The incidence of deep brain stimulator hardware infection: the effect of change in antibiotic prophylaxis regimen and review of the literature

Abstract

The complication of hardware infection related to deep brain stimulator implantation (or revision) varies between 0 and 15.2% in the literature. However, no national guidelines exist at present to define an average or acceptable rate of infection associated with, nor the preferred antibiotic prophylaxis required for, this procedure. The aim of this study was to examine the effect of changing the antibiotic prophylaxis regimen used in a single neurosurgical centre on the incidence and outcome of hardware infection. A prospective cohort of 38 patients undergoing deep brain stimulation (DBS) implantation or internal pulse generator (IPG) replacement and receiving perioperative vancomycin (including intravenous gentamicin on induction) and pouch-installed gentamicin, was compared to a historical cohort of 35 patients receiving perioperative cefuroxime in the same unit. The infection rate over 2 years in the prospective group for DBS surgery was 0 compared to 1 (5.6%) in the historical cohort (p = 0.11, χ(2)); the infection rate for IPG replacements was 1(3.6%) in the prospective cohort, versus 3 (17.6%) in the historical (p = 0.44, χ(2)). In this article, we have also systematically reviewed the literature to date and derived an average infection rate of 4.7% (PI 0.9-22%, Random Effects Meta-analysis, Stata) for 35 studies comprising 3550 patients. There is no significant difference in infection rates between DBS procedures that are primarily internalised (n = 9) compared to those in which there is a period of electrode externalisation (n = 23) (p = 0.9, Meta-regression analysis, Stata).