Prospective randomised trial of computer-aided diagnosis and contrast radiography in acute small bowel obstruction

Abstract

Objective: to compare the ability of computer-aided diagnosis and contrast radiography for the diagnosis of acute mechanical small bowel obstruction.

Design: Prospective randomised trial.

Setting: Kaunas University of Medicine, Lithuania.

Subjects: 80 patients with small bowel obstruction with no clinical evidence of strangulation who were randomised into two groups (n = 40 in each) to be investigated by computer-aided diagnosis and contrast radiography. INTENVENTIONS: 37 patients required operation (46%).

Main outcome measures: specificity, sensitivity, false positive and negative predictive values of the 2 methods; time necessary to make the diagnosis; and morbidity and mortality.

Results: The specificity, sensitivity, positive and negative predictive values in the diagnosis of complete acute small bowel obstruction for the computer-aided group were 100%, 87.5%, 100% and 92.3%, and for the contrast radiography group 100%, 76.9%, 100% and 90%, respectively. The mean time period for making the diagnosis was 1 hour in the computer-aided group and 16 hours in the radiography group (p < 0.001). The overall mortality was 3% and morbidity 9%.

Conclusion: Computer-aided diagnosis had no significant advantage over contrast radiography in the accuracy of diagnosis of the character of small bowel obstruction. However, significantly less time was needed to make the diagnosis in the computer-aided group.