Risk Factors for Infection After Knee Arthroplasty A Register-Based Analysis of 43,149 Cases

Background: Clinical studies have revealed a number of important risk factors for postoperative infection following total knee arthroplasty. Because of the small numbers of cases in those studies, there is a risk of obtaining false-negative results in statistical analyses. The purpose of the present study was to determine the risk factors for infection following primary and revision knee replacement in a large register-based series. Methods: A total of 43,149 primary and revision knee arthroplasties, registered in the Finnish Arthroplasty Register, were followed for a median of three years. The Finnish Arthroplasty Register and the Finnish Hospital Discharge Register were searched for surgical interventions that were performed for the treatment of deep postoperative infections. Cox regression analysis with any reoperation performed for the treatment of infection as the end point was performed to determine the risk factors for this adverse outcome. Results: Three hundred and eighty-seven reoperations were performed because of infection. Both partial and complete revision total knee arthroplasty increased the risk of infection as compared with the risk following primary knee replacement. Male patients, patients with seropositive rheumatoid arthritis or with a previous fracture around the knee, and patients with constrained and hinged prostheses had increased rates of infection after primary arthroplasty. Wound-related complications increased the risk of deep infection. The rate of septic failure was lower after unicondylar than after total condylar primary knee arthroplasty, but the difference was not significant. The combination of parenteral antibiotic prophylaxis and prosthetic fixation with antibiotic-impregnated cement protected against septic failure, especially after revision knee arthroplasty. Following revision total knee arthroplasty, diagnosis and prosthesis type had no effect, but previous revision for the treatment of infection and wound-healing problems predisposed to repeat revision for the treatment of infection. Conclusions: There was an increased risk of deep postoperative infection in male patients and in patients with rheumatoid arthritis or a fracture around the knee as the underlying diagnosis for knee replacement. The results of the present study suggest that the infection rate is similar after partial revision and complete revision total knee arthroplasties. Combining intravenous antibiotic prophylaxis with antibiotic-impregnated cement seems advisable in revision arthroplasty.