Surgical treatment of infected shoulder arthroplasty. A systematic review

Purpose: To investigate the best surgical management of infected shoulder arthroplasty. Methods: A literature review from 1996 to 2016 identified 15 level IV studies that met inclusion criteria. Persistent infection (PI) was considered as treatment failure. Success was regarded as the absence of symptomatic PI without necessity for further treatment. Surgical outcomes were reported according to the mean weighted Constant and Murley score (CMS) for each treatment group. Results: Overall, 287 patients (146 males/141 females) were identified at a mean follow-up of 50.4 (range 32–99.6) months. The PI in the whole population was 11.5%. The pooled mean CMS, available for 218 patients, was 39 ± 13. Twenty-seven patients (9.4%) were treated with debridement (PI 29.6%, CMS 41 ± 12), 52 patients (18.1%) with resection arthroplasty (PI 11.5%, CMS 29 ± 16), 33 patients (11.5%) with permanent spacers (PI 6.1%, CMS 31 ± 14), 98 patients (34.2%) with two-stage revisions (PI 14.3%, CMS 42 ± 12) and 77 patients (26.8%) with one-stage revisions (PI 3.9%, CMS 49 ± 11). Debridement showed the highest PI rate (29.6%) and one-stage revisions reported the lowest PI rate (3.9%). Resection arthroplasty and spacers showed the poorest CMS when compared to the other procedures (p ≤ 0.0001). The debridement PI rate was significantly higher than almost any other procedure. CMS was significantly higher in patients undergoing revision compared to non-revision procedures (45 ± 12 vs. 35 ± 14) (p < 0.0001). One-stage revisions achieved significantly better results in terms of the PI rate compared to two-stage revisions (p = 0.0223), but not in terms of CMS. Conclusion: Debridement showed the highest PI rate (29.6%) and should not be recommended for the management of infected shoulder arthroplasty. Revisions reported better functional outcomes compared to non-revision procedures. The presence of a significantly lower PI rate with comparablely high mean CMS values suggests that one-stage (where technically applicable) could be superior to two-stage revisions. Unfortunately, well-designed randomized controlled trials using validated patient-based outcomes are lacking in this field. Level of evidence: Systematic Review of level IV studies, Level IV.