Quantitative Lung Ultrasonography to Guide Surfactant Therapy in Neonates Born Late Preterm and Later

Abstract

Importance: Surfactant administration may be needed in late preterm through full-term neonates, but the pathophysiology of their respiratory failure can be different from that of early preterm neonates. The lung ultrasonography score (LUS) is accurate to guide surfactant replacement in early preterm neonates, but to our knowledge, it has not yet been studied in the late preterm through full-term neonatal population. Objective: To assess whether LUS is equally accurate to predict surfactant need in late preterm through full-term neonates as in early preterm neonates. Design, Setting, and Participants: This prospective, international, multicenter diagnostic study was performed between December 2022 and November 2023 in tertiary academic neonatal intensive care units in France, Italy, Spain, and the US. Late preterm through full-term neonates (≥34 weeks' gestation) with respiratory failure early after birth were enrolled. Exposure: Point-of-care lung ultrasonography to calculate the neonatal LUS (range, 0-18, with higher scores indicating worse aeration), which was registered in dedicated research databases and unavailable for clinical decision-making. Main Outcomes and Measures: The main outcomes were the area under the curve (AUC) in receiver operating characteristic analysis and derived accuracy variables, considering LUS as a replacement for other tests (ie, highest global accuracy) and as a triage test (ie, highest sensitivity). Sample size was calculated to assess noninferiority of LUS to predict surfactant need in the study population compared with neonates born more prematurely. Correlations of LUS with the ratio of hemoglobin oxygen saturation as measured by pulse oximetry (SpO2) to fraction of inspired oxygen (FiO2) and with the oxygen saturation index (OSI) were assessed. Results: A total of 157 neonates (96 [61.1%] male) were enrolled and underwent lung ultrasonography at a median of 3 hours (IQR, 2-7 hours) of life; 32 (20.4%) needed surfactant administration (pretest probability, 20%). The AUC was 0.87 (95% CI, 0.81-0.92). The highest global accuracy and sensitivity were reached for LUS values higher than 8 or 4 or lower, respectively. Subgroup analysis gave similar diagnostic accuracy in neonates born late preterm (AUC, 0.89; 95% CI, 0.81-0.97; n = 111) and early term and later (AUC, 0.84; 95% CI, 0.73-0.96; n = 46). After adjusting for gestational age, LUS was significantly correlated with SpO2:FiO2 (adjusted β, -10.4; 95% CI, -14.0 to -6.7; P