Chronic hypertension in pregnancy is associated with an increased risk for preeclampsia, eclampsia, preterm delivery, and perinatal mortality.1–3
As a result of the normal physiology of pregnancy, including decreased vascular resistance, systolic and diastolic blood pressures (BPs) fall in early gestation and continue to be 5–10 mm Hg below baseline for most of the second trimester.4 Consequently, many women with chronic hypertension will have normal BPs (less than 140/90 mm Hg) during pregnancy without medications.
Little is known about this particular subgroup of pregnancies complicated by chronic hypertension; the majority of prior studies of chronic hypertension in pregnancy use only diagnostic codes or medication lists to assess the risks associated with chronic hypertension. Whether women with a history of chronic hypertension but who remain normotensive without medication during pregnancy are at increased risk for stillbirth, fetal growth restriction, or other adverse perinatal outcomes during pregnancy is not clear.
The objective of this study is to assess perinatal outcomes in women to compare the perinatal outcomes of normotensive women and women with a history of chronic hypertension with normal BPs before 20 weeks of gestation, stratifying the latter by whether they were receiving antihypertensive medication.
MATERIALS AND METHODS
This study was a retrospective cohort study of all patients with chronic hypertension at the University of Alabama at Birmingham who presented before 20 weeks of gestation for prenatal care from January 1, 2000, to June 1, 2014. Institutional board review was obtained.
Three exposure groups were identified: 1) women without a diagnosis of chronic hypertension (and who had normal BPs before 20 weeks of gestation), 2) women with a history of chronic hypertension but not taking antihypertensive medications who had normal BPs before 20 weeks of gestation, and 3) women with a history of chronic hypertension who were taking antihypertensive medications and had normal BPs before 20 weeks of gestation.
Women with singleton gestations complicated by chronic hypertension were identified through the perinatal database at the University of Alabama at Birmingham using a diagnosis of chronic hypertension. The diagnosis of chronic hypertension was confirmed by chart review based either on maternal history (maternal report of current diagnosis of chronic hypertension) or the use of antihypertensive medication before pregnancy. Women who reported a history of “prehypertension” or who reported resolution of chronic hypertension after lifestyle modification or weight loss were excluded. For the same time period (2000–2014), a randomly selected group of women without a diagnosis of hypertension (n=476) were chosen for comparison using a random number generator. Other than chronic hypertension, the same inclusion and exclusion criteria were used to select the nonhypertensive group of women. Trained chart abstractors reviewed all charts with a standardized chart abstraction form. Data were collected on demographic information, obstetric history, laboratory information, BP and urinalysis at each prenatal visit, labor course, delivery information, and neonatal outcomes until discharge.
Women were excluded from this analysis if BPs before 20 weeks of gestation were greater than 140/90 mm Hg or if they reported major medical problems other than hypertension, diabetes, or baseline renal disease. Women were also excluded if fetal anomalies were identified before neonatal discharge from the hospital.
All women with chronic hypertension were treated per institutional protocol under the supervision of maternal–fetal medicine specialists, with target BPs less than 150/90 mm Hg. Serial fetal growth was assessed by ultrasonography after 28 weeks of gestation, and at least weekly antenatal testing was performed with either contraction stress tests or biophysical profiles starting at 32–34 weeks of gestation.
Outcomes were compared among the three exposure groups. The primary outcome was a perinatal composite of stillbirth, neonatal death, respiratory support at birth, arterial cord pH less than 7, 5-minute Apgar score 3 or less, and seizures. Stillbirth and neonatal death were additionally assessed as secondary outcomes. Respiratory support was defined as receiving continuous positive airway pressure or mechanical ventilation for any amount of time. These outcomes were chosen because they are significantly associated with death or long-term neurologic morbidity.5–9 Additional secondary outcomes were small for gestational age (SGA; defined as birth weight less than the 10th percentile)10 and preterm birth at less than 37 weeks of gestation and preterm birth at less than 34 weeks of gestation. The primary maternal outcome was preeclampsia; secondary maternal outcomes were severe preeclampsia and early-onset severe preeclampsia. Preeclampsia was defined as BPs 140/90 mm Hg or greater with either proteinuria (protein excretion 300 mg or greater in 24 hours or protein-to-creatinine ratio 0.3 or greater), thrombocytopenia (less than 100,000/mL), transaminases (aspartate aminotransferase greater than twice the upper limit of normal), or elevated creatinine (1.2 mg/dL or greater). Severe preeclampsia was defined as preeclampsia plus a severe feature (BP 160/100 mm Hg or greater or a serum laboratory abnormality); neurologic symptoms such as headache were not abstracted as a result of variability in recording and so not included in the definition. Early-onset severe preeclampsia was defined as severe preeclampsia diagnosed before 34 weeks of gestation.
Because hypertension in pregnancy is relatively rare, we collected data on all women with a diagnosis of chronic hypertension from January 1, 2000, to June 1, 2014. To determine the number of nonhypertensive women required, we estimated that approximately 800 women would be included in the study, the primary perinatal composite outcome would occur in 10% of hypertensive women,11,12 and that groups would be evenly distributed between receiving and not receiving antihypertensive medications. To detect a 50% reduction in the risk of the primary perinatal composite outcome in women without hypertension (or an incidence of 5%), with an α of 0.05 and a β of 0.2, 435 women without a diagnosis of chronic hypertension were required. Because we anticipated that the primary perinatal composite would be the least frequent outcome, we anticipated at least 80% power to detect a difference in our secondary outcomes of preterm birth, preeclampsia, and SGA. We randomly selected a group of women from the same time period using the same exclusion criteria as for women with chronic hypertension. To do this, we identified women with singleton gestations from the same time period and no known diagnosis of chronic hypertension and assigned them a number using a random number generator. We then reviewed charts in order of the random number generator. Because we anticipated that 25–30% of women would be ineligible, we identified 625 charts to obtain 435 eligible normotensive women as controls.
Descriptive statistics were compared between the exposure groups using analysis of variance, χ2 test of trend, or Fisher exact, as appropriate. Because we expected that women without a diagnosis of chronic hypertension would have the lowest incidence of adverse outcomes and that women receiving antihypertensive medications would have the highest incidence of adverse outcomes, we used a χ2 test for trend. Potentially confounding variables of the exposure–outcome association were identified in the stratified analyses. Multivariable logistic regression models for the primary outcome were then developed to estimate the effect of the exposure group. Covariates for initial inclusion in multivariable statistical models were selected using the results of the univariable and stratified analyses as well as historic confounding variables (eg, history of preterm delivery for preterm delivery, history of preeclampsia for preeclampsia), and factors were removed in a backward stepwise fashion based on significant changes (10%) in the exposure adjusted odds ratio or significant differences between hierarchical models using the likelihood ratio test. Statistical analysis was performed using STATA 13 Special Edition.
Of 1,478 women with chronic hypertension over the study period, 830 were included in the analysis (Fig. 1). Of these, 212 (26%) had BPs less than 140/90 mm Hg and did not receive antihypertensive medication and 618 (74%) had BPs less than 140/90 mm Hg but were receiving antihypertensive medication before 20 weeks of gestation. We reviewed the charts of 625 women without a diagnosis of chronic hypertension to identify 476 women without chronic hypertension who met inclusion criteria.
The three groups were significantly different regarding several baseline variables (Table 1). Women without chronic hypertension were more likely to be younger, nulliparous, Hispanic, and have lower body mass indexes than either group with chronic hypertension. Women without chronic hypertension were also less likely to have baseline renal disease, gestational diabetes, or pregestational diabetes. As anticipated, women without a diagnosis of chronic hypertension had the lowest average systolic and diastolic BPs throughout all three trimesters, whereas women with chronic hypertension on medications had the highest BPs (Table 2).
The primary composite adverse perinatal outcome (stillbirth, neonatal death, respiratory support at birth, arterial cord pH less than 7, 5-minute Apgar score 3 or less, and seizures) increased in frequency across groups (P<.01; Table 3). After adjusting for prior preterm delivery and nulliparity, women with chronic hypertension and not using antihypertensive medications were at increased risk for the adverse composite outcome compared with women with no chronic hypertension (adjusted odds ratio [OR] 2.9, 95% CI 1.21–6.85) as were women in the with chronic hypertension on antihypertensive medication group (adjusted OR 5.0, 95% CI 2.38–10.54; Table 3). Although the incidence of stillbirth increased across exposure groups (1.3% vs 1.9% vs 2.9%), this trend was not statistically significant (P=.06) nor was it significant in adjusted analyses. The incidence of neonatal death (measured only in those with liveborn neonates) increased across the exposure groups (P=.03), but in adjusted analyses, the increased odds was only significant for women receiving medication. Similarly, the incidence of preterm birth before 34 and 37 weeks of gestation increased across exposure groups, but the increased odds for early preterm birth was significant only for women receiving medication after adjusting for confounding factors. The incidence of SGA was significantly different between groups in unadjusted analyses, but in adjusted analyses, the odds of SGA were significantly higher in those on antihypertensive medication compared with women without chronic hypertension.
Any diagnosis of preeclampsia, severe preeclampsia, and early-onset severe preeclampsia increased across groups (Table 4; P<.01). However, after adjusting for significant confounding variables, women with chronic hypertension not taking antihypertensive medications were not more likely to develop preeclampsia or severe preeclampsia than women in the no hypertension group. Women on medications were more likely to develop preeclampsia or severe preeclampsia.
In our cohort, 25% of women with a history of chronic hypertension were normotensive (BP less than 140/90 mm Hg) before 20 weeks of gestation without receiving antihypertensive medication. Even so, these women were at increased of adverse perinatal outcomes, including perinatal death, cord blood acidemia, seizures, and respiratory support.
Ankumah et al performed a study similar as ours; in their cohort women with untreated hypertension and women using antihypertensive medication were both more likely than women with no history of chronic hypertension to experience preterm delivery, growth restriction, and preeclampsia.13 However, these data did not report BPs during pregnancy. Su et al14 demonstrated that women with untreated hypertension had an increased risk of preterm birth and SGA compared with those without chronic hypertension. However, no data were available on baseline BPs during pregnancy. Ankumah et al13 in a secondary analysis found the lowest risk of adverse outcomes in normotensive chronic hypertensive women at baseline compared with those with elevated baseline BPs, but did not distinguish by whether they were on medication.
The strengths of this study include the a priori determination of sample size needed for the primary outcome. We collected detailed data on a large cohort of women with chronic hypertension, allowing us to clearly identify a population of women with chronic hypertension who have BPs less than 140/90 mm Hg without medications for whom there is a paucity of published information. All patients had documented BPs before 20 weeks of gestation enabling us to distinguish between chronic hypertension and preeclampsia; additionally, all women included in the hypertension groups had a confirmed diagnosis of chronic hypertension. Women with a history of “prehypertension” or hypertension that resolved after weight loss surgery were excluded from the study.
One of the limitations of this study is that the sample is from a single academic institution and therefore may not be generalizable to other institutions. Specifically, our cohort was largely black, obese, and had a high incidence of pregestational diabetes. Additionally, women with chronic hypertension were different than women without chronic hypertension in several important baseline characteristics, including being older, more obese, and more likely to have other medical problems such as diabetes. Therefore, the differences in outcomes may not be solely the result of the presence of chronic hypertension. Although we attempted to control for these factors using logistic regression, the possibility of residual confounding exists. Furthermore, our sample was based on an estimated 10% incidence of the primary outcome with approximately 400 women in the no medication group. However, only 25% of the sample, or 212 women with chronic hypertension, were in the no medication group. Consequently, we had only 57% power to detect a 50% reduction in the risk of the primary perinatal composite outcome in women without hypertension compared with the no medication group. Because the individual components of the primary composite perinatal outcome were less frequent than the composite perinatal outcome, we had limited power to detect a difference in stillbirth, neonatal death, and perinatal death. Similarly, for the secondary outcome of preterm birth before 34 weeks of gestation, we had approximately 35% power to detect a 50% reduction in the risk in women without hypertension compared with the no medication group. We had more than 80% power to detect a difference of a 50% reduction for preeclampsia.
In summary, we found that women with a history of chronic hypertension who are normotensive without antihypertensive medication before 20 weeks of gestation are at increased risk of adverse perinatal outcomes compared with women without a diagnosis of chronic hypertension.
1. Bramham K, Parnell B, Nelson-Piercy C, Seed PT, Poston L, Chappell LC. Chronic hypertension and pregnancy outcomes: systematic review and meta-analysis. BMJ 2014;348:g2301.
2. Cruz MO, Gao W, Hibbard JU. Obstetrical and perinatal outcomes among women with gestational hypertension, mild preeclampsia, and mild chronic hypertension. Am J Obstet Gynecol 2011;205:260.e1–9.
3. Sibai BM, Abdella TN, Anderson GD. Pregnancy outcome in 211 patients with mild chronic hypertension. Obstet Gynecol 1983;61:571–6.
4. Grindheim G, Estensen ME, Langesaeter E, Rosseland LA, Toska K. Changes in blood pressure during healthy pregnancy: a longitudinal cohort study. J Hypertens 2012;30:342–50.
5. Shankaran S, Laptook AR, Ehrenkranz RA, Tyson JE, McDonald SA, Donovan EF, et al. Whole-body hypothermia for neonates with hypoxic-ischemic encephalopathy. N Engl J Med 2005;353:1574–84.
6. Dixon G, Badawi N, Kurinczuk JJ, Keogh JM, Silburn SR, Zubrick SR, et al. Early developmental outcomes after newborn encephalopathy. Pediatrics 2002;109:26–33.
7. Kulak W, Okurowska-Zawada B, Sienkiewicz D, Paszko-Patej G, Krajewska-Kulak E. Risk factors for cerebral palsy in term birth infants. Adv Med Sci 2010;55:216–21.
8. Lie KK, Grøholt EK, Eskild A. Association of cerebral palsy with Apgar score in low and normal birthweight infants: population based cohort study. BMJ 2010;341:c4990.
9. Low JA, Lindsay BG, Derrick EJ. Threshold of metabolic acidosis associated with newborn complications. Am J Obstet Gynecol 1997;177:1391–4.
10. Duryea EL, Hawkins JS, McIntire DD, Casey BM, Leveno KJ. A revised birth weight reference for the United States. Obstet Gynecol 2014;124:16–22.
11. Harper LM, Biggio JR, Anderson S, Tita AT. Gestational age of delivery in pregnancies complicated by chronic hypertension. Obstet Gynecol 2016;127:1101–9.
12. Kuper SG, Tita AT, Youngstrom ML, Allen SE, Tang Y, Biggio JR, et al. Baseline renal function tests and adverse outcomes in pregnant patients with chronic hypertension. Obstet Gynecol 2016;128:93–103.
13. Ankumah NA, Cantu J, Jauk V, Biggio J, Hauth J, Andrews W, et al. Risk of adverse pregnancy outcomes in women with mild chronic hypertension before 20 weeks of gestation. Obstet Gynecol 2014;123:966–72.
© 2018 by The American College of Obstetricians and Gynecologists. Published by Wolters Kluwer Health, Inc. All rights reserved.
14. Su CY, Lin HC, Cheng HC, Yen AM, Chen YH, Kao S. Pregnancy outcomes of anti-hypertensives for women with chronic hypertension: a population-based study. PLoS One 2013;8:e53844.