1989 ARRS President's Award. The validity and utility of sonography in the diagnosis of appendicitis in the community setting.

Two hundred six patients with suspected appendicitis were examined with sonography over a 6-month period in three community teaching hospitals. Of 41 patients in whom the surgeons judged the clinical findings severe enough to warrant immediate surgery (group A), 34 (83%) had appendicitis, and sonography had a sensitivity of 0.76, a specificity of 0.71, and an accuracy of 0.76. Of 165 patients in whom the surgeons judged the clinical findings severe enough to warrant hospitalization for observation but not immediate surgery (group B), 51 (31%) had appendicitis at subsequent surgery. Sonography had a sensitivity of 0.96, a specificity of 0.94, and an accuracy of 0.95. Of 49 surgeons surveyed, the mean testing threshold (i.e., the probability of appendicitis below which they would send the patient home without further tests or observation) was 0.11, and the mean treatment threshold (i.e., the probability of appendicitis above which they would operate immediately) was 0.82. The posttest probability of appendicitis with findings indicating appendicitis present on sonography was 0.93 in group A and 0.88 in group B, and with findings absent on sonography it was 0.62 in group A and 0.02 in group B. We conclude that in group A patients, the use of sonography remains controversial in the diagnosis of appendicitis, but in group B patients it is both valid and useful.