Intermittent CSF drainage and rapid EVD weaning approach after subarachnoid hemorrhage: association with fewer VP shunts and shorter length of stay

Abstract

OBJECTIVE

There is variability and uncertainty about the optimal approach to the management and discontinuation of an external ventricular drain (EVD) after subarachnoid hemorrhage (SAH). Evidence from single-center randomized trials suggests that intermittent CSF drainage and rapid EVD weans are safe and associated with shorter ICU length of stay (LOS) and fewer EVD complications. However, a recent survey revealed that most neurocritical care units across the United States employ continuous CSF drainage with a gradual wean strategy. Therefore, the authors sought to determine the optimal EVD management approach at their institution.

METHODS

The authors reviewed records of 200 patients admitted to their institution from 2010 to 2016 with aneurysmal SAH requiring an EVD. In 2014, the neurocritical care unit of the authors’ institution revised the internal EVD management guidelines from a continuous CSF drainage with gradual wean approach (continuous/gradual) to an intermittent CSF drainage with rapid EVD wean approach (intermittent/rapid). The authors performed a retrospective multivariable analysis to compare outcomes before and after the guideline change.

RESULTS

The authors observed a significant reduction in ventriculoperitoneal (VP) shunt rates after changing to an intermittent CSF drainage with rapid EVD wean approach (13% intermittent/rapid vs 35% continuous/gradual, OR 0.21, p = 0.001). There was no increase in delayed VP shunt placement at 3 months (9.3% vs 8.6%, univariate p = 0.41). The intermittent/rapid EVD approach was also associated with a shorter mean EVD duration (10.2 vs 15.6 days, p < 0.001), shorter ICU LOS (14.2 vs 16.9 days, p = 0.001), shorter hospital LOS (18.2 vs 23.7 days, p < 0.0001), and lower incidence of a nonfunctioning EVD (15% vs 30%, OR 0.29, p = 0.006). The authors found no significant differences in the rates of symptomatic vasospasm (24.6% vs 20.2%, p = 0.52) or ventriculostomy-associated infections (1.3% vs 8.8%, OR 0.30, p = 0.315) between the 2 groups.

CONCLUSIONS

An intermittent CSF drainage with rapid EVD wean approach is associated with fewer VP shunt placements, fewer complications, and shorter LOS compared to a continuous CSF drainage with gradual EVD wean approach. There is a critical need for prospective multicenter studies to determine if the authors’ experience is generalizable to other centers.