Definitions and diagnostic criteria of bronchopulmonary dysplasia: Clinical and research implications

Abbreviations

Bronchopulmonary dysplasia (BPD) is a heterogeneous disease that results from disruption of lung growth secondary to premature birth and lung damage with aberrant post-injury repair. Harmful perinatal and postnatal exposures that contribute to the development of BPD include prolonged respiratory support, infection, and other proinflammatory factors, fetal growth restriction, and maternal smoking. BPD progresses from variable degrees of early respiratory insufficiency that evolve into more chronic forms of persistent lung disease. The impaired lung function associated with BPD frequently endures through childhood and into adulthood. Therefore, it is important that the diagnostic criteria for BPD accurately define the severity of pulmonary dysfunction in preterm infants and provide prognostic indicators of the degree of respiratory impairment anticipated after the initial hospital discharge. Standardized BPD diagnostic and severity criteria are also important to compare outcomes in infants included in clinical trials, for epidemiological identification of risk factors, and to compare outcomes between centers or within centers for quality improvement activities.

In this chapter, we provide a brief update of the epidemiology of BPD, describe the most frequent diagnostic criteria of BPD, and discuss their possible advantages and limitations in the context of changing clinical presentation and management of respiratory disease in very preterm infants.

Clinical presentation of bronchopulmonary dysplasia

The term “bronchopulmonary dysplasia” was introduced by Northway and collaborators to describe a clinical, radiographic, and pathological entity that occurred predominately in moderate and late preterm infants who survived severe respiratory distress syndrome (RDS) after aggressive mechanical ventilation without positive end-expiratory pressure (PEEP) and exposure to high concentrations of inspired oxygen used in the 1960s and 1970s. These infants developed severe respiratory failure shortly after birth, and most required aggressive mechanical ventilation and supplemental oxygen for long periods. The severity of the respiratory failure and the radiographic images from these infants were clear evidence of their serious lung derangement that resulted in poor short- and long-term outcomes. With the introduction of prenatal corticosteroid therapy, postnatal surfactant, and the many advances in respiratory support and neonatal care, the presentation of BPD changed over the years. BPD now predominately affects extremely premature infants for whom survival rates continue to increase but chronic respiratory morbidity is common.

The underlying abnormality in the lungs of the premature infant with BPD is a disruption of the normal processes of alveolar and capillary development. The more severe cases are also associated with postnatal airway and vascular remodeling leading to airway obstruction and pulmonary hypertension that can be accompanied by interstitial edema and fibrous tissue proliferation.

Although the classic severe forms of BPD are less common today, some infants still have a clinical course with significant lung damage, a prolonged need for respiratory support resulting in severe respiratory failure, pulmonary hypertension, and marked alterations in the chest radiographs. These infants offer little diagnostic or prognostic dilemma. Unfavorable outcomes are common in these infants, and enduring deficits in respiratory health are expected. Greater difficulty lays with the diagnosis in infants with less severe forms of BPD who have a milder initial respiratory course and require lower levels of respiratory support. The clinical and radiographic evidence in these infants is less conclusive than that in infants with severe BPD, and most have better long-term outcomes. The chest radiographs in less severely affected infants can show diffuse haziness due to loss of gas volume or fluid accumulation, and denser areas of segmental or lobar atelectasis or pneumonic infiltrates are occasionally observed, but there are no areas of severe overinflation typical of the severe forms of BPD. Lethal, severe cases of BPD are characterized by significant morphologic alterations in lung architecture, including emphysema, atelectasis, fibrosis, and marked epithelial squamous metaplasia and smooth muscle hypertrophy in the airways and in the pulmonary vasculature. These alterations are often associated with airway obstruction, pulmonary hypertension, and cor pulmonale, which contribute to prolonged and severe respiratory failure.

Infants developing BPD today are considerably more immature than the earlier cases of severe BPD, and the incidence among infants born after 32 weeks’ gestation has become negligible in modern neonatal centers except for some survivors with hypoplastic lungs who need long-term ventilatory support. Today, BPD predominantly occurs in infants of less than 29 weeks’ gestation, but the reported incidence varies considerably between centers and studies. This variability may reflect different populations and management strategies, but it is also due to major inconsistencies in how the disease is defined. The incidence of BPD in infants of less than 29 weeks’ gestational age (GA) in the centers of the Eunice Kennedy Shriver National Institute of Child Health and Human Development (NICHD) Neonatal Research Network was 42% during the years 2003 to 2007. According to a recent report from stratified samples of nearly one-fifth of US hospitals, the incidence of BPD among surviving infants of birth weight (BW) below 1500 g declined from 50% to 60% in the 1990s to nearly 40% in the years 2000 to 2006. However, between 1993 and 2012, the incidence of BPD in infants born before 28 weeks in the centers of the NICHD Neonatal Research Network slightly increased. In 2018, 10.3% of infants in the Vermont Oxford Network born less than 30 weeks’ gestations died prior to 36 weeks postmenstrual age (PMA) and 40.6% were diagnosed with BPD based on the use of supplemental oxygen or respiratory support at 36 weeks PMA.

The initial presentation of the infants who develop BPD, even among the least mature infants, is more variable than in earlier years. Most infants who developed BPD 40 to 50 years ago had severe RDS, and it is likely that the incidence of BPD would have been even higher absent the relatively high early mortality resulting from severe respiratory failure in those early years of neonatal intensive care. At present, premature infants who develop BPD may present with no or only mild RDS after birth. Based on autopsy and animal models, BPD today is characterized by a reduction in alveolar septation and vascular development, ^{, , ,} suggesting that underlying lung immaturity and abnormal lung development are key contributors to the pathogenesis and the clinical presentation of BPD.

Bronchopulmonary dysplasia diagnosis

Clinicians are typically confident with a diagnosis of BPD when encountering severe forms of the disease. Such infants often have persistent, severe respiratory failure and require supplemental oxygen and respiratory support for prolonged periods of time. These support needs are consistent with the severe chronic lung injury observed on chest imaging and convey a poor long-term prognosis. In contrast, many infants who develop BPD today have mild initial respiratory distress, and although many may still receive prolonged respiratory support, they typically do not require high levels of assistance and the support is not always continuous. This results in inconsistencies when diagnosing BPD based on the levels of respiratory support.

Because diagnostic criteria for BPD are intended to be used by clinicians not only to diagnose individual infants but also for benchmarking clinical outcomes and as endpoints for clinical and epidemiological studies, such criteria will ideally fulfill several requirements:

• 1.

  They should capture both the presence and severity of lung damage. This is particularly important because the lung dysfunction associated with BPD is a continuum from mild to severe disease rather than a categorical presence or absence of disease.

• 2.

  They should provide meaningful information to clinicians, serve as a reliable outcome to researchers, and ideally predict long-term lung health.

• 3.

  They should be based on objective and verifiable data that can be easily obtained from the medical record. This is necessary for standardization purposes for benchmarking clinical outcomes within and between neonatal centers and in epidemiological or interventional studies.

• 4.

  They should not be affected by distinct confounding pathologies that can produce respiratory failure but are independent of BPD or by therapies that may acutely but transiently improve lung function at the time of assessment.

While some of the current diagnostic criteria of BPD fulfill most of these requirements, none fulfill all of them. This is in part because current criteria rely on treatments for the respiratory failure rather than on assessments of lung structure or function. The diagnostic criteria for BPD are primarily based on the need for supplemental oxygen as the marker of respiratory failure with the duration of supplemental oxygen and the fraction of inspired oxygen (FiO ₂ ) needed for adequate arterial oxygen levels used as surrogates of the severity of BPD. There are some important limitations in using oxygen supplementation as the main criterion to diagnose BPD. Although exposure to oxygen therapy is a simple variable to assess, oxygen is also linked to the pathogenesis of BPD and is the main therapy used to maintain adequate oxygenation in respiratory failure from any cause.

The indications for supplemental oxygen can vary from center to center, but after the oxygen saturation targeting trials, there is now more consistency among clinicians on the optimal arterial oxygen saturation range for these infants. However, there are interventions such as steroids, diuretics, and respiratory stimulants that can acutely change oxygen requirement without necessarily changing long-term respiratory outcomes. Different forms of respiratory support can influence gas exchange and oxygenation, thereby reducing the need for oxygen. For this reason, any diagnostic criteria of BPD must also include the use of other forms of respiratory support.

Existing diagnostic criteria for bronchopulmonary dysplasia