Neonatal Lung Ultrasound and Surfactant Administration: A Pragmatic, Multicenter Study.

ABSTRACT

BACKGROUND: Previous research shows that a lung ultrasound score (LUS) can anticipate CPAP failure in neonatal respiratory distress syndrome. RESEARCH QUESTION Can LUS also predict the need for surfactant replacement? STUDY DESIGN AND METHODS: Multicenter, pragmatic study of preterm neonates who underwent lung ultrasound at birth and those given surfactant by masked physicians, who also were scanned within 24 h from administration. Clinical data and respiratory support variables were recorded. Accuracy of LUS, oxygen saturation to Fio2 ratio, Fio2, and Silverman score for surfactant administration were evaluated using receiver operating characteristic curves. The simultaneous prognostic values of LUS and oxygen saturation to Fio2 ratio for surfactant administration, adjusting for gestational age (GA), were analyzed through a logistic regression model. RESULTS: Two hundred forty infants were enrolled. One hundred eight received at least one dose of surfactant. LUS predicted the first surfactant administration with an area under the receiver operating characteristic curve (AUC) of 0.86 (95% CI, 0.81-0.91), cut off of 9, sensitivity of 0.79 (95% CI, 0.70-0.86), specificity of 0.83 (95% CI, 0.76-0.89), positive predictive value of 0.79 (95% CI, 0.71-0.87), negative predictive value of 0.82 (95% CI, 0.75-0.89), positive likelihood ratio of 4.65 (95% CI, 3.14-6.89), and negative likelihood ratio of 0.26 (95% CI, 0.18-0.37). No significant difference was shown among different GA groups 25 to 27 weeks' GA (AUC, 0.91; 95% CI, 0.84-0.99), 28 to 30 weeks' GA (AUC, 0.81; 95% CI, 0.72-0.91), and 31 to 33 weeks' GA (AUC, 0.88; 95% CI, 0.79-0.95), respectively. LUS declined significantly within 24 h in infants receiving one surfactant dose. When comparing Fio2, oxygen saturation to Fio2 ratio, LUS, and Silverman scores as criteria for surfactant administration, only the latter showed a significantly poorer performance. The combination of oxygen saturation to Fio2 ratio and LUS showed the highest predictive power, with an AUC of 0.93 (95% CI, 0.89-0.97), regardless of the GA interval. INTERPRETATION: LUS is a reliable criterion to administer the first surfactant dose regardless of GA. Its association with oxygen saturation to Fio2 ratio significantly improves the prediction power for surfactant need.