Slowly progressive neuronal death associated with postischemic hyperperfusion in cortical laminar necrosis after high-flow bypass for a carotid intracavernous aneurysm

The authors report a rare case of slowly progressive neuronal death associated with postischemic hyperperfusion in cortical laminar necrosis after radial artery/external carotid artery–middle cerebral artery bypass graft surgery for an intracavernous carotid artery aneurysm. Under barbiturate protection, a 69-year-old man underwent high-flow bypass surgery combined with carotid artery sacrifice for a symptomatic intracavernous aneurysm. The patient became restless postoperatively, and this restlessness peaked on postoperative Day (POD) 7. Diffusion-weighted and FLAIR MR images obtained on PODs 1 and 7 revealed subtle cortical hyperintensity in the temporal cortex subjected to temporary occlusion. On POD 13, 123I-iomazenil (123I-IMZ) SPECT clearly showed increased distribution on the early image and mildly decreased binding on the delayed image with count ratios of the affected–unaffected corresponding regions of interest of 1.23 and 0.84, respectively, suggesting postischemic hyperperfusion. This was consistent with the finding on 123I-iodoamphetamine SPECT. Of note, neuronal density in the affected cortex on the delayed 123I-IMZ image further decreased to the affected/unaffected ratio of 0.44 on POD 55 during the subacute stage when characteristic cortical hyperintensity on T1-weighted MR imaging, typical of cortical laminar necrosis, was emerging. The affected cortex showed marked atrophy 8 months after the operation despite complete neurological recovery. This report illustrates, for the first time, dynamic neuroradiological correlations between slowly progressive neuronal death shown by 123I-IMZ SPECT and cortical laminar necrosis on MR imaging in human stroke.

Reversible Focal Ischemia in the Rat: Effects of Halothane, Isoflurane, and Methohexital Anesthesia

Barbiturates and the volatile anesthetic isoflurane reduce CMR to similar values. If the mechanism of barbiturate protection against focal ischemic injury is due to a reduction in cellular energy requirements, then isoflurane should similarly reduce ischemic injury. To evaluate this, spontaneously hypertensive rats underwent 2 h of reversible middle cerebral artery occlusion (MCAO) while receiving deep methohexital, isoflurane, or halothane anesthesia. Ninety-six hours postischemia, neurologic deficits were present but without a difference between groups. Mean ± SD infarct volume, as assessed by triphenyl tetrazolium chloride staining and computerized planimetry, was significantly less in the methohexital group (n = 8; 166 ± 74 mm3) than in either the halothane (n = 9; 249 ± 71 mm3; p < 0.04) or the isoflurane (n = 9; 243 ± 62 mm3; p < 0.03) groups. One possible explanation for the lack of protective effect for isoflurane might be related to its vasodilative properties, which could result in a cerebral vascular steal. To examine this possibility, rats anesthetized with methohexital or isoflurane underwent autoradiographic determination of CBF with or without MCAO. In isoflurane-anesthetized sham rats (n = 5; no ischemia), CBF was approximately three times greater than in methohexital-treated (n = 5) sham rats. During ischemia, although a regional reduction in flow was noted in both anesthetic groups, mean flow remained greater in the isoflurane group. When the ischemic hemisphere was analyzed for percentage of cross-sectional area where flow was <25 ml/100 g/min, significantly less tissue appeared to be at risk for infarction in the isoflurane group (n = 7; 32.9 ± 19.4%) versus the methohexital group (n = 8; 49.1 ± 12.6%; p < 0.05). These results are consistent with the following conclusions: (a) CMR reduction is not a sufficient criterion for anesthetic-mediated brain protection; (b) isoflurane does not cause cerebrovascular steal; and (c) ischemic flow thresholds for infarction are different for methohexital and isoflurane.

Temporary Vessel Occlusion and Barbiturate Protection in Cerebral Aneurysm Surgery

In a review of 147 patients with intracranial aneurysms surgically treated by one surgeon (FAD) between 1980 and 1987, 36 selected patients received intraoperative barbiturate protection with sodium thiopental during temporary arterial occlusion. Thiopental doses of 5 to 15 mg/kg were used. Twenty-nine of 36 (81%) had ruptured aneurysms. Occlusion times ranged from 3 to 93 minutes, with a mean of 16.2 minutes. Seven patients had new neurological deficit in the immediate postoperative period, but in only two did these persist. Twenty-one patients (72%) with subarachnoid hemorrhage and 6 with incidental aneurysms made a good recovery. Of the 9 patients with significant permanent deficit, all but 2 were related to either the severity of the initial hemorrhage or to delayed vasospasm. In only one instance might temporary arterial occlusion have led to permanent neurological sequelae. Temporary arterial occlusion with barbiturate protection is a safe technique. For aneurysms that are more surgically complex, it allows for complete dissection of the aneurysm neck and identification and preservation of the surrounding vascular anatomy, while reducing the risk of intraoperative rupture and postoperative stroke.