Ultrasound, computed tomography, and laboratory findings in the diagnosis of appendicitis

Abstract

Purpose: To determine the diagnostic accuracy and the clinical impact of ultrasound (US) and computed tomography (CT) in diagnosing appendicitis, and to evaluate the impact of laboratory tests on the treatment of acute appendicitis.

Material and methods: All patients who, during 2005, underwent an acute ultrasound or CT investigation due to suspected appendicitis, or were diagnosed and/or surgically treated for appendicitis at Umeå University Hospital, Umeå, were included. The type of radiological investigation, its findings, the choice of treatment, final diagnosis, C-reactive protein (CRP), leukocyte particle count (LPC), body temperature, age, and sex were recorded for each patient. The histological result from surgery was considered the gold standard.

Results: The material included 305 cases with an overall appendicitis prevalence of 58%. Fifty-two percent of the patients were female. The mean age was 29 years, with a total range of 2-94 years. Twenty percent (60/305) underwent a CT investigation, 40% (123/305) underwent an US investigation, 5% (14/305) underwent both a CT and an US investigation, and 35% (108/305) of patients did not undergo any radiological investigation at all. The sensitivities and specificities were 91% and 94% for CT, and 83% and 98% for US, respectively. The positive likelihood ratio was 15.1 and 45.5 for CT and US, and the negative likelihood ratio was 0.09 and 0.18 for CT and US, respectively. It was not possible to visualize the appendix in 31% of patients examined with US. The prevalence of appendicitis in this group was the same as the prevalence among patients where it was possible to see the appendix, i.e., 35%. The mean CRP for all patients with appendicitis was 59 (95% CI 10-491) mg/l, and the mean LPC was 11.1 (95% CI 2.6-28.1) x10(-9)/l. The mean LPC level was significantly higher for the appendicitis patients. Body temperature could not significantly verify or exclude appendicitis. The overall negative appendectomy rate was 9% (16/176), and it was higher in women, i.e., 11% (9/79). The negative appendectomy rate was slightly higher in the group that was examined by CT and/or US, i.e., 12% (8/69) compared to 7% (8/107) in the group not examined radiologically.

Conclusion: Diagnostic accuracy was high for US as well as for CT. US was better for diagnosing positive findings, while CT was better for excluding diagnosis of appendicitis. The diagnostic accuracy of LPC, CRP, and body temperature was low. By combining findings from the radiological examination with the results from the clinical examination and laboratory values, a low negative appendectomy rate can be achieved.