Topographical distribution of perioperative cerebral infarction associated with transcatheter aortic valve implantation

• Published in: The American heart journal Vol. 197; pp. 113 - 123
• Main Authors: Fanning, Jonathon P., Wesley, Allan J., Walters, Darren L., Wong, Andrew A., Barnett, Adrian G., Strugnell, Wendy E., Platts, David G., Fraser, John F.
• Format: Journal Article
• Language: English
• Published: United States Elsevier Inc 01.03.2018; Elsevier Limited
• Subjects: Anesthesia; Aortic valve; Blood clots; Brain; Brain injury; Brain research; Cerebral infarction; Cerebrovascular system; Cortex; Heart; Hemispheric laterality; Hospitals; Implantation; Infarction; Injuries; Ischemia; Lesions; Magnetic resonance imaging; Medical imaging; Medical personnel; Neuroimaging; Neuropathology; Neuroprotection; NMR; Nuclear magnetic resonance; Patients; Performance evaluation; Stroke; Substantia grisea; Surgery; Thorax; Transplants & implants; Veins & arteries; Vertebrae; Australia
• ISSN: 0002-8703; 1097-6744
• DOI: 10.1016/j.ahj.2017.12.008

Transcatheter aortic valve implantation (TAVI) is associated with a high incidence of cerebrovascular injury. As these injuries are thought to be primarily embolic, neuroprotection strategies have focused on embolic protection devices. However, the topographical distribution of cerebral emboli and how this impacts on the effectiveness of these devices have not been thoroughly assessed. Here, we evaluated the anatomical characteristics of magnetic resonance imaging (MRI)–defined cerebral ischemic lesions occurring secondary to TAVI to enhance our understanding of the distribution of cardioembolic phenomena. Forty patients undergoing transfemoral TAVI with an Edwards SAPIEN-XT valve under general anesthesia were enrolled prospectively in this observational study. Participants underwent brain MRI preprocedure, and 3 ± 1 days and 6 ± 1 months postprocedure. Mean ± SD participant age was 82 ± 7 years. Patients had an intermediate to high surgical risk, with a mean Society of Thoracic Surgeons score of 6.3 ± 3.5 and EuroSCORE of 18.1 ± 10.6. Post-TAVI, there were no clinically apparent cerebrovascular events, but MRI assessments identified 83 new lesions across 19 of 31 (61%) participants, with a median ± interquartile range number and volume of 1 ± 2.8 lesions and 20 ± 190 μL per patient. By volume, 80% of the infarcts were cortical, 90% in the posterior circulation and 81% in the right hemisphere. The distribution of lesions that we detected suggests that cortical gray matter, the posterior circulation, and the right hemisphere are all particularly vulnerable to perioperative cerebrovascular injury. This finding has implications for the use of intraoperative cerebral embolic protection devices, particularly those that leave the left subclavian and, therefore, left vertebral artery unprotected. [Display omitted]