Infection after joint replacement remains a potentially devastating complication. On a national scale, the Centers for Disease Control and Prevention have estimated an annual incidence of 500,000 surgical site infections (SSIs). The Surgical Care Improvement Project (SCIP) is a collaborative effort of national organizations aligned by a common goal: the improvement in surgical care by the reduction of postoperative complications. Implemented in 2005, the SCIP set a goal of reducing the incidence of surgical complications by 25% by 2010. Of a series of ten measures, three apply to the timing of antibiotic initiation (SCIP-inf-1), antibiotic selection (SCIP-inf-2), and antibiotic cessation (SCIP-inf-3). In summary, prophylactic antibiotics should be initiated within one hour of the surgical incision and discontinued within twenty-four hours of the cessation of surgery. Antibiotics should be selected on the basis of likely pathogens, and the recommendations for total hip and knee arthroplasty are for the use of cefazolin or cefuroxime, or for vancomycin or clindamycin in a patient with a beta-lactam allergy.

Ponce et al. present the findings of a retrospective review of the use of cefazolin, vancomycin, cefazolin and vancomycin, or clindamycin as antibiotic prophylaxis in 18,830 patients undergoing elective primary or revision total hip or knee arthroplasty at ninety-four hospitals in the VA (Veterans Affairs) system. The first conclusion drawn was that the use of vancomycin only led to the highest early SSI rate. The SSI rate (combining both superficial and deep infections) at thirty days was 2.3% for vancomycin compared with 1.5% with vancomycin plus cefazolin, 1.3% for cefazolin only, and 1.1% for clindamycin. Vancomycin is a narrow-spectrum antibiotic, and the finding of a higher infection rate seems logical. The use of vancomycin only for antibiotic prophylaxis may not be adequate. The second conclusion was that the antibiotic dosage, and not merely the antibiotic choice, should be specified in the SCIP guidelines. Given that a vancomycin dosage of 15 mg/kg is generally accepted as appropriate but a standard dose of 1 g (appropriate for a patient weight of 67 kg) is often given, many of the patients who received vancomycin were potentially underdosed. The study was not, however, able to address the direct association between the antibiotic dosage used and the subsequent infection rate.