Perioperative testing for joint infection in patients undergoing revision total hip arthroplasty.

BACKGROUND While multiple tests are used to determine the presence of infection at the site of a total hip arthroplasty, few studies have applied a consistent algorithm to determine the utility of the various tests that are available. The purpose of the present study was to evaluate the utility of commonly available tests for determining the presence of periprosthetic infection in patients undergoing revision total hip arthroplasty. METHODS Two hundred and thirty-five consecutive total hip arthroplasties in 220 patients were evaluated by one of two surgeons using a consistent algorithm to identify infection and were treated with reoperation. Receiver-operating-characteristic curve analysis was used to determine the optimal cut-point values for the white blood-cell count and the percentage of polymorphonuclear cells of intraoperatively aspirated hip synovial fluid. Sensitivity, specificity, negative predictive value, positive predictive value, and accuracy were determined. Patients were considered to have an infection if two of three criteria were met; the three criteria were a positive intraoperative culture, gross purulence at the time of reoperation, and positive histopathological findings. RESULTS Thirty-four arthroplasties were excluded because of the presence of a draining sinus, incomplete data, or a preoperative diagnosis of inflammatory arthritis, leaving 201 total hip arthroplasties available for evaluation. Fifty-five hips were judged to be infected. No hip in a patient with a preoperative erythrocyte sedimentation rate of <30 mm/hr and a C-reactive protein level of <10 mg/dL was determined to be infected. Receiver-operating-characteristic curve analysis of the synovial fluid illustrated optimal cut-points to be >4200 white blood cells/mL for the white blood-cell count and >80% polymorphonuclear cells for the differential count. However, when combined with an elevated erythrocyte sedimentation rate and C-reactive protein level, the optimal cut-point for the synovial fluid cell count was >3000 white blood cells/mL, which yielded the highest combined sensitivity, specificity, positive predictive value, negative predictive value, and accuracy of the tests studied. DISCUSSION A synovial fluid cell count of >3000 white blood cells/mL was the most predictive perioperative testing modality in our study for determining the presence of periprosthetic infection when combined with an elevated preoperative erythrocyte sedimentation rate and C-reactive protein level in patients undergoing revision total hip arthroplasty.