Can ultrasound reliably identify complicated appendicitis in children?

Background: The ability of ultrasound to identify specific features relevant to nonoperative management of pediatric appendicitis, such as the presence of complicated appendicitis (CA) or an appendicolith, is unknown. Our objective was to determine the reliability of ultrasound in identifying these features. Methods: We performed a retrospective study of children who underwent appendectomy after an ultrasound at four children's hospitals. Imaging, operative, and pathology reports were reviewed. Sensitivity, specificity, positive predictive value (PPV), and negative predictive value (NPV) of ultrasound for identifying CA based on pathology and intraoperative findings and an appendicolith based on pathology were calculated. CA was defined as a perforation of the appendix. Equivocal ultrasounds were considered as not indicating CA. Results: Of 1027 patients, 77.5% had simple appendicitis, 16.2% had CA, 5.4% had no evidence of appendicitis, and 15.6% had an appendicolith. Sensitivity and specificity of ultrasound for detecting CA based on pathology were 42.2% and 90.4%; the PPV and NPV were 45.8% and 89.0%, respectively. Sensitivity and specificity of ultrasound for detecting CA based on intraoperative findings were 37.3% and 92.7%; the PPV and NPV were 63.4% and 81.4%, respectively. Sensitivity and specificity of ultrasound for detecting an appendicolith based on pathology were 58.1% and 78.3%; the PPV and NPV were 33.1% and 91.0%, respectively. Results were similar when equivoc