Antibiotic Impregnated External Ventricular Drains: Meta and Cost Analysis

OBJECTIVE: To determine whether antibiotic impregnated external ventricular drains (AI-EVDs) are effective in preventing ventriculostomy associated infection (VAI), and to examine their cost effectiveness. METHODS: A comprehensive literature search was performed for published data through May 2014, including randomized controlled trials and observational cohort studies comparing AI-EVDs with nonimpregnated controls. A meta-analysis of included studies was performed using a random effects model. Historical data at the authors' institution were used to estimate both the incremental price of AI-EVDs and the hospital expenses associated with VAI. RESULTS: Three randomized controlled trials and 5 observational studies met inclusion criteria. The analysis demonstrated a statistically significant protective effect of AI-EVDs against VAI (risk ratio = 0.31 [0.15-0.64]; P = 0.002), although there was significant heterogeneity (chi(2) = 18.08; P = 0.01; I-2 = 61%). The number of AI-EVDs needed to prevent one infection (Number needed to treat [NNT]) was 19. Based on $100 as the incremental price, and $30,000 as the estimated expense of one episode of VAI, AI-EVDs would result in an overall savings estimate of $28,100 (range, $26,400-$28,500) per NNT. If a hospital places 150 AI-EVDs annually, savings could range from $109,292 to $278,577 per year. CONCLUSIONS: Meta-analysis demonstrated a significant protective benefit of AI-EVDs against VAI, and this benefit is likely associated with cost savings. However, current data on AI-EVDs are limited, and overall hospital costs will vary among institutions. Although both the efficacy and cost effectiveness of AI-EVDs are supported by this analysis, further study of AI-EVDs is clearly warranted.