Ethnic Differences in Blood Pressure and Hypertensive Complications During PregnancyNovelty and Significance
The Generation R Study
The aim was to investigate ethnic differences in blood pressure levels in each trimester of pregnancy and the risk of gestational hypertensive disorders and the degree to which such differences can be explained by education and lifestyle-related factors. The study included 6215 women participating in a population-based prospective cohort study from early pregnancy onward in Rotterdam. Ethnicity was assessed at enrollment. Blood pressure was measured in each trimester. Information about gestational hypertensive disorders was available from medical charts. Lifestyle factors included smoking, alcohol, caffeine intake, folic acid supplementation, sodium and energy intake, body mass index, and maternal stress. Associations and explanatory pathways were investigated using linear and logistic regression analysis. Dutch pregnant women had higher systolic blood pressure levels as compared with women in other ethnic groups in each trimester of pregnancy. Compared with Dutch women, Turkish and Moroccan women had lower diastolic blood pressure levels in each trimester. These differences remained after adjusting for education and lifestyle factors. Turkish and Moroccan women had a lower risk of gestational hypertension as compared with Dutch women (odds ratio, 0.32 [95% CI, 0.18–0.58] and odds ratio, 0.28 [95% CI, 0.14–0.58]), and Cape Verdean women had an elevated risk of preeclampsia (odds ratio, 2.22 [95% CI, 1.22–4.07]). Differences could not be explained by education or lifestyle. Substantial ethnic differences were observed in blood pressure levels and risk of gestational hypertensive disorders in each trimester of pregnancy, and a wide range of variables could not explain these differences.
Hypertensive disorders during pregnancy complicate ≈7% of all pregnancies and are important causes of maternal and perinatal morbidity and mortality worldwide.1,2 In the Netherlands, eclampsia/preeclampsia is the leading cause of maternal mortality.3 The risk of gestational hypertension and preeclampsia has been demonstrated to differ by ethnic background.3,4 Black descent has been suggested as risk factor for pregnancy-related hypertensive disorders.4–6
Blood pressure levels during pregnancy are important risk factors for gestational hypertensive disorders.7 Little is known about variation in blood pressure development during pregnancy across different ethnic groups. One study found white pregnant women to have the highest blood pressure levels and women with West Indian decent to have the lowest blood pressure levels.8 Another study showed that Nigerian women had higher diastolic blood pressure (DBP) levels as compared with white women.9 The underlying pathways explaining the ethnic differences in blood pressure and gestational hypertensive disorders are largely unknown. Some of these ethnic differences may result from differences in socioeconomic position or in prevalence rates of cardiovascular risk factors, such as obesity, smoking, or physical inactivity.6,10
Ethnic minority groups of non-European origin now form ≈11% of the total population of the Netherlands.11 Studies from the United States, the United Kingdom, and the Netherlands found nonpregnant women from ethnic minority groups to have elevated blood pressure levels and more cardiovascular risk factors compared with white women.10,12,13 Therefore, we hypothesized women from ethnic minority groups to also have higher blood pressure levels during pregnancy and a higher risk of gestational hypertension and preeclampsia. In this study, we assessed the associations between ethnic background and systolic blood pressure (SBP) and DBP levels in each trimester of pregnancy and whether socioeconomic position and lifestyle-related determinants could explain these differences among 6215 pregnant women in Rotterdam. In addition, we examined whether ethnic background was associated with the risk of gestational hypertension and preeclampsia.
This study was embedded within the Generation R Study, a population-based prospective cohort study from early pregnancy onward. Details have been described elsewhere.14–16 Briefly, the cohort includes 9778 mothers and their children living in Rotterdam. All of the children were born between April 2002 and January 2006. Assessments during pregnancy included physical examinations, ultrasound assessments, and questionnaires and were planned in early pregnancy (gestational age <18 weeks), midpregnancy (gestational age 18–25 weeks), and late pregnancy (gestational age ≥25 weeks). The study was conducted in accordance with the guidelines proposed in the World Medical Association of Helsinki and has been approved by the Medical Ethical Committee of the Erasmus Medical Center, University Medical Centre Rotterdam. Written consent was obtained from all of the participating parents.17
Ethnicity was assessed by country of birth of the participating women and their parents and was obtained by questionnaires. Ethnicity was defined according to the classification of Statistics Netherlands. The participant was of non-Dutch origin if one of her parents was born in another country than the Netherlands.18 If both parents were born abroad, the country of the participant's mother decided on her ethnic background. Next, a distinction was made between women of Dutch ethnic background and the non-Dutch minority groups in this study, Turkish, Moroccan, Antillean, Surinamese, and Cape Verdean. Women with a Surinamese ethnicity were further classified into Surinamese-Hindustani, Surinamese-Creole, or Surinamese-other.
Population for Analysis
In total, 8880 women were enrolled during pregnancy. For the present study, we excluded women without information on ethnic background (n=678). Pregnant women with an “other-Western” and “other non-Western” ethnic background were excluded because of small numbers or mixed composition of these populations (n=1715). Of the remaining 6487 women, those without any blood pressure measurement (n=15) and with preexisting hypertension (n=114) were excluded. Furthermore, we excluded pregnancies leading to fetal death (n=66), loss to follow-up (n=9), and twin pregnancies (n=68), because main interest was in low-risk pregnancies. Thus, the cohort for analysis was composed of 6215 women of Dutch, Turkish, Moroccan, Surinamese-Creole, Surinamese-Hindustani, Antillean, and Cape Verdean ethnic backgrounds. A participant flowchart is given in the Figure S1 (available in the online-only Data Supplement).
Blood pressure was measured with the Omron 907 automated digital oscillometric sphygmanometer, which was validated in nonpregnant adults (OMRON Healthcare Europe BV, Hoofddorp, the Netherlands).19 All of the participants were seated in upright position with back support and were asked to relax for 5 minutes. A cuff was placed around the nondominant upper arm, which was supported at the level of the heart, with the bladder midline over the brachial artery pulsation. In case of an upper arm >33 cm, a larger cuff (32–42 cm) was used. The mean value of 2 blood pressure readings over a 60-second interval was documented for each participant.
Pregnancy-Induced Hypertension and Preeclampsia
Information on pregnancy complications was obtained from medical charts. Women suspected of pregnancy complications, based on these records, were crosschecked with the original hospital charts. Details of these procedures have been described elsewhere.14
Briefly, the following criteria were used to identify women with pregnancy-induced hypertension, development of SBP ≥140 mmHg and/or DBP ≥90 mmHg after 20 weeks of gestation in previously normotensive women. These criteria plus the presence of proteinuria (defined as ≥2 dipstick readings of ≥2+, 1 catheter sample reading of ≥1+, or a 24-hour urine collection containing ≥300 mg of protein) were used to identify women with preeclampsia.20
Most effects of ethnicity on blood pressure are probably indirect ones, acting through more proximal determinants of blood pressure.21 We considered the following factors to be such potential explanatory variables in the pathway between ethnicity and blood pressure.
Information on educational level, as one indicator of socioeconomic position, was obtained using questionnaires at enrollment. Highest completed educational level was classified into the following 4 categories: (1) low (no education, primary school, or lower vocational training); (2) midlow (>3 years general secondary school or intermediate vocational training); (3) midhigh (higher vocational training or Bachelor's degree); and (4) high (university degree).
Smoking, alcohol consumption, and caffeine intake were assessed by questionnaires in each trimester. From the first questionnaire, information about folic acid supplementation use was obtained. Prepregnancy weight was established at enrollment through a questionnaire. On the basis of height (in centimeters), measured at enrollment without shoes, and prepregnancy weight, we calculated prepregnancy body mass index (in weight/height2). Maternal distress was measured by a questionnaire in midpregnancy using the Brief Symptom Inventory, which gives a Global Severity Index. Higher Global Severity Index reflected pregnant women to experience more stress.
Daily sodium and energy intake during pregnancy were assessed at enrollment using a quantitative food frequency questionnaire.22 Portion sizes in grams per day were estimated using standardized household measures.23 To calculate average daily nutritional values, the Dutch food composition table 2006 was used.24
We treated maternal age at enrollment, parity, and gestational age at enrollment as potential confounders, because they cannot be considered indisputable explanatory variables.21,25 Parity was obtained through a questionnaire at enrollment. Gestational age was established by fetal ultrasound examination during the first ultrasound visit.15
First, the associations of ethnic background with repeatedly measured SBP and DBP were analyzed using unbalanced repeated-measurement regression analysis.26 These models take the correlation between repeated measurements of the same subject into account and allow for incomplete outcome data and are described in detail in the online-only Data Supplement. Second, we used multivariate linear regression analysis to study the associations of ethnic background with SBP and DBP in each trimester of pregnancy. To study the overall effect of ethnicity on blood pressure, we started with a model that included the potential confounders (model 1). Subsequently, the potential explanatory variables were added to the model, first separately and then simultaneously (adjusted model). Interaction terms between ethnic background and the explanatory variables were tested for significance. If the test was significant, we also stratified the analysis by the variables. Furthermore, we examined the associations of ethnic background with the risks of pregnancy-induced hypertension and preeclampsia using multiple logistic regression models. These models were also adjusted for confounders and the potential explanatory variables. Multiple imputation was used to deal with the missing values in the explaining covariates. Five imputed data sets were created and analyzed together. The repeated-measurement analysis was performed using the Statistical Analysis System version 9.2 (SAS, Institute Inc, Cary, NC). All of the other analyses were performed using the Statistical Package of Social Sciences version 17.0 for Windows (SPSS Inc, Chicago, IL).
Compared with Dutch women, women of the non-Dutch minority populations were younger, more frequently overweight, more frequently lower educated, less frequently consumed alcohol, less often used folic acid supplementation, and had a higher sodium intake (P<0.001; Table 1). Mean SBP levels were significantly higher in Dutch women compared with non-Dutch women in all 3 trimesters of pregnancy, except for Surinamese-Creole women in the first and second trimesters and for Antillean women in the first trimester. Mean DBP levels were significantly higher in Dutch women as compared with Turkish and Moroccan women, except for Turkish women in the first and second trimester. In total, there were 251 cases (4.2%) of pregnancy-induced hypertension and 120 cases (2.1%) of preeclampsia. The highest prevalence of pregnancy-induced hypertension was among Dutch and Surinamese-Creole women (5.2% and 5.5%) and for preeclampsia among Surinamese-Hindustani and Cape Verdean women (3.8% and 4.2%; P<0.001; Table 1).
Figure S2 shows the results of the repeated-measurement analyses of ethnic background and SBP and DBP patterns, respectively. SBP was highest among Dutch pregnant women compared with the non-Dutch pregnant women, except for Surinamese-Creole and Antillean women. In all of the ethnic groups, SBP increased throughout pregnancy (Figure S2A), but the lowest increase was among Surinamese-Creole women (P<0.05). For all of the ethnic groups, DBP showed a midpregnancy dip, with an increase afterward (Figure 2B). The lowest increase in DBP was observed in Moroccan women (P<0.05). The regression coefficients for gestational age–independent (intercept) and gestational age–dependent differences (interaction, ethnic background and gestational age) are given in Table S1.
Ethnicity, Blood Pressure, and Gestational Hypertensive Disorders
Age, parity, gestational age at intake, educational level, body mass index, smoking, alcohol use, caffeine intake, and folic acid use appeared to be significant factors associated with blood pressure levels during pregnancy (Table S2). The amount of women who drank >1 glass alcohol per day was 5.4%, 1.6%, and 1.3%, respectively, in first, second, and third trimesters.
Analysis per trimester showed that, as compared with Dutch women, SBP was lower in the non-Dutch minority groups in each trimester of pregnancy. Turkish and Moroccan pregnant women had lower DBP levels as compared with Dutch women. These differences slightly increased after adjusting for our explanatory variables, mainly after inclusion of education (Table 2).
Significant interaction terms were found between ethnic background and both parity and age in the association with blood pressure levels. Stratified analyses, according to parity and age, are shown in Table S3. In addition, the stratified models were adjusted for the other explanatory variables. This did not alter our previous findings that the association between ethnic background and blood pressure levels during pregnancy remained unexplained (data not shown).
Table 3 shows the associations of ethnic background with gestational hypertensive disorders. Cape Verdean and Surinamese-Hindustani women had a higher risk of preeclampsia as compared with Dutch women in the unadjusted model. In the adjusted model, this higher risk was only observed among Cape Verdean women (odds ratio, 2.06 [95% CI, 1.04–4.09]). As compared with Dutch women, Turkish and Moroccan women had a lower risk of pregnancy-induced hypertension (odds ratio, 0.32 [95% CI, 0.18–0.58]; odds ratio, 0.28 [95% CI, 0.14–0.58]), which did not attenuate after inclusion of the potential explanatory variables.
Our hypothesis that women from ethnic minority groups would have higher blood pressure levels during pregnancy as compared with Dutch women was not confirmed. In contrast, Dutch pregnant women had higher SBP levels than non-Dutch women in each trimester of pregnancy despite more favorable characteristics, such as a higher level of education, lower body mass index, and a lower sodium intake. Turkish and Moroccan women had lower DBP levels and a lower risk of pregnancy-induced hypertension as compared with Dutch women, which could also not be explained by education or lifestyle. Cape Verdean women had an increased risk of preeclampsia as compared with Dutch women.
The strengths of this study are the prospective population-based design and the availability of many important determinants that may explain the association among ethnic background, blood pressure levels, and pregnancy hypertensive complications. In addition, we had a large sample size with 6215 participants with 16615 blood pressure measurements and included the largest ethnic minority groups in the Netherlands.
To various extents, our results may have been influenced by the following limitations. We evaluated a different number of women in each trimester. Furthermore, women with treatment of gestational hypertensive disorders could also have influenced our results. Therefore, we repeated our analyses including only women with BP measured in all 3 trimesters (n=4478), and, second, we repeated the analyses excluding women with gestational hypertensive disorders. Essentially similar results were found as compared with the models with all of the women included (data not shown). The response at baseline for participation in the Generation R cohort was 61%. Pregnant women who participated were higher educated, more healthy, and more frequently of European origin than those who did not participate.16 This selective nonresponse may have resulted in biased effect estimates if the associations would be different between those included and not included in the analyses. Information on many covariates in this study was self-reported, which may have resulted in underreporting of certain adverse lifestyle-related determinants. Furthermore, because of the observational design, residual confounding attributed to unmeasured factors might still be an issue. Another possible limitation is that blood pressure varies during the day according to a circadian rhythm.27 We were unable to account for this, because our study did not include ambulatory blood pressure measurements. This probably introduced some random measurement error. The presence of systematic bias, however, is unlikely, because we do not assume that inaccurate measurements or the influence of the circadian rhythm on blood pressure change differed systematically by ethnic background. Finally, the food frequency questionnaire was used, which is only validated in nonpregnant white women.22 This could have led to an underestimation of the effect of energy and sodium intake on blood pressure levels during pregnancy.
Ethnicity, Blood Pressure, and Gestational Hypertensive Disorders
We observed substantial differences in blood pressure levels in each trimester of pregnancy and the risk of gestational hypertensive disorders between various ethnic groups. In contrast to our study, most studies found black women to have higher blood pressure levels and higher prevalence of hypertension than white women.9,10,12,13,28 These studies, however, were mostly about nonpregnant, older women or conducted in other countries with other cultures, lifestyles, and healthcare systems. In line with our study, another Dutch study among 2413 pregnant women found Dutch women to have the highest SBP levels in pregnancy and the lowest DBP levels for Mediterranean women.6 This may suggest blood pressure differences to reverse during pregnancy and to be higher among white women compared with other ethnic groups. One possible explanation might be that the maternal cardiovascular system of the various ethnic groups adapts differently during pregnancy because of genetic differences or different environmental factors.
Several studies identified black descent as a risk factor for gestational hypertensive disorders.4–6 In our study we found Cape Verdean women but not Surinamese-Creole and Antillean women to have a higher risk of preeclampsia. This is in contrast to the study of Knuist et al,6 who found Surinamese-Creole, Antillean, and West-African women to have a higher risk of preeclampsia. Previously, we have shown that DBP is more strongly associated with the risk of gestational hypertensive disorders and might be a better predictor of gestational hypertensive disorders than SBP.7 In line with these findings, we observed in this study that, although Dutch women had the highest SBP levels, they did not have the highest risk of preeclampsia or pregnancy-induced hypertension. In addition, we found low DBP to be associated with a lower risk of pregnancy-induced hypertension. Several studies described a fall in BP in midpregnancy.27 We observed a midpregnancy dip in DBP but not in SBP. In line with our findings, the study of Nama et al29 also did not find this midpregnancy dip in SBP.
The ethnic blood pressure differences and difference in risks of pregnancy-hypertensive complications found in our study remained largely unexplained, although we included a wide range of potential explanatory variables. It has been suggested that gestational diabetes might partly explain the difference in the risk of gestational hypertensive disorders between ethnic groups. Some studies found that gestational diabetes increased the risk of preeclampsia and that this effect was stronger among black women.4 In our study, there were no significant differences in the prevalence of gestational diabetes per ethnic background, and the effects of gestational diabetes on blood pressure and hypertensive complications were the same among the various ethnic groups (data not shown). Other potential determinants that were not available for the current study, such as physical activity, metabolic factors (eg, cholesterol and fatty acid levels), parameters of endothelial function, genetic factors, or currently unknown risk factors, may contribute to the explanation.30–32 Of note, the percentage of Dutch women using folic acid before pregnancy was higher than the percentage of women who used folic acid during pregnancy. A possible explanation could be a lack of knowledge about adequate folic acid use. Physicians and midwives should be aware of this and should provide information to these women.
In our fully adjusted model we found multiparity, high educational level, and continued alcohol use during pregnancy to be significantly associated with lower blood pressure levels. Most women who used alcohol during pregnancy were low-to-moderate alcohol drinkers (<1 glass of alcohol per day). Previous research also showed that moderate alcohol drinkers had lower BP levels compared with nondrinkers and heavy drinkers.33 However, alcohol use during pregnancy might harm the unborn child and is, therefore, not recommended. Higher body mass index was significantly associated with higher blood pressure levels. These findings are in line with other studies, except for the association between parity and blood pressure, where findings are inconsistent.34,35 We found significant interaction terms between ethnicity and parity. Thus, the effect of parity varied among the different ethnic groups. Repeated pregnancies may negatively influence the risk for cardiovascular disease in the long term36; our findings suggest that this may differ per ethnic group. Furthermore, we observed that older maternal age was associated with a lower SBP among all ethnic groups, but no association was found between maternal age and DBP. However, the effect of maternal age on DBP differed per ethnic group, because Turkish and Moroccan women <35 years of age were observed to have lower DBP levels as compared with Dutch women of the same age.
Our study demonstrated substantial differences in blood pressure levels and pregnancy hypertensive disorders in a multiethnic society. Remarkably, these differences remain largely unexplained, although we included a wide range of know risk factors. Underlying mechanisms for these ethnic differences in blood pressure levels and gestational hypertensive disorders need to be identified, and genes may be considered to play a role. Understanding of these ethnic disparities may lead to entry points for prevention and treatment and ultimately improved maternal and fetal pregnancy outcomes.
Sources of Funding
The first phase of the Generation R Study is made possible by financial support from Erasmus Medical Centre, Erasmus University Rotterdam, and the Netherlands Organization for Health Research and Development.
The Generation R Study is conducted by the Erasmus Medical Centre in close collaboration with the Erasmus University Rotterdam School of Law and Faculty of Social Sciences, the Municipal Health Service Rotterdam area, the Rotterdam Homecare Foundation, and the Stichting Trombosedienst and Artsenlaboratorium Rijnmond (Rotterdam). We gratefully acknowledge the contribution of general practitioners, hospitals, midwives, and pharmacies in Rotterdam.
The online-only Data Supplement is available with this article at http://hyper.ahajournals.org/lookup/suppl/doi:10.1161/HYPERTENSIONAHA.112.194365/-/DC1.
- Received March 7, 2012.
- Revision received March 24, 2012.
- Accepted April 24, 2012.
- © 2012 American Heart Association, Inc.
- Buchbinder A,
- Sibai BM,
- Caritis S,
- Macpherson C,
- Hauth J,
- Lindheimer MD,
- Klebanoff M,
- Vandorsten P,
- Landon M,
- Paul R,
- Miodovnik M,
- Meis P,
- Thurnau G
- Poon LC,
- Kametas NA,
- Pandeva I,
- Valencia C,
- Nicolaides KH
- Gaillard R,
- Bakker R,
- Willemsen SP,
- Hofman A,
- Steegers EA,
- Jaddoe VW
Bevolking; kerncijfers. Centraal Bureau voor de Statistiek. http://statline.cbs.nl/StatWeb/publication/?VW=T&DM=SLNL&PA=37296ned&D1=a&D2=0,10,20,30,40,50,(l-1)-l&HD=111124–1422&HDR=G1&STB=T. Updated October 18, 2011. Accessed May 9, 2012.
- Agyemang C,
- Bindraban N,
- Mairuhu G,
- Montfrans G,
- Koopmans R,
- Stronks K,
- Group SS
- Jaddoe VW,
- Bakker R,
- van Duijn CM,
- van der Heijden AJ,
- Lindemans J,
- Mackenbach JP,
- Moll HA,
- Steegers EA,
- Tiemeier H,
- Uitterlinden AG,
- Verhulst FC,
- Hofman A
Statistics Netherlands. Migrants in the Netherlands 2004 (Allochtonen in Nederland 2004). Voorburg/Heerlen, The Netherlands: Centraal Bureau voor de Statistiek; 2004.
- Donders-Engelen M,
- Heijden van der L,
- Hulshof KF
Netherlands Nutrition Center. Nevo: Dutch Food Composition Database 2006. The Hague: the Netherlands: Netherlands Nutrition Centre; 2006.
- McNamee R
- Goldstein H,
- Goldstein H
- Hermida RC,
- Ayala DE,
- Mojon A,
- Fernandez JR,
- Alonso I,
- Silva I,
- Ucieda R,
- Iglesias M
- Chaturvedi N,
- McKeigue PM,
- Marmot MG
- Solomon CG,
- Carroll JS,
- Okamura K,
- Graves SW,
- Seely EW
- Marcoux S,
- Brisson J,
- Fabia J
- Hansel B,
- Thomas F,
- Pannier B,
- Bean K,
- Kontush A,
- Chapman MJ,
- Guize L,
- Bruckert E
- Erem C,
- Hacihasanoglu A,
- Kocak M,
- Deger O,
- Topbas M
- Dhawan V,
- Brookes ZL,
- Kaufman S
Novelty and Significance
What Is New?
We investigated differences in blood pressure levels in pregnancy among various ethnic groups in Europe.
We looked at the availability of many important determinants that may explain the association among ethnic background, blood pressure levels, and pregnancy-hypertensive complications.
What Is Relevant?
Blood pressure levels during pregnancy are important risk factors for gestational hypertensive disorders.
Low DBP was found to be associated with a lower risk of pregnancy-induced hypertension.
Substantial ethnic differences were found. Dutch women had the highest SBP levels in pregnancy. These ethnic differences in blood pressure levels remained largely unexplained by known risk factors, and genetic background may be considered to play a role.