Evolving practices of spinal drain use for branch/fenestrated endovascular aortic repair patients in the United States Aortic Research Consortium.

Abstract

The use of cerebrospinal fluid drains (CSFDs) for the prevention or mitigation of spinal cord ischemia (SCI) is a subject of debate for patients undergoing branch/fenestrated endovascular aortic repair. We sought to evaluate the practices surrounding CSFD use concurrently with rates of SCI occurrence, recovery, and CSFD complications in the US Aortic Research Consortium. We conducted a retrospective analysis of the US Aortic Research Consortium registry consisting of patients undergoing branch/fenestrated endovascular aortic repair under individual physician-sponsored investigational device exemptions from January 2011 to April 2024. A composite variable consisting of SCI, major CSFD complication, or intracerebral hemorrhage was designated as the primary outcome. Patterns of CSFD use, rates of SCI and recovery, and CSFD complications were analyzed by year. Subanalyses were stratified by Crawford extent (CE) classification based on the length of aortic coverage, with high-risk patients defined as CE I to III. To further characterize the relationship between SCI and CSFD use, patient cohorts were compared by prophylactic CSFD (pCSFD) (n = 949), therapeutic CSFD (tCSFD) (n = 27), and no CSFD (n = 1609). Multivariable logistic regression was performed to determine variables associated with the composite outcome. Survival analysis was conducted using the Kaplan-Meier method. Among 2585 patients, 196 (7.6%) experienced the composite outcome, gradually declining from a maximum of 25.0% (5 patients) in 2011 to 2.9% (11 patients) in 2023. Concurrently, the use of pCSFDs declined from a maximum of 100% in 2011 to 11.0% in 2023 without any substantial increase in tCSFDs. Among high-risk patients (n = 1026), 132 (12.9%) experienced the primary composite outcome, declining from a maximum of 38.5% (5 patients) in 2013 to 3.1% (4 patients) in 2023. PCSFDs in high-risk patients, although nearly universal (92.9%-100%) until 2016, have also been on a continuous decline, reaching a minimum of 23.1% in 2024, without increase in tCSFD use (0%-5.9%). The degree of SCI resolution was not significantly different in the pCSFD (94 patients, 45.7% complete resolution), tCSFD (23 patients, 43.5% complete resolution), or no CSFD (21 patients, 42.9% complete resolution) cohorts (P = .11). Major CSFD complications did not differ between the pCSFD and tCSFD cohorts (13.4% vs 12.0%; P = .85). Multivariable logistic regression revealed CE II (odds ratio [OR], 2.35; 95% confidence interval [CI], 1.48-3.75; P < .001) and III (OR, 2.11; 95% CI, 1.31-3.40; P = .002), staging (OR, 1.66; 95% CI, 1.17-2.35; P = .004), and estimated blood loss of >350 mL (OR, 1.52; 95% CI, 1.11-2.09; P = .009) to be associated significantly with the composite outcome. Composite outcome rates have continued to improve despite declining pCSFD use. SCI recovery was not impacted favorably regardless of pCSFD, tCSFD, or no CSFD use. In the setting of comprehensive SCI prevention protocols, pCSFD use may not be warranted, even in most high-risk patients.