Functional composite spacer (antibiotic cement around a hemiarthroplasty) for the treatment of shoulder infections: minimum 5-year outcomes.

Background: Periprosthetic infections remain a challenging complication after shoulder arthroplasty, and an ideal treatment protocol has yet to be established. Two-stage revision is a common approach. Historically, the first stage entails placement of an all-cement antibiotic spacer. Although prior studies have reported on cement spacers as definitive management, persistent pain and inadequate function often lead many to later request a second-stage procedure. The functional composite spacer consists of a humeral hemiarthroplasty implant with antibiotic cement coated around the stem alone to preserve the metallic humeral head-glenoid articulation. Functional composite spacers have demonstrated improvements in function and motion with high patient satisfaction at 25 months, but longer-term follow-up is needed to better understand the role it may play in the management of shoulder infections. The purpose of this study is to evaluate outcomes at a minimum of 5 years in patients who initially planned to undergo 2-stage revision but elected to retain the functional spacer.

Methods: A retrospective review of a single institution's shoulder surgery repository from 2007 to 2018 identified 30 patients who underwent placement of a composite spacer. Overall, 5 patients underwent second-stage reimplantation and 12 patients did not have 5-year follow-up (6 lost to follow-up and 6 deceased). A total of 13 patients were included who maintained a functional composite spacer and had minimum 5-year follow-up. Patient-reported outcome measures (American Shoulder and Elbow Surgeons score, Simple Shoulder Test, Single Assessment Numeric Evaluation, visual analog scale function, and visual analog score pain), satisfaction, range of motion, and radiographic estimation of glenoid wear were evaluated.

Results: Two of 13 patients (15%) required additional surgery: 1 secondary closure for early superficial wound dehiscence and 1 revision spacer for pain. There were no reinfections. At most recent follow-up, patient satisfaction was high and significant improvements were noted for American Shoulder and Elbow Surgeons score (45.4; P < .001), Simple Shoulder Test (5.3; P = .003), Single Assessment Numeric Evaluation (47.3; P = .002), visual analog scale function (4.9; P = .004), and visual analog score pain (-4.4; P = .007) as well as range of motion including abduction (39.2°; P = .005) and elevation (65.9°; P = .005). There was no significant change in humeral head medialization (P = .11).

Conclusions: Patients who do not undergo an early revision and retain a functional composite spacer maintain good function and range of motion with minimal pain at midterm follow-up.