Sipeng Li, Hongtao Rong, Xueqin Zhang, Zhengshan Zhang, Chao Wang, Rui Tan, Yi Wang, Ting Zheng, Tao Zhu

December 2019, Volume 28, Issue 12, pp 2972 - 2980 Review Article Read Full Article 10.1007/s00586-019-06143-6

First Online: 14 September 2019

Meta-analysis of topical vancomycin powder for microbial profile in spinal surgical site infections


To systematically evaluate the impact of topical vancomycin powder for microbial profile in spinal surgical site infections.


All available literature regarding the topical use of vancomycin powder to prevent postoperative spinal infections was retrieved from the MEDLINE, EMBASE, and Cochrane databases starting from the creation date and up until September 30, 2018.


A total of 21 studies involving 15,548 patients were reviewed. The combined odds ratio showed that topical use of vancomycin powder was effective for reducing the incidence of gram-positive bacterial infections in spinal surgical sites (OR 0.41, P < 0.00001) without affecting its efficacy in the prevention of polymicrobial infections (OR 0.30, P = 0.03). Additionally, it could significantly reduce the infection rate of methicillin-resistant staphylococcus (OR 0.34, P < 0.0001). However, topical vancomycin powder showed no advantage for preventing gram-negative bacterial infections (OR 0.94, P = 0.75).


Topical administration of vancomycin powder may not increase the rates of gram-negative bacterial or polymicrobial infections in spinal surgical sites. On the contrary, it can significantly reduce the infection rates of gram-positive bacteria, methicillin-resistant staphylococcus (MRS) and microorganism. Of course, the topical vancomycin powder cannot change the rates of gram-negative bacterial infections, which may be related to the antimicrobial spectrum of vancomycin. Due to the limited number of articles included in this study, additional large-scale and high-quality studies are needed to provide more reliable clinical evidence.

Graphic abstract

[Figure not available: see fulltext.]

Read Full Article