Angiographic localization of spinal cord blood supply and its relationship to postoperative paraplegia.

Forty-seven patients underwent selective catheterization of middle and lower thoracic intercostal and upper lumbar arteries to define the origin of the artery of Adamkiewicz. One patient had significant atheroembolism, and a second had transient lower extremity paresthesias. No other complications occurred. The origin was found in 26 (55%), and 21 patients underwent thoracoabdominal aneurysm repair with this knowledge. When the critical lumbar or intercostal artery could be included as part of a long proximal or distal anastomosis, all 12 patients could be included as part of a long proximal or distal anastomosis, all 12 patients survived, and one was paralyzed. However, if the aneurysm repair mandated a midgraft anastomosis to intercostal arteries critical to spinal cord perfusion, seven of nine patients either died or were paralyzed (p less than 0.05). In the group of 19 patients operated on in whom spinal cord blood supply was not identified three patients had a technically unsuccessful operation; two died, and one was paralyzed. Twelve of 16 patients who had an adequate, but unsuccessful attempt at localization were treated by intercostal "neglect" and survived. Late paresis developed in two patients, but they are walking now. One of the patients who died had multiple systems failure and awakened paraplegic. She had a patent, enlarged, thoracic radicular artery at T-5 which probably supplied to spinal cord and which was missed angiographically. Paralysis was associated with aneurysm extent (group 2 and III B, dissections vs group 1 & 3, p less than 0.05). Selective intercostal angiography requires further refinement, but it is safe and offers the promise of understanding the mechanisms and risks of spinal cord complications after repair of extensive thoracoabdominal aneurysms.