Background
The risk of preeclampsia increases as maternal body mass index (BMI) increases. The link between increasing maternal BMI and preeclampsia with severe features is less well-established.
Objective
To estimate the effect of increasing severity of obesity on risk of preeclampsia with severe features, stratified by early-onset and late-onset disease.
Study Design
We performed a retrospective cohort study of consecutive singleton live births at a tertiary care facility from 2004 to 2008. Women were included in the cohort if they delivered a singleton live birth and maternal height and weight was measured on admission. The primary exposure was maternal weight category on presentation for delivery, defined as normal (BMI 18.5−24.9; referent group, n = 1473), overweight (BMI 25−29.9, n = 3081), obese (BMI 30−39.9, n = 4196), and morbidly obese (BMI ≥40, n = 1446). The primary outcome was preeclampsia with severe features. Secondary outcome was early-onset preeclampsia with severe features at <34 weeks or late-onset preeclampsia with severe features at ≥34 weeks. Multivariable logistic regression was used to adjust for confounders.
Results
Of the 10,196 patients meeting inclusion criteria, 1119 developed preeclampsia. Of those, 881 (8.6%) women developed preeclampsia with severe features. Overall, the risk of preeclampsia with severe features was not significantly different in the 4 BMI categories. Of the 10,196 women in the cohort, 1072 delivered <34 weeks and 9124 delivered ≥34 weeks. When stratifying by gestational age at delivery, there was a statistically significant increased risk of developing late-onset preeclampsia with severe features at ≥34 weeks in overweight (4.5%, adjusted odds ratio [aOR] 1.4, 95% confidence interval [CI] 1.0−2.1), obese (6.2%, aOR 2.0, 95% CI 1.4−2.8) and morbidly obese (6.8%, aOR 2.0, 95% CI 1.3−2.9) women compared with normal-weight women (2.9%).
Conclusion
Increasing maternal weight was not associated with preeclampsia with severe features in the total cohort; however, overweight, obese, and morbidly obese women are at increased risk of developing late-onset preeclampsia with severe features.
Approximately 10% of pregnancies are complicated by hypertensive disorders. Preeclampsia is the most significant subtype and confers considerable maternal morbidity and mortality. The incidence of severe preeclampsia has increased 30% during the last 20 years and primarily is responsible for the majority of maternal complications from hypertensive disorders of pregnancy. Increasing maternal weight is a well-established risk factor for developing preeclampsia. Obesity now affects more than one-third of reproductive-aged women, and escalating rates of obesity may contribute to an increased prevalence of hypertensive disorders in pregnancy.
Because preeclampsia presents at varied gestational ages and varies in severity, 2 distinct subtypes have been recognized. Early-onset preeclampsia occurs before 34 weeks’ gestation, and late-onset preeclampsia occurs at or beyond 34 weeks’ gestation. These discrete classifications carry different implications for both the mother and fetus. Perinatal death and severe neonatal morbidity are greater in women with early-onset disease than late-onset disease, and rates of severe maternal morbidity and mortality are greater in women with early-onset disease.
Previous studies have suggested that the early- and late-onset phenotypes arise from distinct pathogenesis. Early-onset preeclampsia has been explained as a result of poor placental implantation leading to chronic placental insufficiency and subsequent inflammatory cascade with resultant hypertensive disease. Late-onset preeclampsia has been described as both a placental disease and a result of maternal metabolic and cardiovascular risk factors. Given the suspicion that maternal metabolic factors contribute to the underlying disease, previous studies have examined what effect maternal body mass index (BMI) has on the development of early-onset or late-onset preeclampsia. These studies indicated a greater risk of preeclampsia as maternal BMI increased; however, they included a heterogeneous population both in the definition of early-onset disease (<37 weeks vs <34 weeks) as well as severity of the disease.
The primary aim of our study was to identify the effect of increasing BMI on the risk for developing preeclampsia with severe features. We hypothesized that there would a dose−response relationship between increasing maternal BMI and risk of preeclampsia with severe features. Secondarily, we hypothesized that this effect would be more pronounced with late-onset preeclampsia than early-onset disease.
Methods
We conducted an institutional review board−approved, retrospective cohort study of all consecutive live births occurring during a 4-year period from 2004 to 2008 at Washington University in St. Louis Medical Center. Inclusion criteria consisted of all women with known maternal BMI at the time of admission to labor and delivery who delivered a singleton live birth regardless of gestational age at our tertiary care facility. BMI was calculated from height and weight measured on admission to labor and delivery. Four comparison groups were created according to BMI criteria defined by the World Health Organization: normal weight (BMI 18.5−24.9), overweight (BMI 25−29.9), obese (BMI 30−39.9), and morbidly obese (BMI ≥40). Medical records were reviewed and extracted by trained obstetric research nurses.
The primary outcome was preeclampsia with severe features, defined by American College of Obstetricians and Gynecologists (ACOG) criteria as follows: new-onset severe hypertension (systolic blood pressure ≥160 mm Hg or diastolic blood pressure ≥110 mm Hg, or hypertension requiring antihypertensive therapy) after 20 weeks of gestation (exclusive of chronic hypertension) or preeclampsia in the setting of thrombocytopenia (platelet <100,000/mL), impaired liver function (aspartate aminotransferase or alanine aminotransferase elevated to twice the upper limit of normal; persistent right upper quadrant pain or epigastric pain), renal insufficiency (doubling of a previously normal serum creatinine value or serum creatinine >1.1 mg/dL), pulmonary edema, or central nervous system symptoms (including persistent headache or visual disturbances). Early-onset preeclampsia was defined as preeclampsia with severe features mandating delivery before 34 completed weeks’ gestation. Late-onset preeclampsia was defined as preeclampsia with severe features at or beyond 34 weeks’ gestation. Gestational age was calculated by menstrual dating criteria or ultrasound dating, whichever was deemed more reliable according to ACOG criteria.
Baseline demographics were compared among the groups with one-way analysis of variance for continuous variables and the chi-squared or Fisher exact test for categorical variables as appropriate. Incidence of preeclampsia with severe features was calculated by BMI category. Adjusted odds ratios (aORs) with 95% confidence intervals (CIs) for the association between BMI and preeclampsia with severe features after controlling for pertinent confounding variables were calculated by the use of logistic regression. Women with BMI 18.5−24.9 served as the referent group. Covariates were selected on the basis of biological plausibility, other studies, and results of our univariable analysis. Backward elimination was used to reduce the number of covariates. The final models included only covariates that were statistically significant (maternal age) or of clinical relevance (self-reported tobacco use, chronic hypertension, and pregestational diabetes). The analysis was repeated, stratified by early- or late-onset disease. Sensitivity analysis was conducted for women with chronic hypertension and superimposed preeclampsia. All analyses were performed with the STATA software package, version 12.1, Special Edition (College Station, TX). P values <.05 were considered significant.
Results
Among the 10,564 women delivering during the study, BMI was known for 10,218. After we excluded 22 women with BMI <18.5, 10,196 women were included in the final analysis; 1473 were normal weight (14.4%), 3081 (30.2%) were overweight, 4196 (41.2%) were obese, and 1446 (14.2%) were morbidly obese. A total of 1072 (10.5%) women in the cohort delivered before 34 weeks’ gestation; 9124 (89.5%) delivered at or beyond 34 weeks. Obese and morbidly obese women were more likely to be older, African-American, and multiparous. Increasing BMI also was associated with greater rates of diabetes and chronic hypertension. At baseline, normal-weight women were more likely to use tobacco and illicit drugs. The groups were similar with regards to payor status and alcohol use ( Table 1 ).
| BMI 18.5−24.9, n = 1473 | BMI 25−29.9, n = 3081 | BMI 30−39.9, n = 4196 | BMI >40, n = 1446 | P value | |
|---|---|---|---|---|---|
| Age, y, mean±SD | 23.8 ± 6.2 | 24.6 ± 6.2 | 25.3 ± 6.0 | 26.5 ± 5.8 | .003 |
| Race, n (%) | <.001 | ||||
| African-American | 995 (67.6) | 2014 (65.4) | 3012 (71.8) | 1131 (78.2) | |
| White | 344 (23.4) | 761 (24.7) | 877 (20.9) | 274 (19.0) | |
| Other | 134 (9.0) | 306 (9.9) | 307 (7.3) | 41 (2.8) | |
| Parity, n (%) | <.001 | ||||
| 0 | 618 (42.0) | 1235 (40.1) | 1483 (35.3) | 437 (30.2) | |
| 1 | 401 (27.2) | 882 (28.6) | 1179 (28.1) | 392 (27.1) | |
| > 2 | 454 (30.8) | 964 (31.3) | 1534 (36.6) | 617 (42.7) | |
| Payor, n (%) | .003 | ||||
| Public/uninsured | 1166 (79.2) | 2355 (76.4) | 3336 (79.5) | 1163 (80.4) | |
| Other (private, self) | 307 (20.8) | 726 (23.6) | 860 (20.5) | 283 (19.6) | |
| Gestational age, wk | 36.3 ± 4.3 | 37.3 ± 3.5 | 37.7 ± 3.3 | 37.6 ± 3.6 | <.001 |
| Tobacco, n (%) | 363 (24.6) | 616 (20.0) | 782 (18.6) | 274 (19.0) | <.001 |
| Alcohol, n (%) | 38 (2.6) | 47 (1.5) | 61 (1.5) | 15 (1.0) | .005 |
| Drugs, n (%) | 216 (14.7) | 396 (12.9) | 407 (9.7) | 122 (8.4) | <.001 |
| Pregestational diabetes, n (%) | 2 (0.14) | 21 (0.68) | 54 (1.29) | 50 (3.46) | <.001 |
| Chronic hypertension, n (%) | 22 (1.5) | 57 (1.9) | 188 (4.5) | 175 (12.1) | <.001 |
Among the 1119 (11.0%) patients with preeclampsia, 881 (8.6%) women developed preeclampsia with severe features. Early-onset preeclampsia with severe features occurred in 339 (3.3%) women, whereas 542 (5.3%) women had late-onset preeclampsia with severe features. The relative frequency of preeclampsia with severe features is shown in Figure 1 and demonstrates 8.0% of normal-weight women, 7.4% of overweight women, 9.1% of obese women, and 10.6% of morbidly obese women had preeclampsia with severe features. Early-onset disease occurred in 5.1% of normal-weight women, 2.9% of overweight women, 2.9% of obese women, and 3.7% of morbidly obese women. The incidence of late-onset disease increased with BMI category, occurring in 2.9% of normal-weight women, 4.5% of overweight women, 6.2% of obese women, and 6.8% of morbidly obese women. The relative frequencies of early-onset disease and late-onset disease by BMI category for the total cohort are shown in Figure 1 .
When we examined the total cohort, there was a nonsignificant trend towards increased risk for preeclampsia with severe features and increasing BMI ( Table 2 ); however, when we stratified by early- and late-onset phenotypes, there was a statistically significant relationship between risk for preeclampsia with severe features and increasing BMI in women, specifically with late-onset disease. A total of 4.5% of overweight women (aOR 1.4, 95% CI 1.0−2.1), 6.2% of women with BMI 30−39 (aOR 2.0, 95% CI 1.4–2.8), and 6.8% of women with BMI ≥40 (aOR 2.0, 95% CI 1.3−2.9) had late-onset preeclampsia compared with 2.9% of normal-weight women. No association was seen between overweight and obese women and risk for early-onset severe preeclampsia; however, there was an increased risk for early-onset disease in morbidly obese women compared with normal-weight gravidas (aOR 1.6, 95% CI 1.0−2.6) ( Table 2 ). A total of 442 (4.3%) women in the cohort had chronic hypertension. Of those, 105 (1.0%) developed superimposed severe preeclampsia. Overall, increasing maternal BMI was associated with an increased risk of developing superimposed severe preeclampsia in obese and morbidly obese women (aOR 2.7, 95% CI 1.1−6.8; aOR 6.6, 95% CI 2.6−16.8, respectively) ( Table 2 ).
