Date of Presentation
5-2-2024 12:00 AM
College
Rowan-Virtua School of Osteopathic Medicine
Poster Abstract
Purpose:
Necrotizing enterocolitis (NEC) is an abdominal inflammatory condition that is common in premature neonates. Although abdominal radiograph (AR) remains the imaging standard for NEC, it may miss up to 50% of early signs of NEC and has been described to have a sensitivity as low as 15.4% for detecting pneumoperitoneum. Abdominal ultrasound (US) is portable, non-invasive, and allows real-time bowel integrity, movement, and perfusion assessment. We aim to evaluate the concordance between US and AR in detecting NEC features and the diagnostic performance of both modalities in detecting pneumoperitoneum.
Methods and materials:
We conducted an IRB-approved retrospective, cross-sectional, single-center study. We identified infants with a diagnosis of NEC confirmed by pathology reports that had a bowel US and AR studies obtained before surgery from January 2012 to August 2022. We extracted clinical and demographic data from our electronic chart system. Two pediatric radiologists, blinded to reports, evaluated the images to determine the presence of pneumatosis (PI), portal vein gas (PVG), bowel distension (BD), and pneumoperitoneum on both modalities. A third pediatric radiologist resolved discrepant responses. We calculated the diagnostic performance of both modalities to detect perforation based on the presence of pneumoperitoneum, and the concordance between them utilizing the kappa statistic (κ). We excluded studies with insufficient diagnostic quality.
Results
Our cohort included 9 girls and 22 boys, median age 23 days (IQR 14.5-55 days). Of those, 23 (76%) were born prematurely, and 20 had confirmed intestinal perforation. US demonstrated 35% sensitivity and 90% specificity, while AR demonstrated 15% sensitivity and 100% specificity. Agreement between US and AR was 10/30 (33.3%) for PI (κ=0.01), 22/28 (79%) for PVG (κ=0.2), 19/31 (61%) for BD (κ=0.21), and 24/31 (77%) for pneumoperitoneum (κ=0.34). Each feature was present more frequently on US than AR.
Conclusion:
Our study demonstrated that abdominal US was a valuable complementary tool for detecting NEC features and intestinal perforation. Despite a low to moderate agreement between both modalities, US consistently outperformed AR in identifying NEC features, including pneumoperitoneum. These findings highlight the significance of integrating US into the NEC diagnostic process and the need for revising the current NEC diagnostic algorithm. Future efforts include larger cohorts and a collaborative approach to improve the NEC diagnostic algorithm.
Keywords
Abdomen, Abdominal Radiography, Ultrasound, Necrotizing enterocolitis, Neonate, Premature Birth, Surgical Procedures, Diagnosis
Disciplines
Congenital, Hereditary, and Neonatal Diseases and Abnormalities | Diagnosis | Digestive System Diseases | Health and Medical Administration | Medicine and Health Sciences | Other Analytical, Diagnostic and Therapeutic Techniques and Equipment | Pathological Conditions, Signs and Symptoms | Pediatrics | Radiology | Surgery
Document Type
Poster
DOI
10.31986/issn.2689-0690_rdw.stratford_research_day.169_2024
Included in
Congenital, Hereditary, and Neonatal Diseases and Abnormalities Commons, Diagnosis Commons, Digestive System Diseases Commons, Health and Medical Administration Commons, Other Analytical, Diagnostic and Therapeutic Techniques and Equipment Commons, Pathological Conditions, Signs and Symptoms Commons, Pediatrics Commons, Radiology Commons, Surgery Commons
Ultrasound versus Radiography for Evaluating Surgical Necrotizing Enterocolitis
Purpose:
Necrotizing enterocolitis (NEC) is an abdominal inflammatory condition that is common in premature neonates. Although abdominal radiograph (AR) remains the imaging standard for NEC, it may miss up to 50% of early signs of NEC and has been described to have a sensitivity as low as 15.4% for detecting pneumoperitoneum. Abdominal ultrasound (US) is portable, non-invasive, and allows real-time bowel integrity, movement, and perfusion assessment. We aim to evaluate the concordance between US and AR in detecting NEC features and the diagnostic performance of both modalities in detecting pneumoperitoneum.
Methods and materials:
We conducted an IRB-approved retrospective, cross-sectional, single-center study. We identified infants with a diagnosis of NEC confirmed by pathology reports that had a bowel US and AR studies obtained before surgery from January 2012 to August 2022. We extracted clinical and demographic data from our electronic chart system. Two pediatric radiologists, blinded to reports, evaluated the images to determine the presence of pneumatosis (PI), portal vein gas (PVG), bowel distension (BD), and pneumoperitoneum on both modalities. A third pediatric radiologist resolved discrepant responses. We calculated the diagnostic performance of both modalities to detect perforation based on the presence of pneumoperitoneum, and the concordance between them utilizing the kappa statistic (κ). We excluded studies with insufficient diagnostic quality.
Results
Our cohort included 9 girls and 22 boys, median age 23 days (IQR 14.5-55 days). Of those, 23 (76%) were born prematurely, and 20 had confirmed intestinal perforation. US demonstrated 35% sensitivity and 90% specificity, while AR demonstrated 15% sensitivity and 100% specificity. Agreement between US and AR was 10/30 (33.3%) for PI (κ=0.01), 22/28 (79%) for PVG (κ=0.2), 19/31 (61%) for BD (κ=0.21), and 24/31 (77%) for pneumoperitoneum (κ=0.34). Each feature was present more frequently on US than AR.
Conclusion:
Our study demonstrated that abdominal US was a valuable complementary tool for detecting NEC features and intestinal perforation. Despite a low to moderate agreement between both modalities, US consistently outperformed AR in identifying NEC features, including pneumoperitoneum. These findings highlight the significance of integrating US into the NEC diagnostic process and the need for revising the current NEC diagnostic algorithm. Future efforts include larger cohorts and a collaborative approach to improve the NEC diagnostic algorithm.