We hypothesized that point-of-care ultrasound (POCUS) is as accurate as radiology-performed ultrasound in evaluating children with clinical concern for appendicitis. As part of a staged approach, we further hypothesized that POCUS could ultimately decrease computed tomography (CT) utilization.
This was a prospective, convenience sampling of patients aged 2 to 18 years presenting with abdominal pain to a pediatric emergency department. Those patients with prior abdominal imaging, pregnant, or unable to tolerate the examination were excluded. An algorithm was followed: POCUS was first performed, followed by a radiology-performed ultrasound, and then a CT as necessary. The main outcome measure was the accuracy of the POCUS in diagnosing of appendicitis. This was compared with radiology-performed ultrasound. We also examined whether certain patient or clinical characteristics influenced the performance of POCUS. Lastly, we determined the amount by which CT scans were decreased through this staged algorithm.
Forty patients were enrolled and underwent a POCUS examination. A total of 16 (40%) had pathology-confirmed appendicitis. Point-of-care ultrasound had a sensitivity of 93.8% (95% confidence interval [CI], 69.7%–98.9%) and specificity of 87.5% (95% CI, 67.6%–97.2%). Radiology-performed ultrasound had a sensitivity of 81.25% (95% CI, 54.3%–95.7%) and specificity of 100% (95% CI, 85.6%–100%). The radiology-performed and POCUS examinations had very good agreement (κ = 0.83, P < 0.0005). Patient characteristics including body mass index did not have an affect on the POCUS. However, POCUS identified all patients with an Alvarado score higher than 6. Overall, the reduction in CT examinations was 55%.
In pediatric patients presenting with clinical concern for acute appendicitis, a staged algorithm that incorporates POCUS is accurate and has the potential to decrease CT scan utilization.
From the *Division of Pediatric Emergency Medicine, Division of Emergency Ultrasound, Department of Emergency Medicine, New York Methodist Hospital, Brooklyn, NY; †Department of Emergency Medicine, Stanford University Medical Center, Palo Alto; and ‡Department of Emergency Medicine, University of California, San Francisco, CA.
Disclosure: The authors declare no conflict of interest.
Reprints not available from authors.