A nomogram for predicting reduction loss risk after locking plate fixation for proximal humeral fractures

Abstract Purpose The purpose of our study was to determine the risk factors for reduction loss in patients with proximal humeral fractures after locking plate fixation and establish a nomogram prediction model. Methods We retrospectively analyzed the clinical data of proximal humeral fractures patients who had been surgically treated for locking plate in our institution from January 2016 to December 2018. Perioperative information was obtained through the electronic medial record system, univariate and multivariate analyzes were performed to determine the risk factors of reduction loss, and a nomogram model was constructed to predict the risk of reduction loss. The predictive performance and consistency of the model were evaluated by the consistency coefficient (C-index) and the calibration curve, respectively. Results 115 patients were finally enrolled in our study. Multivariate analysis results showed that age, fracture classification, medial comminution, and calcar screw status were independent risk factors for reduction loss. The accuracy of the contour map for predicting transfusion risk was 0.944. Conclusion We found a correlation between reduction loss and age, fracture classification, medial comminution, and calcar screw status after locking plate fixation for proximal humeral fractures patients. Our nomogram is helpful for clinicians to identify high-risk patients, early intervention and reduce the incidence of reduction loss.