Background Chest X-ray (CXR) has typically been the main investigation in children with suspected respiratory pathology. Recent advances in lung point-of-care ultrasound (POCUS) have shown the potential for it to be comparative, if not better, than CXR. The objective of this study was to compare CXR with lung POCUS in children with respiratory illness in a ward-based setting at a paediatric teaching hospital.

Methods Any child <18 years of age presenting to Southampton Children’s Hospital requiring a CXR for clinical reasons also had lung POCUS performed. CXR was reported by a consultant paediatric radiologist and lung POCUS was reviewed retrospectively by a blinded POCUS clinician, with only the clinical information provided on the CXR request. Comparisons were made between the CXR and lung POCUS findings.

Results 100 paired lung POCUS and CXR were included in the study. 30% of lung POCUS were normal with 97% of these having a normal CXR. 70% of cases had POCUS abnormalities with 96% of POCUS cases identifying comparative lung pathology. Lung POCUS therefore had a sensitivity of 98.51% and a specificity of 87.9% with a diagnostic accuracy of 95% when compared with the CXR report.

Conclusions Lung POCUS has excellent diagnostic accuracy. The diagnosis of normal lung on POCUS when performed by a trained practitioner can reliably reduce the need for a CXR, thus reducing CXR use and radiation exposure in children. An abnormal lung POCUS could then either give the diagnosis or lead to a CXR with the expectation of clinically relevant findings.