Spinal cord blood supply in patients with thoracoabdominal aortic aneurysms

Abstract

Objective: In patients with thoracoabdominal aortic aneurysms (TAAAs), the blood supply to the spinal cord is highly variable and unpredictable because of obstructed intercostal and lumbar arteries. This study was performed for the prospective documentation of patent segmental arteries during TAAA repair and the assessment of their functional contribution to the spinal cord blood supply. Methods: TAAA repair was performed in 184 consecutive patients (68 with type I aneurysm, 91 with type II, and 25 with type III) according to a protocol that included left heart bypass grafting, cerebrospinal fluid drainage, and the monitoring of motor-evoked potentials (MEPs). Patent intercostal and lumbar arteries were documented, and all reattached, selectively grafted, and oversewn segmental arteries were noted. MEP amplitude that decreased to less than 25% of baseline was considered an indication of critical spinal cord ischemia and prompted spinal cord revascularization. Results: Adequate MEP levels were encountered in 183 of 184 patients. One patient had early paraplegia (absent MEPs), two patients had delayed paraplegia develop, and two patients had temporary paraparesis, which accounted for an overall neurologic deficit of 2.7%. The median total number of patent intercostal and lumbar arteries in type I, II, and III aneurysms was three, five, and five, respectively. In eight of 68 type I cases, no segmental arteries were seen between the fifth thoracic vertebrae (T5) and the first lumbar vertebrae (L1) and MEP levels remained adequate because of distal aortic perfusion. In 18 of 91 type II cases, the aortic segment T5 to L1 did not contain patent arteries, and in six of these patients, the segment L1 to L5 did not have lumbar arteries either. In the latter patients, MEP levels depended on the pelvic circulation provided with the left heart bypass graft. In the other 12 of 91 type II cases, the only patent arteries were the lumbar arteries between L3 and L5. The loss of MEPs could be corrected with the reattachment of these arteries. In seven of 25 type III cases, the MEP levels also depended on lumbar arteries L3 to L5 and in three of 25 cases, no segmental arteries were available and MEP levels recovered after the reperfusion of the pelvic circulation. With the combination of the findings of type II and III cases, spinal cord perfusion was directed by lower lumbar arteries in 16% of the cases (19 of 116) and pelvic circulation in 8% of the cases (nine of 116). Conclusion: In patients with TAAA, most intercostal and lumbar arteries are occluded and spinal cord perfusion depends on an eminent collateral network, which includes lumbar arteries and pelvic circulation. The monitoring of MEPs is a sensitive technique for the assessment of spinal cord ischemia and the identification of segmental arteries that critically contribute to spinal cord perfusion. Surgical strategies on the basis of this technique reduced the incidence rate of neurologic deficit to less than 3%. (J Vasc Surg 2002;35:30-7.)