Incidence and Predictors of Neurological Complications Following Thoracic Endovascular Aneurysm Repair in the Global Registry for Endovascular Aortic Treatment

Abstract

Objectives: The aim of this study was to investigate the incidence and predictors of early and mid term neurological complications following thoracic endovascular repair (TEVAR) in the Global Registry for Endovascular Aortic Treatment (GREAT).

Methods: The GREAT is a prospective observational multicentre registry on Gore aortic endografts that was initiated in 2010. Only isolated thoracic aortic pathologies were included (aortic arch and descending thoracic aneurysms, type B dissections, penetrating ulcers, intramural haematomas, pseudoaneurysms, and transections). Thoraco-abdominal aneurysms and concomitant abdominal aneurysms were excluded. Neurological complications were classified as cerebrovascular accidents (CVA) and spinal cord injuries (SCI). Clinical, procedural, and technical data were evaluated for their association with early (30 day) and mid term CVAs and SCIs.

Results: In total, 833 patients were included: 28 with arch aneurysms (3.4%), 329 with descending thoracic aneurysms (39.5%), 273 with type B dissections (32.8%), and 203 (24.4%) with other thoracic pathologies. Altogether, 593 (71.2%) were elective procedures and 240 (28.8%) were urgent. Aortic coverage >20 cm was performed in 42.1% (n = 351); proximal landing zone 0-1-2 was adopted in 267 patients (32.1%) and of these 98 (36.7%) underwent left subclavian artery (LSA) revascularisation. There were 13 early CVAs (1.5%) and the four year freedom from CVA rate was 96.3%. On multivariable analysis, aortic arch aneurysm was the only independent predictor of early CVA (odds ratio 16.7, p = .001). LSA coverage (hazard ratio [HR] 3.31, p = .005) and hypercholesterolaemia (HR 2.96, p = .024) were independent predictors of mid term ischaemic CVAs. There were 15 (1.8%) early SCIs, and the four year freedom from SCI rate was 97.8%. No independent predictors of early SCI were identified, but length of coverage was an independent predictor of SCI at four years (HR 1.24; p = .044).

Conclusions: In this real world registry, the overall rate of neurological complication after TEVAR for isolated thoracic aortic pathologies was low. Aortic arch aneurysms were associated with increased peri-operative CVA risk. Length of coverage was an independent predictor of mid term SCIs, as LSA coverage was associated with late CVAs.

Keywords: Left subclavian artery; Length of coverage; Neurological complications; Spinal cord ischaemia; Stroke; TEVAR.