Size at birth, maternal nutritional status in pregnancy, and blood pressure at age 17: population based analysis

Introduction

Several studies have shown a significant inverse relation between blood pressure and birth weight,1 2 3 4 5 6 which may be a result of fetal adaptations to an adverse environment, particularly undernutrition.7 Data showing a strong association between maternal nutritional status in pregnancy and blood pressure at 10-12 years of age support this contention,8 and further confirmation comes from study of the influence of mother's diet in late pregnancy on their offspring's adult blood pressure.9 However, not all investigators have confirmed this association.10 11 12 13 In addition, the hypothesis that high blood pressure has early nutritional origins was not supported by a study assessing the effect of poor nutrition in infants born before term on later high blood pressure.14

Previous studies examining prenatal effects on blood pressure defined intrauterine growth retardation as birth weight below 2500 g.15 A more accurate assessment of intrauterine growth is provided by size at birth considered in relation to gestational age, rather than birth weight alone. Also, in this study we assessed maternal nutritional status during pregnancy on the basis of body weight before pregnancy corrected for height and weight gain in pregnancy.

We aimed to evaluate the effects of being born small or large for gestational age, and of the mother's nutritional status, on systolic and diastolic blood pressure at 17 years of age. The large computerised dataset allowed adjustment for height and weight of the subjects and their mothers as well as for birth order, ethnic origin, and socioeconomic factors.

Subjects and methods

Data on 11 428 infants born between November 1974 and February 1976 were obtained from the computerised records of the Jerusalem study on oral contraceptive use,16 one of the interlocking epidemiologic studies making up the Jerusalem perinatal study, the design and methods of which have been reported.17 Variables included birth weight and gestational age recorded at the time of delivery, as well as detailed sociodemographic information. Socioeconomic status was assessed by area of residence (classified by municipal tax level areas17) and mother's educational attainment (years of schooling). Ethnic origin was defined according to father's country of birth. Gestational age was determined, in women with regular menstrual cycles,18 according to the date of the last menstrual period acquired through individual interviews with all mothers on the day after delivery.19 The perinatal data were matched to measurements made at age 17 obtained from the computerised records of the Israel defence forces' draft medical examination, using the subject's seven digit identification number assigned at the time of draft.20 The adequacy of the match was confirmed by comparing sex and mother's identity number. Complete information on all variables was available on 10 883 (95.2%) of the subjects. Only 38.6% of our study population (4201 subjects) were women, as women who declare themselves to be orthodox religious are exempted from military service.20 We thus analysed and presented the results separately for men and women. Only a negligible minority (124, <2%) of male subjects (those in hospital because of severe chronic and psychiatric disease and long term prisoners) were not examined and consequently were excluded from the study.

Subjects' birth weights were stratified in fifths for gestational age, on the basis of growth curves determined for our entire study population, to avoid a bias that may be caused by use of growth curves developed for a different population.

Blood pressure was measured in the sitting position in the right arm with a Bauman sphygmomanometer with appropriate cuff size. The end point for diastolic blood pressure was the disappearance of the Korotkoff sounds (phase V). The examiners were not aware of the perinatal data. Standing height was measured without shoes to the nearest centimetre, and body weight without clothes was determined to the nearest 100 g.

Regression coefficients with 95% confidence intervals were calculated by multiple linear regression analysis (SAS Institute, Cary, NC) to control for the effects of the studied independent variables. The variables included in the regression model were birth weight corrected for gestational age, birth order, ethnic origin, social status, height and body mass index (kg/m²) at 17 years of age, and mother's educational attainment, body mass index, and weight gain in pregnancy.

Results

Systolic and diastolic blood pressure at age 17 correlated significantly with body weight and height, body mass index, and mother's body weight and body mass index before pregnancy, but not with birth weight or mother's weight gain in pregnancy (table 1). After adjustment for birth weight, mother's body mass index, body weight at age 17, and ethnic origin by a multiple regression analysis model, systolic and diastolic blood pressure and body weight were positively associated (table 2). Birth weight and Western and Israeli ethnic origin were negatively associated with blood pressure (table 2). Birth order, social status, and mother's and father's educational attainment were not included in the final regression model as they did not influence blood pressure significantly.

The mean systolic and diastolic blood pressures increased with adolescent body mass index group in young women and young men (tables 3 and 4). The fifth of birth weight for gestational age was not associated the mean systolic or diastolic blood pressure at age 17 (tables 3 and 4). Adjustment for body weight at age 17 and mother's weight before pregnancy, body mass index, weight gain in pregnancy, and ethnic origin in a multiple regression model did not change these associations.

After adjustment for the above variables, mother's body mass index was positively associated with blood pressure while weight gain in pregnancy had little influence on blood pressure at age 17 (for details, see extra tables available on the internet at http://www.bmj.com/). Furthermore, subjects born to women with low body mass index before pregnancy, or those who gained little weight in pregnancy, did not have a higher mean systolic or diastolic blood pressure.

Discussion

Adverse intrauterine conditions, suggested by small size for gestational age, were not associated with high blood pressure at age 17. Birth weight was positively, rather than negatively, correlated with blood pressure. This finding is in agreement with our previous study on 33 545 subjects born a decade earlier, in which low birth weight was poorly correlated with blood pressure in late adolescence.10 The present study has an advantage over the previous series, because an accurate estimation of gestational age was available, better reflecting intrauterine growth than birth weight alone. More importantly, maternal nutritional status during pregnancy could be assessed, and low weight and body mass index before pregnancy, as well as poor weight gain in pregnancy, were not associated with high blood pressure. In fact, the mother's body mass index before pregnancy and weight gain in pregnancy had an independent positive effect on blood pressure at 17 years of age.