242 - The Relationship Between Radiographic, Sonographic and Clinical Assessment of Lung Volume in Very Preterm Infants.

Sunday, April 30, 2023

3:30 PM – 6:00 PM ET

Poster Number: 242
Publication Number: 242.339

Arun Sett, Murdoch Children's Research Institute, Melbourne, Victoria, Australia; Sheryle R. Rogerson, Royal Womens Hospital, North Melbourne, Victoria, Australia; Gillian W. Foo, Joan Kirner Women's & Children's, Sunshine Hospital, Melbourne, Victoria, Australia; Jacqui A. Keene, Royal Womens Hospital, Melbourne, Victoria, Australia; Niranjan Thomas, University of Melbourne, Keilor East, Victoria, Australia; Penny Pei Lee Kee, Western Health, Melbourne, Victoria, Australia; Amir M. Zayegh, The Royal Women's Hospital, Melbourne, Victoria, Australia; Susan Donath, University of Melbourne and Murdoch Children's Research Institute, Melbourne, Victoria, Australia; David Tingay, Murdoch Children's Research Institute, Parkville, Victoria, Australia; Peter Davis, University of Melbourne, Melbourne, Victoria, Australia; Brett J. Manley, The Royal Women's Hospital, Melbourne, Australia, Parkville, Victoria, Australia

Presenting Author(s)

AS

Arun Sett, MBBS (Hons) CCPU (Neonatal) FRACP (he/him/his)

Neonatal Research Fellow
The Royal Women’s Hospital, Victoria
Melbourne, Victoria, Australia

Background:

Although lung ultrasound (LUS) use is increasing in neonatal intensive care, chest x-ray (CXR) is still used to assess lung volume. The performance of LUS compared with CXR for this purpose in preterm infants is unknown.

Objective: To determine the accuracy of LUS to detect radiographic under-recruitment and to determine the correlation between CXR and LUS measures of lung volume and oxygenation.

Design/Methods:

LUS was performed on 70 preterm infants before CXR and graded using; 1. A LUS score, 2. An atelectasis score and 3. Measuring atelectasis depth. Accuracy of LUS to predict CXR defined global atelectasis, and the relationship between LUS, CXR and oxygenation was assessed using receiver operator characteristics (ROC) and correlation analysis.

Results:

LUS scores, atelectasis scores and atelectasis depth did not predict radiographic global atelectasis (area under ROC curves [95% CI]: 0.54[0.36,0.71], 0.49[0.34,0.64], 0.47[0.31,0.64], respectively). Radiographic atelectasis of the right upper, right lower, left upper and left lower quadrants was predicted by LUS scores (0.75[0.59,0.92], 0.75[0.62,0.89], 0.69[0.56,0.82], 0.63[0.508,0.751]) and atelectasis depth (0.66 [0.54,0.78], 0.65[0.53,0.77], 0.63[0.50,0.76], 0.56[0.44,0.70]). LUS findings were moderately correlated oxygen saturation index (r=0.52[0.30,0.70]) and lower saturation to fraction of inspired oxygen ratio (r=-0.63[-0.76,-0.46]). The correlation between radiographic lung volume, OSI and S/F ratio was very weak (r=0.36[0.11,0.59] and r=-0.32[-0.53,-0.07] respectively).

Conclusion(s):

There was little relationship between LUS and radiographic global atelectasis. However, LUS predicted radiographic regional atelectasis, and more strongly correlated with oxygenation than CXR. While LUS may not replace all radiographic estimates of lung volume, it more accurately reflects respiratory status in preterm infants.