Prevalence of Hypertensive Phenotypes After PreeclampsiaNovelty and Significance
A Prospective Cohort Study
Preeclampsia is associated with increased cardiovascular and renal risk. The aim of this prospective cohort study was to characterize the early postpartum blood pressure (BP) profile after preeclampsia. We enrolled 115 women with preeclampsia and 41 women with a normal pregnancy in a prospective cohort study. At 6 to 12 week postpartum, we assessed the prevalence of different hypertensive phenotypes using 24-hour ambulatory BP monitoring (ABPM), as well as the risk of salt sensitivity and the variability of BP derived from ABPM parameters. Among patients with preeclampsia, 57.4% were still hypertensive at the office. Daytime ABP was significantly higher in the preeclampsia group (118.9±15.0/83.2±10.4 mm Hg) than in controls (104.8±7.9/71.6±5.3 mm Hg; P<0.01). Differences between groups were similar for nocturnal BP values. Fifty percent of preeclampsia women remained hypertensive on ABPM in the postpartum, of whom 24.3% were still under antihypertensive treatment; 17.9% displayed a white-coat hypertension and 11.6% had masked hypertension. In controls, 2.8% had white-coat hypertension; none had masked hypertension or needed hypertensive treatment. The prevalence of nondippers was similar 59.8% in the preeclampsia group versus 51.4% in controls. High-risk class of salt sensitivity of BP was increased in preeclampsia women (48.6%) compared with controls (17.1%); P<0.01. In conclusion, ABPM 6 to 12 weeks after delivery reveals a high rate of sustained ambulatory, nocturnal, and masked hypertension after preeclampsia. This finding may help identify women who should be included in a postpartum cardiovascular risk management program.
See Editorial Commentary, pp 59–60
Preeclampsia is a hypertensive disorder specific to gestation that affects 5% to 8% of all pregnancies1 and is associated with a greater lifetime risk for cardiovascular and renal complications.2,3 Recently, different international guidelines have underlined that women who had a hypertensive disorder of the pregnancy should benefit from a postpartum screening and management for cardiovascular risk factors, but with no precise recommendations, especially about the timing of the first medical visit.4–7 The existence and the prevalence of specific hypertensive phenotypes after preeclampsia is not known and the place of ambulatory blood pressure monitoring (ABPM) in the early follow-up of preeclampsia needs to be defined.
A blunted nocturnal dip of BP has been described in hypertensive disorders of pregnancy and seems to correlate with the severity of the disease.8,9 Nocturnal hypertension in pregnancy has been associated with more maternal and fetal adverse outcomes, such as preeclampsia, renal insufficiency, severe hypertension, and low birth weight babies.10 However, whether the nondipping pattern of BP persists during the postpartum period has never been investigated systematically.
In the present prospective study, we hypothesized that part of the elevated cardiovascular risk in women with a history of preeclampsia is partly related to the underdiagnosis of hypertension, in particular masked hypertension, and the persistence of abnormalities in the circadian profile of BP after pregnancy. To this purpose, we performed 24-hour ABPM in women with preeclampsia and in women with uncomplicated pregnancies (controls) to assess the different BP phenotypes 6 to 12 weeks after delivery. As abnormal circadian profile has been shown to be associated with increased salt sensitivity of BP,11 we also estimated the risk of salt sensitivity (RSS) and the variability of the BP derived from ABPM parameters.
Study Type and Population
We performed a prospective cohort study in women with preeclampsia and a subgroup without preeclampsia. The main criteria for cases inclusion were preeclampsia during the index pregnancy, consent to participate and have a postpartum ABPM regardless of office BP values, drug treatment, and albuminuria. Main exclusion criteria were the use of anti-inflammatory drugs or any known cardiac, endocrine, or renal diseases. Inclusion criteria for the control group were healthy women who had an uncomplicated normotensive pregnancy, delivered spontaneously at term, consenting to participate and to have a postpartum ABPM.
The Ethics Committee of both institutions approved this study and written informed consent was obtained from each participant. Study was registered and performed according to the Helsinki declaration.
All women were screened after delivery. If the study was accepted, they were evaluated at 6 to 12 week postpartum either at the Hypertension Unit of the University Hospital of Geneva or in the Service of Nephrology and Hypertension at Lausanne University Hospital, Switzerland. The timing of the visit was chosen to coincide with the gynecologic visit routinely performed after delivery. The participants came in the morning after a light snack. Height and weight were measured and body mass index calculated. Relevant maternal and infant medical histories and family factors were collected, and a complete physical examination was performed. Office BP and 24-hour ABPM were performed as described below. Plasma creatinine, spot urine albumin, and creatinine were measured, and kidney function was estimated using the CKD-EPI formula (Chronic Kidney Disease–Epidemiology Collaboration).12
Preeclampsia was defined according to the International Society for the Study of Hypertension in Pregnancy statement of 2014, as new-onset hypertension after 20-week gestation associated with at least 1 end-organ damage.13 Birth weight percentiles were determined according to the Swiss Society of Pediatrics standards, and small for gestational age infants were defined as a birth weight <10th percentile for gestational age at birth.
Hypertension phenotypes were defined as follows: masked hypertension was defined as normal office BP (<140/90 mm Hg) with high diurnal ABP (≥135/85 mm Hg) or nocturnal BP,14 white-coat hypertension referred to high office BP (≥140/90 mm Hg) with normal awake ABP measurement (<135/85 mm Hg)14; sustained hypertension and true normotensive state described situations where both types of BP measurements were concordant. These definitions were used for both treated and untreated patients.
Normal dipping was defined as nocturnal drop of ≥10% to 20% of mean ABP. Nondippers were subjects with 0% to 9% nocturnal fall of mean ABP, extreme dippers with >20% nocturnal fall, and reverse dippers included patients with a nocturnal increase of mean ABP.15
Nocturnal hypertension was defined as night ABP ≥120/70 mm Hg.
BP Measurement Methods
Office BP was measured according to European Society of Hypertension guidelines.14 After 5 minutes of rest, 2 BP measurements were taken at 1 to 2 minute intervals, on the same arm, with a proper cuff size adjusted to the arm diameter and a validated automated device (Omron HEM-907-E; British Hypertension Society Grade A/A). Arithmetic mean of the 2 BP values was used for the present analysis.
A 24-hour ABPM was performed with a validated device (Diasys Integra, Physicor, Switzerland) and programmed to measure BP every 20 minutes from 07:00 to 22:59 and every 30 minutes from 23:00 to 06:59. Calibration of the device was made by 2 simultaneous auscultatory BP measurements, using a calibrated mercury sphygmomanometer connected to the ambulatory device by a Y tube. A difference of >5 mm Hg was considered acceptable.
Subjects were instructed not to smoke or drink alcohol or any caffeine-containing beverage, to record their time of sleep and awakening, as well as their activities in a diary. According to the European Society of Hypertension recommendations, ABPM was considered valid and was included in the analysis when at least 20 BP measurements during the day and 7 readings during the night were recorded.
Assessment of Salt Sensitivity Risk Using ABPM Data
We also estimated the RSS of BP with the formula of Castiglioni et al16 that we validated previously in women with a history of severe preeclampsia.17 Briefly, 3 classes of RSS graded as low, intermediate, and high, were defined based on the dipping status as described above and on the mean heart rate (HR) over 24 hours.
They were then classified as high HR and low HR when the 24-hour mean HR was respectively over or under 70 bpm. Low RSS included subjects classified as dippers and low 24-hour mean HR. Patients identified as nondippers with a high 24-hour mean HR were considered as high RSS class, and the intermediate RSS included all other subjects not otherwise classified.
We compared women with and without preeclampsia. A secondary analysis restricted to white women, nonsmoker, with a body mass index <25 kg/m2, and without past history of chronic hypertension was performed. Results are reported as a mean (SD) for continuous variables and in % (absolute number) for the categorical variables. Normality of the distribution was tested by drawing a histogram and Q-Q plot. Continuous variables were analyzed using a Student t test or Kruskal–Wallis and proportions using either the χ2 or the Fisher test. A 2-sided P value of ≤0.05 was considered as statistically significant. All data analyses were performed with Stata software, version 12.1.
Between July 2010 and December 2013, we screened 250 preeclampsia women at time of delivery, of which 115 gave their consent and had an ABPM done at 6 to 12 weeks postpartum. We also included 41 nonpreeclamptic healthy control women (Figure 1). Eight women were excluded for invalid recording (3 in the preeclampsia group and 5 in the controls). Among preeclampsia women, 57.4% (66/115) were still hypertensive at the office and 41.7% were albuminuric. At time of evaluation, 28 women (24.3%) were still under antihypertensive therapy (labetalol or nifedipine). The baseline characteristics of the groups at the postpartum visit are shown in Table 1. There were no significant differences between women with preeclampsia and controls for age, primiparity, prevalence of pregestational diabetes mellitus, and twin pregnancies. Several characteristics were significantly more prevalent or increased in the group of preeclamptic women: Afro-Caribbean origin (P<0.01), high body mass index (P<0.01), chronic hypertension before the index pregnancy (P=0.01), and smokers (P<0.01). All women were breastfeeding.
Preterm delivery was more frequent among women with preeclampsia compared with normal pregnancy (mean gestational age at birth, 36.0±3.9 versus 39.4±1.6 weeks; P<0.01). Accordingly, birth weight of the newborns in the preeclampsia group was significantly lower (2525.6±898.1 versus 3230.0±449.9 g; P<0.01). However, there was no significant difference between groups in terms of small for gestational age infants (birth weight <10th percentile for gestational age). Of note, 7 neonatal deaths (6.1%) occurred in the group with preeclampsia.
Office BP and 24-hour ambulatory BP values are shown in Table 2. Office BP was significantly higher among women who had preeclampsia than in controls. Daytime systolic and diastolic ABP were significantly higher in the preeclampsia group (118.9±15.0/83.2±10.4 mm Hg) than in the control group (104.8±7.9/71.6±5.3 mm Hg; P<0.01). Nighttime ABP was also significantly higher in the preeclampsia group (111.2±17.6/74.8±11.0 mm Hg) compared with the control group (94.3±7.6/61.7±4.6 mm Hg). The SD of the systolic and diastolic daytime ABP, a proxy of the variability of the BP,18 was significantly higher in the preeclampsia group (13.8±4.7/11.4±6.1 mm Hg) than in controls (10.8±4.5/8.7±2.4 mm Hg; P=0.01), so were the nighttime values (11.0±4.5/9.8±3.3 mm Hg in preeclampsia, and 9.1±2.1/8.0±2.0 mm Hg in controls; P<0.01). Table 3 presents the different BP phenotypes. Among preeclampsia women without treatment, 23.8% (20/84) of them had a confirmed diurnal high ABP (≥135/85 mm Hg) at 6 to 12 week postpartum compared with 0.0% in the controls, P<0.01. Prevalence of white-coat hypertension was also significantly increased (17.9%) compared with controls (2.8%; P<0.01). Masked hypertension was more frequent in preeclampsia group (11.6%) than in the controls (0.0%; P<0.01). Nocturnal hypertension (nighttime ABP ≥120/70 mm Hg) in patient with normal office BP was more frequent in preeclampsia women (33.6% [36/107] versus 2.9% [1/35] in controls; P<0.01). Furthermore, among the 28 patients still requiring antihypertensive treatment 6 to 12 week after preeclampsia, 17.9% had white-coat hypertension and 17.9% masked hypertension. Systolic and diastolic BP patterns are illustrated in Figures 2 and 3.
In preeclampsia women, normal nocturnal dipping was lost in 59.8% of patients, of whom 45.8% were nondippers, 10.3% were reversed dippers, and 3.7% were extreme dippers. In the control group, 51.4% were nondippers but still in the normal range of ABP values, and 11.4% were extreme dippers. We also calculated the salt sensitivity risk index.17 Indeed, 48.6% of preeclampsia women were categorized as high risk for salt sensitivity compared with only 17.1% of the control group (P<0.01).
We performed a secondary analysis restricted to white, non smoker, with a body mass index <25 kg/m2, and no past history of chronic hypertension and found similar results (Tables S1 and S2 in the online-only Data Supplement).
The main finding of this study is that women who had preeclampsia have a high prevalence of persistent diurnal and nocturnal ambulatory hypertension 6 to 12 weeks after delivery. In addition, preeclampsia is associated with an increased prevalence of postpartum white-coat and masked hypertension. The index of salt sensitivity risk and the variability of ambulatory BP as reflected by the SDs, a proxy of variability, were significantly increased in preeclampsia women compared to controls.
Preeclampsia is associated with an increased long-term risk of hypertension, cardiovascular and renal diseases, and diabetes mellitus.19 Recently, guidelines of the American Heart Association and of the European Societies of Hypertension and Cardiology emphasized the need for an appropriate postpartum screening and control of the cardiovascular risk factors for women with a history of hypertensive disorder of pregnancy.14,20 However, specific data on the management of the early follow-up after preeclampsia are lacking and ABPM in the postpartum period has never, to our knowledge, been evaluated. In clinical practice, ABPM has been shown to be a useful and cost effective tool for establishing true diagnosis of hypertension, as well as detecting masked and white-coat hypertension.14,21
Our study shows that ABPM detects a large diversity of BP phenotypes in the postpartum of preeclampsia women. Among women who had preeclampsia, 50% of them have a persistent ambulatory true hypertension when measured 6 to 12 weeks postpartum outside the office. Among the 28 patients under treatment, ABPM revealed that 14 (50%) of them were not controlled and allowed the interruption of 5 (17.9%) treatments.
Studies on hypertensive disorders of pregnancy have focused on antepartum period and unfortunately only few studies addressed the issue of hypertension in the postpartum.22 It is usually believed that many of the postpartum abnormalities of preeclampsia women progressively disappear.
In our study, 17.9% of women with preeclampsia presented with a white-coat hypertension. In some studies, white-coat hypertension was associated with a modest increased risk of stroke and target organ damage in the general population.23–25 In the particular condition of pregnancy, white-coat hypertension identified early during pregnancy has been associated with a slightly increased risk incidence of preeclampsia but with a better pregnancy outcome than pregnancy complicated by chronic hypertension.26,27 If the optimal management of white-coat hypertension is uncertain, especially in the postpartum period, its identification may prevent unnecessary use of antihypertensive medication.
We also found that 11.6% (13/112) of women who had preeclampsia with a normal office BP, had masked hypertension in the postpartum. Masked hypertension has been associated with a higher long-term cardiovascular risk and an increased rate of target organ damage similar to sustained hypertension.23,28–30 Albuminuria is an important cardiovascular risk factor and has been shown to be increased in subjects with masked hypertension.31–33 In our study, ABPM revealed a masked hypertension in 13.0% of patients with albuminuria and in 10.6% nonalbuminuric women (P=0.6). To the best of our knowledge, masked hypertension has never been described after preeclampsia, in particular, in the event of persisting albuminuria. This could represent another link between cardiovascular disease and preeclampsia.23,28,34 Considering the potential long-term consequences for the mother, masked hypertension in the postpartum needs closer attention.
Nondipping profile is also associated with an increased cardiovascular risk and target organ damage. Early studies on circadian variations of BP during preeclampsia showed the loss of the normal nocturnal fall and this was correlated to the severity of the disease.9 In a previous article, we observed the persistence of the nondipper status up to 10 years after preeclampsia.17 In the present study, we found that early after preeclampsia, 59.8% of women had an abnormal nocturnal dipping, and among them, 10.3% had a reversed circadian rhythm. The control group also displayed a nondipper pattern, however absolute nocturnal BP values were low (94.3±7.6/61.7±4.6 mm Hg in the control group versus 111.2±17.6/74.8±11.0 mm Hg in the preeclampsia group). Sleep disruption of the breastfeeding mother may explain the absence of nocturnal BP drop in both groups.
Salt sensitivity of BP is defined as an increase of 5% to 10% of office BP or of ≥4 mm Hg of the mean ambulatory BP after a high salt intake.35 Salt sensitive profile is a known cardiovascular and renal risk factor independent of hypertension and is associated with lack of nocturnal BP fall.36,37 We previously showed that women with a history of severe preeclampsia had an increased salt sensitivity of the ABP and a lack of nocturnal dipping compared with uncomplicated pregnancies, 10 years after the event.17 We also validated the formula derived from ABPM parameters to estimate the RSS of Castiglioni et al,16 in a group of preeclamptic women by assessing the salt sensitivity with a gold standard protocol. In the present study, 48.6% of women who had preeclampsia were classified as having a high RSS in the postpartum period versus 17.1% in the control group. Elevated nocturnal BP may result from a defect in diurnal salt excretory capacity of the kidneys, with a shift of the pressure natriuresis curve to the right at night.38 The kidney is an organ primarily damaged in preeclampsia, and consequently, the ability to maintain sodium balance may be altered in the postpartum period and later in life.17 This suggests that, in addition to lifestyle modifications, a low salt strategy may be beneficial to women who remain hypertensive after preeclampsia.
The increased short-term BP variability on 24-hour ABPM is known to be an independent risk factor for cardiovascular events and has been associated with arterial stiffness, stroke, and heart damage and salt sensitivity of the BP.39 Preeclamptic women in our work displayed a significantly higher SD of the ambulatory daytime and nighttime systolic and diastolic BP suggesting a higher BP variability after preeclampsia when compared with normal pregnancies.
Together, the high ambulatory BP, the lack of nocturnal dipping, and the increased salt sensitivity index, may contribute to the higher morbidity of preeclamptic women observed later in life. Although difficult to evaluate with certainty, a proportion of women with persisting postpartum hypertension might have had unrecognized preexisting hypertension superimposed by preeclampsia. Women who had preeclampsia also seem to share mechanisms usually evoked for BP variability and salt sensitivity with people with increased cardiovascular risk.
We have to acknowledge several limitations in our study. First is the lack of homogeneity in the ethnic origin of patients between groups, as well as for the body mass index, tobacco use, and past history of hypertension. These characteristics are known risk factors for preeclampsia,40,41 except for smoking, and are therefore more represented in the preeclampsia group. However, the secondary analysis excluding women with these potential confounders showed similar results (Tables S1 and S2). Second, the small size of our control group can be explained by a low rate of participation by women who had an uncomplicated pregnancy. As the follow-up was demanding (ABPM, urine and blood tests, medical visits), it was more difficult to motivate them to come back to the hospital for study purposes. However, even in the preeclampsia group, the rate of refusal was important (>50%). Therefore, we cannot exclude a selection bias affecting both groups, but as we showed a large and statistically significant difference, we think it is not necessary to increase the number of unaffected women. Another limitation may be the fact that parameters were measured only at 6 to 12 weeks and not later in the follow-up of preeclampsia women. Indeed, there is still a possibility that some of the BP phenotypes improve up to 6 months after pregnancy.
We present, to the best of our knowledge, the first study on the use of 24-hour ABPM in the early follow-up after preeclampsia. We demonstrate that 24-hour ABPM in the postpartum is useful to discriminate between fixed ambulatory hypertension, requiring close follow-up and possible treatment, from the white-coat hypertension. Moreover, it allows detection of masked hypertension, particularly in patients with persistent albuminuria and offers analysis of nondipper status after preeclampsia, pointing on possible salt sensitivity and renal damage.
The timing of assessment at 6 to 12 weeks postpartum is one of the strengths of our study. Indeed, even though hypertension may resolve during the first year following delivery, there is no reason for letting hypertensive women without care for >1 year. In our opinion, it is not justified to wait for complications or organ damages to happen before initiating effective preventive measures. International expert societies recommend careful follow-up of women with preeclampsia but clear guidelines on the modalities of this follow-up are still lacking.4,6,42 Recently, published works encourage identification of high-risk women early after preeclampsia.43,44 The group of Visser et al44 showed that among 187 women who had a hypertensive disorder of pregnancy, 40% had hypertension at 6 weeks postpartum. Of interest, of these women, 61% still had hypertension 2.5 years postpartum.
In conclusion, we show that 24-hour ABPM is a valuable tool in the follow-up of preeclampsia. It enables to discriminate women with fixed ambulatory hypertension from those with white-coat hypertension and to identify masked hypertension and nondipper status.
As these hypertensive phenotypes are associated with a higher risk for cardiovascular disease later in life, performing a 24-hour ABPM after preeclampsia may be useful to identify women in whom one should implement early lifestyle modifications to prevent the development of hypertension and its complications. The impact of these on future cardiovascular events remains of course to be determined in prospective trials.
In this study, we show that 24-hour ABPM is a valuable tool in the follow-up of preeclampsia. It discriminates between fixed ambulatory hypertension and white-coat hypertension and identifies masked hypertension and nondipper status. The use of 24-hour ABPM helps to identify patients at risk for cardiovascular disease later in life and allows early implementation of lifestyle modifications. Further studies are required to guide clinical management of cardiovascular prevention in this population.
We thank Sylvie Tremblay and Philippe Montillier for their help in data collection and Nicolas Fernandez for his linguistic revision.
Sources of Funding
A. Ditisheim received a grant from the Swiss National Science Foundation and the Swiss Research on Hypertension Foundation; A. Pechère-Bertschi received a grant from the Clinical Research Centre of the University Hospital of Geneva.
The online-only Data Supplement is available with this article at http://hyper.ahajournals.org/lookup/suppl/doi:10.1161/HYPERTENSIONAHA.117.09799/-/DC1.
- Received May 30, 2017.
- Revision received June 16, 2017.
- Accepted September 13, 2017.
- © 2017 American Heart Association, Inc.
- 5.↵National Institute for Health and Clinical Excellence: Guidance. Hypertension in pregnancy: the management of hypertensive disorders during pregnancy. London: RCOG Press; 2010.
- Mounier-Vehier C,
- Amar J,
- Boivin JM,
- Denolle T,
- Fauvel JP,
- Plu-Bureau G,
- Tsatsaris V,
- Blacher J
- Higashi Y,
- Oshima T,
- Ozono R,
- Nakano Y,
- Matsuura H,
- Kambe M,
- Kajiyama G
- Tranquilli AL,
- Dekker G,
- Magee L,
- Roberts J,
- Sibai BM,
- Steyn W,
- Zeeman GG,
- Brown MA
- Castiglioni P,
- Parati G,
- Brambilla L,
- Brambilla V,
- Gualerzi M,
- Di Rienzo M,
- Coruzzi P
- Martillotti G,
- Ditisheim A,
- Burnier M,
- Wagner G,
- Boulvain M,
- Irion O,
- Pechère-Bertschi A
- Bellamy L,
- Casas JP,
- Hingorani AD,
- Williams DJ
- Mosca L,
- Benjamin EJ,
- Berra K,
- et al
- 21.↵National Clinical Guideline Centre (UK). Hypertension: The Clinical Management of Primary Hypertension in Adults: Update of Clinical Guidelines 18 and 34 [Internet]. London: Royal College of Physicians (UK); 2011 Aug. https://www.ncbi.nlm.nih.gov/books/NBK83274/.
- Magee L,
- von Dadelszen P
- Mancia G,
- Bombelli M,
- Brambilla G,
- Facchetti R,
- Sega R,
- Toso E,
- Grassi G
- Gerstein HC,
- Mann JF,
- Yi Q,
- Zinman B,
- Dinneen SF,
- Hoogwerf B,
- Hallé JP,
- Young J,
- Rashkow A,
- Joyce C,
- Nawaz S,
- Yusuf S
- Ishikawa J,
- Hoshide S,
- Eguchi K,
- Schwartz JE,
- Pickering TG,
- Shimada K,
- Kario K
- Weinberger MH,
- Fineberg NS,
- Fineberg SE,
- Weinberger M
- Wolf M,
- Shah A,
- Jimenez-Kimble R,
- Sauk J,
- Ecker JL,
- Thadhani R
- Redman CW
- Levine LD,
- Nkonde-Price C,
- Limaye M,
- Srinivas SK
- Visser VS,
- Hermes W,
- Franx A,
- Koopmans CM,
- van Pampus MG,
- Mol BW,
- de Groot CJ
Novelty and Significance
What Is New?
The use of 24-hour ambulatory blood pressure (BP) monitoring compared with office BP after preeclampsia has never been studied.
What Is Relevant?
Masked hypertension is more frequent in women with a recent history of preeclampsia.
Twenty-four–hour ambulatory BP monitoring is a valuable follow-up tool for pregnancy related hypertensive disorders. It enables detection of women at high risk for cardiovascular disease later in life and to target subgroups of women for whom prevention has to be intensively implemented.
Hypertensive phenotypes have a high prevalence in the early postpartum period after preeclampsia. Postpartum 24-hour ambulatory BP monitoring helps to discriminate between fixed and white-coat hypertension, detects masked hypertension, and identifies nondipper status pointing on possible salt sensitivity pattern. The use of 24-hour ambulatory BP monitoring in the postpartum is a valuable follow-up tool for pregnancy related hypertensive disorders. It allows early identification of phenotypes of BP at risk and helps the management of high-risk women.