Decreasing External Ventricular Drain-Related Infection Rates with Duration-Independent, Clinically Indicated Criteria for Drain Revision: A Retrospective Study.

OBJECTIVE: To lower external ventricular drain (EVD)-related infection rates, in April 2013, our institution enacted a major protocol change, switching from routine EVD replacement every 5 days to EVD replacement only when clinically indicated. In the present study, we evaluated the effect of this change on nosocomial EVD-related infections. METHODS: We performed a retrospective cohort study to compare the EVD-related infection rates between 2 groups (group A, elective EVD replacement; group B, clinically indicated EVD replacement). We analyzed the data from 142 patients (group A, n = 43; group B, n = 99), with a total of 227 EVDs for 5 years and 3 months (1721 catheter days). RESULTS: The overall EVD-related infection rates were elevated in group A (0.14; 32% of patients) compared with group B (0.08; 8%; P = 0.001). The median hospital stay (33 vs. 24 days; P = 0.001) and neurosurgical intensive care unit stay (30.5 vs. 17 days; P < 0.0001) were also longer for group A. The requirement for multiple EVDs was an independent risk factor (P = 0.003), with a 4.6 times greater risk in group A (odds ratio, 4.64; 95% confidence interval, 1.7-12.6). CONCLUSIONS: The findings from our study strengthen an increasing body of evidence suggesting the importance of inoculation of skin flora as a critical risk factor for EVD-related infections, underscoring the importance of drain changes only when clinically indicated and that, as soon as clinically permitted, catheters should be removed.