Clinical meaning of unexpected positive cultures (UPC) in revision shoulder arthroplasty

Purpose This study analyzed the prevalence and clinical meaning of unexpected positive cultures (UPCs) in revision shoulder arthroplasty for causes different than infection. Methods Between 1976 and 2007, 107 consecutive patients with UPCs, no previous suspicion of infection, and fulfilling inclusion criteria were identified. Forty-five partial (PSA) and 62 total shoulder arthroplasties (TSA) with different preoperative diagnoses were reviewed. Cases were classified as true infections, possible infections, contaminants, and undetermined. Mean follow-up was 5.6 ± 5.3 years. Results The prevalence of UPC was 15%. Male sex was a risk factor for UPC. Results of preoperative blood tests and intraoperative pathology were negative in 94 and 97 cases, respectively. Most prevalent bacteria were Propionibacterium acnes (n = 68) and Staphylococcus epidermidis (n = 21). Fifty-three patients received treatment with antibiotics and 54 did not. In 11 patients, a postoperative arthroplasty infection during follow-up was diagnosed by culture and was produced by the same microorganism as the one isolated on the UPC (true infection). Risk factors for true infection vs contamination included revision of a TSA vs a PSA and the number of previous surgeries. Antibiotic use and number of positive cultures did not influence the rate of true infections. Conclusions UPCs are a prevalent condition in revision shoulder arthroplasty for causes different than infection. In at least 25% of cases, UPC had no clinical relevance. In 10% of cases, a persistent infection was demonstrated.