Objective
The purpose of this study was to determine whether superimposed preeclampsia results in worse perinatal outcomes than preeclampsia.
Study Design
We conducted a retrospective cohort study using our perinatal database (1990-2008). Perinatal outcomes among women with chronic hypertension (n = 1032), superimposed preeclampsia (n = 489), and preeclampsia (n = 4217) were compared with outcomes of control subjects (n = 57,103). Outcomes among women with superimposed preeclampsia were also compared with outcomes of women with preeclampsia. Multivariable analysis was used to control for confounders.
Results
Rates of small-for-gestational age, abruption, stillbirth, and eclampsia were not significantly different with superimposed preeclampsia compared with preeclampsia. Delivery at <34 weeks’ gestation (17.3% vs 8.7%; P < .001), cesarean delivery (46.2% vs 36.3%; P < .001), and neonatal intensive care unit admission (16.3% vs 11.4%; P < .002) were significantly higher among women with superimposed preeclampsia. These risks persisted after we controlled for confounders.
Conclusion
Women with superimposed preeclampsia have higher risks of intervention-related events compared with those with preeclampsia.
Chronic hypertension complicates up to 5% of pregnancies, and these rates are expected to rise because obesity and the age of childbearing continue to increase. Pregnancies in women with chronic hypertension are at an increased risk of adverse perinatal outcomes that include preterm birth, intrauterine growth restriction, fetal death, placental abruption, and cesarean delivery. Several studies have demonstrated a higher risk of these adverse outcomes among women with chronic hypertension who developed superimposed preeclampsia compared with women who do not have chronic hypertension. It has been speculated that the underlying vascular abnormalities in patients with chronic hypertension cause an escalation of complications when preeclampsia develops (additive model). Alternatively, the elevated risk of adverse outcomes in women with superimposed preeclampsia may simply be a reflection of the higher risk of adverse outcomes that are associated with preeclampsia, independent of the coexisting chronic hypertension (preeclampsia model).
Although several studies have evaluated perinatal outcomes among women with chronic hypertension, with and without superimposed preeclampsia, data on perinatal outcomes in women with superimposed preeclampsia relative to women with preeclampsia are sparse. The few published studies are further limited by focusing on a single outcome measure (such as small-for-gestational age [SGA]), not adjustment for confounders and the use of population databases, which have inherent limitations such as incomplete detail, missing data, and misclassification. To provide an evidence base for management of pregnancies in women with superimposed preeclampsia, accurate risk estimates are needed. We used a large perinatal database from a single center to evaluate perinatal outcomes in women with preeclampsia, chronic hypertension, and chronic hypertension with superimposed preeclampsia. Specifically, we test the hypothesis that perinatal outcomes among women with superimposed preeclampsia are worse than those with preeclampsia. We anticipate that the results of this study will assist physicians in the counseling and treatment of women with hypertensive disorders in pregnancy.
Materials and Methods
This is a retrospective cohort study that used our computerized perinatal database (January 1990 to December 2008). Approval for the study was obtained from our institutional review board. The Department of Obstetrics and Gynecology keeps a comprehensive perinatal database that is compiled prospectively and maintained by a dedicated data management staff. Pregnancy and delivery information for all patients who receive prenatal care or who are delivered at our medical center is entered into the database. Delivery information for referred patients is obtained from the patients and referring physicians with the use of a standardized delivery record form. This information is validated with the patients’ medical records.
All women in the database with singleton pregnancies and obstetric outcome data were eligible. Multiple pregnancies and pregnancies with known chromosomal or major structural abnormalities were excluded. Study subjects were divided into 4 mutually exclusive groups for comparison. Control subjects consisted of women who had neither chronic hypertension nor preeclampsia (n = 57,103). The chronic hypertension–only group included women with chronic hypertension who did not have superimposed preeclampsia (n = 1032). The preeclampsia group consisted of women without chronic hypertension who subsequently experienced preeclampsia (n = 4217); the superimposed preeclampsia group included women with chronic hypertension who experienced superimposed preeclampsia (n = 489).
Chronic hypertension was defined as previously diagnosed prepregnancy hypertension or systolic blood pressure of ≥140 mm Hg and/or diastolic blood pressure of ≥90 mm Hg before the twentieth week of gestation. Preeclampsia and superimposed preeclampsia were defined according to the guidelines of the International Society for the Study of Hypertension in Pregnancy. With the use of these guidelines, superimposed preeclampsia in women with chronic hypertension was defined as the development of new onset proteinuria (excretion of ≥300 mg of protein over 24 hours). For women without 24-hour urine collection, results of dipsticks analysis were used for the semi-quantitative assessment of proteinuria. A value of ≥1+ protein on the dipstick was considered indicative of 24-hour urine protein value of >300 mg. In women with preexisting proteinuria (baseline, >300 mg/24 hr), the diagnosis of superimposed preeclampsia was based on the worsening of blood pressure and the identification of clinical symptoms (headache, vision changes, upper abdominal pain) or biochemical markers of HELLP (elevated liver enzymes, low platelets, hemolysis) syndrome.
Perinatal outcomes that were examined included SGA (birthweight <10th percentile for gestational age on the Alexander growth curve ), stillbirth (fetal death at ≥20 weeks’ gestation), placental abruption, neonatal intensive care unit (NICU) admission, cesarean delivery, preterm birth (delivery at <34 or <37 weeks’ gestation), and length of maternal hospital stay. Prolonged maternal hospital stay was defined as >3 days for women who had vaginal deliveries and >4 days for women with cesarean deliveries.
Univariable analysis was used to assess differences in baseline demographic and pregnancy characteristics among the 4 groups. χ 2 test was used to compare categoric variables; analysis of variance with Bonferroni post-hoc tests (to adjust for multiple comparisons) was used for continuous outcomes. Unadjusted odds ratios (ORs) with 95% confidence intervals (CIs) were calculated to quantify the risk of each of the individual outcomes in women with chronic hypertension only, superimposed preeclampsia, and preeclampsia compared with control subjects.
Multivariable logistic regression was used to calculate adjusted odds ratios (aORs) for each outcome in the 3 groups with hypertensive disease compared with control subjects. Candidate variables for the logistic regression models were selected on the basis of biologic plausibility, risk factors that were identified in the literature for the various outcomes, and results of our univariable analysis. The number of variables in each model was reduced with the use of backwards elimination. Differences between hierarchic explanatory models, including those with and without relevant interaction terms were assessed with the use of the likelihood ratio test or Wald test. Model fit for each final model was assessed with the Hosmer-Lemeshow lack-of-fit test. To directly compare outcomes in women with superimposed preeclampsia with women with preeclampsia, unadjusted and aORs were calculated for superimposed preeclampsia relative to preeclampsia.
All statistical analysis was performed with STATA software (version 10.0; Stata Corporation, College Station, TX). Tests with probability values of < .05 were considered significant.
Results
The Figure shows the sample selection for the study. A total of 62,841 women met the inclusion criteria. Of these, 57,103 women (90.9%) were control subjects, and 4217 women (6.7%) had a diagnosis of preeclampsia. A total of 1521 women (2.4%) were diagnosed with chronic hypertension, of which 489 women (0.8%) experienced superimposed preeclampsia and 1032 women (1.6%) did not.
Baseline characteristics of the study population are shown in Table 1 . Women with hypertensive disease significantly differed from control subjects in age, parity, body mass index, smoking, and diabetes mellitus. Table 2 shows risk estimates of adverse perinatal outcomes in women with preeclampsia, superimposed preeclampsia, and chronic hypertension only compared with control subjects. Women with preeclampsia were significantly more likely to have an SGA infant, abruption, preterm delivery at <34 weeks’ gestation, preterm delivery at <37 weeks’ gestation, cesarean delivery, NICU admission, and prolonged hospital stay compared with control subjects. On the other hand, the risk of stillbirth was significantly lower in women with preeclampsia compared with control subjects. Superimposed preeclampsia also was associated with a significantly increased risk of the same adverse outcomes compared with control subjects, except for stillbirth and abruption. Chronic hypertension without preeclampsia similarly was associated with a higher risk of adverse outcomes compared with control subjects, except for placental abruption.
| Characteristic | Control subjects (n = 57,103) | Preeclampsia (n = 4217) | Superimposed preeclampsia (n = 489) | Chronic hypertension only (n = 1032) | P value |
|---|---|---|---|---|---|
| Maternal age, y a | 30.2 ± 6.3 | 29.1 ± 6.6 | 31.7 ± 6.1 | 32.7 ± 5.8 | < .001 |
| Race, n (%) | < .001 | ||||
| White | 36,548 (66.5) | 2631 (64.3) | 261 (54.3) | 542 (54.2) | |
| Black | 11,370 (20.6) | 1124 (27.5) | 186 (38.7) | 384 (38.4) | |
| Other | 7344 (13.3) | 335 (8.2) | 34 (7.1) | 74 (7.4) | |
| Parity, n (%) | < .001 | ||||
| Multiparous | 36,231 (63.5) | 1811 (42.9) | 310 (63.4) | 780 (75.6) | |
| Nulliparous | 20,871 (36.5) | 2406 (57.1) | 179 (36.6) | 252 (24.4) | |
| Body mass index, kg/m 2 a | 24.3 ± 8.8 | 27.1 ± 10.2 | 31.0 ± 12.4 | 30.9 ± 12.0 | < .001 |
| Body mass index categories | < .001 | ||||
| Underweight | 4996 (8.8) | 320 (7.7) | 37 (7.6) | 78 (7.6) | |
| Normal weight | 26,504 (46.8) | 1200 (28.8) | 67 (13.8) | 171 (16.7) | |
| Overweight | 12,994 (23.0) | 1101 (26.4) | 108 (22.2) | 171 (16.7) | |
| Obese | 12,094 (21.4) | 1545 (37.1) | 274 (56.4) | 608 (58.9) | |
| Smoking, n (%) | 6419 (11.3) | 452 (10.8) | 68 (13.9) | 144 (14.0) | .007 |
| Fetal sex: female, n (%) | 28,730 (50.8) | 2115 (50.4) | 258 (53.0) | 516 (50.8) | .736 |
| Diabetes mellitus, n (%) | |||||
| Pregestational | 781 (1.4) | 195 (4.6) | 55 (11.3) | 84 (8.1) | < .001 |
| Gestational | 2682 (4.7) | 326 (7.7) | 74 (15.1) | 128 (12.4) | < .001 |
| Premature membranes rupture, n (%) | 1383 (2.4) | 69 (1.6) | 3 (0.6) | 34 (3.3) | < .001 |
| Renal disease, n (%) | 962 (1.7%) | 66 (1.6%) | 6 (1.2%) | 18 (1.7%) | .809 |
Stay updated, free articles. Join our Telegram channel
Full access? Get Clinical Tree
