Objective
We sought to investigate whether previously identified risk factors are associated with eclampsia in a contemporary, heterogeneous cohort of women.
Study Design
Data were collected from birth certificate and hospital discharge records and used to conduct a population-based case-control study among women giving birth to singletons in Washington State from 1987 through 2007. We used multivariable logistic regression to estimate odds ratios and 95% confidence intervals. Multiple imputation procedures were used to address missing data.
Results
Risk of eclampsia was greater in nulliparous compared to parous women. Being a young mother (<20 years) or an older mother (≥35 years) were each associated with elevated eclampsia risk. Longer birth interval, low socioeconomic status, gestational diabetes, prepregnancy obesity, and weight gain during pregnancy above or below recommended guidelines were positively associated with eclampsia. Multiparity and smoking were inversely associated with eclampsia risk.
Conclusion
Exposures identified more than a decade ago continue to be associated with eclampsia in contemporary birth cohorts.
Preeclampsia, defined as the onset of hypertension and the presence of protein in the urine at >20 weeks of gestation in a previously normotensive woman, is a pregnancy complication currently screened for during routine prenatal care. Preeclampsia occurs in 5-8% of pregnancies in developed countries. Eclampsia is one of several severe complications of preeclampsia and includes the above symptoms plus generalized seizures and/or coma in the absence of other neurologic conditions. Estimates for the occurence of eclampsia vary widely, from 1 in 100 pregnancies to approximately 1 in 4000 pregnancies. The etiology of preeclampsia is still not well understood, and limited research is available describing factors that predict which preeclamptic women will eventually develop eclampsia.
Eclampsia can result in maternal morbidity, including fairly severe morbidities such as HELLP syndrome (hemolytic anemia, elevated liver enzymes, and low platelet count), acute kidney injury, coma, pulmonary edema, and disseminated intravascular coagulation. Although these conditions commonly resolve following delivery, eclampsia can also result in ischemic or hemorrhagic stroke in the mother, which can lead to permanent neurologic sequelae or death. Eclampsia also increases the need for a cesarean section and can be an important source of neonatal morbidity and mortality.
Risk factors for eclampsia that have been identified in previous studies include: both young and old maternal age, obesity prior to pregnancy, being unmarried, excessive weight gain during pregnancy, multiple gestations, nulliparity, chronic hypertension, low socioeconomic status, prolonged birth interval, and lack of prenatal care. Studies have also observed a lower risk of eclampsia among current smokers. Few population-based studies designed specifically to identify risk factors for eclampsia in contemporary cohorts have been published, with all but 1 study based on data from the early 1990s or earlier, dates which were characterized by different medical care, conception patterns, and population characteristics (eg, obesity) than the current decade. Additionally, with the exception of 1 study that was conducted in the United States with data from 1984 through 1990, studies have been conducted in fairly homogeneous populations of women, limiting the generalizability of study observations and the ability to investigate variations in maternal characteristics such as education, income, and ethnicity. The aim of the current study was to update and extend our understanding of the associations between social, demographic, and health care–associated characteristics and eclampsia using a contemporary, heterogeneous sample of mothers from Washington State.
Materials and Methods
Study setting and population
We conducted a population-based case-control study among all women giving birth to singletons in Washington State from 1987 through 2007. Washington State collects information for each birth occurring in the state through birth certificates, including: data on demographics for mothers, fathers, and newborns; prenatal care; pregnancy complications; newborn health conditions; and maternal and fetal outcomes. The Birth Events Reporting System links Washington State birth certificate data to the Comprehensive Hospital Abstract Reporting System database, a collection of discharge records for the birth hospitalization that includes information on any procedures or diagnoses associated with the birth in both mother and child. The methods used for linkage have been reported previously. University of Washington institutional review board exemption status was obtained for use of these data.
Women were eligible for this study if they gave birth to a singleton in Washington State from 1987 through 2007. Cases (n = 781) were women who had a discharge diagnosis of eclampsia listed in the delivery hospitalization record. The discharge diagnosis of eclampsia was classified using the International Classification of Diseases, Ninth Revision code 642.6. We then selected as controls a random sample of potentially eligible women without a diagnosis of eclampsia. Controls were frequency matched (4:1) to cases on year of delivery, resulting in a total of 3124 controls. Only the first pregnancy for each case and control was included in the study to avoid the potential bias in SE measurement and inference resulting from correlated data.
Exposure definitions
The following covariates, taken directly from the Washington State birth certificate for each case and control pregnancy in the study, were investigated as risk factors for eclampsia: maternal age in years (<20, 20-34, ≥35), race/ethnicity (Caucasian, African American, Hispanic), prepregnancy body mass index (BMI) in kg/m 2 (<18.5, 18.5-24.9, 25-29.9, ≥30), parity (nulliparous, primiparous, multiparous), presence of gestational diabetes, existence of established diabetes or renal disease, presence of prepregnancy hypertension, mother’s educational level in years of schooling (<9, 9-12, >12), tertiles of maternal annual household income based on the entire cohort (<$32,669, $32,669-44,485, >$44,485), smoking and alcohol use during pregnancy (any vs none), prenatal care (began during first trimester, began after first trimester or not at all), birth interval in years among parous women (<2, 2-5, >5), and weight gain during pregnancy in pounds (<25, 25-35, >35).
Socioeconomic status refers to reported annual household income and level of maternal education. Prepregnancy obesity is defined as BMI ≥30, and inadequate prenatal care refers to prenatal care that either began after the first trimester or never.
Statistical analyses
Odds ratio (OR) estimates and 95% confidence intervals (CIs) for the associations of interest were calculated using logistic regression. Maternal age and parity were considered a priori to be important risk factors based on prior consistent findings in the literature and were therefore included in multivariable models. Three additional confounders were considered in adjusted multivariable models: 2 markers of socioeconomic status (household income and maternal education) and BMI; each factor has been noted in prior studies to be associated with eclampsia, and each was significantly associated (alpha level = 0.05) with multiple other potential risk factors in our study population.
To examine whether associations between characteristics and eclampsia differed according to a mother’s age at birth or parity, multiplicative interaction terms were included in multivariable regression models between each potential eclampsia risk factor and strata of maternal age and parity. Potential interactions between smoking and prepregnancy BMI as well as smoking and preexisting hypertension were also investigated in post hoc analyses. All statistical analyses were completed using software (SAS, version 9.2; SAS Institute, Cary, NC); all P values reported are 2-sided, and statistical significance was considered at the alpha 5% level.
Sensitivity analyses
More than 15% of the data were missing for certain covariates on the birth certificate, including prepregnancy BMI, weight gain, and maternal education. To address the problem of missing data for these variables, we imputed the missing values using the multiple imputation procedure in software (SAS, version 9.2, proc MI). Data were imputed 20 times using the expectation-maximization algorithm option, which imputes values for selected variables based on the mother’s values listed for other variables. Logistic regression estimates using the 20 imputed datasets were combined into 1 effect estimate using software (SAS, proc MIANALYZE), which accounts for not only variation in the dataset but also the added uncertainty due to imputation of missing data when calculating SE. Results were compared to those obtained using the complete case analysis described above.
As a sensitivity analysis, we restricted our investigation to include only births that occurred after 1990 to further eliminate any overlap between our study population and the population from a previous Washington State–based study that was conducted using data from 1984 through 1990.
Results
Compared to mothers in the control group, a greater proportion of mothers who developed eclampsia were <20 years of age, Hispanic, did not have more than a high school education, were in the lowest household income group, were nulliparous, reported receiving inadequate prenatal care, gained >35 lb during pregnancy, and were obese prior to pregnancy ( Table 1 ). Among women who had given birth to >1 child, cases reported a birth interval of >5 years more often than controls. Finally, a greater proportion of control mothers reported smoking or being multiparous compared to mothers who developed eclampsia.
