January 27, 2021


Brumbaugh, J. E., Bell, E. F., Grey, S. F., DeMauro, S. B., Vohr, B. R., Harmon, H. M., Bann, C. M., Rysavy, M. A., Logan, J. W., Colaizy, T. T., Peralta-Carcelen, M. A., McGowan, E. C., Duncan, A. F., Stoll, B. J., Das, A., Hintz, S. R., & Eunice Kennedy Shriver National Institute of Child Health and Human Development Neonatal Research Network. Behavior profiles at 2 years for children born extremely preterm with Bronchopulmonary Dysplasia. The Journal of Pediatrics219, 152-159.e5 (2020). DOI: 10.1016/j.jpeds.2019.12.028. PMID: 32008764


Dana Chemali, MD
Paediatric Resident
Royal Women’s Hospital

Jeanie Cheong MD, FRACP
Professor, Consultant Neonatologist
Royal Women’s Hospital
Murdoch Children’s Research Institute
Department of Obstetrics & Gynaecology, University of Melbourne




What is the behavioral profile at 2 years corrected age of children born extremely preterm (22-26 weeks) with bronchopulmonary dysplasia (BPD)?


  • Design: Multi-centre, secondary analysis of a prospective cohort study from theEunice Kennedy Shriver National Institute of Child Health and Human Development (NICHD) Neonatal Research Network (NRN) premature infant registry and follow-up database.
  • Allocation: Not applicable
  • Blinding: Not available
  • Follow up period: Infants were followed up at 22-26 months corrected age.
  • Setting: Participating centres from the NICHD NRN, USA.
  • Patients:
    • Inclusion: infants born less than 27 weeks gestational age between July 2012-January 2016.
    • Exclusion: children with major syndromes or congenital anomalies.
  • Intervention/Exposure: BPD categorized into 3 grades of severity as defined by treatment with the following support at 36 weeks’ postmenstrual age, regardless of prior or current oxygen therapy: no BPD, no support; grade 1, nasal cannula ≤ 2L/min; grade 2, nasal cannula >2L/min or noninvasive positive airway pressure; grade 3, invasive mechanical ventilation.
  • Outcomes: Neurodevelopmental assessments were done at 22-26 months of corrected age.
    • Primary outcome: Child Behavior Checklist (CBCL) syndrome scales score by BPD grade.
    • Secondary outcome: CBCL problem scores, CBCL scales oriented to the classifications of Diagnostic and Statistical Manual of Mental Disorders (DSM), Bayley Scales of Infant and Toddler Development, 3rd Edition (Bayley-III).
  • Analysis and sample size:
    • Sample size: Sample size was determined by the available participants. No formal sample size calculation was performed. The study screened 4211 infants in the initial cohort. 2439 children were followed up at 2 years corrected age. The CBCL was done on 2310 children by their parents. Of those children, 1208 children did not have BPD, 1102 had BPD (Grade 1 in 806 children, Grade 2 in 177 children, Grade 3 in 119 children)
    • Analysis: Bivariate comparisons were made by BPD grade for maternal, neonatal characteristics and 2-year behavioral outcome.  Comparisons were made using chi-square-tests and ANOVA. To assess how severity of BPD related to behavioral difficulties, ANOVA tests were conducted followed by Tukey adjustments for multiple variables. To control for confounding factors, linear regression models were used to compare CBCL scores by BPD grade. Mediation analysis was done using multi-categorical independent variable. From this, to determine mediators of the relationship between CBCL syndrome scale and BPD grade, a path model was established. Findings were considered statistically significant when P< .05. CBCL standard scores of <65 were considered within normal limits, scores 65-69 were borderline, and scores ≥ 70 (ie, 2 standard deviations or more above the mean) were considered clinically significant.
  • Patient follow up: Not applicable


The study included 2310 children. 52% had no BPD, 35% had grade 1 BPD, 8% had grade 2 BPD, 5% had grade 3 BPD. The reported characteristics of the mothers were similar between the groups with the following exceptions: mothers of children with BPD were more likely to have private health insurance (p=.02) and be of non-black race (p<.001) than those without BPD.

There were differences in infant characteristics between groups. Children with BPD were more likely to be outborn (p=.026), small for gestational age (p<.001), male (p=.002), had greater rates of surfactant use (p<.001), late onset sepsis (p<.001), patent ductus arteriosus (p<.001), severe intraventricular haemorrhage (p=.008), severe retinopathy of prematurity (p<.001) and postnatal corticosteroid use (p<.001) compared with those with no BPD.

Primary outcome:

At the 2-year corrected age follow up, all children had scores within normal limits on the CBCL scale (> 80% in all scales). Children with grade 2 BPD were more likely to score higher in the “borderline” or “clinically significant” category for withdrawn behavior (p=0.07) and pervasive developmental problems (p=0.07) than those with no BPD.

Secondary outcomes:

When looking at the unadjusted comparison on CBCL syndrome scales and DSM oriented scales, children with BPD scored worse for somatic complaints (mean 54.2 ± SD 6.3 vs 53.6 ± 5.7, p=.025), withdrawn behavior (56.9 ± 8.4 vs 55.8 ± 7.5, p<.001), and pervasive developmental problems (57.5 ± 8.5 vs 56.2 ± 7.9, p< .001) compared with those without BPD. Both behaviors worsened with greater severity of BPD.

Children with BPD had better scores for sleep problems (54.1 ± 7.5 vs 54.8 ± 7.8, p=.025) and aggressive behavior (54.4 ± 7.3 vs 55.1± 8.1, p=.0496) compared with those with no BPD. Both problems decreased with worsening BPD grade.

After adjustment for confounding variables, children with grade 3 BPD scored 2.4 points higher on pervasive developmental problems (p<.001) and 2.2 points higher on withdrawn behavior (p<0.001) than those with no BPD. Children with grade 3 BPD scored 2.4 points lower (better) on externalizing problems (p=.051), 2.1 points lower on sleep problems (p=.08), and 1.8 points lower on aggressive behavior (p=.023) than children with no BPD.

When exploring mediators between BPD grade and problem behaviors, cognitive skills were mediators for all CBCL syndrome scales and DSM-oriented scales except for aggressive behavior, sleep problems, and oppositional defiant problems in all grades of BPD. Language and motor skills were significant mediators for attention problems, emotional reactivity, somatic complaints, withdrawn behavior, affective problems, pervasive developmental problems, internalizing problems, and total problems.  The size of the mediation effect increased with BPD severity.


Children with grade 3 BPD displayed increased risk of withdrawn behavior and pervasive developmental problems but decreased risk of sleep problems and aggressive behavior. Cognitive, language, and motor skills mediated the effects of BPD grade on some problem behaviors.  BPD grades were associated with poorer cognitive, language, and motor skills, which were in turn associated with worse behavioral problems.


While medical advances in neonatology have enabled more infants with respiratory distress syndrome to survive, rates of BPD remain high. BPD is a major cause of morbidity in extremely preterm infants (1). They have worse neurodevelopmental outcomes compared to those with no BPD (1, 2). The more severe the BPD grade, the worse their outcomes (3). The proposed basis behind the neurodevelopmental, including behavioral deficits of children with BPD include chronic hypercarbia, hypoxemia, and postnatal steroid exposure (1). However, socioeconomic status, which in some countries is reflected by private insurance status, may have an impact on parenting styles which in turn, may affect child behavior and/or how behaviour is perceived (4, 5). Parental perceptions are an important consideration given assessments such as the CBCL rely on parental perceptions.

This study found that infants with BPD have behavioral difficulties in some domains. The magnitude of effect of BPD on behaviour, however, was small (R2=0.023-0.067) with differences of less than half a SD of the CBCL scale. This must be taken into account when determining the clinical significance of this finding. Importantly, majority of the children scored within the normal range for the CBCL, which is important reassuring information for parents.

The Bayley-4, launched in 2020, will replace the Bayley-III in clinical and research settings. One of the key improvements is the reliability and validity, thus potentially overcoming the main concerns of underestimation of developmental delay using the Bayley-III. Thus, the absolute scores may be higher if the children had been assessed using the Bayley-4. However, as there was no categorization of developmental delay (using mean scores of 100 and SD of 15), the overall conclusions of this study are likely to be unchanged.

As mentioned by the authors, a strength of the study is the use of a new classification of BPD (6), moving away from the frequently used National Institute of Health definition (7). To elaborate on this, the new definition is more applicable to current respiratory support modalities (6). The consensus definition’s (7) validity has diminished as it does not reflect long-term outcomes of current cohorts of preterm infants (8). Jensen et al. reported that the new classification predicted serious respiratory morbidity at 18-26 months in 81% of infants (6).

Brumbaugh et al. highlight the use of the CBCL as a limitation as it is completed by primary caregivers and caregiver mental status was not considered as a potential confounder. Parents with mental health problems report more difficulties in their children compared with those without (9). Another limitation is that they did not include a control group of infants born at term for comparison purposes and to place the findings in context.

This study contributes towards emerging research that the severity of BPD can predict behavioral challenges. Children with BPD should be prioritized for behavioral screening to detect problems as they emerge. Future work needs to determine which interventions are effective in improving their behavioral profile while considering each child’s needs based on their neonatal morbidities.


  1. Cheong JL, Doyle LW, editors. An update on pulmonary and neurodevelopmental outcomes of bronchopulmonary dysplasia. Seminars in Perinatology; 2018: Elsevier.
  2. Anderson P, Doyle LW, Group VICS. Neurobehavioral outcomes of school-age children born extremely low birth weight or very preterm in the 1990s. jama. 2003;289(24):3264-72.
  3. Sriram S, Schreiber MD, Msall ME, Kuban KC, Joseph RM, O’Shea TM, et al. Cognitive development and quality of life associated with BPD in 10-year-olds born preterm. Pediatrics. 2018;141(6).
  4. Hashima PY, Amato PR. Poverty, social support, and parental behavior. Child development. 1994;65(2):394-403.
  5. The Longitudinal Study of Australian Children. Annual statistical report 2010 [Internet]: Australian Institute of Family Studies 2011. Available from:
  6. Jensen EA, Dysart K, Gantz MG, McDonald S, Bamat NA, Keszler M, et al. The diagnosis of bronchopulmonary dysplasia in very preterm infants. An evidence-based approach. American journal of respiratory and critical care medicine. 2019;200(6):751-9.
  7. Jobe AH, Bancalari E. Bronchopulmonary dysplasia. American journal of respiratory and critical care medicine. 2001;163(7):1723-9.
  8. Poindexter BB, Feng R, Schmidt B, Aschner JL, Ballard RA, Hamvas A, et al. Comparisons and limitations of current definitions of bronchopulmonary dysplasia for the prematurity and respiratory outcomes program. Annals of the American Thoracic Society. 2015;12(12):1822-30.
  9. Maoz H, Goldstein T, Goldstein BI, Axelson DA, Fan J, Hickey MB, et al. The effects of parental mood on reports of their children’s psychopathology. Journal of the American Academy of Child & Adolescent Psychiatry. 2014;53(10):1111-22. e5.