Multilevel somatosensory evoked potentials (SEPs) for spinal cord monitoring in descending thoracic and thoraco-abdominal aortic surgery

race-abdominal (11 cases) repair. An aortic dissection was found in 11 cases (acute in 6). Somatosensory evoked potentials were obtained by unilateral left and right posterior tibial nerve (PTN) stimulation at the ankle and recordings were performed on four channels: peripheral nerve, lumbar spinal, brain-stem, and cortical recordings. Our experience led to the following current strategy: the establishment of atrio(aorto)-femoral(aortic) bypass (29 cases), proximal and distal aortic cross-clamping, aortic repair with reimplantation of the culprit artery(ies) as indicated by SEP alterations. Five types of SEP alterations were defined on the basis of the neural level involved: type I (27.7% of cases) = distal spinal ischemia due to proximal aortic cross-clamping in the absence of bypass; type II (21.3%) = PTN ischemia due to left common femoral artery cross-clamping; type III (12.8%)= segmental spinal ischemia due to the exclusion of critical feeding arteries; type IV (4.3%)= ischemia in the left carotid artery territory, type V (4.3%) = global brain hypoperfusion due to systemic hypotension. Forty-five patients survived the operation and could be tested for neurological dysfunction. Three patients presented a postoperative spinal cord deficit, but this deficit was already present preoperatively in one case, so that the actual incidence of a new paraplegia in our series was 2/45 cases (4.4%). One of the two cases was clearly a delayed paraplegia with SEP alterations appearing several hours after the operation. Somatosensory evoked potentials were evaluated on the basis of their sensitivity, specificity, and impact on the surgical strategy. Regarding SEP sensitivity, we did not encounter any unexpected immediate paraplegia, but the critical factor appeared to be the duration of SEP absence due to spinal cord ischemia, which, according to the literature, should never exceed 30 min; after a longer absence, SEP return does not guarantee neurological recovery. Somatosensory evoked potential specificity was also 100%, but only 58% of the abnormalities found were actually consequent to spinal cord ischemia, the rest of the abnormalities being consequent to peripheral nerve or brain ischemia. Finally, SEP monitoring had a significant impact on surgical strategy in 19% of the cases. It is concluded that distal aortic perfusion and multilevel SEP monitoring play a significant role in preventing paraplegia in descending aorta surgery.