Selection pressures of vancomycin powder use in spine surgery: a meta-analysis

Surgical site infection (SSI) is a serious and costly complication of spine surgery. Many surgeons apply vancomycin powder to the surgical wound to prevent SSIs. While multiple studies have reported reduced rates of SSI, others have suggested that widespread use of intrawound vancomycin may increase the incidence of vancomycin-resistant, gram-negative, or polymicrobial spinal infections. To systematically review the current literature on vancomycin powder in spine surgery and its impact on SSI culture profiles. Meta-analysis. We included observational studies, retrospective chart reviews, and randomized controlled trials of patients who underwent spine surgeries with and without vancomycin powder application to surgical wounds and reported SSI rates. The primary outcome was postoperative SSIs. Subgroup analyses compared rates of postoperative SSIs. We performed a comprehensive search of numerous electronic databases and conference proceedings pertaining to this topic. Our meta-analysis was conducted according to Preferred Reporting Items for Systematic Reviews and Meta-analyses guidelines. Inclusion criteria consisted of spine surgeries with or without use of vancomycin powder, case-control design, sample sizes of at least 10 in each arm, SSIs identified with bacterial cultures, and follow-up of at least 1 month. Data extracted from eligible studies included, but was not limited to, SSI rates, cultured organisms, and vancomycin powder dose. Chi-square analyses were used to assess pooled risk-estimates of intrawound vancomycin powder on reducing SSIs and selecting for gram-negative and/or polymicrobial organisms compared to controls. Pooled odds ratios, relative risks, and relative risk increase for observed outcomes were calculated. A meta-analysis was then performed with a forest plot to determine risk estimates’ heterogeneity with I2 index, Q-statistic, and p value under a fixed-effects model. Funnel plot was used to assess publication bias. None of the authors received funding or other support for this review. After reviewing nearly 400 titles and abstracts, 28 articles met inclusion criteria. They included two randomized controlled trials, one observational study, and 25 retrospective analyses. There were 412 cases of SSI (3.8%) in the control group (N=10,846) compared to 197 SSIs (2.3%) in the vancomycin powder group (N=8,456). The pooled odds ratio was 0.60 (95% confidence intervalCI 0.51–0.71, p