Lung Ultrasound for Prediction of Bronchopulmonary Dysplasia

Background and Importance: Despite advances in neonatal care and improved survival among immature infants, bronchopulmonary dysplasia (BPD) remained a serious complication among preterm neonates born at <29 weeks' gestational age. A variety of therapies including antenatal steroid, surfactant administration, high-frequency ventilation, postnatal steroid (systemic and inhaled), and vitamin A supplementation have been studied as means to prevent BPD. Ideally, interventional studies aim to prevent or reduce BPD should include criteria for selecting infants at highest risk for BPD while excluding infants at low risk. Recently, Lung ultrasound severity score (LUSs) has shown in few studies to be a promising tool for assessing lung aeration and for early prediction of preterm infants with evolving BPD. These studies have a major limitation as it includes infants at low risk for BPD (> 28 weeks gestational age). In this study, we will assess the accuracy of LUSs for early identification BPD among extreme preterm infants born at <29 weeks' GA.

Study procedure In this prospective observational study, our aim is to recruit a large cohort of extreme premature neonates (born at ≤ 28+6 weeks gestational age), and sequentially perform an "early-diagnostic assessments "on day 3 of age. Subsequent LUS assessment will be performed at 7 days (+/- 1 day) of age and the third LUS assessment will be at 14 (+/-2 days of age). Assessment will be done in the same manner as described by Cattarossi et.al. Basically, each lung will be divided into 3 zones (upper anterior, lower anterior, and lateral) and will be examined using a L20 linear probe to perform a longitudinal lung scan (Fig 1). Additionally, we will score each zone (score 0-to 3 points) according to the modified LUS score by Brat et.al [20]. Based on the LUS pattern in each lung zone the total score can range from 0 to 18 for the 6 zones of both lungs. As summarized in figure 2 the LUS score will include signs that can be used to diagnose TTN, RDS or identify early changes of BPD.

We will record this data for babies enrolled in the study, along with their relevant medical data. At the end of this study, we will divide the cohort into those who developed BPD and those who did not as per currently used standard criteria and will compare the results of new LUS severity score obtained at earlier time points. Prediction models will be created including variables associated with BPD development at each time point with and without inclusion of lung US severity score.

Regarding LUS assessment a Zonare Ultrasound machine (Z.One PRO ultrasound) with a high-frequency (L20-5) linear probe will be used. This high-frequency probe gives excellent resolution and adequate penetration in these small infants. We decided to start LUS scanning at day 3 of age as most of these babies are at risk for intraventricular hemorrhage and minimal handling especially during the first 3 days is highly recommended.

All LUS assessments will be undertaken at the same time using a standard aseptic technique and without too much handling of the infants. Approval will be obtained from the Research Ethics Board (REB) at Sinai Health System and Health sciences Centre-Winnipeg. Parents will be approached for consent before enrollment.