Beefed-Up Antibiotic Prophylaxis No Help for Post-Arthroplasty Infections

Adding vancomycin to beta-lactam prophylaxis with cefazolin did not help prevent surgical-site infections in patients undergoing arthroplasty mostly for the knees or hips, a randomized trial showed.

In over 4,000 patients without known methicillin-resistant Staphylococcus aureus (MRSA) colonization, surgical-site infections at 90 days occurred in 4.5% of those who received the combination compared with 3.5% of those who received cefazolin plus placebo (relative risk [RR] 1.28, 95% CI 0.94-1.73, P=0.11), reported Trisha N. Peel, MBBS, PhD, of Monash University in Melbourne, Australia, and co-authors in the New England Journal of Medicine.

“Not only did we find that it wasn’t beneficial, but there was some concern that there might be an associated increased risk of infection or harm,” Peel told MedPage Today.

For example, surgical-site infections in patients undergoing knee arthroplasty occurred in 5.7% of the vancomycin group versus 3.7% of the placebo group (RR 1.52, 95% CI 1.04-2.23).

“It’s quite logical to think, well, if we give more antibiotics, we’ll have better outcomes, or it will prevent these infections,” said Peel. “We found that in our cohort, if they didn’t have MRSA on their skin at the time of surgery, we don’t need to be giving everyone vancomycin — that the cefazolin prevents the infections, there’s no added benefit from using the vancomycin.”

Knee and hip surgeries are both common in the U.S., with post-surgery infections responsible for high rates of morbidity and mortality.

Current antibiotic guidelines recommend a first- or second-generation cephalosporin antimicrobial agent, like cefazolin, at the time of surgery. But because this antibiotic may not prevent cases of MRSA and methicillin-resistant Staphylococcus epidermidis infections, adding a glycopeptide antimicrobial like vancomycin has been thought to help provide broader protection. The benefit of this approach, however, was not yet clear, Peel and team noted.

Previous studies have showed that adding a glycopeptide to antimicrobial beta-lactam prophylaxis has yielded either no change in surgical-site infections or a reduced incidence. However, the studies showing a reduction used teicoplanin, not vancomycin, as the glycopeptide.

Kalpana Gupta, MD, MPH, of the VA Boston Healthcare System in West Roxbury, Massachusetts, told MedPage Today that the current study was very well done.

“A lot of places may just go ahead and use two drugs for some of these high-risk procedures, because they’re not looking for MRSA,” she said. “But they know that there’s a small percentage of patients who would have it, and this study suggests that such a strategy would really not have any benefit, and actually has a downside.”

Gupta, who was not involved in the study, also noted that vancomycin can take a long time to administer, and can also lead to adverse effects.

Limiting vancomycin use for this patient population can also have implications for antimicrobial stewardship, she said. “Giving the narrowest spectrum [of antimicrobials] that is indicated has benefits. Those benefits are not only reducing the risk of resistance down the road, but also reducing untoward effects.”

For the multicenter, double-blind Australian Surgical Antibiotic Prophylaxis (ASAP) trial, conducted from 2019 to 2022, Peel and colleagues randomized 4,113 patients from 11 Australian hospitals: 2,233 undergoing knee arthroplasty, 1,850 undergoing hip arthroplasty, and 30 undergoing shoulder arthroplasty. Only patients without known MRSA colonization were included. The primary analysis was performed in a modified intention-to-treat analysis.

Mean age of participants was 66.6 in the vancomycin group (48% men) and 67.1 in the placebo group (47.8% men).

Patients were randomized to receive 1.5 g of intravenous vancomycin or normal saline placebo within 120 minutes of incision. Both groups received prophylaxis with 2 g of cefazolin administered intravenously within 60 minutes of incision.

Adverse events occurred in 1.7% of patients each in the vancomycin and placebo groups, and included hypersensitivity reactions in 1.2% and 0.5%, respectively, and acute kidney injury in 2.1% and 3.6%, respectively.

Peel and team acknowledged that the study was limited by early closures during the COVID-19 pandemic that prevented timely enrollment. They also noted a lack of generalizability to patients with MRSA colonization, revision surgery patients at higher risk for infection from resistant Staphylococcus species, and shoulder arthroplasty patients, because of small numbers. In addition, some treating physicians (4.5%) declined to participate in the trial, which may have contributed to selection bias.