PURPOSE: To evaluate the reliability and accuracy of diagnosis of scapholunate dissociation (SLD) among AO type C (compression articular) fractures of the distal radius. METHODS: A total of 217 surgeons evaluated 21 sets of radiographs with type C fractures of the distal radius for which the status of the scapholunate interosseous ligament was established by preoperative 3-compartment computed tomographic arthrography with direct operative visualization of diagnosed SLD (reference standard). Observers were asked whether SLD was present, and if yes, whether they would recommend operative treatment. Diagnostic performance characteristics were calculated with respect to the reference standard. We assessed interobserver reliability using the Fleiss generalized kappa. RESULTS: The interobserver agreement for radiographic diagnosis of SLD was moderate (κ = 0.44). Correct diagnosis for a given set of radiographs ranged from 8% to 98% (average, 79%) of observers. Diagnostic performance characteristics were: 69% sensitivity, 84% specificity, 84% accuracy, 68% positive predictive value, and 84% negative predictive value. Based on a prevalence of 5%, Bayes adjusted positive and negative predictive values were 18% and 98%, respectively. Raters recommended operative treatment in 74% to 100% of patients diagnosed with SLD. CONCLUSIONS: Radiographs are moderately reliable and are better at ruling out than ruling in SLD associated with type C fracture of the distal radius.
PURPOSE: To evaluate the reliability and accuracy of diagnosis of scapholunate dissociation (SLD) among AO type C (compression articular) fractures of the distal radius. METHODS: A total of 217 surgeons evaluated 21 sets of radiographs with type C fractures of the distal radius for which the status of the scapholunate interosseous ligament was established by preoperative 3-compartment computed tomographic arthrography with direct operative visualization of diagnosed SLD (reference standard). Observers were asked whether SLD was present, and if yes, whether they would recommend operative treatment. Diagnostic performance characteristics were calculated with respect to the reference standard. We assessed interobserver reliability using the Fleiss generalized kappa. RESULTS: The interobserver agreement for radiographic diagnosis of SLD was moderate (κ = 0.44). Correct diagnosis for a given set of radiographs ranged from 8% to 98% (average, 79%) of observers. Diagnostic performance characteristics were: 69% sensitivity, 84% specificity, 84% accuracy, 68% positive predictive value, and 84% negative predictive value. Based on a prevalence of 5%, Bayes adjusted positive and negative predictive values were 18% and 98%, respectively. Raters recommended operative treatment in 74% to 100% of patients diagnosed with SLD. CONCLUSIONS: Radiographs are moderately reliable and are better at ruling out than ruling in SLD associated with type C fracture of the distal radius.
Authors: Shadpour Demehri; Nima Hafezi-Nejad; John N Morelli; Uma Thakur; Scott D Lifchez; Kenneth R Means; John Eng; Jaimie T Shores Journal: Skeletal Radiol Date: 2015-12-14 Impact factor: 2.199
Authors: Jos J Mellema; Wouter H Mallee; Thierry G Guitton; C Niek van Dijk; David Ring; Job N Doornberg Journal: J Digit Imaging Date: 2017-10 Impact factor: 4.056
Authors: Daniel Gonçalves Machado; Sergio Auto da Cruz Cerqueira; Alexandre Fernandes de Lima; Marcelo Bezerra de Mathias; José Paulo Gabbi Aramburu; Rodrigo Ribeiro Pinho Rodarte Journal: Rev Bras Ortop Date: 2016-01-21