Summary

Although the incidence of shoulder arthroplasties is rising, we know little about periprosthetic humeral fractures. Thus, this multicenter study aims to give an insight into this rare entity.

Abstract

Shoulder arthroplasty is a wide-spread treatment option for a variety of shoulder pathologies resulting in pain reduction and improvement of shoulder motion. Thus, the number of arthroplasties has rapidly increased during the last years. One of the major complications after shoulder arthroplasty are periprosthetic humeral fractures. In the literature, the prevalence of periprosthetic humeral fractures is reported between 0.5 and 3%. The aim of this retrospective, multicenter analysis was to collect demographic data, treatment strategies, and associated complications.
Retrospective data of patients with periprosthetic humeral fractures was collected from the database of six trauma centers. A total of 64 patients was identified in a period of 16 years. Demographic parameters were collected and fracture classification was performed according to Wright and Cofield as well as Worland.
The mean age of the 48 female and 16 male patients was 76 years (52-94 years of age). Fracture occurred on average 3.6 years (1 month - 12.8 years) after primary prosthesis implantation, as a result of almost exclusively falls (n=61). 29 patients presented with a reverse total shoulder arthroplasty (RTSA), 32 had a hemiarthroplasty (HA), 2 had an anatomic stemless shoulder prosthesis, and 1 patient had an isoelastic prosthesis. Among the fractures, more than half (n=35) were Wright and Cofield type B fractures. According to Worland, the majority of patients also had a type B fracture (21 type B1, 24 type B2, 8 type B3). The main risk factors were found to be diabetes mellitus in 13 patients and osteoporosis in eight patients. A total of 25 patients were treated conservatively, 27 with plate osteosynthesis, and ten patients with revision prosthetic replacement (6 cemented, 4 cementless). In the decision of plate osteosynthesis, primary palsy of the radial nerve was the main complication in 22% (n=6). The main complication was secondary palsy of the radial nerve in 15% and non-union in 16%. Plexus palsy and arthroplasty dislocation were complications of revision prosthetic replacement. During conservative therapy, only one patient suffered from non-union.
Periprosthetic fractures of the humerus represent a rare entity. The treatment strategy must primarily be tailored to the individual patient. If primary palsy of the radial nerve is diagnosed, plate osteosynthesis with revision of the nerve should be preferred in the vast majority of cases. Regarding the complication rates, conservative therapy is to be preferred. As an alternative with few complications, revision prosthetic replacement can provide good results.