Purpose: We developed and applied a method for providing regional spinal cord hypothermia with epidural cooling (EC) during thoracoabdominal aneurysm (TAA) repair. Preliminary results indicated significant reduction in spinal cord ischemic complications (SCI), compared with historical controls, and a 5-year experience with EC was reviewed.
Methods: From July 1993 to September 1998, 170 patients with thoracic aneurysms (n = 14; 8.2%) or TAAs (types I and II, n = 83 [49%]; type III, n = 66 [39%]; type IV, n = 7 [4.1%]) were treated with EC. An earlier aneurysm resection was noted in 44% of patients, an emergent operation was noted in 20% of patients, and an aortic dissection was noted in 16% of patients. The EC was successful (mean cerebrospinal fluid [CSF] temperature at cross-clamp, 26.4 +/- 3 degrees C) in 97% of cases, with all 170 patients included in an intention-to-treat analysis. The operation was performed with a clamp/sew technique (98% patients) and selective (T(9) to L(1) region) reimplantation of intercostal vessels. Clinical and EC variables were examined for association with operative mortality and SCI by means of the Fischer exact test, and those variables with a P value less than.1 were included in multivariate logistic regression analysis.
Results: The operative mortality rate was 9.5% and was weakly associated (P =.07) with SCI; postoperative cardiac complications (odds ratio [OR], 35. 3; 95% CI, 5.3 to 233; P <.001) and renal failure (OR, 32.2; 95% CI, 6.6 to 157; P <.001) were the only independent predictors of postoperative death. SCI of any severity occurred in 7% of cases (type I/II, 10 of 83 [12%]; all other types, 2 of 87 [2.3%]), versus a predicted (Acher model) incidence of 18.5% for this cohort (P =. 003). Half the deficits were minor, with good functional recovery, and devastating paraplegia occurred in three patients (2.0%). Independent correlates of SCI included types I and II TAA (OR, 8.0; 95% CI, 1.4 to 46.3; P =.021), nonelective operation (OR, 8.3, 95% CI, 1.8 to 37.7; P =.006), oversewn T(9) to L(2) intercostal vessels (OR, 6.1; 95% CI, 1.3 to 28.8; P =.023), and postoperative renal failure (OR, 23.6; 95% CI, 4.4 to 126; P <.001). These same clinical variables of nonelective operations (OR, 7.7; 95% CI, 1.4 to 41.4; P =.017), oversewn T(9) to L(2) intercostal arteries (OR, 9.7; 95% CI, 1.5 to 61.2; P =.016), and postoperative renal failure (OR, 20.8; 95% CI, 3.0 to 142.1; P =.002) were independent predictors of SCI in the subgroup analysis of high-risk patients, ie, patients with type I/II TAA.
Conclusion: EC has been effective in reducing immediate, devastating, total paraplegia after TAA repair. A strategy that combines the neuroprotective effect of regional cord hypothermia, avoiding the sacrifice of potential spinal cord blood supply, and postoperative adjuncts (eg, avoidance of hypotension, CSF drainage) appears necessary to minimize SCI after TAA repair.