Abstract
Objective:
Preterm infants, and especially those with additional comorbidities, are at risk of early life growth failure, which may impact postnatal lung growth and attainment of peak lung function. However, little is known about the early life growth patterns of those with chronic lung disease. The goal of this study was to describe the patterns appreciated in this population and their association with certain clinical characteristics.
Study Design:
Demographic, clinical characteristics, and somatic growth parameters between birth and 3 years were retrospectively reviewed for a cohort of children (n = 616) recruited from an outpatient pulmonary clinic. Group-based trajectory modeling was used to identify unique longitudinal trajectories for each growth parameter. Demographic and clinical characteristics were compared using non-parametric analysis.
Results:
Four distinct trajectories were appreciated in all three somatic growth domains (weight, length, and weight-for-length), which demonstrated a sizable proportion of subjects with a z-score below zero at 36 months of age, suggesting that the traditional preterm paradigm of “catch-up” growth may not be accurate for this population.
Conclusions:
Children with a history of chronic lung disease begin life with somatic growth measurements well below their term peers and display heterogeneous patterns of weight and length growth through the first 3 years of life. Future studies should focus on further understanding the relationship between somatic growth and respiratory outcomes in this population, which will ideally allow for the use of somatic growth measures as surrogate markers to identify individuals at the highest risk of postnatal growth failure and poor respiratory outcomes.
Keywords: bronchopulmonary dysplasia, chronic lung disease of prematurity, growth
1 |. INTRODUCTION
Each year, an estimated 10% of the 4 million infants born in the United States are born prematurely at less than 37 weeks gestational age.1 Bronchopulmonary dysplasia (BPD) is the major respiratory complication of prematurity and affects all components of the developing lung, including the alveoli, airways, and pulmonary vasculature.2 While the survival of preterm infants has improved due to advances in neonatal care including gentler ventilation strategies, exogenous surfactant, parenteral nutrition, and the widespread use of prenatal steroids, BPD continues to affect up to 50,000 infants annually.3
The traditional model of pulmonary function describes lung growth and development beginning in utero and continuing through early adulthood, with maximal lung function attained at 25 years of age followed by a slow decline throughout the remaining lifespan.4 For most individuals who are healthy and avoid significant sources of lung damage, normal age-related decline in lung function is seldom associated with notable clinical morbidity or disability. However, for individuals who are born preterm, failure to achieve postnatal somatic “catch-up” growth, which is historically thought to occur in the first 2 years of life,5 may limit lung growth, resulting in lower peak lung function and lifelong pulmonary repercussions.
Given the reported association between reduced lung function in early life and chronic obstructive pulmonary disease (COPD) in adulthood, there is concern that children with chronic lung disease, who may have impaired lung function in childhood compared to their term peers,6–9 are at increased risk of early-onset COPD.10–12 This highlights the importance of early identification of those individuals at the highest risk of lower lung function trajectories to allow for targeted intervention.
However, identifying preterm infants and young children at greatest risk for long-term respiratory morbidity is hampered by the absence of noninvasive lung function measurements before 6 years of age when traditional spirometry can be reliably performed. Although infant pulmonary function testing is performed in some tertiary care centers13 the technique requires sedation and expertize in children with significant underlying lung disease and is not readily available in most clinical settings. Impulse oscillometry has also been used in children with BPD but longitudinal studies have been limited.14
For children less than 3 years of age, somatic growth measurements (weight, height, and weight-for-length) are potential attractive surrogate measures given the ease in acquisition in virtually all clinical settings. A recent literature review15 demonstrated that there are clear differences in nutrition and growth between infants and children with BPD and those without BPD with infants and children with BPD demonstrating poorer growth, which was attributed to a combination of increased work of breathing,16 growth suppression from chronic stress and inflammation,17 and chronic steroid or diuretic use. To date, there is little information regarding early life somatic growth patterns in infants and children with chronic lung disease; therefore, the goal of this study was to identify distinct trajectories of somatic growth and determine whether patient characteristics were associated with trajectory group membership. We hypothesized that differing somatic growth patterns would be identified in preterm infants and children with chronic lung disease and that clinical factors would differ between trajectory groups.
2 |. METHODS
2.1 |. Population and study design
We conducted a retrospective longitudinal cohort study to describe somatic growth in children (n = 616) recruited from an outpatient pulmonary clinic between January 2008 and January 2018. All subjects were born at less than or equal to 32 weeks gestation and assessed for BPD per the 2001 National Heart, Lung, and Blood Institute (NHLBI) consensus statement.18 Children with at least two outpatient clinical encounters in which somatic growth measurements (weight, length, and weight-for-length) were collected between the ages of 0 and 36 months were included. Although subjects received outpatient pulmonary care at a single center, they received their neonatal care at various neonatal intensive care units (NICUs) across Maryland. All caregivers were consented per study protocol as approved by the Johns Hopkins University Institutional Review Board (Protocol NA_00051884).
2.2 |. Clinical data
Outpatient encounter data, including clinical characteristics and growth parameters from birth through 36 months of age, were obtained through medical record extraction and chart review. Small for gestational age (SGA) (<10th percentile for birth weight corrected for gestational age) was used as a marker of reduced prenatal growth.19 The diagnosis and presence of pulmonary hypertension status was based on echocardiographic data at or after approximately 36 weeks corrected gestational age. Feeding tube status (GT) was defined by the use of any surgically placed feeding tube before 3 years of age. Median household income was based on residential zip code and US Census Data. The primary outcomes of interest, and longitudinally collected growth parameters (height, weight, and weight-for-length) using the patient’s chronological age not corrected for gestation were converted to z-scores using Centers for Disease Control and Prevention (CDC) normative data.20 The decision was made to utilize the CDC growth curves instead of the Fenton growth curves, which assess the growth parameters of preterm infants between 22 and 50 weeks postmenstrual age, to model growth parameters through 36 months postnatal age.21
2.3 |. Statistical methods
Descriptive statistics were used to summarize the demographic and clinical characteristics of the study population. Group-based trajectory modeling (GBTM) was used to identify and characterize the distinct longitudinal trajectories for weight, length, and weight-for-length z-scores from birth to 36 months of age. Briefly, GBTM is a semi-parametric application of finite mixed modeling used to identify unobserved groups, termed classes, that represent unique growth profiles within the distribution of all individual trajectories22,23 and importantly can accommodate the irregular patterns (timing) of growth measurements that were collected. All GBTM models were fitted with a censored normal distribution bounded by −15 and 4 for somatic growth z-scores and used chronological age as the time scale.
To identify the best fitting trajectory models for each of the somatic growth parameters, we first identified the number of classes that characterized each growth pattern by fitting the one class (or one group) solution. Trajectory shapes were then selected from various functional forms (linear, quadratic, and cubic) characterizing each z-score across age and repeated for models with two to six classes. The best fitting model for each somatic growth parameter was chosen based on recommendations of Nagin and Odgers24; the model with the lowest Bayesian Information Criteria and which met the following criteria: average posterior probability of classification of children within each class is greater than 70%, odds of correct classification greater than 5.0 for each trajectory group and having at least 5% of the study population assigned to each class was chosen. Following identification of the best fitting model, children were then categorized into the trajectory group for which they had the highest probability of being classified.
Demographic and clinical characteristics of subjects were summarized and compared between trajectory classes using analysis of variance and χ2 tests for continuous and categorical variables, respectively. Finally, we determined the correlation between the weight and length growth trajectories of all children.
p < 0.05 were considered statistically significant. STATA IC 14 (StatCorp LP) was used for analyses with GBTM performed using the traj plug-in.25
3 |. RESULTS
3.1 |. Demographics and clinical characteristics
A total of 616 children met the inclusion criteria and comprised the study population. Table 1 summarizes the demographic and clinical characteristics of the study population. Overall, 44.2% of the children were female, and 64.6% identified as nonwhite race/ethnicity. Public insurance was common (55.8%), and the median zip code household income was $76,800 ± $27,100. The mean gestational age of children was 26.8 ± 2.3 weeks, and the average birth weight was 919.8 ± 333.0 grams. 12.2% of the study population met the criteria for SGA. At the time of initial discharge from the NICU, 241 (39.1%) infants required supplemental oxygen with an average amount of 0.37 ± 0.34 liters per minute. Six hundred (97.4%) subjects had data available that allowed them to be classified according to the severity of their lung disease with 5.5% with chronic lung disease of prematurity (no BPD), 12.5% with mild BPD, 31.3% with moderate BPD, and 50.7% with severe BPD. GTs were placed in 30.2% of infants and children, 18.2% underwent Nissen fundoplication, 16.9% were diagnosed with pulmonary hypertension, 4.9% of patients had a tracheostomy, and 3.7% required home mechanical ventilation at time of discharge from the NICU.
TABLE 1.
Demographic and clinical characteristics of study population and by weight z-score trajectory group.
| Mean ± S.D. [Range] | Study population (n = 616) | Group 1a (n = 86) 14.3% | Group 2 (n = 275) 44% | Group 3 (n = 218) 35.1% | Group 4 (n = 37) 6.6% | p Valueb | |
|---|---|---|---|---|---|---|---|
| Sex (% female) | 44.2% | 36.0% | 41.1% | 52.3% | 37.8% | 0.021 | |
| Gestational age (weeks) | 26.8 ±2.3 [22.3–32.0] | 27.6 ±2.4 [22.3–32.0] | 26.9 ±2.3 [22.9–32.0] | 26.4 ±2.2 [22.3–32.0] | 26.5 ±2.3 [23.4–32.0] | 0.868 | |
| Birth weight (g, n = 614) | 919.8 ±333.0 [380–2310] | 1175.7 ±371.7 [525–2310] | 948.9 ±323.3 [400–2130] | 822.0 ±271.6 [380– 2070] | 683.6 ±221.9 [409–1417] | 0.001 | |
| Small-for-gestational age (% birth weight percentile <10th%) | 12.2% | 1.2% | 9.8% | 14.2% | 43.2% | 0.001 | |
| Race/ethnicity (% nonwhite) | 64.6% | 79.1% | 67.6% | 57.3% | 51.4% | 0.001 | |
| Median household income ($’000 s) | 76.8 ±27.1 [28.5–186.6] | 76.7 ±33.6 [28.5–186.6] | 74.7 ±25.4 [28.5–156.8] | 79.0 ±26.3 [29.2–156.8] | 79.3 ±27.7 [28.5–156.8] | 0.008 | |
| Public insurance (% yes) | 55.8% | 59.3% | 59.3% | 49.5% | 59.5% | 0.142 | |
| Supplemental oxygen at NICU discharge (% yes) | 39.1% | 32.6% | 38.2% | 41.7% | 45.9% | 0.391 | |
| Oxygen amount at NICU discharge (LPM; n = 241) | 0.37 ±0.34 [0.031–2.00] (n = 241) | 0.31 ±0.21 [0.0625–1.00] (n = 28) | 0.38 ± 0.34 [0.03125–1.5] (n = 105) | 0.39 ±0.38 [0.0625–2.00] (n = 91) | 0.26 ±0.19 [0.125–0.75] (n= 17) | 0.001 | |
| Severity of BPD (n = 600) | None | 5.5% | 10.8% | 6.0% | 3.8% | 0.0% | 0.003 |
| Mild | 12.5% | 20.5% | 10.9% | 13.1% | 2.7% | ||
| Moderate | 31.3% | 33.7% | 34.1% | 27.2% | 29.7% | ||
| Severe | 50.7% | 34.9% | 49.1% | 55.9% | 67.6% | ||
| Gastrostomy tube (% yes) | 30.2% | 12.8% | 32.4% | 33.5% | 35.1% | 0.002 | |
| Nissen fundoplication (% yes) | 18.2% | 8.1% | 20.7% | 19.3% | 16.2% | 0.063 | |
| Pulmonary hypertension (% yes) | 16.9% | 11.6% | 18.5% | 17.0% | 16.2% | 0.523 | |
| Tracheostomy (% yes) | 4.9% | 3.5% | 7.6% | 2.3% | 2.7% | 0.039 | |
| Mechanical ventilation (% yes) | 3.7% | 2.3% | 6.2% | 1.8% | 0% | 0.034 | |
| VP shunt (% yes) | 8.0% | 4.7% | 8.7% | 8.7% | 5.4% | 0.571 | |
| PDA (% yes) | 34.7% | 27.9% | 32.4% | 40.8% | 32.4% | 0.107 | |
| Diuretic use (% yes) | 62.0% | 58.1% | 61.8% | 65.6% | 51.4% | 0.317 | |
Groups arbitrarily labeled 1 through 4 based on weight z-score at 36 months of age with group 1 having the highest z-score and group 4 having the lowest z-score
p Values based on analysis of variance (ANOVA) and student t-tests or Chi-square and Fisher exact tests for continuous and categorical variables, respectively.
3.2 |. Weight
Amongst the 616 children, four distinct weight z-score trajectories were identified based on GBTM procedures (Figure 1), which we labeled one through four based on the group z-score at 36 months with group 1 having the highest weight z-score value and group 4 having the lowest weight z-score. Weight z-score trajectory groups were best characterized by the cubic function with 14.3% of children classified into Group 1, 44.0% classified into Group 2, 35.1% classified into Group 3, and 6.6% classified into Group 4. 94% of the population comprised the top three trajectories, which are within two standard deviations of the mean by age 36 months.
FIGURE 1.
Trajectories of weight z-score during the first 36 months of life. Groups arbitrarily labeled 1 through 4 based on weight z-score at 36 months of age with group 1 having the highest z-score and group 4 having the lowest z-score. Demographic and clinical characteristics of each group are further detailed in the text.
See Table 1 for demographic and clinical characteristics of the children classified into each weight trajectory group. Notably, birth weight significantly differed between trajectory groups, with those classified in group 1 having the highest birth weight (1175.7 ± 371.1 grams) followed by those in group 2 (948.9 ± 323.3 grams), group 3 (822.0 ± 271.6 grams), and group 4 (683.6 ± 221.9 grams). The proportion of infants born SGA demonstrated a similar pattern with the lowest proportion in group 1 (1.2%) followed by 9.8%, 14.2%, and 43.2% for groups 2, 3, and 4. A statistically greater percentage of infants in group 1 identified as non-White (79.1%) as compared to the percentage in group 2 (67.6%), group 3 (57.3%), and group 4 (51.4%). Distribution of BPD severity also differed across weight trajectory groups with the lowest proportion of severe BPD in group 1 (34.9%) and highest in group 4 (67.6%). Frequency of GT placement was lowest in group 1 (12.8%) with similar rates in the other three trajectory groups (32.4%, 33.5%, 35.1%, p = 0.002). History of Nissen fundoplication did not differ between groups.
Further analysis was performed by collapsing the original trajectory groups into two distinct groups: group one which consisted of the original groups one, two, and three who fit the “normal” distribution, within two standard deviations of the general population mean, at 36 months of age, and group two which consisted of the original group 4 (Supporting Information: Table 1).
3.3 |. Length
Four distinct length z-score trajectories were identified amongst the 590 children with at least two length measures and were labeled from Group 1 through Group 4 according to length z-score at age 36 months (Figure 2). Given the variability in birth length measurements appreciated, birth measurements were excluded in this analysis, and the study population included in the length analysis was slightly smaller (590 subjects compared to 616 subjects). Length z-score trajectory groups were best characterized by a cubic function with 19.8% of children classified into Group 1, 57.6% classified into Group 2, 6.9% classified into Group 3, and 15.6% classified into Group 4. The majority of children (84.4%) were categorized into groups 1 through 3 which are all within two standard deviations of the general population mean length by 36 months of age. Similar to the weight trajectories, birth weight differed by trajectory group with the highest average birth weight (1193.7 ± 360.0 grams) in group 1 followed by group 2 (884.8 ± 287.2 grams), group 3 (786.0 ± 234.1 grams), and group 4 (725.8 ± 235.1 grams). SGA also significantly increased across groups with the lowest proportion (6.0%) in group 1. Distribution of BPD severity18 differed between groups with group 1 having the highest proportion with no BPD (11.3%) and the smallest proportion with severe BPD (31.3%). Children in group 1 had the lowest frequency of GT placement (14.5%) and smallest proportion with a Nissen fundoplication performed (7.7%), both of which differed significantly across groups. Overall, subjects classified in group 1 generally represented a healthier profile amongst this population with lower incidence of pulmonary hypertension, ventriculoperitoneal shunts, patent ductus arteriosus, and diuretic use. See Supporting Information: Table 2 for additional demographic and clinical characteristics of the population by length trajectory group.
FIGURE 2.
Trajectories of length z-score during the first 36 months of life. Groups arbitrarily labeled 1 through 4 based on length z-score at 6 months of age with group 1 having the highest (least negative) z-score and group 4 having the lowest (most negative) zscore. Demographic and clinical characteristics of each group are further detailed in the text.
Comparison of subjects’ height and weight trajectory groups demonstrated a moderate positive correlation between the two groups (r = 0.51), such that categorization in less favorable weight trajectory groups is associated with being categorized in a less favorable length trajectory group.
3.4 |. Weight-for-Length
Four distinct weight-for-length trajectories were also identified (Figure 3), all of which are within two standard deviations of the general population mean at age 36 months. The first three trajectories (groups 1, 2, and 3) demonstrate a change of less than one standard deviation between birth and 36 months of age with a slightly upward trend appreciated between 24 and 36 months and z-scores of 1, 0, and −1 respectively by 36 months of age. However, group 4 demonstrated a significant decline with a z-score of 2.5 at birth and a z-score of −1 at 36 months of age. Average birth weight differed between groups with group 4 having the lowest average birth weight (761.3 ± 212.0 g). Group 4 also had the highest proportion of children with severe BPD (72.6%), requiring diuretics (73.7%), and undergoing GT placement (44.2%), indicating greater nutritional failure. See Supporting Information: Table 3 for additional demographic and clinical characteristics of the population by weight-for-length trajectory group.
FIGURE 3.
Trajectories of weight-for-length z-score during the first 36 months of life. Groups arbitrarily labeled 1 through 4 based on weight-for-length z-score at 36 months of age with group 1 having the highest z-score and group 4 having the lowest z-score. Demographic and clinical characteristics of each group are further detailed in the text.
4 |. DISCUSSION
In this study, we described the early somatic growth trajectories in a population of infants and young children with a history of prematurity and chronic lung disease. These results demonstrate that these infants, who begin life with somatic growth measurements well below their term peers, display heterogeneous patterns of weight and length growth trajectories through the first 3 years of life. To our knowledge, this is the first study to describe somatic growth trajectories in children with chronic lung disease of prematurity, therefore, comparison to other studies is limited. Interestingly, our data, in all somatic growth domains, suggests that the traditional preterm paradigm of catch-up growth, achieving an average z-score of 0, occurring in the first 24 months of life, may not accurately describe this population. While some of these findings, the weight trajectories in particular, suggest that there is potential for continued catch-up growth through at least 36 months of age and possibly beyond, it is also possible, especially in terms of linear growth, that those in lower trajectory groups never achieve optimal growth, which highlights the importance of identifying children who demonstrate poor somatic growth early in infancy. This, once again, reinforces the importance of lung-protective strategies, prevention of additional pulmonary insults, and close monitoring of growth throughout early childhood to alleviate as much of the burden of chronic lung disease of prematurity as possible.
In terms of weight, there were four distinct trajectories appreciated in the study population. The majority of subjects (~80%) are in the middle two trajectories (group 2 and group 3) with 14% demonstrating more robust weight gain (group 1) and a small percentage (~6%) making up group 4 exhibiting poorer weight gain. Interestingly, these trajectories all begin at approximately the same z-score (−4) at birth and then diverge, suggesting that there are factors in the NICU that may play a role in establishing an individual’s early life growth trajectory.
Four distinct trajectories of length z-score were also identified; however, unlike the weight trajectories, these trajectories do not all originate at approximately the same z-score and their shapes suggests that there may be potential modifiers of a subject’s linear growth in infancy. The bottom two trajectories, groups 3 and 4, have a mean length z-score of −8 to −10 at six months of age as compared to groups 1 and 2, which have a mean length z-score of −3 to −5. When comparing these two sets of trajectories, groups 3 and 4 have, on average, a lower birth weight, higher percentage of subjects meeting SGA criteria, and overall likely represent a “sicker” population with a greater percentage of subjects with more severe chronic lung disease, GTs, and pulmonary hypertension.
Additionally, when considering the shape of the trajectories, as compared to the weight trajectories, the recumbent length trajectories generally plateau at approximately 18−24 months of age which suggests that the majority of catch-up growth and the greatest potential window for intervention to improve linear growth is within the first 18 months of life. Of note, birth lengths were excluded from the analyses due to increased missing data. This is likely due to subjects’ critical clinical status at birth necessitating immediate resuscitation, including thermoregulation and intubation, which makes obtaining an accurate birth length less of a priority and the measurements that are obtained potentially less accurate.
Weight and length trajectory groups were compared to better understand the pattern of growth in this population. A moderate positive correlation (r = 0.51) was appreciated between an individual’s weight and height trajectories. In general, individuals with the poorest weight gain (group 4) also had the poorest length trajectory (group 4), and individuals with the “highest” weight trajectory (group 1) also had the best length trajectory. It is unknown whether an infant’s weight or length is a better overall measure of somatic growth, and in particular catch-up growth, and it is unclear which somatic growth measures play a more important role in respiratory function. Traditionally, height growth has been associated with increased lung growth; however, length measurements in infants and young children are more susceptible to measurement error, which may make length measurements and overall trajectories less reliable.
Four weight-for-length trajectories were identified, three of which (group 1, group 2, and group 3) were relatively stable over the thirty-six months at a z-score of 1, a z-score of 0, and a z-score of −1, respectively. Group 4, however, demonstrated a “scooping” pattern without an upward trend at 36 months and a z-score less than zero. When stratifying this population in these two groups (groups 1 through 3 vs. group 4), the second group was noted to have a significantly lower average birth weight (p = 0.000), have more severe BPD (p = 0.001), be treated with diuretics (73.7%, p = 0.022), and demonstrate evidence of worsening feeding difficulties with a greater percentage requiring GTs (44.2%, p = 0.01) and Nissen fundoplication (19%, p = 0.036).
5 |. LIMITATIONS
There are several limitations to the present study. Results from a single-center study population may not be generalizable to other centers or other populations of preterm infants. However, similar trajectory data was recently published in young children with cystic fibrosis that illustrated that even annualized growth measurements can be used to create accurate trajectories.26 Infants and children who died before 3 years of age are not represented in the study population; thus, our results only reflect the children who survived to 3 years of age. 50.7% of our study population was classified as having severe BPD based on the 2001 NHLBI consensus statement definition,18 which suggests that our cohort represents a more severe population and our findings may not be generalizable to the greater preterm population. Our study population also only included infants and children that were seen for at least an initial visit in the pediatric pulmonary clinic and then one additional visit, representing a likely selection bias. Additionally, there are several other factors that influence growth, especially in survivors of preterm birth with other significant morbidities, that were not accounted for in this study due to its retrospective design including frequent respiratory infections in early childhood and use of certain medications. Furthermore, while the data suggests that the greatest opportunity for somatic growth occurs in the first 18 months of life, we do not have subjects’ growth datathrough puberty to understand if any further catch-up growth occurs during that period.
6 |. CONCLUSIONS
Appropriate somatic growth is essential for infants and young children with chronic lung disease to complete the delayed alveolarization and lung growth secondary to prematurity. Growth delay is associated with more severe and prolonged respiratory dysfunction, and conversely, improvement in respiratory symptoms is associated with an accelerated growth rate. Nevertheless, the ideal growth pattern to allow for lung development and growth in infants and young children with a history of prematurity and chronic lung disease is not well understood, and there are no current guidelines for optimizing growth in this population. Further understanding of this interplay will ideally allow for the use of somatic growth measures as surrogate markers to help identify and treat individuals at highest risk of poor postnatal growth and respiratory outcomes.
Supplementary Material
ACKNOWLEDGMENTS
The authors wish to thank the families who participated in this study. Funding sources included the National Institutes of Health, Thomas Wilson Foundation, and Children’s Hospital of Philadelphia. This work was supported by the National Center for Advancing Translational Science (NCATS)(BCA: KL2TR003099), National Institutes of Health (Bethesda, MD, USA)(SAM: R01 HL114800), Children’s Hospital of Philadelphia (JMC), and the Thomas Wilson Foundation (JMC). The funding sources had no involvement in the writing of the manuscript or the decision to submit.
Funding information
Children’s Hospital of Philadelphia; National Institutes of Health; Thomas Wilson Foundation; National Center for Advancing Translational Sciences
Abbreviations:
- APPC
average posterior probability of classification
- BPD
bronchopulmonary dysplasia
- CDC
Centers for Disease Control and Prevention
- COPD
chronic obstructive pulmonary disease
- GBTM
group-based trajectory modeling
- GT
gastrostomy tube
- NHLBI
National Heart, Lung, and Blood Institute
- OCC
odds of correct classification
- SGA
small for gestational age
Footnotes
CONFLICT OF INTEREST STATEMENT
The authors have no conflict of interest.
SUPPORTING INFORMATION
Additional supporting information can be found online in the Supporting Information section at the end of this article.
DATA AVAILABILITY STATEMENT
The data that support the findings of this study are available from the corresponding author upon reasonable request.
REFERENCES
- 1.Martin JA, Hamilton BE, Osterman MJK. Births in the United States, 2019. NCHS Data Brief. 2020;387:1–8. [PubMed] [Google Scholar]
- 2.Northway WH Jr., Rosan RC, Porter DY. Pulmonary disease following respirator therapy of hyaline-membrane disease. bronchopulmonary dysplasia. N Engl J Med. 1967;276:357–368. [DOI] [PubMed] [Google Scholar]
- 3.Smith VC, Zupancic JAF, McCormick MC, et al. Trends in severe bronchopulmonary dysplasia rates between 1994 and 2002. J Pediatr. 2005;146:469–473. [DOI] [PubMed] [Google Scholar]
- 4.Kerstjens HA, Rijcken B, Schouten JP, Postma DS. Decline of FEV1 by age and smoking status: facts, figures, and fallacies. Thorax. 1997;52:820–827. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 5.Thurlbeck WM. Postnatal human lung growth. Thorax. 1982;37:564–571. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 6.Kotecha SJ, Edwards MO, Watkins WJ, et al. Effect of preterm birth on later FEV1: a systematic review and meta-analysis. Thorax. 2013;68:760–766. [DOI] [PubMed] [Google Scholar]
- 7.Simpson SJ, Turkovic L, Wilson AC, et al. Lung function trajectories throughout childhood in survivors of very preterm birth: a longitudinal cohort study. Lancet Child Adol Health. 2018;2:350–359. [DOI] [PubMed] [Google Scholar]
- 8.Markestad T, Fitzhardinge PM. Growth and development in children recovering from bronchopulmonary dysplasia. J Pediatr. 1981;98:597–602. [DOI] [PubMed] [Google Scholar]
- 9.Yu VY, Orgill AA, Lim SB, Bajuk B, Astbury J. Growth and development of very low birthweight infants recovering from bronchopulmonary dysplasia. Arch Dis Child. 1983;58:791–794. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 10.Bui DS, Lodge CJ, Burgess JA, et al. Childhood predictors of lung function trajectories and future COPD risk: a prospective cohort study from the first to the sixth decade of life. Lancet Respir Med. 2018;6:535–544. [DOI] [PubMed] [Google Scholar]
- 11.McGrath-Morrow SA, Collaco JM. Bronchopulmonary dysplasia: what are its links to COPD? Therap Adv Respir Dis. 2019;13. doi: 10.1177/1753466619892492 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 12.Moschino L, Bonadies L, Baraldi E. Lung growth and pulmonary function after prematurity and bronchopulmonary dysplasia. Pediatr Pulmonol. 2021;56(11):3499–3508. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 13.Shepherd EG, Clouse BJ, Hasenstab KA, et al. Infant pulmonary function testing and phenotypes in severe bronchopulmonary dysplasia. Pediatrics. 2018;141(5):e20173350. doi: 10.1542/peds.2017-3350 [DOI] [PubMed] [Google Scholar]
- 14.Um-Bergström P, Hallberg J, Thunqvist P, et al. Lung function development after preterm birth in relation to severity of bronchopulmonary dysplasia. BMC Pulm Med. 2017;17(1):97. doi: 10.1186/s12890-017-0441-3 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 15.Bauer SE, Huff KA, Vanderpool CPB, Rose RS, Cristea AI. Growth and nutrition in children with established bronchopulmonary dysplasia: a review of the literature. Nutr Clin Pract. 2022;37(2):282–298. doi: 10.1002/ncp.10841 [DOI] [PubMed] [Google Scholar]
- 16.Lui K, Lloyd J, Ang E, Rynn M, Gupta JM. Early changes in respiratory compliance and resistance during the development of bronchopulmonary dysplasia in the era of surfactant therapy. Pediatr Pulmonol. 2000;30(4):282–290. [DOI] [PubMed] [Google Scholar]
- 17.Balinotti JE, Chakr VC, Tiller C, et al. Growth of lung parenchyma in infants and toddlers with chronic lung disease of infancy. Am J Respir Crit Care Med. 2010;181(10):1093–1097. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 18.Jobe AH, Bancalari E. Bronchopulmonary dysplasia. Am J Respir Crit Care Med. 2001;163:1723–1729. [DOI] [PubMed] [Google Scholar]
- 19.Oken E, Kleinman KP, Rich-Edwards J, Gillman MW. A nearly continuous measure o fbirth weight for gestational age using a United States national reference. BMC Pediatr. 2003;3:6. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 20.Flegal KM, Cole TJ. Construction of LMS parameters for the centers for disease control and prevention 2000 growth charts. National health statistics reports. 2013;63:1–3. [PubMed] [Google Scholar]
- 21.Fenton TR, Kim JH. A systematic review and meta-analysis to revise the fenton growth chart for preterm infants. BMC Pediatr. 2013;13:59. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 22.Nagin DS. Group-Based trajectory modeling: an overview. Ann Nutr Metab. 2014;65(2-3):205–210. doi: 10.1159/000360229 [DOI] [PubMed] [Google Scholar]
- 23.Nagin D, Tremblay RE. Trajectories of boys’ physical aggression, opposition, and hyperactivity on the path to physically violent and nonviolent juvenile delinquency. Child Develop. 1999;70(5):1181–1196. doi: 10.1111/1467-8624.00086 [DOI] [PubMed] [Google Scholar]
- 24.Nagin DS, Odgers CL. Group-based trajectory modeling in clinical research. Annu Rev Clin Psychol. 2010;6(1):109–138. doi: 10.1146/annurev.clinpsy.121208.131413 [DOI] [PubMed] [Google Scholar]
- 25.Jones BL, Nagin DS. A note on a stata plugin for estimating group-based trajectory models. Sociol Methods Res. 2013;42(4):608–613. doi: 10.1177/0049124113503141 [DOI] [Google Scholar]
- 26.Psoter KJ, Dickinson KM, Riekert KA, Collaco JM. Early life growth trajectories in cystic fibrosis are associated with lung function at age six. J Cyst Fibros. 2023;S1569-1993(23):00060–00067. [DOI] [PMC free article] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.
Supplementary Materials
Data Availability Statement
The data that support the findings of this study are available from the corresponding author upon reasonable request.