Similar to other methods of organ preservation, "spinoplegia" may protect the spinal cord from the effects of oxygen desaturation during aortic cross-clamping. In porcine experiments, spinal cord O2 saturation was studied during intraoperative localization of the blood supply to the spinal cord using hydrogen; division of arteries not supplying the spinal cord; aortic cross-clamping for 60 minutes; and 60 minutes after unclamping. In 5 animals, 120 mL of cold saline solution with lidocaine (100 mg/dL) was infused into the aorta during aortic cross-clamping. During sequential localization, O2 saturation dropped by 40.02% (standard deviation, 20.16%) for T-14 artery testing versus a decrease of 17.27% (standard deviation, 11.88%; p = 0.0075) for L-5 artery segment testing in the control animals and returned to baseline thereafter. During aortic cross-clamping maximal O2 desaturation was 5% of baseline (15.7%; p less than 0.0001), which improved slightly by 30 minutes after clamping (48% of baseline +/- 37.37%; p = 0.048 versus maximum) and then returned to baseline (97.1% of baseline +/- 41%) with unclamping; 5 minutes later, hyperoxygenation occurred with a progressive decline thereafter (68% of baseline +/- 29.3%; p = 0.025, 45 minutes after unclamping versus baseline). The decrease in spinal motor evoked potentials was significantly less (p less than 0.02) in the treated group. Intraoperative hydrogen testing in 8 patients was demonstrated to be safe. It accurately localized reattached arteries, and O2 saturation of the spinal cord fell by 56% (standard deviation, 29%; p = 0.0025) with aortic cross-clamping. We conclude that spinal cord ischemia occurs with aortic cross-clamping in both animals and humans.(ABSTRACT TRUNCATED AT 250 WORDS)