Implications of the changing pattern of bacterial infections following total joint replacements.

PURPOSE To retrospectively review the causative organisms isolated from infected hip and knee replacements requiring revision. METHODS We reviewed a consecutive series of 36 patients who underwent revision hip or knee arthroplasties due to bacterial infection at Pamela Youde Nethersole Eastern Hospital in Hong Kong between 1995 and 2003. The male to female ratio was 1:2, and the mean age of patients was 70 years (range, 54-82 years). The mean duration of follow-up was 3.8 years (range, 1.1-8.3 years). No patient was lost to follow-up. All 14 revision knee patients had previously undergone cemented and patella-resurfacing total knee arthroplasties. Of the 22 revision hip patients, 9 had cementless, 6 had cemented, and 7 had hybrid total hip arthroplasties previously. RESULTS None of the bacteria isolated from 1995 to 1996 were multiple-drug resistant. Subsequently, however, most of the isolates were multiple-drug resistant, with methicillin-resistant Staphylococcus aureus (MRSA) being the most common. Half of the isolates of Staphylococcus epidermidis and Escherichia coli demonstrated multiple-drug resistance. The incidence of positive culture in revision hip patients was 59%, 46% of which were MRSA. All 13 revision hips with positive cultures showed chronic sepsis: 4 occurred within one year and 10 occurred 2 or more years after the index arthroplasty. The incidence of positive culture in revision knee patients was 57%, 46% of which were MRSA. All 8 revision knees with positive cultures showed chronic sepsis: 3 occurred within one year, 5 occurred 2 or more years after the index arthroplasty. Only one patient, who was infected with gram-negative bacilli, required a second revision for residual sepsis. Harris hip scores for the revision hip patients improved from a mean of 65 (range, 55-75) to 85 (range, 75-90). Knee Society knee scores of the revision knee patients improved from a mean of 68 (range, 55-75) to 80 (range, 70-85). There were no radiological signs suggestive of loosening in the 2 groups at the latest follow-up. CONCLUSION This study found a definite increase in multiple-drug-resistant bacteria isolated from periprosthetic infections around total hip and knee prostheses.