation of preeclampsia as a risk factor for cardiovascular disease in women by the American Heart Association Effectiveness-Based Guidelines for the Prevention of Cardiovascular Disease in Women—2011 Update.29
1. Smith GC, Pell JP, Walsh D. Pregnancy complications and maternal risk of ischaemic heart disease: a retrospective cohort study of 129,290 births. Lancet 2001;357:2002–6.
2. McDonald SD, Malinowski A, Zhou Q, Yusuf S, Devereaux PJ. Cardiovascular sequelae of preeclampsia/eclampsia: a systematic review and meta-analyses. Am Heart J 2008;156:918–30.
3. Craici I, Wagner S, Garovic VD. Preeclampsia and future cardiovascular risk: formal risk factor or failed stress test? Ther Adv Cardiovasc Dis 2008;2:249–59.
4. Hannaford P, Ferry S, Hirsch S. Cardiovascular sequelae of toxaemia of pregnancy. Heart 1997;77:154–8.
5. Lin YS, Tang CH, Yang CY, Wu LS, Hung ST, Hwa HL, et al.. Effect of pre-eclampsia–eclampsia on major cardiovascular events among peripartum women in Taiwan. J Cardiol 2011;107:325–30.
6. Arnidottir GA, Geirsson RT, Arngrimsson R, Jonsdottir LS, Olafsson Ö. Cardiovascular death in women who had hypertension in pregnancy: a case–control study. BJOG 2005;112:286–92.
7. Ladner HE, Danielsen B, Gilbert WM. Acute myocardial infarction in pregnancy and the puerperium: A population-based study. Obstet Gynecol 2005;105:480–4.
8. Tang CH, Wu CS, Lee TH, Hung ST, Yuan C, Yang C, et al.. Preeclampsia-eclampsia and the risk of stroke among peripartum in Taiwan. Stroke 2009;40:1162–8.
9. Brown DW, Dueker N, Jamieson DJ, Cole JW, Wozniak MA, Stern BJ, et al.. Preeclampsia and the risk of ischemic stroke among young women results from the Stroke Prevention in Young Women Study. Stroke 2006;37:1055–9.
10. Wilson BJ, Watson MS, Prescott GJ, Sunderland S, Campbell DM, Hannaford P, et al.. Hypertensive diseases of pregnancy and risk of hypertension and stroke in later life: results from cohort study. BMJ 2003;326:845–9.
11. Irgens HU, Reisæter L, Irgens LM, Lie RT. Long term mortality of mothers and fathers after pre-eclampsia: population based cohort study. BMJ 2001;323:1213–7.
12. Ray JG, Vermeulen MJ, Schull MJ, Redelmeier DA. Cardiovascular health after maternal placental syndromes (CHAMPS): population-based retrospective cohort study. Lancet 2005;366:1797–803.
13. Funai EF, Friedlander Y, Paltiel O, Tiram E, Xue X, Deutsch L, et al.. Long-term mortality after preeclampsia. Epidemiology 2005;16:206–15.
14. Haukkamaa L, Salminen M, Laivuori H, Leinonen H, Hiilesmaa V, Kaaja R. Risk for subsequent coronary artery disease after preeclampsia. Am J Cardiol 2004;93:805–8.
15. Ness RB, Hubel CA. Risk for coronary artery disease and morbid preeclampsia: a commentary. Ann Epidemiol 2005;15:726–33.
16. Kostis JB, Wilson AC, O'Dowd K, Gregory P, Chelton S, Cosgrove NM, et al.. Sex differences in the management and long-term outcome of acute myocardial infarction. A statewide study. MIDAS Study Group. Myocardial Infarction Data Acquisition System. Circulation 1994;90:1715–30.
17. Shao YH, Croitor SK, Moreyra AE, Wilson AC, Kostis WJ, Cosgrove NM, et al.. Comparison of hospital versus out of hospital coronary death rates in women and men. Am J Cardiol 2010;106:26–30.
18. Campbell KM. Rule your data with the Link King (a SAS/AF application for record linkage and unduplication). SUGI 30 Proceedings. Philadelphia (PA): SUGI 30; 2005.
19. James AH, Jamison MG, Biswas MS, Brancazio LR, Swamy GK, Myers ER. Acute myocardial infarction in pregnancy: a United States population-based study. Circulation 2006;113:1564–71.
20. Chesley SC, Annitto JE, Cosgrove RA. The remote prognosis of eclamptic women. Sixth periodic report. Am J Obstet Gynecol 1976;124:446–59.
21. Poh CL, Lee CH. Acute myocardial infarction in pregnant women. Ann Acad Med Singapore 2010;39:247–53.
22. Redman CW, Sargent IL. Latest advances in understanding preeclampsia. 2005;308:1592–4.
23. Hausvater A, Giannone T, Sandoval YH, Gomez YH, Doonan RJ, Antonopoulos CN, et al.. The association between preeclampsia and arterial stiffness. J Hypertens 2012;30:17–33.
24. Heitzer T, Schlinzig T, Krohn K, Meinertz T, Münzel T. Endothelial dysfunction, oxidative stress, and risk of cardiovascular events in patients with coronary artery disease. Circulation 2001;104:2673–8.
25. Mongraw-Chaffin ML, Cirillo PM, Cohn BA. Preeclampsia and cardiovascular disease death prospective evidence from the Child Health and Development Studies Cohort. Hypertension 2010;56:166–71.
26. Gastrich MD, Cohen D, Pitchika A. Common markers of preeclampsia and CVD sequelae. Clin Lab Int V 2010;34:8–12.
27. Mann JI, Doll R, Thorogood M, Vessey MP, Waters WE. Risk factors for myocardial infarction in young women. Br J Prev Soc Med 1976;30:94–100.
28. Valensise H, Vasapollo B, Gagliardi G, Novelli GP. Early and late preeclampsia: two different maternal hemodynamic states in the latent phase of the disease. Hypertension 2008;52:873.
29. Mosca L, Benjamin EJ, Berra K, Bezanson JL, Dolor RJ, Lloyd-Jones DM, et al.. Effectiveness-based guidelines for the prevention of cardiovascular disease in women—2011 update: a guideline from the American Heart Association. Circulation 2011;123:1243–62.
30. Bellamy L, Casas JP, Hingorani AD, Williams DJ. Pre-eclampsia and risk of cardiovascular disease and cancer in later life: systematic review and meta-analysis. BMJ 2007;335:974.
31. Banerjee M, Cruickshank JK. Pregnancy as the prodrome to vascular dysfunction and cardiovascular risk. Nat Clin Pract Cardiovasc Med 2006;3:596–603.
32. Lewallen CS, Courtright P. Epidemiology in practice: case-control studies. Comm Eye Health 1998;11:57–8.
Supplemental Digital Content
© 2012 by The American College of Obstetricians and Gynecologists. Published by Wolters Kluwer Health, Inc. All rights reserved.