Kacl, G. M., Zanetti, M., Amgwerd, M., Trentz, O., Seifert, B., Stucki, H., and Hodler, J.
European Radiology. March 1997, Vol. 7 Issue 2, p187, 5 p.
Rapid prototyping -- Analysis
The purpose of this study was to evaluate and compare the diagnostic performance of stereolithography vs workstation-based three-dimensional (3D) reformations in intra-articular calcaneal fractures. A total of 30 intra-articular calcaneal fractures were examined using standard radiographs, coronal CT scans, and 2D and 3D reformations. The CT data were transferred to an outside institution, and stereolithograms were produced from photopolymer resin employing a laser beam system. 3D reformations and stereolithograms were analyzed in a blinded fashion by two staff radiologists. Receiver-operating-characteristic (ROC) curves were obtained for six clinically significant fracture components. Standard radiographs, coronal CT scans, and 2D reformations served as the standard of reference. The area under the ROC curves for 3D reformations and stereolithograms were 1.0 and 0.98 for abnormal tuber angles, 0.91 and 0.91 for anterior and middle talo-calcaneal joint involvement, 0.90 and 0.95 for involvement of the posterior talo-calcaneal joint, 0.65 and 0.78 for the presence of a lateral bulge, 0.80 and 0.81 for the involvement of the calcaneocuboidal joint, and 0.62 and 0.67 for the presence of a 'tongue-type' fracture. No statistically significant difference was demonstrated for the two methods (Wilcoxon signed-rank test, p = 0.138). Based on our results stereolithograms did not prove to be statistically superior to workstation-based 3D reformations. Stereolithograms may still be useful for teaching purposes and for surgical planning at a thinking-efficacy level.