I. DEFINITION. A 2001 National Institutes of Health (NIH) consensus conference proposed definitions for bronchopulmonary dysplasia (BPD) (also known by the more general term chronic lung disease [CLD] of prematurity)

A. For infants born at <32 weeks’ gestation who received supplemental oxygen for their first 28 days, the NIH defined BPD at 36 weeks’ postmenstrual age (PMA) as

1. Mild: no supplemental O₂ requirement

2. Moderate: supplemental O₂ requirement <30%

3. Severe: supplemental O₂ requirement ≥30% and/or continuous positive airway pressure (CPAP) or ventilator support

B. For infants born at ≥32 weeks, the NIH defined BPD as supplemental O₂ requirement for the first 28 days with severity level based on O₂ requirement at 56 days.

C. Physiologic definition of BPD. The need for supplemental oxygen is based on oxygen saturation (SpO₂) during a room air challenge performed at 36 weeks’ PMA (or 56 days for infants >32 weeks’ PMA) or before hospital discharge. Persistent SpO₂ <90% is the cutoff below which supplemental O₂ should be considered.

D. Operationally, many clinicians simply define BPD as requirement for oxygen supplementation at 36 weeks’ PMA. Lung parenchyma usually appears abnormal on chest radiographs. This definition can also apply to term infants who require chronic ventilatory support following meconium aspiration syndrome, pneumonia, and certain cardiac and gastrointestinal (GI) anomalies. BPD is associated with the development of chronic respiratory morbidity (CRM).

II. EPIDEMIOLOGY. Approximately 10,000 to 15,000 new cases of BPD occur in the United States each year. The incidence of BPD increases with decreasing gestational age at birth. Infants <28 weeks’ gestation or <1,000 g birth weight are most susceptible, with incidence rates of 35% to 50%. Differences in populations (race/ethnicity/socioeconomic status), clinical practices, and definitions account for wide variation in the rate reported among centers. The relative risk is decreased in African Americans and females. Of infants with BPD, 44% develop CRM (defined as a requirement for pulmonary medications at 18 months corrected age). Of similar preterm infants who do not require O₂ at 36 weeks’ PMA, 29% also develop CRM.

A. Etiology. A number of factors have been associated with BPD, some of which may be causal.

1. Immature lung substrate. The lung is most susceptible before alveolar septation begins. Injury at this stage may lead to an arrest of alveolarization and simplified lung structures that are the hallmark of new BPD.

2. Volutrauma and lung injury from mechanical ventilation or bag-andmask ventilation

3. Oxygen toxicity. Insufficient production of the antioxidant enzymes superoxide dismutase, catalase, glutathione peroxidase, and/or deficiency of free radical sinks such as vitamin E, glutathione, and ceruloplasmin may predispose the lung to O₂ toxicity. Similarly, inadequate antiprotease protection may predispose the lung to injury from the unchecked proteases released by recruited inflammatory cells.

4. Genetic factors may contribute to BPD risk, but the mechanism is uncertain.

5. Excessive early intravenous fluid administration, perhaps by contributing to pulmonary edema

6. Persistent left-to-right shunt through the patent ductus arteriosus (PDA). Although prophylactic PDA ligation or administration of indomethacin or ibuprofen does not prevent BPD, persistent leftto-right shunt and late PDA closure appear to be associated with increased BPD risk. However, surgical PDA closure is also associated with increased BPD risk.

7. Intrauterine or perinatal infection, with cytokine release, may contribute to the etiology of BPD or modify its course. Ureaplasma urealyticum
has been associated with BPD in premature infants, although whether this relationship is causal is uncertain. Intrauterine Chlamydia trachomatis and other viral infections have also been implicated.

8. Intrauterine growth restriction has been linked to later development of BPD, although whether this is a causal mechanism for disease or just an association is uncertain.

9. Increased inositol clearance may lead to diminished plasma inositol levels and decreased surfactant synthesis or impaired surfactant metabolism.

10. An increase in vasopressin and a decrease in atrial natriuretic peptide release may alter pulmonary and systemic fluid balance in the setting of obstructive lung disease.

1. Acute lung injury is caused by the combination of O₂ toxicity, barotrauma, and volutrauma from mechanical ventilation. Cellular and interstitial injury results in the release of proinflammatory cytokines (interleukin 1β [IL-1β], IL-6, IL-8, tumor necrosis factor alpha [TNF-α]) that cause secondary changes in alveolar permeability and recruit inflammatory cells into interstitial and alveolar spaces; further injury from proteases, oxidants, and additional chemokines, and chemoattractants cause ongoing inflammatory cell recruitment and leakage of water and protein. Airway and vascular tone may be altered. Sloughed cells and accumulated secretions not cleared adequately by the damaged mucociliary transport system cause inhomogeneous peripheral airway obstruction that leads to alternating areas of collapse and hyperinflation and proximal airway dilation. Inflammation may also alter critical molecular pathways required for lung development leading to impaired alveolarization and emphysematous changes in the lung. In the original report by Northway in 1967 of the “old” BPD affecting infants with mean gestational age of 33 weeks and birth weight of 2,000 g, pathology of nonsurvivors showed predominantly small airway injury, fibrosis, and emphysema. In contrast, in the postsurfactant therapy era, “new” BPD affects mostly extremely preterm infants and the most significant pathologic finding in nonsurvivors is decreased alveolarization.

2. In the chronic phase of lung injury, the interstitium may be altered by fibrosis and cellular hyperplasia that results from excessive release of growth factors and cytokines, leading to dysregulated repair. Interstitial fluid clearance is disrupted, resulting in pulmonary fluid retention. Airways develop increased muscularization and hyperreactivity. The physiologic effects are decreased lung compliance, increased airway resistance, and impaired gas exchange with resulting ventilation-perfusion mismatching and air trapping.

3. Histopathology. Arrest of alveolarization and dilated simplified terminal airspaces are characteristic histologic features of “new” BPD seen at lower gestational ages. The resultant emphysematous changes and impairment in alveolar development leads to diminished surface area for gas exchange. In severe cases, pathology may reflect that seen in “old” BPD with detectable changes observed within the first few days
after birth. In these cases, necrotizing bronchiolitis, obstruction of small airway lumens by debris and edema, and areas of peribronchial and interstitial fibrosis are present. Changes in both large airways (glandular hyperplasia) and small airways (smooth muscle hyperplasia) likely form the histologic basis for reactive airway disease. Pulmonary vascular changes associated with pulmonary hypertension (PH) may be seen.

A. Physical examination typically reveals tachypnea, retractions, and rales on auscultation.

B. Arterial blood gas (ABG) analysis shows hypoxemia and hypercarbia with eventual metabolic compensation for the respiratory acidosis.

C. The chest radiograph appearance changes as the disease progresses. With “new” BPD, the initial appearance is often diffuse haziness, increased density, and normal-to-low lung volumes. In more severe disease, chronic changes may include inhomogeneous regions of opacification and hyperlucency with superimposed hyperinflation.

D. Cardiac evaluation. Nonpulmonary causes of respiratory failure should be excluded. Electrocardiogram (ECG) can show persistent or progressive right ventricular hypertrophy if cor pulmonale develops. Left ventricular hypertrophy may develop with systemic hypertension. Two-dimensional echocardiography may be useful in excluding left-to-right shunts (see Chapter 41) and PH. Biventricular failure is unusual when good oxygenation is maintained, and the development of PH is avoided.

E. Infant pulmonary function testing (iPFT). Increased respiratory system resistance (Rrs) and decreased dynamic compliance (Crs) are hallmarks of BPD. In the first year after birth, iPFTs reveal decreased forced expiratory flow rate, increased functional residual capacity (FRC), increased residual volume (RV), and increased RV/total lung capacity ratio and bronchodilator responsiveness, with an overall pattern of mild-to-moderate airflow obstruction, air trapping, and increased airway reactivity. Although such testing is feasible, it is not typically used in clinical practice.

V. INPATIENT TREATMENT. The goals of treatment during the neonatal intensive care unit (NICU) course are to prevent or minimize further lung injury (barotrauma and volutrauma, O₂ toxicity, inflammation), maximize nutrition, and diminish O₂ consumption.

1. Vitamin A (5,000 IU intramuscular [IM], three times weekly for the first 28 days of age) reduced the incidence of chronic lung disease (CLD) in extremely low birth weight (ELBW) infants by 10%. Although some centers routinely treat ELBW infants with vitamin A using this protocol, the impact on long-term outcomes is uncertain.

2. Caffeine citrate (20 mg/kg loading dose and 5 mg/kg daily maintenance) started during the first 10 days after birth in infants 500 to 1,250 g birth weight reduced the rate of BPD from 47% to 36% and
improved the rate of survival without neurodevelopmental disability at 18 to 21 months corrected age.

a. Inhaled nitric oxide (iNO). In animal models of BPD, iNO may act to relax airway and pulmonary vascular tone and diminish lung inflammation. Several multicenter clinical trials assessed the potential efficacy of iNO in attenuating or preventing BPD using different treatment regimens. One trial found that BPD was reduced in infants >1,000 g although not for the overall group; another found overall benefit that was limited to those treated at 7 to 14 days. Because benefit is unclear and both safety and long-term impact have not been established, an NIH consensus panel recommended that use of iNO to prevent or treat BPD is not supported by available evidence.

b. In <27-week gestation infants, intratracheal recombinant human Cu/Zn superoxide dismutase administered intratracheally every 48 hours while intubated resulted in an approximately 50% reduction in use of asthma medications, emergency room visits, and hospitalizations in the first year of life. Recombinant human club cell protein 10, a natural innate anti-inflammatory protein abundant in the lung, is undergoing evaluation for intratracheal administration for prophylaxis against CRM.

c. Whether azithromycin may decrease the risk of developing BPD in infants with documented Ureaplasma colonization or infection is under investigation.

1. Acute phase. Volume targeted compared to pressure limited ventilation appears to reduce the incidence of the combined outcome of BPD or death and of BPD, as well as air leak. We initially target 3 to 5 mL/kg/breath while providing adequate gas exchange (see Chapter 29). It is possible that use of patient-controlled ventilator modalities such as patient-triggered breaths and pressure-supported spontaneous breaths may lower BPD risk. Early use of nasal CPAP may avoid the need for intubation and surfactant therapy, although this has not clearly reduced the risk of BPD and some studies suggest increased rates of pneumothorax. Nasal intermittent positive pressure ventilation (NIPPV) does not appear to be any more effective than standard nasal CPAP in avoiding need for intubation and surfactant therapy.

In most circumstances, we avoid hyperventilation and target arterial carbon dioxide tension (PaCO₂) at ≥55 mm Hg, with pH ≥7.25, and target SpO₂ at 90% to 95% and arterial oxygen tension (PaO₂) 55 to 80 mm Hg. Although routine use of high-frequency oscillatory ventilation (HFOV) does not prevent BPD, follow-up at 11 to 14 years of age of infants enrolled in a large trial found better lung function in those treated with HFOV compared to conventional ventilation. We sometimes use heated, humidified high-flow nasal cannula (HHHFNC) for postextubation care in preterm infants >28 weeks’ gestation. HH-HFNC therapy may decrease the risk of extubation failure with the additional benefit of inducing less nasal trauma than CPAP. The impact of HHHFNC use on BPD risk has not been evaluated.

2. Chronic phase. Baseline ventilator settings are maintained with an aim to keep PaCO₂ <70 mm Hg with a compensated respiratory acidosis. Although an effort to transition to CPAP or HHHFNC as soon as possible is encouraged, subsequent support is not aggressively weaned until a pattern of steady weight gain is established.

C. Supplemental oxygen is supplied to maintain the PaO₂ >55 mm Hg. The Surfactant, Positive Pressure, and Oxygenation Randomized Trial (SUP-PORT) of low (85% to 89%) versus high (91% to 95%) SpO₂ targets in infants <28 weeks’ gestation revealed a higher mortality rate and no reduction in BPD rate (physiologic definition) in the low SpO₂ group, although severe ROP was less frequent in survivors. One approach for infants who receive supplemental oxygen is to target SpO₂ at 92% to 95% with alarm limits at 84% to 96%. Another is to adjust the target saturations according to gestational age or PMA (Table 34.1). Oximeter alarm limits may be set 0% to 2% outside the appropriate target range.

When end-expiratory pressure is no longer needed and FiO₂ is <0.3, we supply O₂ by nasal cannula (NC). We use a flow meter that is accurate at low rates and gradually decrease the flow of 100% O₂ while maintaining the appropriate SpO₂. Alternatively, flow can be decreased to the lowest marking on the flow meter as tolerated, and then O₂ concentration can be decreased. Estimates of the actual concentration of O₂ delivered to the lungs by NC at different flows of 100% O₂ have been generated by hypopharyngeal measurements (see Fig. 34.1). Once the infant remains stable on a low flow rate, we attempt a trial of withdrawal of NC support with close monitoring of O₂ saturation to determine if continued O₂ supplementation is required. In general, SpO₂ should remain >90% during sleep, feedings, and active periods before supplemental O₂ is discontinued. An “oxygen challenge test” can be performed at 36 weeks’ PMA to confirm whether an infant requires supplemental oxygen to maintain SpO₂ >90% and thus meets the physiologic definition of BPD.