DOS 2017

· DOS Abstracts

Risk of revision or clinical failure in 2,418 patients with

stemmed hemiarthroplasty for acute proximal humeral

fracture

Alexander Amundsen, Jeppe V. Rasmussen, Bo S. Olsen, Stig Brorson

Department of Orthopaedic Surgery, Herlev Hospital

Background:

Revision rates are commonly used as primary outcome after

shoulder arthroplasty for proximal humeral fractures. However, revision rates

do not necessarily reflect the clinical outcome as some failures are never revised

due to patient or surgery related factors.

Purpose / Aim of Study:

The aim was to report revision rates, seven-year

prosthesis survival and to determine the proportion of patients that are not

revised despite an unsatisfactory clinical outcome.

Materials and Methods:

The Danish Shoulder Arthroplasty Registry was used

to obtain patients’ demographics, surgical information and one-year Western

Ontario Osteoarthritis of the Shoulder (WOOS) index on all patients who un-

derwent a stemmed hemiarthroplasty for acute proximal humeral fracture be-

tween 2006 and 2012. Revision rate was used as primary outcome and the

WOOS at one year as secondary outcome. The WOOS score was converted

to percentages of a maximum score, with 100 being the best. A WOOS score

below 30 was arbitrarily defined as a clinical failure.

Findings / Results:

Mean age was 71.9 ± 11.3 years. 1,873 (77.5%) were

women. 106 (4.4%) arthroplasties were revised. Luxation and rotator cuff fail-

ure were the most common reasons for revision. The cumulative seven-year

survival rate was 93.8 %. 154 patients died (6.4%) and 45 were revised (1.9%)

before answering WOOS, leaving 2,219 patients available for follow-up. 1,581

(71.2%) patients completed WOOS with a mean score of 54.9 ± 26.1. A WOOS

below 30 was reported by 314 (13.0%) patients.

Conclusions:

We reported a revision rate of 4.4%. However, 13.0% of patients

reported a functional outcome which was regarded as a clinical failure. This indi-

cates that a substantial number of clinical failures are not revised.

No conflicts of interest reported

28.