Predictors of Mortality in Head-Preserving Treatment for Dislocated Proximal Humerus Fractures: A Retrospective Analysis of 522 Cases with a Minimum Follow-Up of 5 Years

Abstract

Purpose: Proximal humerus fractures (PHFs) are among the most common fractures in elderly patients, but there is still inadequate knowledge about mortality risk factors after such injuries. In order to provide the best possible therapy, individual risk factors have to be considered and evaluated thoroughly. There is still controversy regarding treatment decisions for proximal humerus fractures, particularly for the elderly. Methods: In this study, patient data from 522 patients with proximal humerus fractures were obtained from 2004 to 2014 at a Level 1 trauma centre. After a minimum follow-up of 5 years, the mortality rate was assessed, and independent risk factors were evaluated. Results: A total of 383 patients (out of 522) were included in this study. For our patient collective, the mean follow-up was at 10.5 ± 3.2 years. The overall mortality rate was 43.8% in our respondent group and was not significantly impacted by concomitant injuries. The binary logistic regression model showed an increased risk for mortality by 10% per life year, a 3.9 times higher mortality risk for men and a 3.4 times higher risk for conservative treatment. The most powerful predictor was a Charlson Comorbidity Index of more than 2, with a 20 times higher mortality risk. Conclusions: Outstanding independent predictors of death in our patient collective were serious comorbidities, male patients, and conservative treatment. This patient-related information should influence the process of decision making for the individual treatment of patients with PHFs.