Reply to Letter: “Glycopeptides Versus Beta-lactams for the Prevention of Surgical Site Infections in Cardiovascular and Orthopedic Surgery
It is with great pleasure that we read the letter by Dr Dellinger regarding prevention of Staphylococcal infections. Studies aimed at infection prevention are particularly challenging to design and conduct. On one hand, prospective, randomized, and double-blinded clinical trials offer the highest quality of evidence, but they fail at providing sufficient power to detect statistical difference in relatively infrequent incidents such as surgical site infection (SSI). Population-based studies do provide a large sample size and enough statistical power, but they introduce significant bias. This is particularly problematic in observational studies when evaluating a multifaceted approach to preventing SSI. Introducing preoperative nasal screening and decolonization protocols, as well as different prophylactic antibiotic regimens, make it difficult to decide with certainty on the attributing factor to the reduction in SSI. Furthermore, specifics of each intervention can influence its efficacy, such as administration time of preoperative antibiotics or time between decolonization and time of surgery as mentioned by Dr Dellinger. These factors are hard to capture in observational studies and are likely to introduce bias. For those reasons, we decided to conduct a meta-analysis to attempt to capture high-quality evidence with enough power to evaluate SSI and causative pathogens.1
We do believe that investigating each pathogen separately is necessary in evaluating prophylactic antibiotic efficacy. Although overall SSI is the ultimate outcome, knowing the different causative species is helpful in identifying resistance patterns, and adjusting future antibiotic regimens. Furthermore, different clean surgical procedures have different microbial profile, and different surgical centers have different MRSA prevalence. Although susceptible Staphylococci SSIs are generally more frequent than resistant ones, this is certainly hospital-dependent. The large prospective clinical trial by Finkelstein et al2 comparing vancomycin with cefazolin did show relatively equal incidence of both resistant and susceptible Staphylococci, as they considered their center to have a high prevalence of MRSA. Furthermore, Patrick et al3 conducted a prospective randomized clinical trial in a low-risk population of patients undergoing vascular surgery and found MRSA to be the most common causative pathogen. Therefore, there may be a benefit from the use of vancomycin as antibiotic prophylaxis in patients undergoing cardiovascular procedures in centers with high prevalence of MRSA. Miller et al4 developed a decision analysis model to determine what MRSA prevalence might benefit from the use of vancomycin and found that vancomycin should be considered in populations with MRSA prevalence greater than 3%. Perhaps the solution to the heterogeneity found between different studies comparing antibiotic efficacy is to tailor prophylactic antibiotic regimens to the institution and surgical population.
Our study, however, did show that β-lactams were superior to glycopeptides in preventing susceptible Staphylococcal SSIs.1 This supports infection prevention guidelines that if vancomycin is to be considered, it should not be used as the sole prophylactic agent due to its inferiority against MSSA and gram negatives. We agree with Dr Dellinger that future research should compare dual antibiotic prophylaxis (vancomycin plus cefazolin) with single agent prophylaxis (cefazolin). The study by Patrick et al3 was a prospective trial of low-risk patients undergoing vascular surgery, comparing cefazolin with cefazolin and vancomycin for preoperative prophylaxis in a center with high prevalence of MRSA (60%). Interestingly, they did not find a difference in the rate of MRSA SSI between the groups, but their analysis was limited by the small sample size.
Moving forward, multicenter randomized clinical trials are needed for larger sample size. Efforts should be made to ensure a uniform infection prevention protocol that may include preoperative screening and decolonization.