Bedside external ventricular drain placement: can multiple passes be predicted on the computed tomography scan before the procedure?

OBJECTIVE

Bedside external ventricular drain (EVD) placement is less than perfect and often requires multiple passes to achieve cerebrospinal fluid flow. We conducted this prospective study to understand why multiple passes are necessary and whether this affects the incidence of hemorrhage.

METHODS

We compared the number of passes in 47 EVD placement procedures to the incidence of hemorrhage after the procedure. We also analyzed computed tomography scans before the procedure to identify variables that correlate with multiple passes.

RESULTS

Of the procedures analyzed, 72% (34/47) were single pass whereas 28% (13/47) required multiple passes. Average number of passes was 1.85 (± 1.8), but average number of passes when multiple passes were made was 4.1 (± 2.29; range, 2-9). Incidence of tract hemorrhage was 10.6% (5/47). Of those, 11.8% (4/34) were in the single-pass group and 7.7% (1/13) from the multiple-pass group. There was no statistical relationship between the number of passes and hemorrhage (P > 0.99). Subarachnoid hemorrhage, intraventricular hemorrhage, and midline shift were not found to be statistically significant in relation to the number of passes. The presence of midline rostral hematoma significantly correlated with multiple passes. One of 34 patients (2.9%) needed a single pass and 5/13 (38.5%) needed multiple passes in the presence of midline rostral hematoma (P = 0.0011). The average targeted frontal horn volume was larger in patients who needed single pass EVD (12.4 ± 6.3 cm(2) vs. 8.0 ± 4.7 cm(2); P = 0.035).

CONCLUSIONS

Multiple passes are inherent to the bedside EVD procedure, but did not increase the rate of intracranial hemorrhage.