Predicting the risk of paraplegia after thoracic and thoracoabdominal aneurysm repair

Background: Endovascular repair of descending thoracic and thoracoabdominal aortic aneurysms is an appealing alternative to the standard surgical approach, but precludes revascularization of segmental arteries (SAs). For safer surgical and endovascular repairs, an accurate prediction of the risk of paraplegia in relation to the extent of SA sacrifice is needed. Methods: From January 1994 to October 2008, 609 patients (mean age, 63 ± 14 years) underwent surgical descending thoracic or thoracoabdominal aortic aneurysm repair without SA reimplantation. Three hundred seventy-six patients (62%) were male; 159 (26%) had urgent or emergent operation; 199 (33%) had previous aortic surgery. Somatosensory- or motor-evoked potential monitoring and cerebrospinal fluid drainage were routinely performed. Results: Hospital mortality was 10.7% (65 patients). Spinal cord injury (SCI) occurred in 3.4% (21 patients). The extent of resectionexpressed as the number of SAs sacrificed (p = 0.007)and the need for visceral artery reimplantation (p = 0.03) were independent risk factors for paraplegia. Further analysis identified four risk groups (p <0.0001): fewer than 8 SAs sacrificed (group A, SCI = 1.2%); sacrifice of 8 to 12 SAs with proximal origin in the upper thorax (group B, SCI = 3.7%); 8 to 12 SAs sacrificed beginning in the lower thorax (group C, SCI = 15.4%); and 13 or more SAs sacrificed (group D, SCI = 12.5%). This four-group model more accurately predicts SCI risk than the Crawford classification (goodness of fit c statistic: 0.748 versus 0.640). Conclusions: The extent of SA sacrifice is the most powerful predictor of paraplegia risk. For aneurysms of moderate extent, a more distal location involving the abdominal aorta increases the risk of spinal cord injury. Sacrifice of fewer than 8 SAs is associated with a very low paraplegia risk regardless of location.