Medical Student Icahn School of Medicine at Mount Sinai New York, New York, United States
Introduction: External ventricular drain (EVD) placement is a life-saving procedure that can be limited by device malfunction. Despite this limitation, little data exists regarding the predictors and impacts of bedside EVD irrigation (“flushing”) performed by providers.
Methods: All patients undergoing EVD placement from October 1, 2021-September 30, 2022 at a single institution were retrospectively reviewed. Univariate analyses at a significance level of p≤0.05 were conducted to determine predictors of EVD flushing and prolonged intensive care unit (ICU) length of stay (LOS). LOS was dichotomized at the median number of days in this cohort.
Results: Out of 168 EVDs placed over 12 months, 58 (34.5%) EVDs were recorded to be flushed by the neurosurgery team and 13 (7.7%) EVDs were replaced. On univariate analysis, primary diagnosis was significantly associated with EVD flushing (flushed: 55% intracerebral hemorrhage, 31% subarachnoid hemorrhage, 2% tumor, 12% other; not flushed: 51% intracerebral hemorrhage, 16% subarachnoid hemorrhage, 14% tumor, 19% other, p=0.01). Other examined variables, including patient age (p=0.36), antiplatelet/anticoagulant status (p=0.36), tract hemorrhage (p=0.23), number of passes during EVD placement (p=0.91), catheter size (p=0.57), and ventriculitis (p>0.99) were not significantly associated with EVD irrigation. EVD flushing was significantly associated with prolonged ICU LOS (>14.58 days). 46% of EVDs in the prolonged ICU LOS cohort were flushed compared to 22% in the normal ICU LOS cohort (p=0.004).
Conclusion : A high number of patients with EVDs require bedside flushing, particularly after hemorrhagic stroke. The observed association between EVD flushing and ICU LOS is likely due to the presence of higher grade subarachnoid and intracerebral hemorrhages that require more intensive neurocritical management. Flushing was not associated with ventriculitis.
