Pulmonary Hypertension in Premature Infants. Sharpening the Tools of Detection

Bronchopulmonary dysplasia (BPD), the most common chronic lung disease of infancy, is characterized by alveolar simplification and developmental arrest of the lungs. BPD is a major complication of prematurity that affects one-third of extremely low-birth-weight infants, or 10,000 babies annually in the United States, and results in substantial long-term pulmonary and neurodevelopmental morbidity (1, 2). Improved survival of extremely premature babies has led to increased recognition of pulmonary hypertension (PH) and right ventricular failure in 12–25% of infants with BPD (3–8). Risk factors for BPD-associated PH include low birth weight, intrauterine growth restriction, oligohydramnios, and duration of mechanical ventilation with oxygen (5, 8). Once infants develop PH, little is known about how to best treat them. Poor outcomes are common, including up to 50% mortality by 2 years of life, increased risk for tracheostomy, and a 4.6-fold risk for death versus matched BPD infants without PH (4, 6, 8, 9). For those who do survive, the long-term consequences remain to be seen, as the oldest survivors are just reaching young adulthood. Adolescent survivors of BPD have been described to have persistent airway obstruction and reductions in airflow (10), but there are scant long-term data on infants discharged with PH in addition to BPD. It seems likely that the developmental arrest of the lungs and the pulmonary vasculature that accompanies BPD and PH will be accompanied by further cardiopulmonary morbidity as infants grow into childhood and adolescence.

The underlying pathophysiology of BPD-associated PH is complex and poorly understood. After birth, pulmonary arterial pressure decreases more slowly in preterm relative to term infants, especially in the presence of respiratory failure (11). Alterations of NO-cyclic guanosine monophosphate signaling disrupt lung and pulmonary vascular development in animal models of prematurity and lung injury (12–14), but these findings have not been successfully translated into therapeutic advances. Vascular agents such as iNO and sildenafil may acutely improve pulmonary artery pressure, but their long-term efficacy remains uncertain and poorly studied (15–17). A major barrier in conducting larger scale studies of promising therapies has been the identification of appropriate patients for inclusion. Moreover, until recently, there has been little evidence to guide the practitioner in when and how to perform PH screening in preterm babies. Part of the reason for this knowledge gap is that echocardiography is technically challenging in the face of patient agitation or concurrent illness and may not correlate well with PH severity, as measured by cardiac catheterization (18). In contrast, cardiac catheterization carries increased risk for procedure-related complications in vulnerable small infants. Serum biomarkers such as B-type natriuretic peptide or asymmetric dimethylarginine have not yet been well validated as predictors of disease severity in infants with BPD-associated PH (19–21).

In this issue of the Journal, the report by Mourani and colleagues (pp. 87–95) is the first to prospectively examine rates of both early (at 7 days) and late (at 36-week corrected gestational age) PH in a large population of premature infants (22). This study provides a comprehensive look at how the choice of echo parameters affects the incidence of PH. Using the “strictest” criteria of pulmonary artery pressure, as estimated by the tricuspid valve regurgitant jet, the authors found rates of early PH comparable to those found in other single-center prospective studies (7, 23). However, rates were considerably higher when the diagnostic criteria were expanded to include any of the following: right ventricle systolic pressure higher than 40 mm Hg, right ventricle systolic pressure/systemic systolic blood pressure ratio higher than 0.5, any cardiac shunt with bidirectional or right-to-left shunting, or any degree of ventricular wall septal flattening. In particular, findings of septal wall flattening and RV dilation occurred in up to 40% of preterm babies at 7 days of age and were most predictive of BPD and late PH. These used to be considered subtle findings of unclear significance, but they should now be viewed as important markers of PH in preterm babies.

Similar to previous reports, Mourani and colleagues found that early PH did not universally predict BPD and later PH (7). Thus, later screening remains necessary to identify those infants needing long-term follow-up after discharge. Mourani and colleagues also noted an association between lower birth-weight Z-scores and risk for PH, reinforcing other reports suggesting that infants with intrauterine growth restriction are a high-risk population of infants for future screening protocols (8). Perhaps the most surprising finding was that 10% of infants with no or mild BPD had detectable pulmonary hypertension at 36-week corrected gestational age. This novel observation suggests that a primary vascular injury may occur in some infants, independent of the lung disease. Whether this PH is clinically significant and portends the same degree of long-term morbidity and mortality as in infants with more severe associated lung disease will require further study.

Adding these prospective data to other recent reports, we recommend using Mourani and colleagues’ screening algorithm to assess for the presence of PH in all premature infants with moderate and severe BPD at 36-week corrected gestational age. The use of earlier testing, and of screening premature infants in the no and mild BPD categories, should be prioritized as an area of future investigation. Future studies should also build on Mourani and colleagues’ findings by using these screening tools to select high-risk populations for prevention and therapeutic intervention trials. We hope that Mourani and colleagues’ study will help unite neonatology, pulmonary, and critical care physicians in identifying the best strategies to prevent and reverse this devastating disease.