Improving Diagnostic Interpretability of Abdominal Ultrasound for Neonates with Suspected Intestinal Injury

Yogen Singh, MD, FRCPCH, MA (Cantab)^1*Trevor Robert Bushong^2,3*Eric Diaz⁴Elim Man⁵Belinda Chan^1*

• ¹Neonatology Division, Department of Pediatrics, University of California Davis Health Children’s Hospital, Sacramento, United States
• ²University of California, Davis, Davis, United States
• ³St Joseph's Medical Center Stockton, Stockton, United States
• ⁴Department of Radiology, University of California Davis Health Children’s Hospital, Sacramento, United States
• ⁵Department of Paediatrics & Adolescent Medicine, Hong Kong Children’s Hospital and School of Clinical Medicine, The University of Hong Kong, Hong Kong, China

Background: Abdominal ultrasound (AUS) is increasingly utilized as a diagnostic adjunct in neonates undergoing evaluation for intestinal injuries such as necrotizing enterocolitis (NEC), spontaneous intestinal perforation (SIP), volvulus, and intestinal obstruction, which need urgent surgical evaluation and often emergent intervention. However, the interpretability of AUS—defined as the number of explicit documentations of high-risk ultrasound findings (HRF)—varies in radiology reports, potentially influenced by clinical and technical factors. Objective: To identify clinical and technical factors associated with increased interpretability of neonatal AUS in the evaluation of suspected intestinal injury needing surgical intervention. Methods: This retrospective, single-center case series reviewed AUS exams performed from 2022-2024 at a level IV neonatal intensive care unit. All neonates who had AUS performed prior to exploratory laparotomy were included in the study. For this project "interpretability of AUS" was defined as the number of explicit reporting of eight predefined HRF indicative of surgical need: pneumoperitoneum, increased or decreased bowel wall thickness, reduced intestinal perfusion on color Doppler, absent or decreased peristalsis, bowel dilation, complex intra-abdominal fluid collections, and reversed orientation of the superior mesenteric artery and vein. Clinical and technical factors that may have potentially influenced interpretability were analyzed. Results: Twenty-eight AUS exams from 18 neonates were analyzed. The median gestational age at birth was 34+2 weeks, and the median birth weight was 1.93 kg. The median HRF of all AUS exams were 2 (range 0-8). Higher-frequency ultrasound transducers (>10 MHz) improved interpretability of AUS images (HRF 4 vs 1), particularly in neonates weighing < 2 kg. Serial ultrasound evaluations within seven days of surgery were associated with greater interpretability compared to a single isolated exam (HRF 6 vs 3). Clinical symptoms with hypotension or abdominal discoloration and examinations ordered with comprehensive clinical details for the attention of radiology team showed trends towards improved interpretability. Conclusions: In our pilot study, interpretability of neonatal AUS images was strongly influenced by using higher-frequency transducers (>10 MHz) with better resolution, particularly in neonates weighing <2 kg. Obtaining serial imaging improved subsequent interpretability.