6226 Impact of point-of-care lung ultrasound in undifferentiated respiratory presentations in the paediatric emergency department

Objectives Respiratory symptoms are prevalent presentations to the paediatric emergency department (PED) and represent a significant disease burden globally. Point-of-care ultrasound (POCUS) shows promise as a radiation-sparing diagnostic tool in respiratory presentations. This study aims to explore the impact of POCUS in undifferentiated respiratory presentations in the PED by assessing its feasibility, evaluating the diagnostic accuracy in novice scanners, and investigating the potential to predict disease severity using POCUS. The objectives of the study are; To assess the feasibility and accuracy of POCUS by comparing the proportion of children diagnosed with different respiratory conditions via POCUS with that via standrard diagnostic modalities. Estimate the proportion of children presenting with respiratory symptoms who have abnormalities detected by lung POCUS. Estimate the diagnostic accuracy of a lung ultrasound (LUS) in a novice scanner. Investigate the potential of POCUS in predicting disease severity and guiding management decisions. Methods A prospective, cross-sectional study was conducted in a PED with an annual consensus of 60,000 patients. A trained student sonologist performed lung POCUS on patients with respiratory symptoms before clinical assessment. A six-zone scanning protocol was followed, scanning the anterior, lateral and posterior chest. The clinical team was blinded to the results of the scan. Results 47 participants completed the study protocol; (72.3% males, median age: 1.75 years). Difficulty in breathing was the most common presenting complaint (76.0%). Final clinical diagnoses included viral induced wheeze (VIW) (27.7%), bronchiolitis (27.7%), asthma exacerbation (8.5%), and lower respiratory tract infection (LRTI) (6.4%). Patients admitted had a higher rate of positive LUS findings (95.2%) compared to those that were discharged (69.2%). LUS demonstrated a high sensitivity in identifying pathology (97.1%) and predicting admission (95.2%), with lower specificities (66.7% and 30.8% respectively). For the diagnosis of specific clinical conditions, the sensitivity of LUS to VIW was 75.0%, 100% for asthma, 78.6% for bronchiolitis and 57.1% for LRTI. A notable finding included a patient who was diagnosed with consolidation and pleural effusion on LUS but received a misdiagnosed of VIW by the clinical team. Conclusion POCUS is an effective tool, with a high diagnostic accuracy, and is feasible to be incorporated into clinical practice to aid early diagnosis, reduce unnecessary interventions, and improve patient outcome. Although LUS demonstrated high sensitivities to identifying pathology and predicting admission in the hands of a novice scanner, the low specificity implies over-diagnosis, which could lead to unnecessary admissions. Further evaluation is necessary to assess its implementation in primary care and to address the challenges such as limited training and lack of guidelines.