As methicillin resistant Staph aureus (MRSA) infections became more prevalent, surgeons have considered different perioperative antibiotic regimens. Intravenous (IV) cefazolin has been the traditional choice for perioperative antibiotic prophylaxis, typically given within 1 hour prior to skin incision and continued for 24 hours post-operatively. However, cefazolin does not cover MRSA, so the natural consideration has been to add vancomycin. The use of intra-wound vancomycin powder has been the most publicized approach to this, with many papers suggesting intra-wound vancomycin powder significantly decreased the infection rate in patients undergoing instrumented posterior thoracic or lumbar fusion. Less has been published about the addition of IV vancomycin to the perioperative antibiotic regimen, despite it being an obvious alternative or addition to intra-wound vancomycin. The department of orthopaedic surgery at Brigham and Women's Hospital in Boston changed their policy for perioperative antibiotics for all cases involving implants in 2010, such that patients received both IV vancomycin and IV cefazolin within 1 hour prior to incision and for 24 hours after surgery. Prior to this, patients received only IV cefazolin for the same duration. Dr. Lopez and colleagues performed a pre-post analysis in which the rates of revision surgery for infection following instrumented spinal fusion were compared from 2005-2009 to 2011-2015. The cefazolin-only cohort included almost 1,300 patients and the cefazolin + vancomycin cohort almost 2,000 patients. At baseline, the cefazolin + vancomycin cohort was 4 years older, included more smokers, and fewer trauma patients. Eight percent of the cefazolin + vancomycin cohort also had documentation of intra-wound vancomycin powder application. The overall revision rate for infection decreased from 4% in the cefazolin-only cohort to 2% in the cefazolin + vancomycin cohort, and these results were essentially unchanged when controlling for baseline patient and surgical characteristics. The overall revision rate also decreased from 14% to 7% over the same period. There was no difference in the proportion of different organisms (i.e. MRSA, MSSA, others) isolated on intra-operative cultures between the two cohorts.
The authors have done a nice job looking at a natural experiment performed at their institution, and the results suggest that the addition of IV vancomycin to the typical perioperative antibiotic regimen might help reduce infection rate. The biggest threat to the validity of this study, and any pre-post study, is the potential for unmeasured secular trends acting as confounders. Secular trends represent other changes in practice over time not included in the analysis, and in this case could include the use of intra-wound vancomycin powder, changes in dressings, closure materials or irrigation techniques, or a change in the threshold to return to the operating room to address surgical site infection. While the authors did report that 8% of the cases in the cefazolin + vancomycin cohort also received intra-wound vancomycin powder, this could not be controlled for as no patients received intra-wound vancomycin in the cefazolin-only cohort. Given that the use of vancomycin powder was widely adopted from 2011-2015, the authors may have also not captured all of the cases in which it was used. The analysis would have been cleaner if anterior procedures were not included in the analyses as these procedures (especially ACDF) have much lower infection rates than posterior instrumented fusions. While the only way to convincingly demonstrate that perioperative IV vancomycin reduces surgical site infection would be an RCT, this paper demonstrates that the entire suite of infection control measures that were likely adopted in the later timeframe significantly reduced infection rates. Given that IV vancomycin was a key part of their infection control protocol, I would not change a thing if I were them. However, it is not clear if IV vancomycin should be adopted by other institutions that have seen significantly lower infection rates associated with their infection control protocols that do not include it. The use of IV vancomycin has some downsides, including the logistical challenges associated with its long duration of infusion and the potential for adverse events like red man syndrome. Adverse reactions to vancomycin seem much less common when administered as a powder in the surgical site. Individual institutions and surgeons will have to judge the available evidence to determine if it makes sense to add IV vancomycin to an infection control protocol that already includes intra-wound vancomycin.
Please read Dr. Lopez's article in the March 15 issue. Will this prompt you to add IV vancomycin to your perioperative antibiotic protocol? Let us know by leaving a comment on The Spine Blog.
Adam Pearson, MD, MS
Associate Web Editor