Accuracy of ultrasound in the diagnosis of acute appendicitis compared with the surgeon's clinical impression

Abstract

Objective: To compare the accuracy of the surgeon's clinical diagnosis of acute appendicitis with that of an ultrasonographic examination of the abdomen.

Design: Prospective trial.

Setting: US Navel Hospital, San Diego, Calif.

Patients: One hundred ten patients admitted to the hospital with suspected appendicitis from May 1990 to June 1992.

Intervention: Symptoms and signs for each patient were recorded, along with the surgeon's clinical impression of immediate surgery or observation. The patient then underwent an ultrasound examination performed by a staff radiologist. On the basis of the ultrasound findings the patient was placed into one of three categories: appendicitis, normal examination results, or other conditions. Patients with an ultrasound-based diagnosis of appendicitis proceeded to the operation, regardless of the surgeon's clinical impression. Those with other conditions diagnosed with ultrasonography were treated as was appropriate for the condition.

Results: The ultrasound-derived diagnosis of appendicitis had a sensitivity of 85.5%, a specificity of 84.4%, a positive predictive value of 88.3%, a negative predictive value of 80.1%, and an overall accuracy of 85.0%. The surgeon's clinical impression at the time of admission had a sensitivity of 62.9%, a specificity of 82.2%, a positive predictive value of 82.9%, a negative predictive value of 61.7%, and an overall accuracy of 71.2%.

Conclusion: The overall accuracy of ultrasonography in the diagnosis of appendicitis was statistically superior to that of the surgeon's clinical impression (P < .0001). However, 24% of the patients with normal ultrasound findings were ultimately found to have appendicitis at operation, emphasizing the point that ultrasonography cannot be relied on to the exclusion of the surgeon's careful and repeated evaluation.