Question

• Should plain abdominal radiographs be routinely ordered in the evaluation of the patient with clinically suspected appendicitis?

Clinical Bottom Lines

1. Plain abdominal x-ray is neither sensitive nor specific in the suspected diagnosis of appendicitis.
2. In a review of 821 consecutive patients hospitalized for appendicitis, radiographic findings were noted in 51% with and 47% without appendicitis.(1)
3. No specific x-ray finding was sensitive or specific.
4. Specific conditions were suggested in 10 % of impressions, which failed to correlate with clinical diagnosis over half (57%) of the time.
5. Hospital cost per radiograph was estimated at $67; hospital cost per correct diagnosis was $1593.
6. In comparison to a similar study by the same authors, hospital cost per appendiceal computed tomography scan was $270. (2)

Summary of Key Evidence

1. The study group was composed of 821 patients admitted to hospital through an Emergency Department for suspected appendicitis.(1)
2. 423 male, 398 female patients ranging from 1 year to 89 years.
3. Of the 821 patients, 642 (78%) received an abdominal radiographic series as part of their initial evaluation.
4. Radiographs on the 642 patients were interpreted without knowledge of final clinical diagnosis, and a review of the reports was conducted blinded to each patient's clinical diagnosis.  Reports were reviewed for findings, impressions related to appendicitis, and overall impressions.
5. Final diagnosis was either appendicitis, a specific alternate diagnosi,s or nonspecific abdominal pain and was determined based on review of charts for clinical diagnosis at discharge, not ascertained via surgery and pathological examination. However, 524 (64%) patients had a diagnosis of appendicitis confirmed by pathology.
6. 51% (CI 44.2-54.3%) of patients with appendicitis had findings of some kind on abdominal radiograph.
7. 47% (CI 46.6 to 59.4%) of patients with diagnosis of appendicitis excluded had findings on abdominal radiograph.
8. No individual radiographic finding was statistically more likely to occur in patients with appendicitis compared to those without.
9. Overall radiographic impressions were normal 50 % of the time in patients with appendicitis, and 60 % of the time in patients without (p=0.0075).
10. Radiographic impression suggested a diagnosis (appendicitis or other) in 10% of the cases; failed to correlate 57 % of the time.
11. Average cost of a x-ray estimated at $67; average cost per correct diagnosis was $1593 (642 X $ 67 divided by total number of correct and specific diagnoses=27).

Additional Comments

• The study is limited by the fact that it is a retrospective chart review only of patients admitted to the hospital with suspected appendicitis; those patients evaluated with x-rays and sent home were not included.
• 78 % of the patients admitted got radiographs; 22% did not.
• Patients that did not get x-rays were 71 % more likely to have appendicitis.
• Gold standard (pathologic examination) not applied to ALL patients.
• Initial radiograph evaluated with clinical suspicion in mind.

Citation

1. Rao P, Rhea J, et al. Plain abdominal radiography in clinically suspected appendicitis: diagnostic yield, resource use, and comparison to CT. American Journal of Emergency Medicine, 1999; 17:325-328.
   
2. Rao P, Rhea J, Novelline RA, et al. Effect of computed tomography of the appendix on treatment of patients and use of hospital resources. New England Journal of Medicine, 1998; 338:141-146.

CAT Author: Megan Spohr, MD

CAT Appraisers: John G. Frohna, MD

Date appraised: March 27, 2000