Surgeon Accuracy in Prediction of Stability of Posterior Wall Acetabular Fractures. A Survey of Orthopaedic Trauma Surgeons

Abstract

Objectives: This study aims to investigate surgeon accuracy in prediction of the stability of posterior wall acetabular fractures by comparing “examination under anesthesia” findings to submitted estimations on the basis of radiograph and computed tomography (CT) imaging across a range of experience in orthopaedic surgeons and trainees. Methods: Records of patients who underwent examination under anesthesia after presenting with posterior wall acetabular fractures at 2 different institutions were pooled for data collection, totaling 50 cases. Radiographs, CT images, and information regarding the presence of a hip dislocation requiring procedural reduction were provided to participants for review. A survey was generated for submission of impressions of stability for each individual case and disseminated among orthopaedic trainees and surgeons in practice. Results: The submissions of 11 respondents were analyzed. Mean accuracy was calculated to be 0.70 (SD = 0.07). Sensitivity and specificity of respondents were 0.68 (SD = 0.11) and 0.71 (SD = 0.12), respectively. Positive predictive value and negative predictive value for respondents were 0.56 (SD = 0.09) and 0.82 (SD = 0.04), respectively. There was poor correlation of accuracy with years of experience with R2 calculated to be 0.0004 and poor agreement between observers with Kappa measurement of interobserver reliability of 0.46. Conclusions: Our study suggests that surgeons are unable to consistently differentiate between stable and unstable patterns on the basis of assessments based on x-ray and CT. Years of experience in training or practice was not found to be associated with improved accuracy of stability prediction.