The management of infected elbow arthroplasty by two-stage revision.

BACKGROUND: Deep prosthetic infection is a potentially devastating complication after total elbow arthroplasty, with an incidence of up to 12%. This study examined the demographics, microbiologic profile, and outcomes of infected total elbow arthroplasty treated with 2-stage revision in a tertiary referral unit. METHODS: We identified 19 consecutive patients (mean age, 65 years) undergoing revision arthroplasty for deep prosthetic infection. All patients underwent a first-stage procedure with removal of implants, débridement, and insertion of an antibiotic-loaded cement spacer, followed by at least 6 weeks of intravenous antibiotics. Fourteen patients required a second-stage revision. RESULTS: Five patients did not undergo a second-stage procedure because of patient choice (n = 2), medical or surgical risk factors (n = 2), and death from an unrelated cause (n = 1). Of the 19 patients undergoing a first-stage procedure, 16 (84%) remained infection free, and 11 of the 14 patients (79%) undergoing reimplantation of an elbow prosthesis remained infection free. Six patients required further surgery (3 for recurrent infection, 3 for noninfective indications). The commonest infecting organism was Staphylococcus aureus (47%). A degree of postoperative ulnar nerve dysfunction occurred in 37% of patients, but all resolved fully without further treatment. CONCLUSIONS: Management of prosthetic joint infection using 2-stage revision can result in high rates of eradication, although rates of reoperation and transient ulnar nerve dysfunction are high.

An experimental glenoid rim strain analysis for an improved reverse anatomy shoulder implant fixation.

Loosening of glenoid components in TSA is a main cause of failure. In reverse anatomy TSA designs used for unstable joints, fixation is particularly demanding. Strains developed around the glenoid rim of biomechanical sawbone scapulae implanted with (a) the original fixed-fulcrum Bayley-Walker glenoid prosthesis in current clinical use, and (b) a revised version with conical cross-section, were compared. The conical shape of the revised design was hypothesized to produce greater strains in the glenoid rim than the original tapered screw design. The 2D strain field at three accessible locations around the rim of each scapula was measured with three-element rosette strain gauges for two types of simulated cancellous bone fill under applied physiologically relevant loads. The average strain energy densities around the rim for the conical design were greater than for the original design by a factor of 1.55-2.25 for all loading conditions. Results indicate that a significantly greater proportion of load was directed toward cortical bone in the conical design, thus promoting cortical bone loading.