Letter to the Editor concerning "Intrawound vancomycin to prevent infections after spine surgery: a systematic review and meta-analysis" by Evaniew N, Khan M, Drew B, Peterson D, Bhandari M, Ghert M (2014) Eur Spine J; DOI 10.1007/s00586-014-3357-0.

Abstract

We read with great interest the article by Evaniew et al. [1]. The authors conducted a systematic meta-analysis of published clinical studies investigating the use of intrawound vancomycin powder with spine surgery. Evaniew et al. identified eight observational comparative studies and one randomized control trial. The pooled analyses from the observational studies demonstrated a clinically significant risk reduction in surgical site infections (SSIs) with intrawound vancomycin. However, the randomized controlled trial failed to demonstrate a clinical difference. Evaniew et al. demonstrated a high level of clinical bias and poor quality of studies, when assessing the literature with the Methodological Index for Non-Randomized Studies (MINORS) and the Cochrane Collaboration’s Risk of Bias tool. The authors concluded that there is a lack of highquality evidence to support the use of intrawound vancomycin. The clinical heterogeneity in the included studies should be emphasized, as they may help guide future investigations. Most of the observational comparative studies investigated study populations undergoing specified surgical procedures; however, the RCT investigated all spine surgical procedures. Six of the observational studies, one of which contained the largest cohort available [2], demonstrated a significant reduction in SSIs when specifically investigating patients receiving posterior instrumentation. Martin et al. [3] demonstrated no difference with the use of vancomycin when investigating thoracolumbar and lumbar fusion for spinal deformity, a study group with an inherently higher risk of SSIs. It is unclear if the RCT reported by Tubakiet al. [4] included both anterior and posterior surgical approaches, which has been shown to result in different rates of SSIs. Further, Tubaki et al. did not evaluate risk factors such as BMI and smoking, which have been shown to contribute to infections. We agree that there is a lack of evidence to use vancomycin in all cases, but we cannot conclude that no group would benefit with this intervention. The variation in the application of vancomycin under different anti-septic protocols makes it difficult to draw firm conclusions from pooled results. While Pahys et al. [5] demonstrated a significant difference in the use of intrawound vancomycin (1.86 vs. 0 %, p = 0.048), this was not observed when the control anti-septic regimen included alcohol foams and drains (0.3 vs. 0 %). Furthermore, the definition and methodology in assessing SSI are neither uniform nor clearly identified in the studies selected in this review. The authors are highly commended for highlighting the controversial, yet informative, evidence in the unlabeled use of intrawound vancomycin for SSI prophylaxis in spine surgery. The use of vancomycin powder in patients with increased risk of surgical site infection is currently under prospective investigation (NCT01566422, NCT01977989, NCT01764750) and this study may help guide both future use as well as the development of clear clinical recommendations.