Suboptimal maternal and fetal nutrition might lead to fetal cardiovascular developmental adaptations and subsequent CVD in later life(Reference Barker1–Reference Jaddoe3). Support for this hypothesis is largely based on experimental animal studies and historical cohort studies in human subjects showing associations of maternal exposure to extreme famine during pregnancy with the development of hypertension in later life(Reference Roseboom, van der Meulen and Ravelli4–Reference Ozaki, Nishina and Hanson6). An imbalance in nutrient intake during pregnancy might also be an explanation for the development of hypertension(Reference Roseboom, van der Meulen and Ravelli4). Not much is known about the associations of less extreme variations in maternal dietary intake with CVD in the offspring. Several studies have focused on the associations of maternal micro- or macronutrient intake with childhood blood pressure but the results are inconclusive(Reference Aaltonen, Ojala and Laitinen7–Reference McGarvey, Zinner and Willett17). For example, two previous studies suggested that higher maternal fat intake during pregnancy was associated with diastolic blood pressure in the offspring, although the studies observed opposite effects(Reference Aaltonen, Ojala and Laitinen7, Reference Adair, Kuzawa and Borja8). Also, maternal intake of micronutrients during pregnancy, such as Ca, Na and Fe, has been suggested to be associated with blood pressure in children, but results were not consistent(Reference Bakker, Rifas-Shiman and Kleinman9–Reference Belizan, Villar and Bergel11, Reference Brion, Leary and Smith13–Reference Koleganova, Piecha and Ritz15). Other micronutrients, including folate, homocysteine and vitamin B12, might affect vascular development(Reference Oosterbaan, Steegers and Ursem18). High homocysteine and low folate levels are associated with endothelial dysfunction(Reference Martin, Lindblad and Norman19, Reference Genser, Prachar and Hauer20) and might subsequently lead to higher blood pressure in later life. Since vitamin B12 lowers homocysteine levels, this might also influence blood pressure.
We assessed in a large population-based prospective cohort study among 2863 Dutch mothers and their children the associations of maternal daily total energy intake and intake of macronutrients and micronutrients in the first trimester of pregnancy with blood pressure in children at the age of 6 years. We also assessed whether first-trimester folate, homocysteine and vitamin B12 concentrations in maternal blood were associated with childhood blood pressure.
Subjects and methods
Design
The present study was embedded in the Generation R Study, a population-based prospective cohort study from fetal life onwards in Rotterdam, The Netherlands(Reference Jaddoe, Bakker and van Duijn21, Reference Jaddoe, van Duijn and van der Heijden22). Enrolment in the study was aimed at early pregnancy, but was allowed until the birth of the child. Information about maternal diet and other lifestyle-related variables during pregnancy was collected at enrolment. At the age of 6 years, all participating children and their mothers were invited to a dedicated research centre, to participate in detailed hands-on measurements. The study has been approved by the Medical Ethics Committee of the Erasmus Medical Center, Rotterdam. Written informed consent was obtained from all parents of the participants. The present analysis was limited to Dutch mothers, since the FFQ was validated for the assessment of dietary intake in a Dutch population. We selected mothers who were enrolled in the study before a gestational age of 25 weeks, because we aimed to assess dietary intakes in the first trimester of pregnancy. In total, 4032 Dutch mothers were enrolled before a gestational age of 25 weeks with a median of 13·6 weeks of gestation (90 % range 11·5–20·2 weeks; see Fig. 1). Data on maternal dietary intake or biomarker concentrations were available in 3960 (98 %) of these mothers. We excluded multiple births (n 104) and stillbirths (n 29) from the analyses, leaving 3827 Dutch mothers. Of the 3827 singleton live-born children, 2949 (77 %) children attended the follow-up visit at the age of 6 years. Blood pressure was successfully measured in 2863 (97 %) of the children who visited the research centre.
Fig. 1 Flow chart of the participants included for analysis.
Maternal daily dietary intake
We assessed maternal dietary intake at enrolment in the study in Dutch mothers using a modified version of the validated semi-quantitative FFQ of Klipstein-Grobusch et al. (Reference Klipstein-Grobusch, den Breeijen and Goldbohm23). The FFQ considered food intake over the prior 3 months, thereby covering the dietary intake in the first trimester of pregnancy. The FFQ consisted of 293 items structured according to the meal pattern. Questions included consumption frequency, portion size, preparation method and additions. Portion sizes were estimated using Dutch household measures and photographs of foods showing different portion sizes(Reference Donders-Engelen, van der Heijden and Hulshof24). We used the Dutch food composition table for calculating daily intake of nutritional values(25).
Folic acid supplement intake
Information on folic acid supplement use (0·4–0·5 mg) and the initiation of supplementation was obtained by questionnaires at the enrolment of the study. We categorised folic acid supplement use into three groups: (1) periconceptional use; (2) start when pregnancy was known; (3) no use during pregnancy. Self-reported folic acid use was validated in a subgroup by serum folate levels in the first trimester, i.e. before 12 weeks of gestation. Within the group of mothers who reported using folic acid supplements (n 204), the median of serum folate was 23·5 (range 4·3–45·3) nmol/l, whereas the median serum folate concentration of mothers who did not report folic acid supplement use (n 68) was 11·1 (range 4·7–29·6) nmol/l. The difference in distribution function (Mann–Whitney test) was statistically significant (P< 0·001)(Reference Timmermans, Jaddoe and Hofman26). Information about folic acid supplement use was available in 2505 subjects (87·5 %).
Folate, homocysteine and vitamin B12 concentrations
In early pregnancy (median 12·9 weeks of gestation, 90 % range 10·6–16·8), venous samples were drawn and stored at room temperature before being transported to the regional laboratory for processing and storage for future studies. Processing was aimed to be completed within a maximum of 3 h after venous puncture. The samples were centrifuged and thereafter stored at − 80°C(Reference Jaddoe, Bakker and van Duijn21). To analyse folate, homocysteine and vitamin B12 concentrations, EDTA plasma samples (folate and homocysteine) and serum samples (vitamin B12) were picked and transported to the Department of Clinical Chemistry at the Erasmus University Medical Center, Rotterdam in 2008. Folate, homocysteine and vitamin B12 concentrations were analysed using an immunoelectrochemoluminence assay on the Architect System (Abbott Diagnostics B.V.). The between-run CV for plasma folate were 8·9 % at 5·6 nmol/l, 2·5 % at 16·6 nmol/l and 1·5 % at 33·6 nmol/l, with an analytic range of 1·8–45·3 nmol/l. The same CV for plasma homocysteine were 3·1 % at 7·2 μmol/l, 3·1 % at 12·9 μmol/l and 2·1 % at 26·1 μmol/l, with an analytic range of 1–50 μmol/l. This CV for serum vitamin B12 was 3·6 % at 142 pmol/l, 7·5 % at 308 pmol/l and 3·1 % at 633 pmol/l, with an analytic range of 44–1476 pmol/l. Plasma concentrations of maternal folate, homocysteine and serum concentrations of vitamin B12 in the first trimester of pregnancy were available in 2305 (80·5 %) of the mothers of the children included in the present study. Missing data were mainly due to logistical reasons.
Blood pressure measurements
Blood pressure measurements in children were conducted around the age of 6 years in a dedicated research centre in the Erasmus Medical Center, Rotterdam, The Netherlands. The child was lying quietly, while systolic blood pressure and diastolic blood pressure were measured at the right brachial artery in a supine position, four times with 1 min intervals. A cuff was selected with a cuff width approximately 40 % of the arm circumference and long enough to cover 90 % of the arm circumference. We used the validated automatic sphygmomanometer Datascope Accutorr Plus™ (Mindray DS USA, Inc.)(Reference Wong, Tz Sung and Leung27). Of all children visiting the research centre, 91·3 % had four successful blood pressure measurements available.
Covariates
Information on maternal age, pre-pregnancy BMI, parity, alcohol use and smoking habits during pregnancy, and educational level was obtained from questionnaires. Maternal education was defined as highest followed education according to the classification of Statistics Netherlands and categorised into primary, secondary and higher(28). Child sex, gestational age at birth and birth weight were obtained from midwife and hospital registries. Breast-feeding (ever/never) was assessed using questionnaires. Current height and weight were measured without shoes and heavy clothing at the visit at 6 years, and BMI (kg/m2) was calculated.
Statistical methods
Differences in child characteristics at the age of 6 years between boys and girls were assessed using the t test and χ2 tests for independent samples. Maternal dietary intake variables were categorised into quintiles. This approach was used for all dietary exposures (total energy, carbohydrate, fat, protein intake and protein:carbohydrate ratio; Ca, Fe and Na intake; folate, homocysteine and vitamin B12 concentrations). We used mixed models to assess the associations between predictors and blood pressure(Reference Laird and Ware29). The mixed-model method fits each of the as many as four blood pressure measurements of every child as repeated outcome measures. An advantage of this modelling approach over using the average measure of blood pressure for each child as an outcome is that children with more measurements and less variability in their measurements are assigned more weight than those with fewer measurements, more variability or both(Reference Bakker, Rifas-Shiman and Kleinman9, Reference Gillman and Cook30). We used similar mixed models to assess the association of folic acid supplement use with blood pressure at the age of 6 years. All analyses were adjusted for child's sex and age at blood pressure measurement (crude model). Potential covariates were selected based on previous literature(Reference Lawlor, Najman and Sterne31, Reference Lawlor and Smith32). We assessed crude associations, adjusted for age and sex, of possible covariates with childhood blood pressure. Only the covariates that were significantly associated with systolic or diastolic blood pressure in the present study population were included in our fully adjusted model. The fully adjusted model included maternal age, pre-pregnancy BMI, alcohol use and smoking during pregnancy, educational level, gestational age at birth, birth weight, current BMI and month in which blood pressure measurement was taken. Tests for trends were conducted by using maternal dietary intake variables as the continuous variable in the linear mixed models. We provided the effect per standard deviation increase in the dietary intake variable. In the macronutrient analyses, we used the energy partition method to adjust for energy intake of the other macronutrients, since total energy intake and macronutrient intakes were strongly correlated(33). The actual energy derived from the macronutrients was used in the analyses. In the micronutrient analyses, we used the residual nutrient method, and additionally adjusted for total energy intake(33).
Missing values in covariates (ranging from 0 to 15 %) were multiple-imputed to reduce potential bias associated with missing data(Reference Sterne, White and Carlin34). We created five imputed datasets and each dataset was analysed separately to obtain the effect sizes and standard errors. The results of all the five imputed analyses were pooled and are presented in the present study. We investigated maternal macronutrient intake, micronutrient intake and maternal biomarker concentrations in pregnancy. Within these exposure groups, variables are correlated (Table S1, available online). To take into account multiple testing, we applied a Bonferroni correction and considered a P value lower than 0·017 (0·05/3 (three exposure groups)) as statistically significant. The mixed models were fitted using the SAS version 9.2 (SAS Institute, Inc.). All other statistical analyses were performed using SPSS version 17.0 for Windows (SPSS, Inc.).
Results
Table 1 presents the maternal and birth subject characteristics. The mean intake of total energy was 8982·8 (sd 2117·3) kJ/d, Ca 1222 (sd 420) mg/d, Fe 12·1 (sd 3·3) mg/d and Na 3314 (sd 931) mg/d. The mean birth weight was 3497 (sd 548) g. Child characteristics at the age of 6 years for boys and girls separately are shown in Table 2. Systolic blood pressure and diastolic blood pressure were significantly higher in girls compared with boys (P< 0·05).
Table 1 Maternal and birth characteristics: the Generation R Study Cohort, Rotterdam, The Netherlands* (Mean values and standard deviations; medians, 95 % ranges and percentages)
* Only covariates with missing data were multiple-imputed.
Table 2 Child characteristics: the Generation R Study Cohort, Rotterdam, The Netherlands† (Mean values and standard deviations; medians, 95 % ranges and percentages)
* Mean values were significantly different from those of boys (P< 0·05).
† Only covariates with missing data were multiple-imputed.
Maternal total daily energy intake and daily intake of carbohydrates, fat and proteins were not associated with childhood systolic and diastolic blood pressure (Table 3). The protein:carbohydrate intake ratio was not associated with childhood blood pressure (Table 3). After adjustment for confounders, maternal intake of Fe tended to be inversely associated with childhood systolic blood pressure ( − 0·30 mmHg per standard deviation increase in Fe intake (95 % CI − 0·61, 0·01), P for trend 0·06), but not with diastolic blood pressure (Table 4). Maternal intake of Ca and Na was not consequently associated with childhood systolic or diastolic blood pressure.
Table 3 Associations of maternal macronutrient intake during pregnancy with blood pressure in children at the age of 6 years in a Dutch population† (Regression coefficients and 95 % confidence intervals)
* P< 0·05.
† Values reflect the difference in blood pressure (mmHg) at the age of 6 years for each quintile of macronutrient intake (kJ derived from macronutrients). The crude model is adjusted for child sex and age at blood pressure measurement. The adjusted model is additionally adjusted for maternal age, pre-pregnancy BMI, alcohol use and smoking during pregnancy, educational level, gestational age at birth, birth weight, current BMI and month in which blood pressure measurement was taken and additionally adjusted for energy from the other macronutrients following the energy partition method(Reference Barker1).
‡ Tests for trends were conducted using the maternal dietary intake variables as the continuous variable in the linear mixed models. Values reflect the difference in blood pressure (mmHg) per standard deviation increase in macronutrient intake.
Table 4 Associations of maternal micronutrient intake during pregnancy with blood pressure at the age of 6 years in a Dutch population† (Regression coefficients and 95 % confidence intervals)
* P< 0·05, **P< 0·01.
† Values reflect the difference in blood pressure (mmHg) at the age of 6 years for each quintile of micronutrient intake. The crude model is adjusted for child sex and age at blood pressure measurement. The adjusted model is additionally adjusted for maternal age, pre-pregnancy BMI, alcohol use and smoking during pregnancy, educational level, gestational age at birth, birth weight, current BMI and month in which blood pressure measurement was taken.
‡ Tests for trends were conducted using the maternal dietary intake variables as the continuous variable in the linear mixed models. Values reflect the difference in blood pressure (mmHg) per standard deviation increase in micronutrient intake.
We did not find an association of folic acid supplement use during pregnancy with childhood systolic or diastolic blood pressure (Table 5). The associations of maternal first-trimester folate, homocysteine and vitamin B12 concentrations with childhood blood pressure are presented in Table 6. We found no consistent associations of maternal first-trimester folate and homocysteine concentrations with childhood systolic blood pressure and diastolic blood pressure in 6-year-old children. Lower maternal first-trimester vitamin B12 concentrations were associated with lower childhood diastolic blood pressure (difference 0·31 mmHg per standard deviation increase in vitamin B12 concentration (95 % CI 0·06, 0·56), P value for trend 0·02), but not with systolic blood pressure. However, the associations between Fe intake and vitamin B12 concentrations with childhood blood pressure did not reach the significance threshold after adjustment for multiple testing.
Table 5 Associations of maternal folic acid supplement use during pregnancy with blood pressure at the age of 6 years in a Dutch population* (Regression coefficients and 95 % confidence intervals)
* Values reflect the difference in blood pressure (mmHg) at the age of 6 years for each group of maternal folic acid supplement use. The crude model is adjusted for child sex and age at blood pressure measurement. The adjusted model is additionally adjusted for maternal age, pre-pregnancy BMI, alcohol use and smoking during pregnancy, educational level, gestational age at birth, birth weight, current BMI and month in which blood pressure measurement was taken.
Table 6 Associations of maternal first-trimester folate, homocysteine and vitamin B12 levels with blood pressure at the age of 6 years in a Dutch population† (Regression coefficients and 95 % confidence intervals)
* P< 0·05.
† Values reflect the difference in childhood blood pressure (mmHg) at the age of 6 years for each quintile of biomarker concentrations. The highest quintile in maternal folate and vitamin B12 levels and the lowest quintile in homocysteine levels were used as reference groups. The crude model is adjusted for child sex and age at blood pressure measurement. The adjusted model is additionally adjusted for maternal age, pre-pregnancy BMI, alcohol use and smoking during pregnancy, educational level, gestational age at birth, birth weight, current BMI and month in which blood pressure measurement was taken.
‡ Tests for trends were conducted using the maternal dietary intake variables as the continuous variable in the linear mixed models. Values reflect the difference in blood pressure (mmHg) per standard deviation increase in micronutrient intake.
Discussion
The results from the present population-based prospective cohort study suggest that maternal daily macronutrient and micronutrient intake in the first trimester of pregnancy is not associated with childhood blood pressure at the age of 6 years. We observed that higher Fe intake and vitamin B12 concentrations may be associated with a lower childhood systolic and higher diastolic blood pressure, respectively, but we should also consider the effect of multiple testing. After multiple testing correction, none of the results remained significant.
Strengths and limitations
The main strength of the present study is the prospective design from early life onwards and the large sample size of this population-based cohort. To our knowledge, this is one of the largest prospective studies that examined the associations between the daily maternal intake of micronutrients and macronutrients in the first trimester of pregnancy with childhood blood pressure. We used a FFQ, previously validated in a Dutch population(Reference Klipstein-Grobusch, den Breeijen and Goldbohm23), to assess dietary intake of the mothers. Although we have used a validated questionnaire, misclassification of dietary intake can still occur, which might have led to an underestimation of the effect estimates. We measured blood pressure in 6-year-old children using a validated automatic sphygmanometer(Reference Wong, Tz Sung and Leung27) and acquired multiple blood pressure measurements to minimise measurement error. Furthermore, information about a large number of potential confounders was available. However, some limitations need to be addressed. Of all mothers included before a gestational age of 25 weeks, information on maternal dietary intake was missing for only 1·4 % of all mothers. This non-response would lead to biased effect estimates if the associations of maternal dietary intake and childhood blood pressure would be different between mothers included and not included in the analyses. This seems unlikely because biased estimates in large cohort studies mainly arise from loss to follow-up rather than from a non-response at baseline(Reference Nohr, Frydenberg and Henriksen35). Of all Dutch singleton live-born children with available data on maternal dietary intake during the first trimester of pregnancy, 74·3 % participated in the follow-up measurements at the age of 6 years. Overall, mothers who did not visit the research centre were younger, less well educated, smoked more frequently and used less alcohol during pregnancy. This selective loss to follow-up might have led to biased effect estimates. We also have to acknowledge that the present study population included in the analyses is comprised of relatively healthy women with a higher proportion of folic acid supplement use compared with other populations. This might have resulted in smaller observed differences and might limit generalisability to other populations. Also, maternal dietary intake was only assessed in the first trimester of pregnancy, while dietary intake might change during pregnancy. Although it has been demonstrated that maternal nutritional intake did not change significantly during pregnancy(Reference Cuco, Fernandez-Ballart and Sala36), second- or third-trimester maternal diet might be associated with cardiovascular development, and influence childhood blood pressure. Unfortunately, we were not able to assess these associations in the present study. Finally, although we have performed adjustment for various potential confounders, residual confounding might still be an issue due to the observational design of the study.
Maternal macronutrient intake
Low birth weight is associated with higher blood pressure and CVD in later life(Reference Huxley, Shiell and Law37, Reference Eriksson, Forsen and Tuomilehto38). It has been suggested that these associations might be explained by suboptimal maternal diet(Reference Barker1–Reference Jaddoe3). Historical cohort studies in The Netherlands and China showed that maternal exposure to famine is associated with a higher blood pressure in adult offspring(Reference Roseboom, van der Meulen and Ravelli4, Reference Li, Jaddoe and Qi39). The effect of the normal variation in maternal daily energy intake and macronutrient intake on childhood blood pressure has been examined in several studies, but showed inconsistent results(Reference Aaltonen, Ojala and Laitinen7, Reference Adair, Kuzawa and Borja8, Reference Leary, Ness and Emmett16). Consistent with previous studies, we did not observe associations of maternal total energy intake during pregnancy and childhood blood pressure(Reference Adair, Kuzawa and Borja8, Reference Roseboom, van der Meulen and van Montfrans40). As suggested by Roseboom et al. (Reference Roseboom, van der Meulen and Ravelli4), it might be that an imbalance of maternal macronutrients, rather than total energy intake, is associated with childhood blood pressure. Using the energy partition method(33), we did not observe associations of maternal daily carbohydrate, fat or protein intake with childhood blood pressure. Previously, two smaller studies from Finland and the Philippines showed conflicting results regarding the associations of maternal fat intake with childhood blood pressure(Reference Aaltonen, Ojala and Laitinen7, Reference Adair, Kuzawa and Borja8). The present results suggest that there is no association of maternal fat intake with childhood blood pressure. Also, a previous study showed that a higher maternal protein intake is associated with a lower blood pressure in boys, but these findings were not confirmed in other studies(Reference Adair, Kuzawa and Borja8, Reference Roseboom, van der Meulen and van Montfrans40, Reference Webb, Conlisk and Barnhart41). The inconsistent results might be explained by differences in study populations.
Maternal calcium, iron and sodium intake
Several studies, including follow-up studies of randomised clinical trials, have shown an association of Ca supplement use during pregnancy with a lower blood pressure in children(Reference Belizan, Villar and Bergel11, Reference Gillman, Rifas-Shiman and Kleinman14). However, maternal Ca intake from normal dietary intake was not associated with infant and childhood blood pressure(Reference Bakker, Rifas-Shiman and Kleinman9, Reference Gillman, Rifas-Shiman and Kleinman14). In line with these studies, we did not find an association between maternal Ca intake in the first trimester of pregnancy and offspring blood pressure. It should be noted that our population had a high dietary Ca intake. This could have affected our power to investigate the effects of low Ca intake on childhood blood pressure. Another explanation might be that mothers who used Ca supplements also used other micronutrient supplements, and the combined effect of these supplements might lead to a lower blood pressure in the offspring(Reference Gillman, Rifas-Shiman and Kleinman14).
Animal studies have shown that maternal Fe restriction during pregnancy was associated with a higher blood pressure in the offspring(Reference Lewis, Petry and Ozanne42, Reference Gambling, Dunford and Wallace43). In human subjects, Brion et al. (Reference Brion, Leary and Smith13) observed in a prospective cohort study among 7484 subjects that maternal Fe supplement use and Fe intake from food sources, assessed in the third trimester of pregnancy, were not associated with offspring blood pressure. Belfort et al. (Reference Belfort, Rifas-Shiman and Rich-Edwards10) showed in a prospective longitudinal cohort study among 1098 American children an association of maternal Fe supplement use with a higher systolic blood pressure in children. However, maternal Fe dietary intake from food sources, assessed in the first and second trimesters of pregnancy, was not associated with offspring blood pressure at both time points(Reference Belfort, Rifas-Shiman and Rich-Edwards10). We observed an association of higher maternal Fe intake from food with a lower blood pressure, although the effect size was small and borderline significant. We assessed dietary intake in the first trimester of pregnancy. It might be that part of the differences in results between the present study and other studies can be explained by different timings of the exposure assessment. Animal studies have shown that offspring of Fe-restricted mothers have a higher blood pressure in later life, possibly mediated by cardiovascular adaptations in response to anaemia, which is in line with the present findings(Reference Lewis, Petry and Ozanne42, Reference Crowe, Dandekar and Fox44). Previous studies in animals described the associations of maternal Na intake with childhood blood pressure and suggested that high Na intake was associated with a higher blood pressure(Reference Koleganova, Piecha and Ritz15, Reference Porter, King and Honeycutt45). In the present study, maternal Na intake was not associated with childhood blood pressure. However, Na intake is poorly assessed by the FFQ, which could have led to an underestimation of the effect. Further experimental studies in animals and observational studies in human subjects are needed to assess the associations of maternal micronutrient intake with offspring blood pressure in later life, and to investigate the possible underlying mechanisms.
Maternal folate, homocysteine and vitamin B12 concentrations
Low folate and high homocysteine concentrations during pregnancy have been associated with lower birth weight and pregnancy complications, such as spontaneous abortion and pre-eclampsia(Reference George, Mills and Johansson46–Reference Ray and Laskin49). Low folate and vitamin B12 levels and high homocysteine levels might affect vascular development and subsequently lead to endothelial dysfunction and higher blood pressure(Reference Oosterbaan, Steegers and Ursem18). A previous study described the associations of supplement intake, including Fe, folic acid and vitamin B12, during pregnancy with a lower systolic blood pressure in children at the age of 2 years(Reference Vaidya, Saville and Shrestha50). However, since the supplement contained a combination of these nutrients, they could not infer which nutrient contributed to the effect.
We did not observe the associations of maternal folate and homocysteine levels during the first trimester of pregnancy with childhood blood pressure in their offspring. Since higher vitamin B12 levels are associated with lower homocysteine levels, and lower homocysteine levels are associated with a lower blood pressure in adults, we expected that higher maternal vitamin B12 concentrations in the blood would be associated with a lower childhood blood pressure. However, the present results show an opposite effect; higher maternal vitamin B12 levels were associated with a higher childhood diastolic blood pressure, although the effect size was small and borderline significant. However, after correction for multiple testing, this association was not statistically significant. To the best of our knowledge, this association has not been described before and the underlying mechanisms are not known.
Conclusions
The present study shows within a population-based cohort that the normal variation in the maternal intake of macronutrients and micronutrients during the first trimester of pregnancy is not associated with childhood blood pressure. We found some indications that high maternal Fe intake and low maternal vitamin B12 levels seemed to be associated with a lower blood pressure in children at the age of 6 years. However, we investigated multiple exposures, the effect sizes are small and the associations were not significant after correction for multiple testing. Further studies are needed to replicate these findings, to elucidate the underlying mechanisms and to assess whether these differences persist in later life.
Supplementary material
To view supplementary material for this article, please visit http://dx.doi.org/10.1017/S0007114513000676
Acknowledgements
The Generation R study was conducted by the Erasmus Medical Center in close collaboration with the School of Law and Faculty of Social Sciences of Erasmus University Rotterdam, the Municipal Health Service Rotterdam area, Rotterdam, the Rotterdam Homecare Foundation, Rotterdam and the Stichting Trombosedienst & Artsenlaboratorium Rijnmond, Rotterdam. We gratefully acknowledge the contributions of general practitioners, hospital, midwives and pharmacies in Rotterdam. The first phase of the Generation R Study is made possible by financial support from the Erasmus Medical Center, Rotterdam, Erasmus University Rotterdam and The Netherlands Organization for Health Research and Development (ZonMw). Additional support was provided by a grant from the Dutch Kidney Foundation (C08.2251) and a grant from the Dutch Heart Foundation (no. 2008B114). V. W. V. J. received an additional grant from The Netherlands Organization for Health Research and Development (ZonMw 90700303, 916.10159). L. C. L. v. d. H. and H. R. T. were responsible for the statistical analyses, the interpretation of the data, and the revisions of the manuscript. L. C. L. v. d. H. and H. R. T. also wrote the first draft of the manuscript. V. W. V. J. supervised L. C. L. v. d. H. with data-analysis, interpretation of the data, and writing of the manuscript. L. L. d. J., D. H. M. H., E. A. P. S. and A. J. v. d. H. contributed to the interpretation of the data, and critically revised the manuscript. A. H. and V. W. V. J. initiated and designed the study, were responsible for the infrastructure in which the study was conducted, contributed to the original data collection, and critically revised the manuscript for important intellectual content. All authors read and approved of the final version of the manuscript. All the authors declare that there are no competing interests.
