• George A Taylor, Michael J Callahan, Diana Rodriguez, and Douglas S Smink.
• Department of Radiology, Children's Hospital Boston, Harvard Medical School, Boston, MA, USA. george.taylor@childrens.harvard.edu
• Pediatr Radiol. 2006 Apr 1;36(4):331-7.

BackgroundMistakes have been made by the use of CT in diagnosing children with suspected appendicitis. Although others have reported the frequency of diagnostic errors, we were unable to find any studies that addressed the specific situations in which diagnostic errors occurred in children with suspected appendicitis.ObjectiveTo investigate the frequency and type of diagnostic errors resulting from CT of children with suspected appendicitis when compared to surgical and pathological diagnosis.Materials And MethodsWe reviewed imaging, clinical and pathological data on 1,207 consecutive pediatric patients who underwent CT examination for suspected appendicitis. Imaging findings were categorized as false-positive, false-negative, or indeterminate. Errors were classified as interpretative, technical or unavoidable. Concordance between surgical and pathological findings was also evaluated.ResultsThe imaging findings of 34 patients (2.8%) were discrepant with the pathological examination or clinical follow-up. The errors in 23 cases were classified as interpretive (68%) and 11 as unavoidable (32%), and no errors were classified as technical. There were 23 false-positive errors (68%), 6 false-negative errors (18%), and 5 indeterminate imaging studies (15%). Isolated CT findings of an enlarged (greater than 6 mm) appendix, fat stranding, thickened bowel or non-visualization of the distal appendix were the most common false-positive CT findings. Of these 34 patients, 22 underwent appendectomy, with 10 (45%) having discordant surgical and pathological findings.ConclusionsIsolated CT findings of an appendicolith, an enlarged appendix, or minimal fat stranding are not sufficient signs for the diagnosis of appendicitis. Pathological diagnosis rather than surgical findings should be used as the reference standard of true-positive imaging findings.

27 keywords...

hide…