O3295: Predictors of Non-diagnostic ultrasound leading to Overutilization of Cross-sectional imaging for the Diagnosis of Acute Appendicitis

Background: Acute appendicitis accounts for the majority of pediatric surgical disorders evaluated in the emergency department. While computed tomography(CT) scans enhance diagnostic accuracy, their overuse poses detrimental radiation exposure risks in this population, especially in rural areas that lack pediatric imaging capabilities. This study evaluates the utilization of cross-sectional imaging vs ultrasound (US) used to diagnose acute appendicitis at regional community hospitals in our referral network compared to our tertiary children’s hospital with pediatric imaging expertise and determine which factors (demographic or disease severity) led to an increase in nondiagnostic US read and subsequent cross-sectional imaging.

Methods: A single-center, retrospective study was conducted from January 2018 to December 2022 of pediatric surgical patients aged < 18 years diagnosed with acute appendicitis who presented to the emergency department. All patients with imaging that demonstrated acute appendicitis were included, whether from an outside hospital or the tertiary center. All imaging studies were re-read by the tertiary center radiologist if obtained at an outside hospital. Patient-specific demographic, imaging, and clinical data were collected. Collected data was compared between the two groups using nonparametric statistical analyses and multivariable logistic regression.

Results: 301 patients met the inclusion criteria. 200 patients were transferred from a regional hospital, and 101 initially presented to the pediatric tertiary care center. 92.1% of the children who presented to the pediatric tertiary care center had a US as their initial imaging compared to 31.0% who presented as a transfer. The rate of non-diagnostic US was lower at the pediatric tertiary center compared with community hospitals (29.0% vs 49.3%, p=0.08). More patients from community hospitals received cross-sectional than those from the pediatric tertiary center (80.0% vs 37.6%, p< 0.001). The probability of having a nondiagnostic US was not associated with patient BMI, degree of symptoms, age, insurance status, or lab values. African American race and having the US completed and read at the pediatric tertiary care center was significantly associated with the nondiagnostic US. The odds of a nondiagnostic US at a transferring center were 2.4 (95% CI 1.2-4.7, p=0.011) times that of a nondiagnostic US at the specialized pediatric center. The odds of nondiagnostic US for African Americans were 5.4 (95% CI 1.2-38.7, p=0.046) times the odds of nondiagnostic US in other races.

Conclusion: This work aligns with the current literature on increased utilization of cross-sectional imaging at centers without pediatric imaging expertise. African American race demonstrated a significant difference in the rates of nondiagnostic odds that requires further investigation. This work is instrumental in creating and implementing collaborative imaging guidelines outreach programs to decrease utilization at these regional facilities and potentially mitigate some of the radiation-associated malignancy risks in this population of patients.