Two-stage revision for periprosthetic infection in shoulder arthroplasty: an institutional experience

Abstract

BackgroundPeriprosthetic joint infection involving shoulder arthroplasty is often treated with a 2-stage approach, involving resection and placement of an antibiotic cement spacer followed by staged reimplantation after eradication of the infection. While there are several studies examining the near-term clinical course and outcomes of this sequence, there is still ambiguity surrounding the diagnosis and treatment of periprosthetic joint infection, including the nature of the infection as well as treatment failure rates, and we hypothesized that trends may be observed in an institutional cohort of periprosthetic shoulder arthroplasty infections. MethodsA retrospective review of all shoulder arthroplasty cases performed at a tertiary care health system between May 2013 and September 2021 was conducted, revealing 32 patients who underwent a 2-stage treatment for periprosthetic infection, including reimplantation. Cases were reviewed for basic demographic factors, preoperative inflammatory markers, intraoperative frozen specimens and cultured organisms, and antibiotic regimen. Inpatient length of stay, discharge location, and unplanned 90-day readmissions were recorded, as well as subsequent clinical course including further post-reimplantation revisions. ResultsAll 32 cases underwent reimplantation, with a median time of 3.3 months (14.3 weeks) between stages and a post-reimplantation median follow-up of 0.6 years (minimum 1 year: 40%). Twenty-four (75%) cases required no further revision surgery, and in no cases was there a recurrence of deep infection or repeat 2-stage treatment after reimplantation. Four patients subsequently underwent aseptic revision, 3 sustained a periprosthetic fracture, and 1 underwent an irrigation and débridement for superficial cellulitis. At spacer placement, 17 cases (57%) were culture-negative infections, while 6 cases (20%) grew Cutibacterium acnes. Intraoperative frozen sections were taken in 17 cases, but only 3 (18%) were positive for acute inflammation. ConclusionTwo-stage treatment for periprosthetic infection in shoulder arthroplasty appears to have a high efficacy in preventing recurrence of deep infection, but patients should be counseled regarding its high all-cause revision rate following reimplantation. Further, many infections can be culture-negative with no intraoperative evidence of acute inflammation.