Association of preoperative spinal drain placement with spinal cord ischemia among patients undergoing thoracic and thoracoabdominal endovascular aortic repair.

ABSTRACT

OBJECTIVE: Spinal cord ischemia (SCI) is among the most devastating complications of thoracic endovascular aortic repair (TEVAR). Spinal fluid drainage has been proposed as a viable means to reduce SCI, but few data exist to support its routine use. This study investigated the association of preoperative spinal fluid drainage with the risk of SCI after TEVAR. METHODS: The Vascular Quality Initiative TEVAR module was queried for adult patients (≥18 years) undergoing TEVAR (coverage of zones 0-5) between September 2014 and March 2018 (inclusive). Patients with preoperative spinal malperfusion, aortic rupture on presentation, and connective tissue disorders were excluded. One-to-one propensity matching was used to balance patients on 44 separate dimensions by the nearest neighbor principle to compare those with vs those without preoperative spinal drainage. The primary end point was SCI still present at the time of discharge. Secondary outcomes were 30-day mortality and prolonged intensive care unit stay (>7 days). RESULTS: Among 4287 patients who underwent TEVAR (mean age, 67.1 [standard deviation, 13.7] years; 1665 [38.8%] women and 2622 [61.2%] men), 2076 had a spinal drain placed. Propensity matching yielded 1292 pairs with adequate covariate balance (all 44 absolute standardized differences <0.1). In the 2584 propensity-matched patients, spinal drain placement was associated with a reduced risk of SCI (1.5% vs 2.5%; risk-adjusted odds ratio [OR], 0.47; 95% confidence interval [CI], 0.24-0.89; P = .02). The rates of 30-day mortality (4.5% vs 5.0%; risk-adjusted OR, 0.67; 95% CI, 0.44-1.01; P = .05) and prolonged intensive care unit stay (7.0% vs 5.7%; risk-adjusted OR, 1.10; 95% CI, 0.84-1.45; P = .48) did not differ on the basis of spinal drain placement. The crossover rate was 10% (127/1292), and those with postoperative drain placement had a 20% (26/127) SCI rate on discharge. CONCLUSIONS: Among patients undergoing thoracic and thoracoabdominal endovascular aortic repair, preoperative placement of a spinal drain, compared with no drain, was associated with reduced risk of SCI. Cerebrospinal fluid drainage as a rescue measure does not provide the same protection offered by routine preoperative placement. Further investigation, including randomized controlled trials, is needed to more definitively determine the role for spinal drainage in TEVAR.