Background: We report our experience in treating infected shoulder arthroplasty and primary shoulder sepsis using a commercially produced antibiotic-impregnated cement spacer. Materials and methods: We treated 16 shoulders in 15 patients for infected arthroplasty or osteomyelitis of the proximal humerus with irrigation and debridement, hardware removal, or humeral head resection, or both, and placement of an interval articulating hemiarthroplasty with a commercially made gentamicin-impregnated cement spacer. Results: Mean follow-up was 20.5 months after spacer placement. At the time of debridement, 12 shoulders had positive cultures; the most common organisms were methicillin-resistant Staphylococcus aureus (n = 3) and S. epidermidis (n = 3). Twelve patients underwent revision. Four refused revision and have retained antibiotic spacers. White blood cell counts returned to within normal ranges in all patients at the time of revision, the erythrocyte sedimentation rate in 5 of 12 patients, C-reactive protein in 8 of 12 patients, and interleukin-6 in 9 of 11 patients. Mean visual analog pain scale score decreased from 8.4 before spacer placement to 0.5 at the final follow-up. Active forward flexion increased from a mean of 65 degrees to 110 degrees, and active external rotation from -5 degrees to 20 degrees. Mean University of California Los Angeles (UCLA) Shoulder Rating Scale score increased from 7 to 26, Simple Shoulder Test (SST) from 1.2 to 6.6, American Shoulder and Elbow Surgeons (ASES) Standardized Shoulder Assessment Form score from 16 to 74, and Constant score from 16 to 57. There was no recurrence of infection. Conclusions: Treatment of glenohumeral sepsis with a commercially produced antibiotic-impregnated cement spacer appears to be an effective treatment modality, and serum interleukin-6 level appears to be useful in the evaluation of shoulder infection. Level of evidence: Level IV, Case Series, Treatment Study. (C) 2010 Journal of Shoulder and Elbow Surgery Board of Trustees.