A New Algorithm for Classifying Acetabular Fracture Patterns on Three-dimensional Computed Tomography Reconstructions Markedly Improves Residents' Ability to Correctly Classify Fractures.

Abstract

CT and three-dimensional (3D) CT reconstructions have been shown to improve the understanding of acetabular fractures. With the increased availability of 3D pelvic CT reconstructions, our goal for this study was to develop an algorithm to aid residents in the classification of acetabular fractures. We hypothesized that the use of a stepwise algorithm will markedly enhance the trainees' ability to correctly identify acetabular fracture patterns. This was a multicenter study that included 33 residents. Residents reviewed 15 sets of 3D reconstructions of the 10 acetabular fracture patterns. Residents completed the first round, and the results were collected electronically. Three weeks later, they were asked to classify the fractures a second time with the use of the algorithm. The number of correct responses from the two sessions was analyzed to determine if the algorithm improved residents' ability to correctly classify fracture patterns. Thirty-three residents classified 15 fractures which yielded 495 unique responses. Residents correctly classified 52.5% (260/495) of fractures without the algorithm, which significantly increased to 77.5% (384/495) (P = 0.001) with the algorithm. When stratified by year in residency, all residents were able to correctly classify markedly more fractures with the algorithm. Overall, we believe this method is a reproducible diagnostic tool that will assist residents in classifying acetabular fractures. We were able to demonstrate that with the use of this algorithm, residents' ability to correctly classify acetabular fractures is markedly enhanced, regardless of year in training. This algorithm will be a useful adjunct to assist and advance trainees' education and understanding of a complex topic.