Bronchopulmonary Dysplasia as a Determinant of Respiratory Outcomes in Adult Life

Abstract

Respiratory disease is unfortunately common in preterm infants with the archetype being bronchopulmonary dysplasia (BPD). BPD affects approximately 50,000 preterm infants in the U.S. annually with substantial morbidity and mortality related to its pathology (alveolar, airway, and pulmonary vasculature maldevelopment). Predicting the likelihood and severity of chronic respiratory disease in these children as they age is difficult and compounded by the lack of consistent phenotyping. Barriers to understanding the actual scope of this problem include few longitudinal studies, information limited by small retrospective studies and the ever-changing landscape of therapies in the NICU that affect long-term respiratory outcomes. Thus, the true burden of adult respiratory disease caused by premature birth is currently unknown. Nevertheless, limited data suggest that a substantial percentage of children with a history of BPD have long-term respiratory symptoms and persistent airflow obstruction associated with altered lung function trajectories into adult life. Small airway disease with variable bronchodilator responsiveness, is the most common manifestation of lung dysfunction in adults with a history of BPD. The etiology of this is unclear however, developmental dysanapsis may underlie the airflow obstruction in some adults with a history of BPD. This type of flow limitation resembles that of aging adults with chronic obstructive lung disease (COPD) with no history of smoking. It is also unclear whether lung function abnormalities in people with a history of BPD are static or if these individuals with BPD have a more accelerated decline in lung function as they age compared to controls. While some of the more significant mediators of lung function such as tobacco smoke and respiratory infections have been identified, more work is necessary to identify the best means of preserving lung function for individuals born prematurely throughout their lifespan.

Background

One of the most common and potentially profound injuries to the respiratory system in early life is premature birth. The archetype of post prematurity respiratory disease (PPRD) is bronchopulmonary dysplasia (BPD), a phenotype defined by oxygen use and respiratory support at 36 weeks post menstrual age (PMA).¹ The initial description of BPD (“old” BPD) was published over 50 years ago, and included features of emphysematous alveoli, airway muscle hypertrophy, pulmonary arteriole lesions, and right ventricular hypertrophy on autopsy.² Exogenous surfactant and improvements in respiratory support during the late 1980’s led to a modified phenotype (“new” BPD), which featured alveolar hypoplasia and vascular simplification.^3,4 BPD is estimated to affect 50,000 infants annually (~1/80 live births in the United States).⁵ Infants born extremely premature (< 28 weeks gestation) continue to be at highest risk for developing BPD with gestational age being a major driver for development of BPD.^6,7

While increased severity of BPD is associated with later lower function, this phenotype alone may not capture the effects of premature birth on lung function in later life. Indeed, low birthweight by itself may adversely affect adult lung function. One study examined lung function in men born between 1911 and 1930, and found that low birth weight was associated with lower adult lung function and higher mortality from COPD.⁸ Nevertheless children with a history of BPD are more likely to have lower lung function in adult life than preterm infants without BPD.⁹ Recently, the term post-prematurity respiratory disease (PPRD) has been used in reference to preterm infants and children with a range of chronic or intermittent respiratory symptoms following initial hospital discharge, and includes BPD. Limited data suggests that a substantial percentage of children with PPRD will have long-term respiratory symptoms and persistent airflow obstruction associated with altered lung function trajectories into adult life.¹⁰ However, predicting the likelihood and severity of chronic respiratory disease in children with PPRD throughout childhood and adult life has been difficult, and is compounded by the paucity of longitudinal phenotyping as the child ages. Additionally, a history of upper airway lesions, tracheobronchomalacia, small airway disease, and alveolar or parenchymal disease in early life may influence adult lung function in children with PPRD. Furthermore, the effects of environmental exposures, both prenatal and postnatal life¹¹ and genetic and epigenetic influences¹² likely contribute to lung function trajectories in children with PPRD. For instance, cord blood from several childhood cohorts demonstrated associations between differentially methylated regions (DMRs) in childhood and later pulmonary outcomes (childhood asthma and COPD).¹² These studies suggest that epigenetic changes during early life can influence respiratory function in adult life.

Several longitudinal studies, have revealed that lung function peaks in early adulthood and declines with aging, and within that model there are many trajectories of lung function.^13–15 These trajectories may have different peaks as well as different patterns of decline over time. Recent studies have suggested that early life events can profoundly affect lung function in adult life by altering the achieved maximal FEV₁ peak. For example, the Tucson Children’s Respiratory Study revealed that children with radiologically ascertained pneumonia within the first three years of life had lower lung function through 26 years of age compared to their peers without such a history.¹⁶ Other studies have indicated the presence of distinct populations of individuals with low lung function in preadolescence who continue to have lower adult trajectories.¹⁷ For example, children with persistent wheezing during the first few years of life are more likely to have lower lung function in adult life.¹⁸ However, other factors in addition to airway caliber likely contribute. It has been noted that children with mild asthma symptoms may or may not have lower lung function trajectories as adults.¹⁹ Conversely, although smoking is a major risk factor for adult-onset COPD; others who do not smoke may develop COPD due to airway dysanapsis.^14,20

Furthermore, the view that PPRD resolves within early childhood or that respiratory phenotypes due to BPD are static during adult life is not necessarily accurate. Tepper et al., reported that although clinical symptoms improved during the first year of life in infants with BPD; these same children had significant expiratory airflow limitation by pulmonary function testing.²¹ Several studies have also shown that manifestations of PPRD can persist into early adulthood, if not longer, while others with a history of PPRD may be at increased risk for more rapid decline in lung function as they age.^22,23 Decline in FEV1 was noted in a cohort of formerly extreme low birth weight infants in Japan between the ages of 8 and 12 years of age, ²² indicating that factors outside of infancy can influence lung function during childhood in these children, including an absence of catch-up lung growth. Others have theorized that infants with a history of BPD are more susceptible to COPD as adults.^24,25 Previously, we have described the epidemiology of lung disease associated with prematurity.⁵ In this review we will focus on the burden of lung disease and potential effects on lung function trajectories in adults caused by premature birth (being born at less than 37 weeks gestation). The major barriers to understanding the true scope of this problem include information limited by small retrospective studies²³ and the ever changing landscape of therapies in the NICU that effect long-term respiratory outcomes. To this end however we will outline the common manifestations of PPRD (including BPD), their persistence into adult life, possible modifiers that can worsen or ameliorate lung function outcomes, and lastly, gaps in our understanding of the effects of PPRD (and BPD) on long term respiratory outcomes.

Alveolar disease

Alveolar disease is common in infants with PPRD,⁵ which may present as hypercarbia or hypoxemia. Compared to full-term controls, young children with a history of BPD at a mean corrected age of 17.4 months demonstrate reduced DLCO measurements suggesting impaired or disrupted alveolar development.²⁶ While catch-up growth may aid in alveolarization, the majority of which occurs within the first several years of life,²⁷ a number of exercise studies in former preterm children and young adults suggest limitations in pulmonary reserve.^28–31 Additionally, other reasons for decreased exercise tolerance in these children could include deconditioning from reduced activity or expiratory flow limitation. ³² Other manifestations of alveolar disease have also been detected, including evidence of restrictive lung disease on plethysmography. Restrictive lung disease seems to be more limited to children with a history of severe BPD. Interestingly, one study in school age children found that alveolar size, as measured by helium-3 magnetic resonance, was similar between prematurity and term-born children suggesting that alveolar catch-up growth was possible in children with prematurity.³³ Nevertheless the children with BPD in this study had significantly lower FEV1 values compared to preterm children without BPD and controls, indicating that although alveolar growth occurs in children with BPD, significant residual lung disease often persists. Severe alveolar disease can also include a cystic component on imaging, which can be associated with impaired lung function.³⁴ However the association between cysts and emphysematous changes on imaging and risk of COPD is not known in children with a history of BPD.⁵

Upper Airway Abnormalities

Patients with prematurity and BPD are at higher risk for having sleep disordered breathing which may improve with age.³⁵ This may be due to a variety of factors, including upper airway lesions and/or alterations in airway tone; both of which can adversely affect gas exchange and pulmonary function.³⁶ Additionally, abnormalities in upper airway tone may act synergistically with other lower airway and parenchymal features of severe BPD; further increasing gas exchange abnormalities and work of breathing, particularly during sleep and with acute respiratory illnesses. Infants with a history of prematurity may also acquire subglottic narrowing or stenosis as a function of traumatic, repeated, or prolonged intubation.³⁷ Vocal cord lesions and paralysis can occur with laryngeal nerve injuries from birth trauma or cardiac surgery.³⁸ These lesions can present as fixed obstruction visible on flow-volume loops during pulmonary function testing.

Tracheobronchomalacia

Large airway malacia (tracheomalacia and/or bronchomalacia) is relatively uncommon among infants with BPD (~5%).³⁹ Nevertheless, it is a lung phenotype that has been associated with more severe BPD in extremely low-birth infants; ⁴⁰ and young adults with a history of severe BPD are more likely to have abnormal lung function. ²³ Although, symptomatically, airway malacia often resolves in early childhood, it is unclear whether it persists into later life, since long-term bronchoscopic studies are usually not performed. Tracheomalacia has been associated with COPD and worse quality of life for patients with COPD.^41,42 Large airway malacia can appear as an obstructive lesion on flow-volume loops⁴³ and along with parenchymal and airway dysfunction may contribute to lung function abnormalities in people with a history of BPD. Although rare, tracheal stenosis is a known complication of tracheostomies, which patients with severe PPRD requiring home ventilation may require.⁴⁴ Laryngotracheal stenosis is associated with alterations in spirometry in adults.⁴⁵

Small Airway Disease

Small airway disease is a common manifestation of PPRD leading to an obstructive lung disease presentation on spirometry or evidence of air trapping on plethysmography. This obstructive component may be due to airway inflammation and smooth muscle hypertrophy, responsive to standard asthma management, but also may have a fixed or structural component (i.e., dysanapsis). Higher exhaled nitric oxide levels are commonly found in people with an eosinophilic asthma phenotype.⁴⁶ However, a meta-analysis of studies that measured exhaled nitric oxide (eNO) levels in preterm children with and without chronic lung disease found similar eNO levels as found in term controls.⁴⁷ These findings indicate that the etiology of airflow obstruction in preterm individuals with PPRD is likely different from individuals with eosinophilic asthma phenotypes. Neutrophilic inflammation drives small airway disease in many people with chronic obstructive pulmonary disease (COPD).⁴⁸ Longitudinal studies are needed to better understand the relationship between neutrophilic airway inflammation and small airway dysfunction in children and adults with PPRD.

Developmental dysanapsis resulting from preterm birth may underlie the persistent airflow abnormalities found in people with PPRD and a history of BPD. Dysanapsis can be defined as a mismatch between airway caliber and lung size, which may have implications for airway resistance and dead space ventilation.⁴⁹ Measures of dysanapsis can be derived from standard spirometric measurements or CT imaging, although is not commonly done clinically.^50–52 Clinically, dysanapsis could lead to small airway disease symptoms that are poorly responsive to standard asthma or COPD therapies owing to the airways being smaller, rather than bronchoconstricted.^5,51 Developmental dysanapsis likely accounts for some of the abnormal spirometric values and FEV1 measurements reported in early and later life in children born extremely premature.^21,53 Dysanapsis may have relevance for adult lung function outcomes in aging adults with PPRD as developmental dysanapsis is a risk factor for adult-onset COPD.²⁰

In addition, it is unclear to what extent pre-existing airway inflammation associated with BPD versus inflammation triggered by environmental factors, impacts small airway function.⁵⁴ It is also uncertain whether modulating therapies for inflammation (e.g., corticosteroids) alter long-term outcomes,⁵⁵ a meta-analysis suggested that the use of inhaled corticosteroids within the NICU may decrease the incidence of BPD.⁵⁶ A 2013 meta-analysis that employed several decades of cross-sectional studies demonstrated that the severity of small airway disease (as measured by FEV₁) is lessening in more recent birth cohorts, possibly from improved strategies in the NICU that lessen airway injury.¹⁰ However, in other meta-analyses, it is variable whether obstructive lung disease persists or improves over the course of the lifespan for individuals born prematurely.^57,58 Mechanistic phenotyping of small airway dysfunction and inflammation in children and adults with PPRD could help direct therapeutic decision making towards more effective treatments.

Exogenous Factors Modifying Lung Function

Preterm children with a history of BPD are at increased risk for hospitalization secondary to respiratory viral illnesses. Most notably this is true with respiratory syncytial virus (RSV) and rhinovirus.^59,60 Children hospitalized for RSV in early life have also been reported to have a higher risk for developing recurrent wheezing.⁶¹ In a retrospective study, it was shown that preterm infants hospitalized with bronchiolitis were more likely to have asthma at 6–9 years of age.⁶² Aeroallergen sensitization was another risk factor for asthma in this study. Interestingly, however a meta-analysis of observational studies did not support a causal relationship between lower respiratory tract viral illnesses and wheezing and airflow obstruction in later childhood.⁶³

Additionally, little is known if preterm birth and the cumulative effects of recurrent lung insults can alter lung function in later life through changes in lung immune responses and injury repair. Chronic damage to airway epithelium can impair mucociliary clearance, disrupt epithelial cell junctions, create areas of local airway inflammation, and cause chronic pathogen colonization that may contribute to long-term changes in lung function.⁶⁴ Disruption of airway epithelium could lead to abnormal responses to airway pathogens and aeroallergens. Although airflow obstruction from dysanaptic airway growth caused by prematurity likely contributes to the severity of respiratory symptoms caused by viral illnesses; neutrophilic airway inflammation and release of pro-inflammatory cytokines may also contribute to disease severity in children with PPRD. Premature birth by itself may potentially cause reprogramming of immune responses, however little direct evidence currently supports this. Gene and environment interactions have been shown to influence susceptibility to asthma during childhood and adult life.⁶⁵ Living on a farm and being exposed to a diverse array of microbes early in life has been associated with a lower risk of asthma.⁶⁶ More studies are needed to examine the effects of early life microbe exposures on airway responsiveness in children and adults with PPRD.

Preclinical studies suggest that early life exposure may alter immune responses to respiratory viruses in later life. One study in mice reported that exposure to neonatal hyperoxia, altered the lungs immune response to influenza towards a more fibrotic response.⁶⁷ Other preclinical studies found age-related changes in lung immune responses to pneumonia and injury-induced aberrant changes in lung cell phenotypes and cell migration.^68,69 More mechanistic studies are needed to understand the impact of premature birth and BPD on subsequent immune responses and lung defense mechanisms in later life. Additionally, further understanding of gene and environmental interactions may help to explain the variabilities in FEV₁ which can range from normal to severe obstructive lung disease, among preterm children of similar gestational ages.

Improving Adult Outcomes

While prematurity and its associated respiratory disease are risk factors for worse adult lung function, there are steps that can be taken to mitigate its effects in childhood as well as adulthood. As would be expected, many of these interventions revolve around avoiding a “second hit” to the respiratory system, if prematurity is considered the “first hit.”

As mentioned previously, respiratory infections can have profound effects on adult lung function. Indeed, respiratory viral infections in infancy can result in intermittent wheezing for years (e.g., respiratory syncytial virus).⁷⁰ Accordingly, it is important to adhere to recommended immunizations, including the influenza vaccine as well as palivizumab if eligible.⁷¹ The effects of SARS-CoV-2 (COVID-19) remain unknown at this time on long-term PPRD outcomes. Other environmental factors may also increase the risk for contracting these infections, such as daycare attendance.⁷²

For both children and adults, inhalation of particulate matter and other toxicants can have profound changes in lung function. The best characterized of these is tobacco smoke where active (or even former) smoking is associated with a more rapid decline in lung function in the general population compared to never smokers.⁷³ We noted that secondhand smoke in infants with more severe BPD was associated with increased hospital admissions and activity limitations compared to non-exposed infants.⁷⁴ The data are more mixed as to whether adults with exposure to secondhand smoke experience a more rapid decline than those not exposed.^75,76 If this is the case, decline could be accentuated in preterm individuals already at propensity for lower lung function. A similar emerging exposure is electronic cigarettes and other vaping devices, again with both active use and secondhand exposures. Owing to the rapidly changing nature of these devices and their relative novelty there are limited data on their effects on the developing lung or on long term effects on pulmonary function testing. Murine models suggest that exposure to developing lung could result in impaired alveolar growth.⁷⁷ Studies of the lung function of adult users of electronic cigarettes or other vaping devices as well as those exposed secondhand are lacking.

Similar to smoke exposure in terms of effects on lung function are indoor and outdoor pollutants. Outdoor pollutants may include industrial sites, but more commonly traffic. A study of estimated traffic exposure in preterm infants demonstrated an association between proximity to a major roadway and chronic respiratory symptoms.⁷⁸ In adults increasing levels of traffic pollution are associated with a more rapid decline in FEV₁.⁷⁹ Indoor pollution encompasses secondhand smoke or vaping exposure, but also includes any source of particulate matter such as combustion sources for indoor cooking. Data for the effects of indoor air pollution on preterm infants are more limited, but suggest that chronic respiratory symptoms can be associated with indoor air pollution, although the effects may be attenuated by air purifiers.⁸⁰ Certainly adult lung function is detrimentally affected by indoor air pollution, specifically biomass smoke.⁸¹

Other potential modifiers for preterm infants may ameliorate poor adult lung function outcomes, including somatic growth and medication management. Adequate nutrition for preterm infants is critical within the first several years of life, and maintaining this may aid in adult outcomes as the majority of alveolar growth occurs within the first 2 years of life.²⁷ Additionally, it is likely that adult lung function in former preterm infants is related to height velocity;⁸² preterm infants within the first several years of life may be subject to increases in weight but lagging height velocity, potentially due to the use of corticosteroids and aggressive supplemental nutrition.⁸³

We have previously mentioned that a meta-analysis suggested the use of inhaled corticosteroids within the NICU may decrease the incidence of BPD.⁵⁶ COPD studies have suggested that the use of salmeterol/fluticasone and/or tiotropium may provide modest slowing of the rate of pulmonary function decline in COPD.⁸⁴ It is unknown whether the empiric use of common asthma or COPD therapies would slow any accelerated lung function decline associated with PPRD, but these studies remain intriguing.

Role of Adult Healthcare Providers

A common misconception of respiratory disease associated with prematurity is that it resolves in infancy or early childhood. As outlined above, manifestations of PPRD may persist into adulthood, some of them unrecognized by patient or provider. Screening for such disease may start with obtaining a history of prematurity and respiratory symptoms. Periodic pulmonary function testing for those with a history of prematurity may be helpful in diagnosing subtle changes in lung function. For those with accelerated declines in lung function, imaging could be considered to assess for cystic changes or potential bronchiectasis. Providers should encourage avoidance of respiratory infections through influenza and pneumococcal immunizations as appropriate, and when respiratory infections do occur, prompt treatment may limit further declines in lung function. Providers can also be instrumental in prescribing appropriate respiratory maintenance medications and ensuring that inhaled medications are optimally used (e.g., use of aerochambers, etc.).

Gaps in Knowledge

Although neonatology care for extremely preterm infants has been gradually advancing over time, the incidence of BPD (and likely PPRD) has not decreased over the past 2 decades,⁷ suggesting that this is an influence on adult function that providers will continue to face. It also should be noted that the presentations of PPRD will not be static over time as therapies and management improve, and it is hoped that improvements in early life care continue to mitigate adverse adult outcomes. As such, ongoing studies are needed to determine the effects of therapeutic decisions (e.g., oxygen use, saturation levels, inhaled therapies) on long term outcomes. For example, a shift in histopathology of BPD (and subsequent improvement in outcomes) was observed in the 1990s with the use of pulmonary surfactant preparations.^3,4 However, even with improvements in care, it remains unclear whether prematurity alone is a risk factor for reduced adult lung function, independent of any PPRD manifestations.

Both neonatology and pediatric pulmonary medicine tend to define preterm respiratory disease as BPD, typically referring to the definition from a NIH workshop in 2001.¹ In this schema, BPD is defined as respiratory support (FiO₂ and/or positive pressure ventilation). While it is problematic to define disease by therapy,⁸⁵ a more pertinent concern for adult patients is relying on a single marker of respiratory disease that is measured once at 36 weeks corrected gestational age before the preterm infant has reached full term age (≥37 weeks gestation). Standardized phenotypes that can extend through the lifespan are sorely needed. For example, the use of spirometry or imaging measures to track dysanapsis could be helpful in identifying preterm young adults at risk for COPD.

While data from NICU settings informs our knowledge concerning the development of PPRD in infancy, in particular BPD, our understanding of the complex interplay of preterm lung injury, lung development in childhood/adolescence, environmental factors, and immune responses is limited and somewhat speculative (Table 1). Future studies are needed to identify novel environmental factors that alter lung function for those born prematurely, such as socioeconomic status and psychosocial stressors, and to determine whether the preterm respiratory and immune systems are potentially more sensitive to alteration from already identified and novel factors. Other studies focusing on the links between specific respiratory viruses and subsequent immune responses, and equally important, how to modulate these responses. Studies such as the Prematurity and Respiratory Outcomes Program, which aims to develop objective biomarkers and outcome measures of respiratory morbidity in the <29 week gestation population beyond just the NICU hospitalization is an important first step.⁸⁶

TABLE 1.

Potential determinants of adult lung function and their timing in people with PPRD

	Fetal Life	Infancy	Childhood	Adolescence
Preterm Birth		√	√	√
Genetic Variants (Susceptibility genes, etc.)	√	√	√	√
Secondhand Smoke Exposure	√	√	√	√
Primary Smoking/Vaping				√
Air Pollution		√	√	√
Viral Respiratory Infections		√	√
Pneumonia		√	√	√
Nutrition	√	√	√
Epigenetics	√	√

While we have posited modifiers of PPRD as “second hits” or mitigating factors for adult lung function outcomes, an alternative model is that premature birth is the “second hit” on a genetic background that may predispose to adult lung disease. Twin studies have demonstrated heritability for the phenotype of BPD, but similar studies of later respiratory outcomes in preterm individuals have not been performed.^87,88 Genetic variants that may lead to an increased risk of adult disease potentially in heterozygote carriers could include variants in the causative genes for alpha-1 antitrypsin, cystic fibrosis, primary ciliary dyskinesia, and/or interstitial lung disease. Little is known about the combination of PPRD and carrier states for these disorders, and the findings from three genome-wide association studies of BPD have not identified any common genetic variants between them.^89–91 Additionally, while there are limited data to suggest that different epigenetic profiles exist between preterm infants with and without BPD,^92,93 there are no studies on the long term effects of any epigenetic changes that may be associated with prematurity.

Conclusions

Advances in neonatal care have led to increased survival for preterm infants, but may not have necessarily resulted in decreases in respiratory disease in infancy. Ongoing studies continue to confirm that manifestations of this respiratory disease may persist through the lifespan and can be detected on various measures of lung function. While some of the more significant mediators of lung function such as tobacco smoke and respiratory infections have been identified, more work is necessary to identify the best means of preserving lung function for individuals born prematurely.