Decision-making and treatment results of complex proximal humeral fractures in geriatric patients: retrospective study from a level 1 trauma centre

Abstract

Listen

Translate article

Proximal humeral fractures are common in elderly, but despite the high incidence, optimal treatment is still discussed and remains a topic of controversy. Nonoperative treatment continuous to be the main modality. However, due to advancements in surgical technology with new techniques and implants, operative treatment could lead to better outcomes and less complications, even in older patients. Decision-making in elderly should incorporate comorbidities, activity level and patient expectations. This study was performed with the intention to find out, if there is a significant difference in treatment strategy and number of operations, in the last five years. Patients older than 65 years with proximal humeral fractures were included. Retrospective analysis of radiographic material and post-injury data was performed, from patients treated in 2015, 2019 and 2020. Last two years were also compared separately to exclude the effect of Coronavirus disease 2019 (COVID-19) pandemic. Epidemiological data assessment, fracture type and treatment strategy were analysed for corresponding years. Statistical analysis was focused on complex three-and four-part fractures. There were no statistically significant differences regarding incidence between the analysed years. Low energy fall was the mechanism of injury in majority of patients. Patients with tuberosity fractures were in average younger than patients in other groups. Although there were more computed tomography (CT) scans done in younger elderly patients, there was no significant difference in number of CTs compared to older patients (year 2015: p = 0.246; year 2019: p = 0.710, year 2020: p = 0.849). The number of operative interventions was the lowest in 2019 (p = 0.498) and the same was for the osteosynthesis using intramedullary nails (p = 0.014). Frequency of reversed shoulder arthroplasty surgeries is increasing, but the difference is not significant (p = 0.390). Both operative and nonoperative treatment result in similar range of motion (ROM) measurements (p = 0.164 for anteflexion. p = 0.163 for abduction), however the groups were not comparable regarding exact fracture types. In the analysed period of 5 years, epidemiology and treatment strategy of proximal humeral fractures did not change. Nonoperative approach remained the main treatment modality. No significant difference was noted in number of interventions or implants used, although there seemed to be an increased trend towards treatment with reverse shoulder arthroplasty (RSA) in complex fractures. A strong correlation was observed between radiographic indications for conservative treatment and actual implementation of it. However, when surgical treatment was indicated using the same radiological criteria, there were more than half of patients, who were not operated on. Radiologic indications are thus not enough for decision-making in treatment of three- and four-part fractures, and patient factors, such as comorbidities and pre-injury activity level, play a major role.