Background: The objective of our study was to evaluate the results of surgical treatment by reverse shoulder arthroplasty (RSA) compared with nonsurgical treatment after 2 years of follow-up in patients aged 70 years or older with displaced 3-part or 4-part proximal humeral fractures. Methods: Two groups were formed: the RSA group (n = 28) and the nonsurgical group (n = 32). Minimum follow-up was 2 years. We included patients with 3-part or 4-part fractures according to the Neer classification. Main outcome measures were the Constant-Murley score, the 11-item version of the Disabilities of the Arm, Shoulder and Hand score, and the Subjective Shoulder Value score. An autonomy score, a cognitive assessment score, and a pain score were also measured. Results: Clinical features in the 2 groups were not significantly different. The Constant-Murley mean score was significantly higher for the RSA group (82.1% vs. 76.8%; P = .03). Amplitudes were all higher in favor of the RSA group (P < .02). There was no significant difference in mean score on the 11-item version of the Disabilities of the Arm, Shoulder and Hand score, the preoperative and postoperative Subjective Shoulder Value score, the autonomy score, the cognitive assessment score, or the pain score in both groups at the last follow-up. The complication rate was higher for the RSA group. Conclusions: These results suggest that RSA treatment of proximal humeral fractures with 3-part or 4-part displaced fragments in elderly patients provides better functional outcomes than does nonsurgical treatment. However, the observed clinical difference was relatively small. This solution must be proposed only to patients who have a significant functional demand.

4.