50 - Clinical Suspicion Compared to PECARN Prediction Rule for Intra-Abdominal Injury Undergoing Acute Intervention (IAIai) in Pediatric Patients with Blunt Torso Trauma

Poster Number: 50
Publication Number: 50.104

Kenneth Yen, University of Texas Southwestern Medical School, Dallas, TX, United States; Pradip P. Chaudhari, Children's Hospital Los Angeles, Los Angeles, CA, United States; Paul Ishimine, UC San Diego, San Diego, CA, United States; Grant Tatro, Virginia Commonwealth University School of Medicine, Richmond, VA, United States; Kevan A. McCarten-Gibbs, University of California, San Francisco, School of Medicine, Oakland, CA, United States; Irma Ugalde, McGovern Medical School at the University of Texas Health Science Center at Houston, Houston, TX, United States; Mohamed Badawy, UT Southwestern Medical Center, Dallas, TX, United States; Nathan Kuppermann, University of California, Davis, School of Medicine, Sacramento, CA, United States; James F. Holmes, University of California, Davis, School of Medicine, sacramento, CA, United States

Background: The Pediatric Emergency Care Applied Research Network (PECARN) Intra-abdominal Injury (IAI) prediction rule identifies children at very low risk of IAI undergoing acute intervention (IAIai). For a prediction rule to be useful, it should be more accurate than clinician suspicion. 
 

Objective: Our objective was to compare the test characteristics of clinician suspicion with the PECARN prediction rule for IAIai.

Design/Methods: We analyzed the data from a prospective multicenter cohort study in children ( < 18 years) with blunt torso trauma. Clinicians documented their suspicion for IAIai before CT (if performed< ![if !supportAnnotations] >[PI1]< ![endif] > ) as < 1%, 1-5%, 6-10%, 11-50%, or >50%. IAIai was defined as IAI with laparotomy, angiographic embolization, blood transfusion for abdominal hemorrhage, or IV fluid administration for ≥2 days in those with pancreatic or gastrointestinal injuries. We compared test characteristics with 95% confidence intervals (CI) of clinician suspicion > 1% of IAIai versus having at least one PECARN prediction rule variable present.

Results: Clinician suspicion was documented in 6,274/6,327 (99%) of enrolled children and were analyzed. IAIai was diagnosed in 126 (2%). Abdominal CTs were obtained in 841/4,563 (18.5%; 95% CI 17.3, 19.6%) of those with clinician suspicion < 1%, 677/1,061 (63.8% 95% CI 60.8, 66.7%) with suspicion 1-5%, 297/338 (87.9% 95% CI 83.9, 91.1%) with suspicion 6-10%, 205/218 (94.0% 95% CI 90.0, 96.8%) with suspicion 11-50%, and 82/94 (87.2% 95% CI 78.8, 93.2%) with suspicion >50%. Prediction rule sensitivity (126/126, 100% CI 97.7,100%) was higher than clinician suspicion ≥1% (115/126; 91.2%, 95% CI 84.9, 95.6%; difference 8.7%, 95% CI 3.8, 13.7%). Prediction rule specificity (2,840/6,148; 46%, 95% CI 45, 47%) was lower than clinician suspicion (4,552/6,148; 74%, 95% CI 73, 75%; difference -28%, 95% CI 26, 30%). In the 11 patients with IAIai and suspicion < 1%, seven (63%) had laparotomy. In the 841 with clinician suspicion < 1% who underwent CT, the most frequent reasons for CT were severe mechanism n=421, abnormal abdominal exam n=237, and trauma surgery request n=198.  Conclusion(s): The PECARN IAIai prediction rule had a significantly higher sensitivity but lower specificity than clinician suspicion. The higher specificity of suspicion, however, did not translate into lower CT rates for those very low risk for IAIai.