Background: Late preterm and term infants may develop respiratory issues with severe outcomes. Early identification of these diseases shortly after infants’ birth can improve their management. Lung ultrasound (LUS) has been used to diagnose neonatal respiratory diseases. However, few LUS methods have been reported to predict the need for respiratory support, the basis of infant respiratory disease management.
Methods: We conducted a prospective diagnostic accuracy study following the STARD (Standards for the Reporting of Diagnostic Accuracy Studies) guidelines at a tertiary academic hospital between 2019 and 2020. A total of 310 late preterm and term infants with mild respiratory symptoms were enrolled. The LUS assessment was performed for each participant at one of the following times: 0.5 h, 1.0 h, 2.0 h, or 4.0 h after birth. Predictive reliability was tested by ROC analysis. The main outcome was the need for any respiratory support determined according to international guidelines.
Results: Seventy-four infants needed respiratory support, and 236 were healthy according to a 3-day follow-up confirmation. Six LUS imaging patterns were found. Two “high-risk” patterns were strongly correlated with respiratory support needs (AUC = 0.95; 95% CI: 0.92-0.98, p<0.001). The optimal cut-off value for “high-risk” patterns was 2 (sensitivity=87.8% and specificity=91.1%). The predictive value of LUS was greater than that of a symptom-based method (the ACoRN score) (AUCs’ p value <0.01).
Conclusions: LUS can be used to predict the need for respiratory support in late preterm and term infants and is more reliable than tools based on respiratory symptoms.