Surgical management of periprosthetic shoulder infections.

Abstract

The treatment of periprosthetic joint infection is a difficult challenge in shoulder arthroplasty. This study investigated 1-stage modular component exchange vs. 1-stage complete removal and reimplantation (CRR) vs. 2-stage revision arthroplasty for periprosthetic joint infection. Between January 1, 2004, and December 31, 2012, 79 patients received a component exchange (n = 15), CRR (n = 45), or a 2-stage (n = 19) revision for infection. A binary logistic regression analysis was performed to determine factors presenting the greatest risk of reinfection. Complications and functional outcomes were also evaluated. Overall, 4 of 15 (27%) component exchanges, 2 of 45 (4%) CRRs, and 4 of 19 (21%) 2-stage procedures required a reoperation for infection with a minimum of 1 year of follow-up. The difference between the CRR group and exchange group was significant (P = .030); however, the difference between the CRR group and 2-stage group did not reach statistical significance (P = .059). No preoperative and intraoperative selection bias between the groups was found. Binary logistic regression predicted that reinfection was highest in patients whose cultures grew Staphylococcus aureus (P = .004) or coagulase-negative Staphylococcus species (P = .041) or those treated with a component exchange (P = .015). The difference between groups for noninfection-related complications was not significant (P = .703). All procedures provided improved functional outcomes and pain relief. Patients with infection caused by Staphylococcus aureus or coagulase-negative Staphylococcus species may require additional operations to treat the infection. Although effective in some cases, component exchange presents an increased risk for reinfection. A 1-stage CRR procedure had similar reinfection rates as a 2-stage procedure in our patient population.