|Home | About | Journals | Submit | Contact Us | Français|
Rationale: Few studies have investigated childhood respiratory outcomes of intrauterine growth retardation (IUGR), and it is unclear if catch-up growth in these children influences lung function.
Objectives: We determined if lung function differed in 8- to 9-year-old children born at term with or without growth retardation, and, in the growth-retarded group, if lung function differed between those who did and those who did not show weight catch up.
Methods: Caucasian singleton births of 37 weeks or longer gestation from the Avon Longitudinal Study of Parents and Children (n = 14,062) who had lung spirometry at 8–9 years of age were included (n = 5,770).
Measurements and Main Results: Infants with gestation-appropriate birthweight (n = 3,462) had significantly better lung function at 8–9 years of age than those with IUGR (i.e., birthweight <10th centile [n = 576] [SD differences and confidence intervals adjusted for sex, gestation, maternal smoking during pregnancy, and social class: FEV1, −0.198 (−0.294 to −0.102), FVC, −0.131 (−0.227 to −0.036), forced midexpiratory flow between 25 and 75% of vital capacity −0.149 (−0.246 to −0.053)]). Both groups had similar respiratory symptoms. All spirometry measurements were higher in children with IUGR who had weight catch-up growth (n = 430) than in those without (n = 146), although the differences were not statistically significant. Both groups remained significantly lower than control subjects. Growth-retarded asymmetric and symmetric children had similar lung function.
Conclusions: IUGR is associated with poorer lung function at 8–9 years of age compared with control children. Although the differences were not statistically significant, spirometry was higher in children who showed weight catch-up growth, but remained significantly lower than the control children.
Some evidence suggests that intrauterine growth retardation (IUGR) is associated with poor lung function in infancy and beyond. Furthermore, weight catch-up early in infancy in term babies may be associated with less than optimal lung function in infancy.
This study provides evidence that term infants born without IUGR defined by weight have better lung spirometry at 8 to 9 years of age than term infants born with IUGR. All measures of lung function at age 8 to 9 were better in the children with weight catch-up than in those without, but the differences were not statistically significant.
Increasing evidence suggests that low birthweight (LBW) is associated with long-term morbidity, including adverse cardiovascular, endocrine, and respiratory outcomes (1–3). Intrauterine growth retardation (IUGR) represents a significant worldwide problem (4). The relationship between LBW and lung function has been studied in adults, a meta-analysis of eight such studies reporting an increase of 48 ml in FEV1 for each kilogram increase in birthweight after adjustments for age, smoking, and height (5). Associations between IUGR and lung function in childhood have been studied less extensively. Interpretation is often complicated by combining LBW and prematurity (6), which affect lung function abnormalities by different mechanisms. The literature is further complicated by different definitions of IUGR; the most commonly used definition is based on birthweight alone, as stated by the World Health Organization (4), but length is only rarely accounted for. Lower lung function has been reported in LBW infants compared with normal-birthweight infants (7–9). Rona and colleagues studied children aged 5–11 years and reported that lung function measurements, except forced midexpiratory flow between 25 and 75% of vital capacity (FEF25–75%), were positively associated with birthweight, but respiratory symptoms were associated with prematurity (10). Wjst and colleagues (11) reported decreased total lung capacity and peak expiratory flow rates, but increased bronchial reactivity and asthma in 5–14 year olds who were born with LBW.
IUGR represents fetal growth restraint due to an adverse intrauterine environment, or may be due to genetic factors. Catch-up growth may represent “release” from adversity, such as maternal smoking, which may lead to improved lung function later in life. Those not showing catch-up growth may represent healthy infants who have reached their growth potential, and thus should have normal lung function later in life when adjusted for height. Weight catch-up in childhood is associated with improved respiratory outcomes in adulthood (12, 13). However, Lucas and colleagues (8) reported that gaining weight quickly in the first few weeks of life is linked to suboptimal lung development in 5- to 14-week-old infants born at term. Another study of term infants reported that lung function assessed at 1 and 12 months was inversely related to weight gain in infancy (14).
We defined IUGR on the basis of birthweight and ponderal index (PI), which includes both weight and length at birth and tested two hypotheses: children aged 8–9 years, born at term with IUGR, have poorer lung function compared with infants born without IUGR; and, in the IUGR group, differences in lung function exist between those who did and those who did not show weight catch-up growth.
We used data from the Avon Longitudinal Study of Parents and Children (ALSPAC), which has been described previously (15). Additional detail is provided in the online supplement. Ethical approval was obtained from the ALSPAC Law and Ethics Committee and the local research ethics committees.
The best estimate of gestational age was based on maternal reporting and medical assessments (16). The recorded birthweight and PI (PI = birthweight/length3 [kg/m3]) were standardized and converted to z scores, adjusted for gestational age and gender in babies born at 37 weeks gestation or later. Details of the methods for standardization are provided in the online supplement. IUGR was defined by birthweight, and separately by PI at birth, including babies with a z score of less than −1.28 (i.e., <10th centile) (4); infants between the 20th and 80th centiles for each variable were selected as control subjects. This range was chosen to avoid emerging obesity or borderline IUGR influencing the results. The Child Growth Foundation growth charts (17) were not used, as the data on which they were based were acquired between 1978 and 1990, and the ALSPAC data showed a trend toward earlier maturity and growth (data not shown). Infants below the 10th centile for both birthweight and PI were classified as “asymmetric IUGR,” and those below the 10th centile for birthweight and at or above the 10th centile for PI were classified as “symmetric IUGR” (4). Weight at the time of lung function testing was converted to a z score (details are given in the online supplement). Catch-up growth for weight was defined as an increase in z score of at least 0.67 between birth and ages 8–9 years, as previously described (18).
Spirometry was performed at ages 8–9 years using identical equipment and protocol, and following American Thoracic Society standards (19). FEV1, FEV0.5, FEF25–75%, FEF25%, FEF50%, and FEF75% were converted to SD scores and adjusted for age, sex, and also for height (20). Children, unselected for asthma status or wheezing, had a bronchial challenge test with methacholine following the method of Yan and colleagues (21). Additional details of the methods are given in the online supplement. Previous respiratory illnesses/symptoms were recorded. Current asthma, from a questionnaire study at 91 months, was based on reported doctor-diagnosed asthma ever with symptoms and/or treatment in previous 12 months.
Smoking during pregnancy, socioeconomic status (see online supplement for definitions) and exclusive breast feeding longer than 3 months were considered as potential confounders, but there was no evidence for breast feeding being a confounder. More details of confounders are provided in the online supplement.
Lung function comparisons between the IUGR and control groups were performed using t tests. Comparisons were adjusted for potential confounders using a general linear model (22). The modeling was adjusted simultaneously for all of the confounders. Similar methods were employed for the comparison within the IUGR group between those with and those without catch-up growth. One-way analysis of variance was used to compare both these groups with the control children. All statistical analysis was performed using SPSS 15 (SPSS Inc., Chicago, IL). A P value less than 0.05 was considered significant.
Participants' characteristics are shown in Table 1. From 14,049 live births, 7,394 had lung spirometry at 8–9 years of age. After exclusions of preterm births of less than 37 weeks, multiple births, deaths, and non-whites, 5,770 remained, with 576 classified as IUGR and 3,462 as control subjects. From the IUGR group, 430 showed catch-up growth and 146 did not, with 94 remaining below the 10th centile. Table E1 in the online supplement shows additional group characteristics, and Table E2 shows characteristics of those who did not attend lung spirometry (see online supplement). As expected in a longitudinal study, loss to follow up was associated with lower social class; thus, the prevalence of maternal smoking in those who attended for lung spirometry was lower. The difference in proportions for maternal smoking between those attending and those not attending for lung spirometry was 14.2% (95% confidence interval [CI], 12.6–15.8%). The difference in proportions for manual social class between those attending and those not attending for lung spirometry was 14% (95% CI, 12.1–15.9%). Maternal smoking in pregnancy was reported for 18% of the control group and 29% of the IUGR group, and was associated with both IUGR and catch-up status (Tables E3 and E4); 32 and 19% of mothers smoked during pregnancy in the catch-up growth and no catch-up groups, respectively.
For spirometry, Table 2 shows the mean and SD in each group, the difference in means between subjects with IUGR and control subjects, with a 95% CI, both unadjusted and after adjustment for confounders of maternal smoking and social status, and a P value for the adjusted comparison. More details of confounders are shown in Tables E3–E8. There were clear differences between the IUGR and control populations for all spirometry measures, except FEF25%. Approximate means of these differences were −35 ml for FEV0.5, −50 ml for FEV1, −40 ml for FVC, and −80 ml for FEF25–75%, adjusted for height, age, and sex. No significant differences between the groups were noted for previous respiratory illnesses, including current asthma at 91 months (Table 1), self-reported wheezing, and any respiratory infections or symptoms.
Although all lung function measures were greater in the catch-up IUGR group compared with the no catch-up group, the differences were not statistically significant, but remained significantly lower than the control group (Table 3). The results remained essentially unchanged when we regressed the lung function measurements on the actual change in weight z scores, increasing our confidence that the conclusions are robust (data not shown). Approximate mean differences, adjusted for height, age, and sex, were −35 ml for FEV0.5, −30 ml for FEV1, −25 ml for FVC, and −50 ml for FEF25–75%. No differences were noted between the groups for previous respiratory illnesses, including asthma, self-reported wheezing, or for reported respiratory infections or symptoms.
Infants with IUGR who had both weight and length at birth were divided into 335 (73%) symmetrical and 126 (27%) asymmetrical infants based on their birth PI, with catch up occurring in 241 (72%) of the former and 106 (84%) of the latter group. The symmetrical and asymmetrical groups had similar lung function measurements at 8–9 years of age (Table 4). The catch-up infants in the symmetrical group appeared to have slightly higher values for spirometry than those who did not show catch-up. However, these differences were not statistically significant.
A total of 356 (61.8%) in the IUGR group and 2,229 (64.4%) of control children underwent methacholine challenge; 7.3% and 5.9%, respectively, were excluded, as their FEV1 was less than 70% predicted. Of those who did not start a methacholine challenge, the prevalence of current asthma was similar (16 and 17.7% in IUGR and control groups, respectively), being similar to the study groups (Table 1). Similar percentages of children who completed the methacholine challenge (14.3 and 11.4% for the IUGR and control groups, respectively) had current asthma at 91 months. Response to methacholine challenge (≥20% fall in FEV1 at less than maximal cumulative dose) occurred in 15.7 and 14.4% of the IUGR and control groups, respectively; the remainder were classed as nonresponders. Bronchial responsiveness in responders was similar between the IUGR and control groups (geometric mean PD20, 2.12 and 1.68 μmol, respectively). For the whole group undergoing bronchial challenge, the PD20 geometric means were also similar (IUGR, 9.24 μmol; controls, 9.14 μmol—with maximum methacholine dose allocated to nonresponders). Furthermore, the PD20 and rates of asthma for the catch-up and no catch-up groups were similar to each other and to the control groups.
After using similar criteria, definitions, and exclusions as for the IUGR defined by birthweight, 445 children had standardized PI at or below the 10th centile, and 2,684 between the 20th and 80th centile. Lung function at 8–9 years did not differ significantly between those with and those without IUGR by this definition (Table 5).
In this, the largest study to our knowledge of accurately defined IUGR, we have shown that term infants with IUGR had significantly reduced lung function at 8–9 years of age compared with term infants of normal birthweight. Differences were noted for both small and large airway function, but did not translate into increased clinical symptoms. Although the differences were not statistically significant, all measures of lung function at ages 8–9 years appeared to be higher in children showing catch-up growth compared with those without catch-up growth. These measures, however, remained significantly lower than the control group. Furthermore, asymmetric and symmetric children with IUGR had similar lung function. Symmetric infants with IUGR with catch up appeared to show better spirometry than those without, although overall differences were small and not statistically significant.
Like most other similar studies, we defined IUGR by birthweight, but included length in defining PI (which indicates proportionality) and symmetry/asymmetry. However, by including length or PI, the population defined as IUGR changed considerably compared with the population identified by the commonly used birthweight definition (4). For example, a low PI may indicate a “normal” birthweight on the 25th centile, but a length on the 90th centile, which is very different to a child born below the 10th centile for birthweight although the former may represent relative fetal growth restraint. We chose to use as our primary exposure a definition of IUGR based on birthweight. Haggarty and colleagues (23) reported that body weight alone was a better predictor than PI of factors thought to be associated with IUGR. Furthermore, birthweight is more frequently and more accurately measured than length at birth. Height at the time of spirometry was taken into account, as lung function values were standardized for height, in addition to sex and age. Nevertheless, we subdivided the infants with IUGR into symmetrical and asymmetrical infants to include birth length via PI in order to assess the influence of birth length on later lung function.
Our study focused entirely on infants born at term, and thus avoided combining LBW term and premature infants in an IUGR group. This distinction is important, as mechanisms leading to abnormal lung development are possibly different in these two groups, with little known about the former, but evidence suggesting dysregulation of alveolar development in the latter (24).
Our results are in agreement with a recent meta-analysis in adults reporting a positive association between FEV1 and birthweight (5), and also with three studies that reported that infants born at or near term with LBW had lower lung function than normal-birthweight infants (7–9). Rona and colleagues (10) reported that most lung spirometry parameters in school-aged children were associated with birthweight, but respiratory symptoms were associated more with prematurity. We also noted an association of lower lung function with LBW, but reported respiratory symptoms were not.
Studies of adults generally conclude that catch-up growth, or gaining weight faster in early childhood, is associated with improved pulmonary function outcomes (12, 13), but it has been suggested that catch-up growth in infancy may be detrimental to lung function (8, 14). Our data demonstrate that spirometry at age 8–9 years was nonsignificantly higher in IUGR children who showed catch-up growth than those who did not, although their lung function was significantly lower than in control infants. Despite ours being one of the largest studies of term infants with IUGR, the numbers in this category are relatively small, reducing the power for these comparisons. It is important to confirm or refute this finding, given the importance of lung function tracking during childhood and into adulthood. Stern and colleagues (25) suggested, “poor airway function shortly after birth should be recognized as a risk factor for airflow obstruction in young adults.”
An explanation for lung function differences in childhood between infants with and those without catch-up growth is that they were different soon after birth due to intrauterine development. Infants of maternal smokers were smaller at birth compared with infants of nonsmokers, as previously reported in ALSPAC (26). The incidence of catch-up growth in infants with IUGR was greater if the mothers had smoked during pregnancy than if they had not, suggesting that the effects on growth of an adverse intrauterine environment may be compensated by rapid postnatal growth. However, developmental changes associated with fetal growth restraint may persist in spite of catch-up somatic growth. Infants with genetic or other constitutional determinants of IUGR may have programmed reduced airway and lung development during fetal life, with little opportunity to improve after birth. We found asymmetric and symmetric children with IUGR defined by PI at birth had similar lung function at 8–9 years of age; although there was limited evidence of differences between catch-up and no catch-up children in the symmetric group. If this was demonstrated in larger studies, it would be consistent with generalized growth restraint in the symmetrically IUGR group being compensated in part by postnatal growth.
The magnitudes of lung function differences between the IUGR and control groups were similar for markers of both large and small airway function, with only FEF25% not showing significant differences between the two groups. The population that attended for spirometry was represented by fewer maternal smokers and lower social class. This is unlikely to affect the results, as greater attendance of manual socioeconomic class is most likely to enhance the differences. A meta-analysis of adult lung function in relation to birth weight (5) suggested an increase of 48 ml in FEV1 for each kilogram increase in birthweight after adjustments. Our data suggest a difference of approximately 50 ml for FEV1 that is already established at 8–9 years of age in children with IUGR (mean birthweight, 2,745 g) compared with control subjects (mean birthweight, 3,491 g). These data suggest that IUGR at term may have a greater effect on lung function than previously reported. In addition, children born with IUGR at term who did not have catch-up growth have a decrement in FEV1 of approximately 30 ml compared with those who had catch-up growth, suggesting that there may be different life-course implications for different mechanisms of IUGR as currently defined.
Differences observed in lung function did not translate into clinical symptoms, perhaps unsurprisingly given the respiratory reserve in childhood and the relatively small effect sizes. Our data did not show that the children born with IUGR had greater airway reactivity, nor did they report an increased incidence of asthma or other respiratory illnesses. Identifying decreased lung function is important, as public health measures, such as improved nutrition, may potentially improve lung development, and future prevention of smoking may prevent or retard the development of later chronic obstructive pulmonary disease.
In summary, our data show a clear difference for all spirometric measures of lung function between children born at term with IUGR and control term infants, but without associated increased clinical symptoms. Catch-up growth was associated with a nonsignificantly higher lung function, suggesting that nutritional improvement after fetal growth restraint may modify postnatal lung function development. Studies of larger numbers of IUGR infants with early life lung function measurements are required to confirm these findings.
The authors are extremely grateful to all the families who took part in this study, the midwives for their help in recruiting them, and the whole Avon Longitudinal Study of Parents and Children team, which includes interviewers, computer and laboratory technicians, clerical workers, research scientists, volunteers, managers, receptionists, and nurses. This publication is the work of the authors, and Professor Sailesh Kotecha will serve as guarantor for the contents of this paper.
Supported by a grant from Nutricia Research Foundation (S.J.K.); The Medical Research Council, the Wellcome Trust, and the University of Bristol provide core support for the Avon Longitudinal Study of Parents and Children.
This article has an online supplement, which is accessible from this issue's table of contents at www.atsjournals.org
Originally Published in Press as DOI: 10.1164/rccm.200906-0897OC on January 21, 2010
Conflict of Interest Statement: S.J.K. does not have a financial relationship with a commercial entity that has an interest in the subject of this manuscript; W.J.W. does not have a financial relationship with a commercial entity that has an interest in the subject of this manuscript; J. Heron does not have a financial relationship with a commercial entity that has an interest in the subject of this manuscript; J. Henderson received up to $1,000 from MSD in advisory board fees; F.D.D. does not have a financial relationship with a commercial entity that has an interest in the subject of this manuscript; S.K. does not have a financial relationship with a commercial entity that has an interest in the subject of this manuscript.