[
Original Research Pulmonary Physiology
]
Impact of Childhood Anthropometry Trends on Adult Lung Function Sadasivam Suresh, MBBS; Michael O’Callaghan, MD; Peter D. Sly, DSc; and Abdullah A. Mamun, PhD
Poor fetal growth rate is associated with lower respiratory function; however, there is limited understanding of the impact of growth trends and BMI during childhood on adult respiratory function.
BACKGROUND:
The current study data are from the Mater-University of Queensland Study of Pregnancy birth cohort. Prospective data were available from 1,740 young adults who performed standard spirometry at 21 years of age and whose birth weight and weight, height, and BMI at 5, 14, and 21 years of age were available. Catch-up growth was defined as an increase of 0.67 Z score in weight between measurements. The impact of catch-up growth on adult lung function and the relationship between childhood BMI trends and adult lung function were assessed using regression analyses.
METHODS:
Lung function was higher at 21 years in those demonstrating catch-up growth from birth to 5 years (FVC, men: 5.33 L vs 5.54 L; women: 3.78 L vs 4.03 L; and FEV1, men: 4.52 L/s vs 4.64 L/s; women: 3.31 L/s vs 3.45 L/s). Subjects in the lowest quintile of birth (intrauterine growth retardation) also showed improved lung function if they had catch-up growth in the first 5 years of life. There was a positive correlation between increasing BMI and lung function at 5 years of age. However, in the later measurements when BMI increased into the obese category, a drop in lung function was observed.
RESULTS:
These data show evidence for a positive contribution of catch-up growth in early life to adult lung function. However, if weight gain or onset of obesity occurs after 5 years of age, an adverse impact on adult lung function is noted. CHEST 2015; 147(4):1118-1126 CONCLUSIONS:
Manuscript received March 23, 2014; revision accepted September 29, 2014; originally published Online First October 23, 2014. ABBREVIATIONS: FEF25-75 5 forced expiratory flow, midexpiratory phase; IUGR 5 intrauterine growth retardation; MUSP 5 Mater-University of Queensland Study of Pregnancy AFFILIATIONS: From the School of Population Health (Drs Suresh and Mamun), Mater Children’s Hospital, Mater Research Institute (Dr Suresh), Queensland Children’s Medical Research Institute (Drs Suresh and Sly), and the Department of Paediatrics and Child Health, School of Medicine (Dr O’Callaghan), University of Queensland, Brisbane, QLD, Australia.
1118 Original Research
FUNDING/SUPPORT: The core study was funded by the National Health
and Medical Research Council [NHMRC ID 631507] of Australia. Sadasivam Suresh, MBBS, Department of Respiratory and Sleep Paediatrics, Mater Children’s Hospital, South Brisbane, 4101, QLD, Australia; e-mail: [email protected] © 2015 AMERICAN COLLEGE OF CHEST PHYSICIANS. Reproduction of this article is prohibited without written permission from the American College of Chest Physicians. See online for more details. DOI: 10.1378/chest.14-0698 CORRESPONDENCE TO:
[
147#4 CHEST APRIL 2015
]
Increasing evidence suggests that low birth weight is associated with long-term morbidity, including adverse cardiovascular, endocrine, and respiratory outcomes.1 The relationship between birth weight and lung function has been studied in different groups, and a small but direct association has been noted.2,3 The strength of the association declines through adulthood, suggesting that environmental influences contribute to the decline in lung function with age.2 There are few reports in the literature on the impact of intrauterine growth retardation (IUGR) on adult lung function and the role of catch-up growth in early life and how it influences adult lung function.4,5 Hancox et al4 reported a small, direct impact on lung function that did not reach statistical significance. Kotecha et al5 reported lower lung function at school age in those with IUGR and a trend toward higher lung function in those with catch-up growth that was not statistically significant. However, the impact on adult lung function of weight gain or catch-up growth that occurs in childhood has not been studied in detail. Postnatal growth has the potential to contribute to eventual lung function, and a longitudinal study in which details about IUGR, catch-up growth, and adult lung function are available gives us the opportunity to explore this further. Various research is available in the literature on the impact of BMI on lung function; but we found only two
Materials and Methods The Study The MUSP is a prospective study of 8,556 pregnant women interviewed after their first clinic visit in pregnancy, with 7,223 singleton infants constituting the birth cohort. They were followed up at 3 to 5 days, 6 months, and 5, 14, and 21 years.8 At the 21-year follow-up, 5,185 young adults who were singleton babies and whose mothers had agreed at the 14-year follow-up to be contacted again were sent a questionnaire. Almost 3,800 completed the questionnaire, of whom a subsample of 2,612 young adults attended for physical assessment including lung function tests. Because of limited funding, those living outside Brisbane or who were unable to make an appointment for a face-to-face interview completed a mailed questionnaire and did not undergo the physical assessment. Ethics approval was obtained from the institution (University of Queensland Ethics Committee, B/660/SS/01/NHMRC, Mater Hospital 506A), and written consent was appropriately obtained from all participants. For this analysis, we included data obtained from 1,740 young adults who underwent spirometry at 21 years and for whom anthropometric data at birth and at 5, 14, and 21 years were available. The study sample used in these analyses was comparable to the original cohort on most aspects. Loss to follow-up was related primarily to social disadvantage (Table 1). Measurements Lung Function Testing: Lung function testing was performed at the 21-year follow-up using a Spirobank G spirometer system attached to a laptop computer. Qualified and trained interviewers familiar with
journal.publications.chestnet.org
studies that looked at BMI trends through childhood and how this impacts adult lung function.6,7 Bua et al,6 in their study of a Danish cohort, reported advantages of higher BMI during childhood toward higher lung function. However, this association with BMI was not noted in the second decade or during adulthood. Curry et al,7 in their Australian longitudinal cohort, reported that a positive association between higher BMI and lung function was possibly related to lean body mass and that adiposity adversely impacted the adult lung function attained. The aims of the current study were (1) to investigate the association of catch-up growth in childhood at the ages of 5 and 14 years with lung function at 21 years of age using prospectively collected data, (2) to assess the strength of this association in babies born with IUGR, and (3) to describe how BMI at different ages during childhood impact adult lung function. We hypothesized that catch-up growth improves lung function at 21 years and that the onset of obesity in childhood reduces the eventual lung function at 21 years. We used data from the Mater-University of Queensland Study of Pregnancy (MUSP), a large community-based birth cohort study, to investigate these aims. Our study is unique in that it explores all these three aims in a longitudinal cohort.
the instrument performed standard spirometry, in accordance with American Thoracic Society guidelines.9 A minimum of three and a maximum of five trials were attempted. If testing was unsatisfactory for any reason, the reason(s) were noted on the record sheets. For the purpose of this study FVC; FEV1; and forced expiratory flow, midexpiratory phase (FEF25-75), were considered as outcomes of interest. The all-age reference10,11 ranges for spirometry were used to compute Z scores of our lung function values. Anthropometry and Early Life Variables: Birth weight (in grams) and gestational age were recorded at delivery. The following variables were selected from the first clinical visit questionnaire for further analysis: maternal history of smoking, maternal education, and height of both mother and father. These variables were selected in view of the reported associations with lung function in the literature12-15 and their availability in our dataset. Subjects’ smoking history was obtained from the 21-years questionnaire. Maternal history of smoking during pregnancy was classified into the following groups: nonsmokers, mild smokers (1-9 cigarettes per day), and heavy smokers (ⱖ 10 cigarettes per day). The level of mother’s education was assessed at entry to the study, and answers were divided into three groups: incomplete high school, high school completers, and tertiary education. Personal smoking at 21 years was classified as nonsmoker, mild smoker, or heavy smoker (as described previously). Height and weight measurements during the 5-year, 14-year, and 21-year follow-up were documented. Height was measured without shoes using a portable stadiometer to the nearest 0.1 cm, and weight was measured in light clothing with a scale accurate to 0.2 kg. Two measures of weight and height were taken, and the mean of these two measures
1119
TABLE 1
] Early Life Variables in Study Sample and in Total Cohort
Cohort Variables
Initial Cohort, No. (%)
Study Sample (n 5 1,740), No. (%)
P Value
Total MUSP cohort (n 5 7,223) Male
3,758 (52.0)
869 (49.9)
Female
3,465 (48.0)
871 (50.1)
Smokers during pregnancy (n 5 7,152)
NS NS , .001
Nonsmokers
4,395 (61.5)
1,178 (68.4)
Light smokers (, 10/d)
1,191 (16.6)
260 (15.1)
Heavy smokers
1,566 (21.9)
285 (16.5)
Maternal education (n 5 7,170)
, .001
Incomplete high school
1,309 (18.3)
256 (14.8)
Completed high school
4,609 (64.3)
1,105 (64.0)
Post high school
1,252 (17.5)
367 (21.2)
1,886 (50.0)
866 (50.2)
21-y questionnaire (n 5 3,775) Current smoking history Nonsmokers
NS
Light smokers
922 (24.4)
444 (25.8)
NS
Heavy smokers
967 (25.6)
414 (24.0)
NS
Birth weight, g (SD)
3,385 (519)
3,403 (511)
NS
P value for comparisons between the study sample and the total MUSP cohort; x2 test for categorical variables and F test for continuous variables. MUSP 5 Mater-University of Queensland Study of Pregnancy; NS 5 not significant.
was used in all analyses. Weight measurements at birth and at 5, 14, and 21 years were converted into Z scores. For the purposes of analysis, the lowest quintile of birth weight was considered to be the IUGR group. Catch-up growth for weight across the whole group was defined as in previous work5,16 as an increase in Z score of at least 0.67 between two measuring points: birth to 5 years, birth to 14 years, and 5 to 14 years.
linear regression for the IUGR group and the whole cohort. Catch-up growth through different measuring points was analyzed to assess the impact of growth through certain periods of childhood on lung function achieved. Maternal smoking in pregnancy, socioeconomic status, and personal smoking status were adjusted for in the analyses.
BMI was calculated at 5, 14, and 21 years of age using the formula of (weight in kilograms)/(height in meters)2. BMI was then grouped into categories of normal, overweight, and obese based on internationally published standards for childhood BMI.17
The association between BMI in childhood and eventual lung function in young adulthood was analyzed using analysis of variance at each age group: 5, 14, and 21 years. Further analysis was undertaken taking into consideration evolving obesity trends from 5 to 21 years. The analyses were conducted separately for men and women.
Statistical Analyses The outcome variables were FVC, FEV1, and FEF25-75. Primary exposure variables were growth through childhood years. The impact of catch-up growth on lung function attained at 21 years was analyzed using multiple
For categorical variables, we primarily used x2 tests to examine the association between the baseline characteristics of the participants and loss to follow-up at 21 years. All analyses were undertaken using Stata version 11.0 (StataCorp).
Results
Lung function values, expressed as group mean (SD), and international Z scores are presented in Table 2. The percentages of abnormal lung function ⱕ 2 SD by international Z scores were 5.84% for FVC (men, 5.96%; women, 5.72%) and 3.8% for FEV1 (men, 4.33%; women, 3.5%). The lung function values of our cohort are comparable to internationally published norms.10,18
Descriptive Findings
A total of 2,612 young adults underwent lung function testing at the 21-year follow-up. From these, 1,740 datasets (50% men) in which all the anthropometric data were available at birth and at 5, 14, and 21 years of age constituted our primary dataset. The loss-to-follow-up data are presented in Table 1. There was no statistically significant difference in maternal lifestyle factors between the male and female offspring in the cohort. The men were heavier at birth and taller and heavier at the age of 21 years (all P , .001). 1120 Original Research
The relationship between catch-up growth and lung function is presented in Table 3. There were statistically significant increases in FVC and FEV1 in subjects with catch-up growth between birth and 5 years. This association remains for FVC for catch-up growth between birth and 14 years in both men and women, but the
[
147#4 CHEST APRIL 2015
]
TABLE 2
] Lung Function at Age 21 y
Lung Function Values
Men (n 5 869)
Women (n 5 871)
Mean (SD)
Range
Mean Z
Z Range
Mean (SD)
Range
Mean Z
Z Range
4.55 (0.67)
1.76-6.81
20.11
25.33 to 3.21
3.33 (0.45)
1.14-4.85
20.17
25.86 to 3.9
FVC,a L
5.38 (0.81)
2.24-9.37
20.28
25.44 to 6.38
3.84 (0.59)
1.42-6.61
20.19
26.13 to 6.31
FEF25-75,a L/s
4.82 (1.13)
0.9-8.14
0.19
24.8 to 2.48
3.76 (0.86)
0.73-7.24
20.17
24.82 to 3.82
FEV1,a L
Z scores are from all-age reference ranges in Stanojevic et al.10 FEF25-75 5 forced expiratory flow, midexpiratory phase. aP , .001 for difference between men and women.
significance of the association with FEV1 drops out in men. Significant catch-up growth from 5 to 14 years was associated with lower lung function (both FVC and FEV1) in men but greater lung function in women. TABLE 3
The IUGR cohort who showed catch-up growth also demonstrated higher values of FVC at age 21 years, regardless of whether catch-up growth was seen at 5 years or at 14 years. The increase in FVC was seen in
] Lung Function Values With and Without Catch-up Growth for the Whole Cohort and the IUGR (Lowest Quintile) With Adjusted Regression Values
IUGR (Lowest Quintile) Lung Function Values
Whole Cohort, Mean (SD)
No Catch-up Growth
Catch-up Growth
No Catch-up Growth
Catch-up Growth
0-5 y
5.33 (5.26-5.39)
5.54 (5.44-5.65)a
4.84 (4.60-5.07)
5.35 (5.19-5.51)a
0-14 y
5.34 (5.27-5.40)
5.52 (5.41-5.62)
4.97 (4.73-5.21)
5.28 (5.11-5.45)a
5-14 y
5.41 (5.35-5.48)
5.25 (5.12-5.38)b
…
…
0-5 y
4.52 (4.46-4.57)
4.66 (4.57-4.75)a
4.16 (3.96-4.37)
4.46 (4.32-4.60)a
0-14 y
4.54 (4.49-4.59)
4.60 (4.51-4.69)
4.28 (4.10-4.50)
4.40 (4.24-4.52)
5-14 y
4.58 (4.53-4.63)
4.43 (4.33-4.54)b
…
…
0-5 y
4.79 (4.70-4.88)
4.93 (4.78-5.08)
4.59 (4.24-4.94)
4.69 (4.45-4.94)
0-14 y
4.83 (4.74-4.92)
4.83 (4.68-4.98)
4.63 (4.28-4.99)
4.67 (4.42-4.91)
5-14 y
4.85 (4.76-4.93)
4.74 (4.56-4.92)
…
…
0-5 y
3.78 (3.74-3.83)
4.03 (3.95-4.12)a
3.55 (3.44-3.65)
3.86 (3.75-3.96)a
0-14 y
3.80 (3.76-3.85)
3.95 (3.86-4.03)
3.57 (3.45-3.68)
3.82 (3.72-3.92)a
5-14 y
3.81 (3.76-3.86)
3.95 (3.86-4.03)a
…
…
0-5 y
3.31 (3.27-3.34)
3.45 (3.38-3.51)a
3.16 (3.07-3.24)
3.32 (3.24-3.40)a
0-14 y
3.32 (3.28-3.35)
3.41 (3.35-3.48)
3.15 (3.06-3.24)
3.31 (3.23-3.40)a
5-14 y
3.31 (3.28-3.35)
3.41 (3.35-3.47)a
…
…
0-5 y
3.76 (3.69-3.82)
3.78 (3.65-3.91)
3.68 (3.51-3.85)
3.68 (3.52-3.85)
0-14 y
3.74 (3.68-3.81)
3.84 (3.70-3.94)
3.62 (3.45-3.80)
3.73 (3.57-3.89)
5-14 y
3.73 (3.66-3.80)
3.84 (3.72-3.97)
…
…
Men (n 5 869) FVC, L
FEV1, L/s
FEF25-75
5.38 (0.81)
a
4.55 (0.67)
4.82 (1.13)
Women (n 5 871) FVC, L
FEV1, L/s
FEF25-75
3.84 (0.59)
a
3.33 (0.45)
a
3.76 (0.87)
Difference between mean values was analyzed by analysis of variance and F test for significance. Multiple linear regression analysis was used for adjusting maternal smoking in pregnancy, maternal education, and current smoking status. IUGR 5 intrauterine growth retardation. See Table 2 legend for expansion of other abbreviation. aStatistically significant increase in lung function. bStatistically significant decrease in lung function.
journal.publications.chestnet.org
1121
both men and women. Greater values of FEV1 were also seen in female subjects with catch-up growth between birth and 5 years and birth and 14 years; however, this was only significant in male subjects for catch-up growth occurring between birth and 5 years.
the obese category, drop in lung function is noted (Figs 1A, 1B). Adjusting for evolving obesity clearly shows a negative correlation with the lung function achieved at 21 years of age (e-Fig 1).
The data presented in Table 3 are adjusted for early life factors. This did not alter the positive associations between adult lung function and catch-up growth in early childhood (unadjusted values are shown in e-Table 1).
In this longitudinal birth cohort study in which growth parameters were available through childhood, we have shown that catch-up growth in early childhood is associated with increased lung function at the age of 21 years. This is in addition to the association between birth weight and adult lung function that has been reported from longitudinal cohorts, including our cohort.2,3 The catch-up growth potentially confers additional gain in lung function.
The BMI trend from 5 years to 21 years and lung function values are presented in Table 4. There was a positive correlation with increasing BMI at 5 years of age and adult lung function. This trend diminishes with increasing age. In the older age groups 14 years and 21 years, the positive correlation with increasing BMI is limited to the overweight category and when the BMI increases to TABLE 4
Discussion
As data were gathered at 5, 14, and 21 years in our cohort, we looked at catch-up growth between the ages
] Lung Function Values at 21 y Grouped by BMI at Different Age Groups (5, 14, and 21 y)
Lung Function Values
Whole Cohort
P Value
Normal
Overweight
Obese
BMI at 5 y
5.35 (0.80)
5.47 (0.89)
5.76 (0.79)
.028a
BMI at 14 y
5.34 (0.80)
5.52 (0.87)
5.47 (0.73)
.028a
BMI at 21 y
5.36 (0.82)
5.50 (0.80)
5.27 (0.81)
.032b
BMI at 5 y
4.53 (0.67)
4.65 (0.65)
4.78 (0.57)
.048a
BMI at 14 y
4.52 (0.67)
4.66 (0.67)
4.57 (0.57)
.060
BMI at 21 y
4.53 (0.69)
4.68 (0.63)
4.42 (0.61)
.002b
BMI at 5 y
4.79 (1.14)
5.04 (1.07)
4.95 (1.06)
.088
BMI at 14 y
4.78 (1.16)
4.96 (1.04)
4.84 (1.08)
.187
BMI at 21 y
4.75 (1.18)
5.06 (1.06)
4.71 (0.98)
.003b
BMI at 5 y
3.80 (0.58)
4.06 (0.60)
3.95 (0.73)
, .001b
BMI at 14 y
3.81 (0.58)
3.89 (0.62)
4.05 (0.66)
.012a
BMI at 21 y
3.81 (0.60)
3.91 (0.55)
3.89 (0.61)
.078
BMI at 5 y
3.31 (0.45)
3.45 (0.45)
3.37 (0.59)
.005b
BMI at 14 y
3.32 (0.44)
3.37 (0.47)
3.42 (0.56)
.154
BMI at 21 y
3.32 (0.45)
3.36 (0.44)
3.35 (0.47)
.564
BMI at 5 y
3.77 (0.85)
3.73 (0.85)
3.76 (1.14)
.909
BMI at 14 y
3.75 (0.85)
3.79 (0.79)
3.75 (1.19)
.867
BMI at 21 y
3.76 (0.85)
3.76 (0.85)
3.79 (0.87)
.938
Men (n 5 869) FVC, mean (SD), L
FEV1, mean (SD), L/s
FEF25-75
5.38 (0.81)
4.55 (0.67)
4.82 (1.13)
Women (n 5 871) FVC, mean (SD), L
FEV1, mean (SD), L/s
FEF25-75
3.84 (0.59)
3.33 (0.45)
3.76 (0.87)
See Table 2 legend for expansion of abbreviation. Statistically significant increase in lung function. bStatistically significant decrease in lung function. a
1122 Original Research
[
147#4 CHEST APRIL 2015
]
Figure 1 – A, Lung function (SD) at 21 y with BMI trends at 5, 14, and 21 y. B, Lung function at 21 y with BMI trends at 5, 14, and 21 y.
of 0 to 5 years, 0 to 14 years, and 5 to 14 years. The strongest association was noted in the 0- to-5-year weight-gain data. This is consistent with current journal.publications.chestnet.org
thinking that most postnatal lung growth occurs in the first few years of life.19,20 As described previously,5,16 we defined catch-up growth as an increase in weight by a 1123
Z score of 0.67 (quartile), and the increasing lung function effect was noted across the whole cohort. We agree that this would include the biologic catch-up growth in the IUGR quintile as well as the rapid growth in the heavier quintiles (2-5 quintiles). The platform for growth in infants with IUGR will include (1) the mechanism and timing of the cause of the IUGR and (2) likely epigenetic effects on metabolism. In addition, the factors (hormonal, nutrition, and social) that drive growth in the appropriately grown infants also likely contribute to the growth in IUGR. We acknowledge that differences in biologic mechanisms may exist because of these factors. We also analyzed the data for the lowest quintile separately because this is likely to reflect the IUGR cohort, and catch-up growth was again strongly associated with improved lung function, suggesting that postnatal growth in the early years contributes to the increased lung function noted in this group compared with the subjects without catch-up growth. When the trend was compared with the 0-to-14-year time period, the effect was less marked. More importantly where there was a significant increase in weight Z scores between 5 and 14 years, reduced lung function was seen, when compared with the group without significant weight gain. This finding suggests that increase in weight gain in later childhood does not impart any positive gains in relation to lung function and is consistent with other adult literature that inversely associates lung function with weight.7,21-23 Our findings vary with evidence suggesting rapid weight gain early in infancy adversely impacts childhood lung function24,25 measured in infancy or at the age of 5 years. Longitudinal follow-up of these cohorts is required to determine whether these early findings translate to an impact on adult lung function. The association noted between weight gain and lung function was marked with FVC but was less so with FEV1 (Table 3). This may suggest that childhood growth has a limited impact on the caliber of the airway measure (FEV1), because this is determined predominately at a fetal level. The other small-airway measure (FEF25-75) also showed no statistical significance. This measure has been reported to have no impact on the clinical decisionmaking process26; however, it is widely used in epidemiologic studies. Consistent with previously published literature, BMI in the younger age group was positively correlated with increasing adult lung function.6,7 However, with increasing age, the BMI-lung function relationship was
1124 Original Research
not linear. Overall, BMI that was at the high end of normal or in the overweight category was associated with the best lung function. It appears that increasing BMI beyond the overweight values does not offer any somatic advantage and produces adverse impacts and reduced lung function. This finding is comparable to those of similar adult obesity studies.21,22 The “inverse U relationship” appears to be more pronounced with increasing age in men (Fig 1). The relationship in women is similar by the time they reach adulthood. However, the positive correlation that is found only at 5 years in male subjects is also found at 14 years in female subjects. The mechanism for this relationship is poorly understood. One possible explanation is that the early onset of puberty in women may augment lung function values in early adult life. Further studies are needed to test this hypothesis. The trends of evolving obesity suggest that a degree of consistent somatic growth confers a positive influence on lung function (e-Fig 1). However, if the progression of obesity is swift during childhood, the eventual lung function at 21 years appears to be less than that which was achieved by subjects with normal BMI. Obesity that develops in later childhood also has the potential to have an impact on chest wall dynamics and to alter the outcome of forced expiratory maneuver and the lung volumes obtained. We also acknowledge that BMI may be limited in expressing the adiposity/lean body mass trends through childhood.7 The strength of the current study comes from the prospectively recorded anthropometric data that have given us the opportunity to visit the impact of catch-up growth and growth trends on adulthood lung function. Although we have adjusted for some of the confounders (maternal smoking during pregnancy, socioeconomic status by maternal education, and current smoking history), there are other potential confounders that could have impacted the observations but could not be taken into account. Another limitation of the study is the attrition rate of subjects who eventually performed the lung function and the lack of lung function measurements at 5 and 14 years. Missing data because of attrition and item nonresponse may cause bias in the analyses, loss of power, or both.27 The most severe case is if the data are missing not at random (ie, where the probability of missing data depends on the outcome of interest). Although this is impossible to test, it seems unlikely that the probability that the data are missing is dependent on the outcome (lung function), once other exposures are
[
147#4 CHEST APRIL 2015
]
taken into account. A variety of modeling strategies have been used in MUSP studies to adjust for attrition, although the use of these methods has not resulted in a marked alteration of findings.28,29 However, the birth weight of the sample that performed the lung function is very comparable to the birth weight of the primary cohort and is unlikely to have influenced our findings, which were also adjusted for socioeconomic status. The association between lung function and growth trends was not adjusted for factors such as childhood illnesses.
Conclusions Our longitudinal cohort study provides evidence of links between early somatic growth and adult lung function at the age of 21 years. The roles of childhood respiratory illness and pubertal status on eventual lung function need to be explored further.
Acknowledgments Author contributions: A. A. M. is the guarantor of this manuscript. S. S. and A. A. M. contributed to the formulation of the analysis plan; M. O. and P. D. S. contributed to the review of the analysis plan; S. S. contributed to the performance of the analysis; A. A. M. contributed to the assessment of the longitudinal dataset and statistical support; and S. S., M. O., P. D. S., and A. A. M. contributed to the preparation of the manuscript. Financial/nonfinancial disclosures: The authors have reported to CHEST the following conflicts of interest: Dr Mamun is supported by a Career Development Award from the NHMRC (ID 1026598). Drs Suresh, O’Callaghan, and Sly have reported that no potential conflicts of interest exist with any companies/organizations whose products or services may be discussed in this article. Role of sponsors: The views expressed in the paper are those of the authors and not necessarily those of any funding body and no funding body influenced the way in which the data were analyzed and presented. Additional information: The e-Figure and e-Table can be found in the Supplemental Materials section of the online article.
4.
5.
6.
7.
8.
References 1. Barker DJ, Godfrey KM, Fall C, Osmond C, Winter PD, Shaheen SO. Relation of birth weight and childhood respiratory infection to adult lung function and death from chronic obstructive airways disease. BMJ. 1991;303(6804):671-675. 2. Suresh S, Mamun AA, O’Callaghan M, Sly PD. The impact of birth weight on peak lung function in young adults. Chest. 2012;142(6):1603-1610. 3. Lawlor DA, Ebrahim S, Davey Smith G. Association of birth weight with adult
journal.publications.chestnet.org
9.
10.
11.
The following is already known: • Lung function in adulthood is predicted by birth weight. • Catch-up growth during childhood is associated with improved lung function but is not statistically significant. • Obesity in adulthood accelerates lung function decline. This study adds the following: • Catch-up growth during early childhood (first 5 years of life) confers improved lung function at the age of 21 years. This is present across all ranges of birth weight. • Early somatic growth and higher BMI at 5 years is associated with higher lung function at 21 years. However, obesity that appears after 5 years of age has an adverse impact on eventual lung function at 21 years.
lung function: findings from the British Women’s Heart and Health Study and a meta-analysis. Thorax. 2005;60(10): 851-858. Hancox RJ, Poulton R, Greene JM, McLachlan CR, Pearce MS, Sears MR. Associations between birth weight, early childhood weight gain and adult lung function. Thorax. 2009;64(3):228-232. Kotecha SJ, Watkins WJ, Heron J, Henderson J, Dunstan FD, Kotecha S. Spirometric lung function in schoolage children: effect of intrauterine growth retardation and catch-up growth. Am J Respir Crit Care Med. 2010;181(9):969-974. Bua J, Prescott E, Schack-Nielsen L, et al. Weight history from birth through childhood and youth in relation to adult lung function, in Danish juvenile obese and non-obese men. Int J Obes (Lond). 2005;29(9):1055-1062. Curry BA, Blizzard CL, Schmidt MD, Walters EH, Dwyer T, Venn AJ. Longitudinal associations of adiposity with adult lung function in the Childhood Determinants of Adult Health (CDAH) study. Obesity (Silver Spring). 2011; 19(10):2069-2075. Najman JM, Bor W, O’Callaghan M, Williams GM, Aird R, Shuttlewood G. Cohort Profile: the Mater-University of Queensland Study of Pregnancy (MUSP). Int J Epidemiol. 2005;34(5):992-997. American Thoracic Society. Standardization of spirometry, 1994 update. Am J Respir Crit Care Med. 1995;152(3): 1107-1136. Stanojevic S, Wade A, Stocks J, et al. Reference ranges for spirometry across all ages: a new approach. Am J Respir Crit Care Med. 2008;177(3):253-260. Stanojevic S. Lung function in growth and aging: a united worldwide approach. Global Lung Function Initiative website.
12.
13.
14.
15.
16.
17.
18.
19.
20.
http://lungfunction.org/growinglungs. Accessed July 22, 2011. Lawlor DA, Ebrahim S, Davey Smith G. Association between self-reported childhood socioeconomic position and adult lung function: findings from the British Women’s Heart and Health Study. Thorax. 2004;59(3):199-203. Hayatbakhsh MR, Sadasivam S, Mamun AA, Najman JM, Williams GM, O’Callaghan MJ. Maternal smoking during and after pregnancy and lung function in early adulthood: a prospective study. Thorax. 2009;64(9):810-814. Edwards CA, Osman LM, Godden DJ, Campbell DM, Douglas JG. Relationship between birth weight and adult lung function: controlling for maternal factors. Thorax. 2003;58(12):1061-1065. Menezes AM, Dumith SC, Perez-Padilla R, et al. Socioeconomic trajectory from birth to adolescence and lung function: prospective birth cohort study. BMC Public Health. 2011;11:596. Ong KK, Ahmed ML, Emmett PM, Preece MA, Dunger DB. Association between postnatal catch-up growth and obesity in childhood: prospective cohort study. BMJ. 2000;320(7240):967-971. Cole TJ, Bellizzi MC, Flegal KM, Dietz WH. Establishing a standard definition for child overweight and obesity worldwide: international survey. BMJ. 2000;320(7244):1240-1243. Equations and tools. Global Lung Function Initiative website. http://www. lungfunction.org/tools.html. Accessed October 23, 2014. Cooney TP, Thurlbeck WM. The radial alveolar count method of Emery and Mithal: a reappraisal 1—postnatal lung growth. Thorax. 1982;37(8):572-579. Thurlbeck WM. Postnatal human lung growth. Thorax. 1982;37(8):564-571.
1125
21. Thyagarajan B, Jacobs DR Jr, Apostol GG, et al. Longitudinal association of body mass index with lung function: the CARDIA study. Respir Res. 2008;9:31. 22. Canoy D, Luben R, Welch A, et al. Abdominal obesity and respiratory function in men and women in the EPICNorfolk Study, United Kingdom. Am J Epidemiol. 2004;159(12):1140-1149. 23. Pistelli F, Bottai M, Carrozzi L, et al. Changes in obesity status and lung function decline in a general population sample. Respir Med. 2008;102(5):674-680. 24. van der Gugten AC, Koopman M, Evelein AM, Verheij TJ, Uiterwaal CS,
1126 Original Research
van der Ent CK. Rapid early weight gain is associated with wheeze and reduced lung function in childhood. Eur Respir J. 2012;39(2):403-410. 25. Lucas JS, Inskip HM, Godfrey KM, et al. Small size at birth and greater postnatal weight gain: relationships to diminished infant lung function. Am J Respir Crit Care Med. 2004;170(5): 534-540. 26. Quanjer PH, Weiner DJ, Pretto JJ, Brazzale DJ, Boros PW. Measurement of FEF25-75% and FEF75% does not contribute to clinical decision making. Eur Respir J. 2014;43(4):1051-1058.
27. Little RJA, Rubin DB. Statistical Analysis With Missing Data. New York, NY: John Wiley; 1987. 28. Alati R, Al Mamun A, O’Callaghan M, Najman JM, Williams GM. In utero and postnatal maternal smoking and asthma in adolescence. Epidemiology. 2006;17(2):138-144. 29. Al Mamun A, Lawlor DA, Alati R, O’Callaghan MJ, Williams GM, Najman JM. Does maternal smoking during pregnancy have a direct effect on future offspring obesity? Evidence from a prospective birth cohort study. Am J Epidemiol. 2006;164(4):317-325.
[
147#4 CHEST APRIL 2015
]
Original Research Pulmonary Physiology
]
Impact of Childhood Anthropometry Trends on Adult Lung Function Sadasivam Suresh, MBBS; Michael O’Callaghan, MD; Peter D. Sly, DSc; and Abdullah A. Mamun, PhD
Poor fetal growth rate is associated with lower respiratory function; however, there is limited understanding of the impact of growth trends and BMI during childhood on adult respiratory function.
BACKGROUND:
The current study data are from the Mater-University of Queensland Study of Pregnancy birth cohort. Prospective data were available from 1,740 young adults who performed standard spirometry at 21 years of age and whose birth weight and weight, height, and BMI at 5, 14, and 21 years of age were available. Catch-up growth was defined as an increase of 0.67 Z score in weight between measurements. The impact of catch-up growth on adult lung function and the relationship between childhood BMI trends and adult lung function were assessed using regression analyses.
METHODS:
Lung function was higher at 21 years in those demonstrating catch-up growth from birth to 5 years (FVC, men: 5.33 L vs 5.54 L; women: 3.78 L vs 4.03 L; and FEV1, men: 4.52 L/s vs 4.64 L/s; women: 3.31 L/s vs 3.45 L/s). Subjects in the lowest quintile of birth (intrauterine growth retardation) also showed improved lung function if they had catch-up growth in the first 5 years of life. There was a positive correlation between increasing BMI and lung function at 5 years of age. However, in the later measurements when BMI increased into the obese category, a drop in lung function was observed.
RESULTS:
These data show evidence for a positive contribution of catch-up growth in early life to adult lung function. However, if weight gain or onset of obesity occurs after 5 years of age, an adverse impact on adult lung function is noted. CHEST 2015; 147(4):1118-1126 CONCLUSIONS:
Manuscript received March 23, 2014; revision accepted September 29, 2014; originally published Online First October 23, 2014. ABBREVIATIONS: FEF25-75 5 forced expiratory flow, midexpiratory phase; IUGR 5 intrauterine growth retardation; MUSP 5 Mater-University of Queensland Study of Pregnancy AFFILIATIONS: From the School of Population Health (Drs Suresh and Mamun), Mater Children’s Hospital, Mater Research Institute (Dr Suresh), Queensland Children’s Medical Research Institute (Drs Suresh and Sly), and the Department of Paediatrics and Child Health, School of Medicine (Dr O’Callaghan), University of Queensland, Brisbane, QLD, Australia.
1118 Original Research
FUNDING/SUPPORT: The core study was funded by the National Health
and Medical Research Council [NHMRC ID 631507] of Australia. Sadasivam Suresh, MBBS, Department of Respiratory and Sleep Paediatrics, Mater Children’s Hospital, South Brisbane, 4101, QLD, Australia; e-mail: [email protected] © 2015 AMERICAN COLLEGE OF CHEST PHYSICIANS. Reproduction of this article is prohibited without written permission from the American College of Chest Physicians. See online for more details. DOI: 10.1378/chest.14-0698 CORRESPONDENCE TO:
[
147#4 CHEST APRIL 2015
]
Increasing evidence suggests that low birth weight is associated with long-term morbidity, including adverse cardiovascular, endocrine, and respiratory outcomes.1 The relationship between birth weight and lung function has been studied in different groups, and a small but direct association has been noted.2,3 The strength of the association declines through adulthood, suggesting that environmental influences contribute to the decline in lung function with age.2 There are few reports in the literature on the impact of intrauterine growth retardation (IUGR) on adult lung function and the role of catch-up growth in early life and how it influences adult lung function.4,5 Hancox et al4 reported a small, direct impact on lung function that did not reach statistical significance. Kotecha et al5 reported lower lung function at school age in those with IUGR and a trend toward higher lung function in those with catch-up growth that was not statistically significant. However, the impact on adult lung function of weight gain or catch-up growth that occurs in childhood has not been studied in detail. Postnatal growth has the potential to contribute to eventual lung function, and a longitudinal study in which details about IUGR, catch-up growth, and adult lung function are available gives us the opportunity to explore this further. Various research is available in the literature on the impact of BMI on lung function; but we found only two
Materials and Methods The Study The MUSP is a prospective study of 8,556 pregnant women interviewed after their first clinic visit in pregnancy, with 7,223 singleton infants constituting the birth cohort. They were followed up at 3 to 5 days, 6 months, and 5, 14, and 21 years.8 At the 21-year follow-up, 5,185 young adults who were singleton babies and whose mothers had agreed at the 14-year follow-up to be contacted again were sent a questionnaire. Almost 3,800 completed the questionnaire, of whom a subsample of 2,612 young adults attended for physical assessment including lung function tests. Because of limited funding, those living outside Brisbane or who were unable to make an appointment for a face-to-face interview completed a mailed questionnaire and did not undergo the physical assessment. Ethics approval was obtained from the institution (University of Queensland Ethics Committee, B/660/SS/01/NHMRC, Mater Hospital 506A), and written consent was appropriately obtained from all participants. For this analysis, we included data obtained from 1,740 young adults who underwent spirometry at 21 years and for whom anthropometric data at birth and at 5, 14, and 21 years were available. The study sample used in these analyses was comparable to the original cohort on most aspects. Loss to follow-up was related primarily to social disadvantage (Table 1). Measurements Lung Function Testing: Lung function testing was performed at the 21-year follow-up using a Spirobank G spirometer system attached to a laptop computer. Qualified and trained interviewers familiar with
journal.publications.chestnet.org
studies that looked at BMI trends through childhood and how this impacts adult lung function.6,7 Bua et al,6 in their study of a Danish cohort, reported advantages of higher BMI during childhood toward higher lung function. However, this association with BMI was not noted in the second decade or during adulthood. Curry et al,7 in their Australian longitudinal cohort, reported that a positive association between higher BMI and lung function was possibly related to lean body mass and that adiposity adversely impacted the adult lung function attained. The aims of the current study were (1) to investigate the association of catch-up growth in childhood at the ages of 5 and 14 years with lung function at 21 years of age using prospectively collected data, (2) to assess the strength of this association in babies born with IUGR, and (3) to describe how BMI at different ages during childhood impact adult lung function. We hypothesized that catch-up growth improves lung function at 21 years and that the onset of obesity in childhood reduces the eventual lung function at 21 years. We used data from the Mater-University of Queensland Study of Pregnancy (MUSP), a large community-based birth cohort study, to investigate these aims. Our study is unique in that it explores all these three aims in a longitudinal cohort.
the instrument performed standard spirometry, in accordance with American Thoracic Society guidelines.9 A minimum of three and a maximum of five trials were attempted. If testing was unsatisfactory for any reason, the reason(s) were noted on the record sheets. For the purpose of this study FVC; FEV1; and forced expiratory flow, midexpiratory phase (FEF25-75), were considered as outcomes of interest. The all-age reference10,11 ranges for spirometry were used to compute Z scores of our lung function values. Anthropometry and Early Life Variables: Birth weight (in grams) and gestational age were recorded at delivery. The following variables were selected from the first clinical visit questionnaire for further analysis: maternal history of smoking, maternal education, and height of both mother and father. These variables were selected in view of the reported associations with lung function in the literature12-15 and their availability in our dataset. Subjects’ smoking history was obtained from the 21-years questionnaire. Maternal history of smoking during pregnancy was classified into the following groups: nonsmokers, mild smokers (1-9 cigarettes per day), and heavy smokers (ⱖ 10 cigarettes per day). The level of mother’s education was assessed at entry to the study, and answers were divided into three groups: incomplete high school, high school completers, and tertiary education. Personal smoking at 21 years was classified as nonsmoker, mild smoker, or heavy smoker (as described previously). Height and weight measurements during the 5-year, 14-year, and 21-year follow-up were documented. Height was measured without shoes using a portable stadiometer to the nearest 0.1 cm, and weight was measured in light clothing with a scale accurate to 0.2 kg. Two measures of weight and height were taken, and the mean of these two measures
1119
TABLE 1
] Early Life Variables in Study Sample and in Total Cohort
Cohort Variables
Initial Cohort, No. (%)
Study Sample (n 5 1,740), No. (%)
P Value
Total MUSP cohort (n 5 7,223) Male
3,758 (52.0)
869 (49.9)
Female
3,465 (48.0)
871 (50.1)
Smokers during pregnancy (n 5 7,152)
NS NS , .001
Nonsmokers
4,395 (61.5)
1,178 (68.4)
Light smokers (, 10/d)
1,191 (16.6)
260 (15.1)
Heavy smokers
1,566 (21.9)
285 (16.5)
Maternal education (n 5 7,170)
, .001
Incomplete high school
1,309 (18.3)
256 (14.8)
Completed high school
4,609 (64.3)
1,105 (64.0)
Post high school
1,252 (17.5)
367 (21.2)
1,886 (50.0)
866 (50.2)
21-y questionnaire (n 5 3,775) Current smoking history Nonsmokers
NS
Light smokers
922 (24.4)
444 (25.8)
NS
Heavy smokers
967 (25.6)
414 (24.0)
NS
Birth weight, g (SD)
3,385 (519)
3,403 (511)
NS
P value for comparisons between the study sample and the total MUSP cohort; x2 test for categorical variables and F test for continuous variables. MUSP 5 Mater-University of Queensland Study of Pregnancy; NS 5 not significant.
was used in all analyses. Weight measurements at birth and at 5, 14, and 21 years were converted into Z scores. For the purposes of analysis, the lowest quintile of birth weight was considered to be the IUGR group. Catch-up growth for weight across the whole group was defined as in previous work5,16 as an increase in Z score of at least 0.67 between two measuring points: birth to 5 years, birth to 14 years, and 5 to 14 years.
linear regression for the IUGR group and the whole cohort. Catch-up growth through different measuring points was analyzed to assess the impact of growth through certain periods of childhood on lung function achieved. Maternal smoking in pregnancy, socioeconomic status, and personal smoking status were adjusted for in the analyses.
BMI was calculated at 5, 14, and 21 years of age using the formula of (weight in kilograms)/(height in meters)2. BMI was then grouped into categories of normal, overweight, and obese based on internationally published standards for childhood BMI.17
The association between BMI in childhood and eventual lung function in young adulthood was analyzed using analysis of variance at each age group: 5, 14, and 21 years. Further analysis was undertaken taking into consideration evolving obesity trends from 5 to 21 years. The analyses were conducted separately for men and women.
Statistical Analyses The outcome variables were FVC, FEV1, and FEF25-75. Primary exposure variables were growth through childhood years. The impact of catch-up growth on lung function attained at 21 years was analyzed using multiple
For categorical variables, we primarily used x2 tests to examine the association between the baseline characteristics of the participants and loss to follow-up at 21 years. All analyses were undertaken using Stata version 11.0 (StataCorp).
Results
Lung function values, expressed as group mean (SD), and international Z scores are presented in Table 2. The percentages of abnormal lung function ⱕ 2 SD by international Z scores were 5.84% for FVC (men, 5.96%; women, 5.72%) and 3.8% for FEV1 (men, 4.33%; women, 3.5%). The lung function values of our cohort are comparable to internationally published norms.10,18
Descriptive Findings
A total of 2,612 young adults underwent lung function testing at the 21-year follow-up. From these, 1,740 datasets (50% men) in which all the anthropometric data were available at birth and at 5, 14, and 21 years of age constituted our primary dataset. The loss-to-follow-up data are presented in Table 1. There was no statistically significant difference in maternal lifestyle factors between the male and female offspring in the cohort. The men were heavier at birth and taller and heavier at the age of 21 years (all P , .001). 1120 Original Research
The relationship between catch-up growth and lung function is presented in Table 3. There were statistically significant increases in FVC and FEV1 in subjects with catch-up growth between birth and 5 years. This association remains for FVC for catch-up growth between birth and 14 years in both men and women, but the
[
147#4 CHEST APRIL 2015
]
TABLE 2
] Lung Function at Age 21 y
Lung Function Values
Men (n 5 869)
Women (n 5 871)
Mean (SD)
Range
Mean Z
Z Range
Mean (SD)
Range
Mean Z
Z Range
4.55 (0.67)
1.76-6.81
20.11
25.33 to 3.21
3.33 (0.45)
1.14-4.85
20.17
25.86 to 3.9
FVC,a L
5.38 (0.81)
2.24-9.37
20.28
25.44 to 6.38
3.84 (0.59)
1.42-6.61
20.19
26.13 to 6.31
FEF25-75,a L/s
4.82 (1.13)
0.9-8.14
0.19
24.8 to 2.48
3.76 (0.86)
0.73-7.24
20.17
24.82 to 3.82
FEV1,a L
Z scores are from all-age reference ranges in Stanojevic et al.10 FEF25-75 5 forced expiratory flow, midexpiratory phase. aP , .001 for difference between men and women.
significance of the association with FEV1 drops out in men. Significant catch-up growth from 5 to 14 years was associated with lower lung function (both FVC and FEV1) in men but greater lung function in women. TABLE 3
The IUGR cohort who showed catch-up growth also demonstrated higher values of FVC at age 21 years, regardless of whether catch-up growth was seen at 5 years or at 14 years. The increase in FVC was seen in
] Lung Function Values With and Without Catch-up Growth for the Whole Cohort and the IUGR (Lowest Quintile) With Adjusted Regression Values
IUGR (Lowest Quintile) Lung Function Values
Whole Cohort, Mean (SD)
No Catch-up Growth
Catch-up Growth
No Catch-up Growth
Catch-up Growth
0-5 y
5.33 (5.26-5.39)
5.54 (5.44-5.65)a
4.84 (4.60-5.07)
5.35 (5.19-5.51)a
0-14 y
5.34 (5.27-5.40)
5.52 (5.41-5.62)
4.97 (4.73-5.21)
5.28 (5.11-5.45)a
5-14 y
5.41 (5.35-5.48)
5.25 (5.12-5.38)b
…
…
0-5 y
4.52 (4.46-4.57)
4.66 (4.57-4.75)a
4.16 (3.96-4.37)
4.46 (4.32-4.60)a
0-14 y
4.54 (4.49-4.59)
4.60 (4.51-4.69)
4.28 (4.10-4.50)
4.40 (4.24-4.52)
5-14 y
4.58 (4.53-4.63)
4.43 (4.33-4.54)b
…
…
0-5 y
4.79 (4.70-4.88)
4.93 (4.78-5.08)
4.59 (4.24-4.94)
4.69 (4.45-4.94)
0-14 y
4.83 (4.74-4.92)
4.83 (4.68-4.98)
4.63 (4.28-4.99)
4.67 (4.42-4.91)
5-14 y
4.85 (4.76-4.93)
4.74 (4.56-4.92)
…
…
0-5 y
3.78 (3.74-3.83)
4.03 (3.95-4.12)a
3.55 (3.44-3.65)
3.86 (3.75-3.96)a
0-14 y
3.80 (3.76-3.85)
3.95 (3.86-4.03)
3.57 (3.45-3.68)
3.82 (3.72-3.92)a
5-14 y
3.81 (3.76-3.86)
3.95 (3.86-4.03)a
…
…
0-5 y
3.31 (3.27-3.34)
3.45 (3.38-3.51)a
3.16 (3.07-3.24)
3.32 (3.24-3.40)a
0-14 y
3.32 (3.28-3.35)
3.41 (3.35-3.48)
3.15 (3.06-3.24)
3.31 (3.23-3.40)a
5-14 y
3.31 (3.28-3.35)
3.41 (3.35-3.47)a
…
…
0-5 y
3.76 (3.69-3.82)
3.78 (3.65-3.91)
3.68 (3.51-3.85)
3.68 (3.52-3.85)
0-14 y
3.74 (3.68-3.81)
3.84 (3.70-3.94)
3.62 (3.45-3.80)
3.73 (3.57-3.89)
5-14 y
3.73 (3.66-3.80)
3.84 (3.72-3.97)
…
…
Men (n 5 869) FVC, L
FEV1, L/s
FEF25-75
5.38 (0.81)
a
4.55 (0.67)
4.82 (1.13)
Women (n 5 871) FVC, L
FEV1, L/s
FEF25-75
3.84 (0.59)
a
3.33 (0.45)
a
3.76 (0.87)
Difference between mean values was analyzed by analysis of variance and F test for significance. Multiple linear regression analysis was used for adjusting maternal smoking in pregnancy, maternal education, and current smoking status. IUGR 5 intrauterine growth retardation. See Table 2 legend for expansion of other abbreviation. aStatistically significant increase in lung function. bStatistically significant decrease in lung function.
journal.publications.chestnet.org
1121
both men and women. Greater values of FEV1 were also seen in female subjects with catch-up growth between birth and 5 years and birth and 14 years; however, this was only significant in male subjects for catch-up growth occurring between birth and 5 years.
the obese category, drop in lung function is noted (Figs 1A, 1B). Adjusting for evolving obesity clearly shows a negative correlation with the lung function achieved at 21 years of age (e-Fig 1).
The data presented in Table 3 are adjusted for early life factors. This did not alter the positive associations between adult lung function and catch-up growth in early childhood (unadjusted values are shown in e-Table 1).
In this longitudinal birth cohort study in which growth parameters were available through childhood, we have shown that catch-up growth in early childhood is associated with increased lung function at the age of 21 years. This is in addition to the association between birth weight and adult lung function that has been reported from longitudinal cohorts, including our cohort.2,3 The catch-up growth potentially confers additional gain in lung function.
The BMI trend from 5 years to 21 years and lung function values are presented in Table 4. There was a positive correlation with increasing BMI at 5 years of age and adult lung function. This trend diminishes with increasing age. In the older age groups 14 years and 21 years, the positive correlation with increasing BMI is limited to the overweight category and when the BMI increases to TABLE 4
Discussion
As data were gathered at 5, 14, and 21 years in our cohort, we looked at catch-up growth between the ages
] Lung Function Values at 21 y Grouped by BMI at Different Age Groups (5, 14, and 21 y)
Lung Function Values
Whole Cohort
P Value
Normal
Overweight
Obese
BMI at 5 y
5.35 (0.80)
5.47 (0.89)
5.76 (0.79)
.028a
BMI at 14 y
5.34 (0.80)
5.52 (0.87)
5.47 (0.73)
.028a
BMI at 21 y
5.36 (0.82)
5.50 (0.80)
5.27 (0.81)
.032b
BMI at 5 y
4.53 (0.67)
4.65 (0.65)
4.78 (0.57)
.048a
BMI at 14 y
4.52 (0.67)
4.66 (0.67)
4.57 (0.57)
.060
BMI at 21 y
4.53 (0.69)
4.68 (0.63)
4.42 (0.61)
.002b
BMI at 5 y
4.79 (1.14)
5.04 (1.07)
4.95 (1.06)
.088
BMI at 14 y
4.78 (1.16)
4.96 (1.04)
4.84 (1.08)
.187
BMI at 21 y
4.75 (1.18)
5.06 (1.06)
4.71 (0.98)
.003b
BMI at 5 y
3.80 (0.58)
4.06 (0.60)
3.95 (0.73)
, .001b
BMI at 14 y
3.81 (0.58)
3.89 (0.62)
4.05 (0.66)
.012a
BMI at 21 y
3.81 (0.60)
3.91 (0.55)
3.89 (0.61)
.078
BMI at 5 y
3.31 (0.45)
3.45 (0.45)
3.37 (0.59)
.005b
BMI at 14 y
3.32 (0.44)
3.37 (0.47)
3.42 (0.56)
.154
BMI at 21 y
3.32 (0.45)
3.36 (0.44)
3.35 (0.47)
.564
BMI at 5 y
3.77 (0.85)
3.73 (0.85)
3.76 (1.14)
.909
BMI at 14 y
3.75 (0.85)
3.79 (0.79)
3.75 (1.19)
.867
BMI at 21 y
3.76 (0.85)
3.76 (0.85)
3.79 (0.87)
.938
Men (n 5 869) FVC, mean (SD), L
FEV1, mean (SD), L/s
FEF25-75
5.38 (0.81)
4.55 (0.67)
4.82 (1.13)
Women (n 5 871) FVC, mean (SD), L
FEV1, mean (SD), L/s
FEF25-75
3.84 (0.59)
3.33 (0.45)
3.76 (0.87)
See Table 2 legend for expansion of abbreviation. Statistically significant increase in lung function. bStatistically significant decrease in lung function. a
1122 Original Research
[
147#4 CHEST APRIL 2015
]
Figure 1 – A, Lung function (SD) at 21 y with BMI trends at 5, 14, and 21 y. B, Lung function at 21 y with BMI trends at 5, 14, and 21 y.
of 0 to 5 years, 0 to 14 years, and 5 to 14 years. The strongest association was noted in the 0- to-5-year weight-gain data. This is consistent with current journal.publications.chestnet.org
thinking that most postnatal lung growth occurs in the first few years of life.19,20 As described previously,5,16 we defined catch-up growth as an increase in weight by a 1123
Z score of 0.67 (quartile), and the increasing lung function effect was noted across the whole cohort. We agree that this would include the biologic catch-up growth in the IUGR quintile as well as the rapid growth in the heavier quintiles (2-5 quintiles). The platform for growth in infants with IUGR will include (1) the mechanism and timing of the cause of the IUGR and (2) likely epigenetic effects on metabolism. In addition, the factors (hormonal, nutrition, and social) that drive growth in the appropriately grown infants also likely contribute to the growth in IUGR. We acknowledge that differences in biologic mechanisms may exist because of these factors. We also analyzed the data for the lowest quintile separately because this is likely to reflect the IUGR cohort, and catch-up growth was again strongly associated with improved lung function, suggesting that postnatal growth in the early years contributes to the increased lung function noted in this group compared with the subjects without catch-up growth. When the trend was compared with the 0-to-14-year time period, the effect was less marked. More importantly where there was a significant increase in weight Z scores between 5 and 14 years, reduced lung function was seen, when compared with the group without significant weight gain. This finding suggests that increase in weight gain in later childhood does not impart any positive gains in relation to lung function and is consistent with other adult literature that inversely associates lung function with weight.7,21-23 Our findings vary with evidence suggesting rapid weight gain early in infancy adversely impacts childhood lung function24,25 measured in infancy or at the age of 5 years. Longitudinal follow-up of these cohorts is required to determine whether these early findings translate to an impact on adult lung function. The association noted between weight gain and lung function was marked with FVC but was less so with FEV1 (Table 3). This may suggest that childhood growth has a limited impact on the caliber of the airway measure (FEV1), because this is determined predominately at a fetal level. The other small-airway measure (FEF25-75) also showed no statistical significance. This measure has been reported to have no impact on the clinical decisionmaking process26; however, it is widely used in epidemiologic studies. Consistent with previously published literature, BMI in the younger age group was positively correlated with increasing adult lung function.6,7 However, with increasing age, the BMI-lung function relationship was
1124 Original Research
not linear. Overall, BMI that was at the high end of normal or in the overweight category was associated with the best lung function. It appears that increasing BMI beyond the overweight values does not offer any somatic advantage and produces adverse impacts and reduced lung function. This finding is comparable to those of similar adult obesity studies.21,22 The “inverse U relationship” appears to be more pronounced with increasing age in men (Fig 1). The relationship in women is similar by the time they reach adulthood. However, the positive correlation that is found only at 5 years in male subjects is also found at 14 years in female subjects. The mechanism for this relationship is poorly understood. One possible explanation is that the early onset of puberty in women may augment lung function values in early adult life. Further studies are needed to test this hypothesis. The trends of evolving obesity suggest that a degree of consistent somatic growth confers a positive influence on lung function (e-Fig 1). However, if the progression of obesity is swift during childhood, the eventual lung function at 21 years appears to be less than that which was achieved by subjects with normal BMI. Obesity that develops in later childhood also has the potential to have an impact on chest wall dynamics and to alter the outcome of forced expiratory maneuver and the lung volumes obtained. We also acknowledge that BMI may be limited in expressing the adiposity/lean body mass trends through childhood.7 The strength of the current study comes from the prospectively recorded anthropometric data that have given us the opportunity to visit the impact of catch-up growth and growth trends on adulthood lung function. Although we have adjusted for some of the confounders (maternal smoking during pregnancy, socioeconomic status by maternal education, and current smoking history), there are other potential confounders that could have impacted the observations but could not be taken into account. Another limitation of the study is the attrition rate of subjects who eventually performed the lung function and the lack of lung function measurements at 5 and 14 years. Missing data because of attrition and item nonresponse may cause bias in the analyses, loss of power, or both.27 The most severe case is if the data are missing not at random (ie, where the probability of missing data depends on the outcome of interest). Although this is impossible to test, it seems unlikely that the probability that the data are missing is dependent on the outcome (lung function), once other exposures are
[
147#4 CHEST APRIL 2015
]
taken into account. A variety of modeling strategies have been used in MUSP studies to adjust for attrition, although the use of these methods has not resulted in a marked alteration of findings.28,29 However, the birth weight of the sample that performed the lung function is very comparable to the birth weight of the primary cohort and is unlikely to have influenced our findings, which were also adjusted for socioeconomic status. The association between lung function and growth trends was not adjusted for factors such as childhood illnesses.
Conclusions Our longitudinal cohort study provides evidence of links between early somatic growth and adult lung function at the age of 21 years. The roles of childhood respiratory illness and pubertal status on eventual lung function need to be explored further.
Acknowledgments Author contributions: A. A. M. is the guarantor of this manuscript. S. S. and A. A. M. contributed to the formulation of the analysis plan; M. O. and P. D. S. contributed to the review of the analysis plan; S. S. contributed to the performance of the analysis; A. A. M. contributed to the assessment of the longitudinal dataset and statistical support; and S. S., M. O., P. D. S., and A. A. M. contributed to the preparation of the manuscript. Financial/nonfinancial disclosures: The authors have reported to CHEST the following conflicts of interest: Dr Mamun is supported by a Career Development Award from the NHMRC (ID 1026598). Drs Suresh, O’Callaghan, and Sly have reported that no potential conflicts of interest exist with any companies/organizations whose products or services may be discussed in this article. Role of sponsors: The views expressed in the paper are those of the authors and not necessarily those of any funding body and no funding body influenced the way in which the data were analyzed and presented. Additional information: The e-Figure and e-Table can be found in the Supplemental Materials section of the online article.
4.
5.
6.
7.
8.
References 1. Barker DJ, Godfrey KM, Fall C, Osmond C, Winter PD, Shaheen SO. Relation of birth weight and childhood respiratory infection to adult lung function and death from chronic obstructive airways disease. BMJ. 1991;303(6804):671-675. 2. Suresh S, Mamun AA, O’Callaghan M, Sly PD. The impact of birth weight on peak lung function in young adults. Chest. 2012;142(6):1603-1610. 3. Lawlor DA, Ebrahim S, Davey Smith G. Association of birth weight with adult
journal.publications.chestnet.org
9.
10.
11.
The following is already known: • Lung function in adulthood is predicted by birth weight. • Catch-up growth during childhood is associated with improved lung function but is not statistically significant. • Obesity in adulthood accelerates lung function decline. This study adds the following: • Catch-up growth during early childhood (first 5 years of life) confers improved lung function at the age of 21 years. This is present across all ranges of birth weight. • Early somatic growth and higher BMI at 5 years is associated with higher lung function at 21 years. However, obesity that appears after 5 years of age has an adverse impact on eventual lung function at 21 years.
lung function: findings from the British Women’s Heart and Health Study and a meta-analysis. Thorax. 2005;60(10): 851-858. Hancox RJ, Poulton R, Greene JM, McLachlan CR, Pearce MS, Sears MR. Associations between birth weight, early childhood weight gain and adult lung function. Thorax. 2009;64(3):228-232. Kotecha SJ, Watkins WJ, Heron J, Henderson J, Dunstan FD, Kotecha S. Spirometric lung function in schoolage children: effect of intrauterine growth retardation and catch-up growth. Am J Respir Crit Care Med. 2010;181(9):969-974. Bua J, Prescott E, Schack-Nielsen L, et al. Weight history from birth through childhood and youth in relation to adult lung function, in Danish juvenile obese and non-obese men. Int J Obes (Lond). 2005;29(9):1055-1062. Curry BA, Blizzard CL, Schmidt MD, Walters EH, Dwyer T, Venn AJ. Longitudinal associations of adiposity with adult lung function in the Childhood Determinants of Adult Health (CDAH) study. Obesity (Silver Spring). 2011; 19(10):2069-2075. Najman JM, Bor W, O’Callaghan M, Williams GM, Aird R, Shuttlewood G. Cohort Profile: the Mater-University of Queensland Study of Pregnancy (MUSP). Int J Epidemiol. 2005;34(5):992-997. American Thoracic Society. Standardization of spirometry, 1994 update. Am J Respir Crit Care Med. 1995;152(3): 1107-1136. Stanojevic S, Wade A, Stocks J, et al. Reference ranges for spirometry across all ages: a new approach. Am J Respir Crit Care Med. 2008;177(3):253-260. Stanojevic S. Lung function in growth and aging: a united worldwide approach. Global Lung Function Initiative website.
12.
13.
14.
15.
16.
17.
18.
19.
20.
http://lungfunction.org/growinglungs. Accessed July 22, 2011. Lawlor DA, Ebrahim S, Davey Smith G. Association between self-reported childhood socioeconomic position and adult lung function: findings from the British Women’s Heart and Health Study. Thorax. 2004;59(3):199-203. Hayatbakhsh MR, Sadasivam S, Mamun AA, Najman JM, Williams GM, O’Callaghan MJ. Maternal smoking during and after pregnancy and lung function in early adulthood: a prospective study. Thorax. 2009;64(9):810-814. Edwards CA, Osman LM, Godden DJ, Campbell DM, Douglas JG. Relationship between birth weight and adult lung function: controlling for maternal factors. Thorax. 2003;58(12):1061-1065. Menezes AM, Dumith SC, Perez-Padilla R, et al. Socioeconomic trajectory from birth to adolescence and lung function: prospective birth cohort study. BMC Public Health. 2011;11:596. Ong KK, Ahmed ML, Emmett PM, Preece MA, Dunger DB. Association between postnatal catch-up growth and obesity in childhood: prospective cohort study. BMJ. 2000;320(7240):967-971. Cole TJ, Bellizzi MC, Flegal KM, Dietz WH. Establishing a standard definition for child overweight and obesity worldwide: international survey. BMJ. 2000;320(7244):1240-1243. Equations and tools. Global Lung Function Initiative website. http://www. lungfunction.org/tools.html. Accessed October 23, 2014. Cooney TP, Thurlbeck WM. The radial alveolar count method of Emery and Mithal: a reappraisal 1—postnatal lung growth. Thorax. 1982;37(8):572-579. Thurlbeck WM. Postnatal human lung growth. Thorax. 1982;37(8):564-571.
1125
21. Thyagarajan B, Jacobs DR Jr, Apostol GG, et al. Longitudinal association of body mass index with lung function: the CARDIA study. Respir Res. 2008;9:31. 22. Canoy D, Luben R, Welch A, et al. Abdominal obesity and respiratory function in men and women in the EPICNorfolk Study, United Kingdom. Am J Epidemiol. 2004;159(12):1140-1149. 23. Pistelli F, Bottai M, Carrozzi L, et al. Changes in obesity status and lung function decline in a general population sample. Respir Med. 2008;102(5):674-680. 24. van der Gugten AC, Koopman M, Evelein AM, Verheij TJ, Uiterwaal CS,
1126 Original Research
van der Ent CK. Rapid early weight gain is associated with wheeze and reduced lung function in childhood. Eur Respir J. 2012;39(2):403-410. 25. Lucas JS, Inskip HM, Godfrey KM, et al. Small size at birth and greater postnatal weight gain: relationships to diminished infant lung function. Am J Respir Crit Care Med. 2004;170(5): 534-540. 26. Quanjer PH, Weiner DJ, Pretto JJ, Brazzale DJ, Boros PW. Measurement of FEF25-75% and FEF75% does not contribute to clinical decision making. Eur Respir J. 2014;43(4):1051-1058.
27. Little RJA, Rubin DB. Statistical Analysis With Missing Data. New York, NY: John Wiley; 1987. 28. Alati R, Al Mamun A, O’Callaghan M, Najman JM, Williams GM. In utero and postnatal maternal smoking and asthma in adolescence. Epidemiology. 2006;17(2):138-144. 29. Al Mamun A, Lawlor DA, Alati R, O’Callaghan MJ, Williams GM, Najman JM. Does maternal smoking during pregnancy have a direct effect on future offspring obesity? Evidence from a prospective birth cohort study. Am J Epidemiol. 2006;164(4):317-325.
[
147#4 CHEST APRIL 2015
]
