| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
(Hypertension. 2007;50:572.)
© 2007 American Heart Association, Inc.
Original Articles |
From the Julius Center for Health Sciences and Primary Care (C.C.G., D.E.G., B.M.d.J., C.S.P.M.U.) and Department of Pediatric Pulmonology (C.K.v.d.E., B.M.d.J., M.M.v.d.Z., N.v.P.-K., J.L.L.K.), Wilhelmina Childrens Hospital, University Medical Center Utrecht, Utrecht, The Netherlands.
Correspondence to Cuno S.P.M. Uiterwaal, Julius Center for Health Sciences and Primary Care, University Medical Center Utrecht, PO Box 85500, 3508 GA Utrecht, The Netherlands. E-mail c.s.p.m.uiterwaal{at}umcutrecht.nl
| Abstract |
|---|
|
|
|---|
2 months of age. Smoking exposure of mothers in pregnancy was obtained by questionnaire. Of 456 mothers whose infants had blood pressure measured, 363 (79.6%) were not exposed to tobacco smoke in pregnancy, 63 (13.8%) did not smoke in pregnancy but were exposed by others, and 30 (6.6%) smoked. Infant offspring of mothers who had smoked during pregnancy had 5.4 mm Hg (95% CI: 1.2 to 9.7; P=0.01) higher systolic blood pressure levels than offspring of mothers who were not exposed to tobacco smoke in pregnancy, taking account of birth weight, infant age, gender, nutrition, and age of mother. No associations were found between maternal exposure to tobacco smoke in pregnancy and diastolic blood pressure. A positive association between maternal exposure to tobacco smoke and heart rate was largely explained by confounding. It can be concluded that maternal exposure to tobacco smoke in pregnancy has a substantial increasing effect on systolic blood pressure in early infancy.
Key Words: prenatal exposure newborn tobacco infant blood pressure maternal smoking WHISTLER birth cohort
| Introduction |
|---|
|
|
|---|
Smoking by the mother in pregnancy is related to other postnatal outcomes in the offspring that could underlie a relation with childhood blood pressure. Particularly, there are consistent reports showing that maternal smoking in pregnancy is associated with overweight or obesity in childhood and adolescence.4,9–12 Relative weight is a strong determinant of blood pressure, also in childhood.13 Therefore, it is, eg, not clear whether higher blood pressure in offspring of mothers who smoked in pregnancy results from higher relative weight.4
Tobacco exposure in utero causes an adverse fetal environment, and such environment, in turn, may permanently affect the fetus.14 Blood pressure measurements in healthy newborn infants should help distinguish between constitutional and environmental explanations, because at that age there is the lowest possibility of blood pressure changes induced by longer-lasting postnatal exposures. To our knowledge, there have been very few studies on the relation between prenatal smoking and blood pressure in infancy, but with rather different designs and results. One study in a hospital population showed in the neonatal period a minor effect of maternal smoking on diastolic blood pressure only.15 Another hospital-based study showed very large effects on both systolic and diastolic blood pressure in the neonatal period.16 Yet another study used blood pressure change after neonatal head tilting as a marker for autonomic function.17
We have measured blood pressure levels in healthy newborn infants of an unselected population of parents, with reported data on the gestational and postnatal period and with a parental cardiovascular health profile. Our aim was to assess whether maternal exposure to tobacco smoke in pregnancy is associated with blood pressure in their young infant offspring.
| Methods |
|---|
|
|
|---|
Leidsche Rijn currently has an estimated average birth rate of 40 per 1000 inhabitants per year. The present study is part of the Wheezing Illnesses Study Leidsche Rijn (WHISTLER) that was initiated in December 2001, of which design and rationale are extensively described elsewhere.19 Briefly, WHISTLER is an ongoing birth cohort that is embedded in UHP and uses the same source population. Its focus is on early life determinants of respiratory disease in childhood. All of the parents in this area who had a baby are invited by telephone within 14 days after birth to participate by visiting the ambulatory clinic of UHP with their child at around week 4 of life. Exclusion criteria are gestational age <36 weeks, major congenital abnormalities, and neonatal respiratory disease. A record is kept of all of the newborns with exclusion criteria and of those not consenting. Information on the child is gathered by a questionnaire filled out by the parents during baseline examination with regard to prenatal, perinatal, and postnatal factors, including general characteristics (gestational age and gender), maternal smoking habits, breast or bottle feeding, and recorded birth weight and height. Body weight at the visit was measured using a standard electronic scale and body length using an infant stadiometer. Maximal occipital-frontal and thoracic circumference were assessed using a tape measure. With regard to the original research question of WHISTLER, children further undergo a lung function measurement that is described in detail elsewhere.20
Currently, some 1300 infants are included in WHISTLER. The current response rate of WHISTLER is 78%. Blood pressure measurements in WHISTLER were started in January 2003. After that date, 1014 infants were included in WHISTLER, of whom 791 (78%) had their lung function successfully measured. Of these 791 infants, 456 (58%) had successful blood pressure measurements. The present study deals with these 456 infants with blood pressure measurements and prenatal, perinatal, and postnatal data. Of these, 293 mothers and 244 fathers participated in UHP.
Determinant Measurement
In the WHISTLER cohort, the smoking status during pregnancy was assessed by questionnaire. This enabled formation of 3 major categories for tobacco smoke exposure in pregnancy: (1) mothers who did not smoke and who were not exposed to smoke by others; (2) mothers who did not smoke but were exposed to smoke by others; and (3) mothers who smoked. These categories were used in the main analysis.
Outcome Measurement
Infant blood pressure was measured after successful lung function measurement, which required the infant to sleep naturally (>75% of children). Blood pressure measurements were performed during natural sleep 3 times at the lower leg using an electronic "vital signs" device (Critikon Dinamap, model 1846SX). Cuffs of a minimum 7-cm and maximum 13-cm circumference were used (Johnson & Johnson, Medical BV, ref 2524). The average of 3 blood pressure measurements was used for analysis.
Confounders
In the analysis of maternal smoking in pregnancy and offspring blood pressure, some factors were to be considered possible confounders because of their relation with our determinant or outcome. Maternal smoking in pregnancy is associated with lower birth weight,21 whereas birth weight, in turn, has been associated with infant blood pressure.22 Periconceptual smoking has been suggested to be associated with a reduced male:female ratio.23 Socioeconomic status is associated with maternal smoking and breast feeding rates,21,24 whereas breastfeeding is shown to have an effect on systolic blood pressure.25 Young maternal age may be a determinant of smoking in pregnancy,21 whereas higher maternal age has been associated with higher systolic blood pressure of the offspring.26 Therefore, in further analyses we particularly evaluated birth weight, infant gender, infant feeding, and maternal age as possible confounders.
Data Analysis
Central estimators and variance measures of general characteristics of parents and children were calculated. To evaluate possible confounders of the association between maternal smoking and blood pressure, we tested for differences in general characteristics of newborns between the groups of mothers who were and were not exposed to smoke during pregnancy.
2 tests were used for proportional data and Student t tests for continuous data. We evaluated separately which of the newborns general characteristics were associated with their blood pressure or heart rate using linear regression. General linear regression models were used to assess the relation between maternal smoking in pregnancy and blood pressure and heart rate. The same models were used to adjust this association for possible confounders. Confounder-adjusted mean blood pressure levels across categories of tobacco smoke exposure in pregnancy were calculated using blood pressure as outcome, smoke exposure categories as factors, and confounders as covariates in the model. All of the analyses were performed using SPSS for Windows, version 14.0.
| Results |
|---|
|
|
|---|
|
|
Of the 293 mothers, 93 (31.7%) had taken their individual health profile in the UHP after and 200 (68.3%) before baseline measurements of their child in WHISTLER, and 13 indicated being pregnant at profile measurement. We were not able to find differences in parental blood pressure between the groups of mothers who smoked during pregnancy and the mothers who were not exposed.
Table 3 shows that age of the infant at measurement was strongly positively associated with systolic blood pressure and with lower heart rate. Higher birth weight was associated with higher systolic and diastolic blood pressure and with lower heart rate. Being taller at birth was particularly related to lower heart rate. Longer gestation was associated with higher systolic and diastolic blood pressure and with lower heart rate. Higher head and thorax circumference were associated with higher systolic blood pressure and higher thorax circumference with higher diastolic blood pressure. Nutrition was not related to blood pressure, but breastfeeding was associated with lower infant heart rate than bottle feeding, although not statistically significant. Using individual health profile data from UHP showed no associations between diastolic or systolic blood pressure of the mothers and the fathers and blood pressure levels in their offspring.
|
Not all of the characteristics that could be considered possible confounders were associated with the determinant, maternal smoking exposure, and the outcome variables, systolic and diastolic blood pressure and heart rate. From previous knowledge (see Confounders section) combined with findings in Tables 2 and 3
, we decided that birth weight, gender, infant age at measurement, infant feeding, and maternal age were the possible confounders to be considered for each of the 3 outcome variables.
Table 4 provides the main analysis showing that there was a statistically significant relation between maternal smoking in pregnancy and systolic blood pressure. Newborns of smoking mothers had a 5.4 mm Hg higher systolic blood pressure than offspring of mothers who were not exposed to tobacco smoke in pregnancy, an association that was not explained by confounding. There was no relation between maternal smoking and diastolic blood pressure or heart rate. A separate analysis for gender showed that male offspring of smoking mothers had 8.6 mm Hg (95% CI: 0.5 to 16.8; P=0.04) higher systolic blood pressure than male offspring of mothers who were not exposed to tobacco smoke, taking account of birth weight, infant age, nutrition, and age of mother. In female offspring of smoking mothers, there was no significant difference in systolic blood pressure.
|
| Discussion |
|---|
|
|
|---|
A slightly lower proportion of males among the offspring of maternal smokers, although not statistically significant, may be consistent with recent findings suggesting a lower male:female ratio in the offspring of parents who smoked around the period of conception.23 Among the most well-known associations of maternal smoking in pregnancy is that with low birth weight, which seems to be even transmitted through generations.28 Moreover, exposure of nonsmoking pregnant women to tobacco smoke in their environment was found to have such an effect on birth weight.29 In turn, birth weight is considered an important determinant of infant blood pressure and adult blood pressure.22,30,31 Whether a lower frequency of male gender and lower birth size among offspring of mothers who smoked in pregnancy are by any means related phenomena is currently unknown. Breastfeeding has been associated with slightly lower systolic blood pressure in childhood.32 Higher maternal age was shown to be a predictor of higher systolic blood pressure in newborn offspring.26 However, birth weight, infant age, gender, infant nutrition, and maternal age did not explain our findings concerning systolic blood pressure.
To our knowledge, there have been very few studies in infants on maternal smoking in pregnancy and infant blood pressure. In a study on 3- and 4-day-old neonates born in the hospital, smoking of mothers was associated with somewhat higher (1.7 mm Hg) diastolic blood pressures only.15 In that study, a very high proportion (45%) of women were reported to smoke, but it was not clearly specified that all of these women actually did smoke during pregnancy. Given the very high smoking prevalence in that community and if, indeed, passive smoking does have an influence on infant blood pressure, we speculate that the exposure contrast in that study may have been smaller than in our study. In another study performed in Greece, 333 term neonates who were appropriate in size for gestational age were pooled with a further 18 term neonates of smoking mothers, as well as a control group of 18 matched term neonates born to nonsmoking mothers.16 That study showed higher systolic and diastolic blood pressures in neonates of smoking versus neonates of nonsmoking mothers, particularly if they smoked
15 cigarettes per day. Comparing such heavy smokers versus nonsmokers, these differences ranged from
5 mm Hg at age 1 hour to >10 mm Hg at age 72 hours for systolic blood pressure and with ranges for diastolic blood pressure from 4 mm Hg at 1 hour to almost 9 mm Hg at 72 hours. In none of these associations were adjustments made for possible confounding factors. The authors further claimed that these differences disappeared in later follow-up at 12 and 24 months, but those observations pertained to only very small followed subgroups, which were most likely too small to detect blood pressure differences. In another study it was found that maternal smoking in pregnancy disturbed autonomic function as a possible mechanism underlying sudden infant death syndrome.17 Head tilting yielded different responses of systolic blood pressure in children of smoking mothers than in children of nonsmoking mothers, and effects in the neonatal period differed from effects in infancy.
Smoking by mothers during pregnancy induces fetal hypoxia and changes in placenta morphology, thus raising the risk of intrauterine growth retardation resulting in low birth weight.21,33 How such effects translate to higher blood pressure levels in the offspring is basically unknown. It has been shown that, in spontaneously hypertensive rat breeds that are genetically susceptible to high blood pressure, exposure to nicotine in utero led to higher blood pressure, higher cholesterol levels, and lower kidney weight.34 This is compatible with the suggestion that an adverse fetal environment as manifested through low relative birth weight has consequences for organ growth in general,21 including fetal kidney growth restriction, with low nephron numbers leading to high blood pressure and kidney disease.35
Alternatively, maternal smoking leads to lower concentrations of catecholamines in neonates,36 which may indicate that hormonal changes are directly involved. Furthermore, there are indications that maternal smoking in pregnancy induces changes in autonomic function in infants.17
We did find that the smoking-associated rise in blood pressure was predominant in males. We can only speculate on the reasons for this. A study in young adults showed that males react significantly differently on pain with an increase in systolic blood pressure and higher cortisol levels.37 Perhaps gender is a modifier of stress responses in general, including tobacco smoke exposure in pregnancy.
Overall, our finding indicates that there is a direct causal influence of maternal smoking on infant blood pressure. This is underlined by recent findings suggesting that blood pressure levels are similar in young offspring whose mothers had quit smoking during pregnancy to levels in offspring of women who had never smoked.1 Other effects of maternal smoking, such as childhood overweight,4,9–12 are unlikely to explain our finding but may of course further contribute to raised blood pressure in offspring of mothers who smoked during pregnancy.
If our findings are true, they would indicate that maternal smoking during pregnancy has a substantial impact on systolic blood pressure in early infancy, as it amounted to half an SD. Notably, blood pressure measured in the actual neonatal phase15–17 may have different determinants than blood pressure measured in early infancy, such as in our study. An example may be our previous work showing a positive association between birth weight and neonatal blood pressure, which reverses at 3 months of age.22 The question is whether it is this smoking-induced blood pressure elevation in infancy that persists into the smoking-associated elevated blood pressure in later childhood.1–5 Our findings in very young infants are in agreement with findings in childhood, as the associations in childhood were also found particularly with systolic blood pressure and not diastolic blood pressure.1,4 Because blood pressure tracking is stronger from early infancy than from the neonatal phase onward,38 it is likely that our findings do have an impact for later-life blood pressure. Moreover, we have recently shown persistent effects of maternal smoking in pregnancy through an excess rise in total cholesterol levels in the offspring throughout adolescence.39
It is claimed from nonrandomized observations that multidisciplinary prenatal interventions targeted toward specific risks, such as smoking, inadequate weight gain, and psychosocial problems, successfully improved infant birth weight.40 However, randomized evaluation of motivational interviewing by trained midwives aimed to persuade pregnant women to quit or cut smoking showed that such intervention was not effective.41 Our finding may contribute to further awareness of the offspring risks involved with maternal exposure to tobacco smoke. We conclude from our study that maternal exposure to tobacco smoke in pregnancy has a substantial increasing effect on systolic blood pressure in early infancy.
Perspectives
The consequences of smoking during pregnancy on blood pressure in healthy newborn offspring have not been studied extensively. Our study shows that tobacco exposure in fetal life is independently related to an increase in systolic blood pressure in the early postnatal period. The results suggest that this association is already set in utero and it is not because of postnatal sharing of environments. Maternal smoking in pregnancy leads to adverse developmental changes. To see whether the effect on infant blood pressure will track into childhood, a follow-up of the relation between smoking during pregnancy and blood pressure in childhood is necessary.
| Acknowledgments |
|---|
Sources of Funding
WHISTLER is funded by the Health Research and Development Council of The Netherlands grant 2001-1-1322 and by an unrestricted grant from Glaxo Smith Kline Netherlands.
Disclosures
None.
Received March 28, 2007; first decision April 29, 2007; accepted July 5, 2007.
| References |
|---|
|
|
|---|
Find additional patient-related information at:
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
|
Hypertension Home | Subscriptions | Archives | Feedback | Authors | Help | AHA Journals Home | Search Copyright © 2007 American Heart Association, Inc. All rights reserved. Unauthorized use prohibited. |