Evaluation of abdominal gas by plain abdominal radiographs

Abstract

Background Our aim was to determine the reliability of plain abdominal radiographs for the evaluation of abdominal gas content in patients with functional digestive symptoms. Methods Abdominal CT scan scout views, mimicking a conventional plain abdominal radiograph, were obtained from 30 patients both during episodes of abdominal distension and basal conditions. Physicians (n = 50) were instructed to rate the estimated volume of gas in the 60 images presented in random sequence using a scale graded from 0 to ≥600 ml. Key Results The gas volumes estimated in the scout views differed from those measured by CT by a median of 90 (95% CI 70–102) ml, and the misestimation was not related to the absolute volume in the image. The accuracy of the observers, measured by their mean misestimation, was not related to their specialty or the training status (misestimation by 96 (95% CI 85–104) ml in staff vs 78 (70–106) ml in residents; p = 0.297). The accuracy was independent of the order of presentation of the images. Gas volume measured by CT in the images obtained during episodes of abdominal distension differed by a median of 39 (95% CI 29–66) ml from those during basal conditions, and this difference was misestimated by a median of 107 (95% CI 94–119) ml. The accuracy of these estimations was not related to the absolute gas volumes (R = −0.352; p < 0.001) or the magnitude of the differences. Conclusions & Inferences Plain abdominal radiographs have limited value for the evaluation of abdominal gas volume in patients with functional gut disorders.