Abstract:
HTML
Purpose: To evaluate the diagnostic performance of conventional screen-film radiography, storage phosphor radiography and a digital flat panel detector system in detecting fractures of the forearm and the carpal bones.
Methods and Materials: 66 cadaver arms from our Department of Anatomy (mean age of the donors ± SD: 80 ± 10 years) were loaded in a materials˘ testing machine simulating a fall onto the outstretched hand. Thus different fracture types of the proximal and distal radius and ulna as well of the carpal bones were created. Then radiographs of the complete forearm were obtained with an amorphous silicon flat panel detector, with conventional screen-film and storage phosphor radiography at an identical exposure dose. Four experienced radiologists evaluated the images concerning the presence or absence of fractures according to a five level confidence scale. In addition an interobserver analysis was performed rating image quality subjectively using a four level scale. Then anatomical preparation of the forearm was performed to depict the exact extent of the fractures. Receiver operating characteristics (ROC) analysis was performed assessing the diagnostic performance of the readers for the different imaging systems. The subjective image quality was evaluated using an analysis of variance.
Results: While the radiologists obtained a high diagnostic performance in the distal radius fractures (Area under curve (AZ)=0.95), performance was limited in the carpal bone fractures (AZ=0.74). However, compared to the anatomical preparation, depiction of the true extent of the radial fractures was limited using radiography. There were no statistically significant (p>0.05) differences between the three investigated radiography systems in the detection of the fractures. Concerning subjective image quality, conventional screen-film radiography and the digital flat panel detector obtained the highest scores.
Conclusion: In this in vitro study a significant difference between the three imaging systems could not be demonstrated concerning the diagnostic performance in forearm fractures (p>0.05). Using anatomical preparation as a standard of reference limitations were found in depicting carpal bone fractures and the true extent of radial fractures.
Questions about this event email: smetz@roe.med.tu-muenchen.de
Metz MD, S,
Comparison of Three Different Radiography Systems for Detection of Fractures in an Experimental Study. Radiological Society of North America 2003 Scientific Assembly and Annual Meeting, November 30 - December 5, 2003 ,Chicago IL.
http://archive.rsna.org/2003/3105279.html
