Profound hypothermia protects . cerebral function during total circulatory arrest(TCA) in the surgical treatment of a variety of cardiac and aortic diseases. Despite its importance, there is no ideal technique to monitor the brain injury from ischemia. Since 1994, we have developed compressed spectral array(CSA) of electroencephalography(EEG) and monitored cerebral activity to reduce ischemic injury. The purposes of this study are to analyse the efficacy of CSA and to establish objective criteria to consistently identify the safe level of temperature and arrest time. We studied 6 patients with aortic dissection(AD, n=3) or aortic arch aneurysm(n=3, ruptured in 2). Body temperatures from rectum and esophagus and the EEG were monitored continuously during cooling and rewarming period. TCA with cerebral ischemia was performed in 3 patients and TCA with selective cerebral perfusion was performed in 3 patients. Total ischemic time was 30, 36 and 56 minutes respectively for TCA group and selective perfusion time was 41, 56 and 92 minutes respectively for selective perfusion group. The rectal temperatures for flat EEG were between 16.1 and 22. $1^{circ}C$ (mean: 18.4 $pm$ 2.0): the esophageal temperatures between 12.7 and $16.4^{circ}C$ (mean $14.7pm1.6).$ The temperatures at which EEG reappeared $5~15.4^{circ}C$ for esophagus. There was no neurological defic t and no surgical mortality in this series. In summary, the electrical cerebral activity Teappeared within 23 minutes at the temperature less than $16^{circ}C$ for rectum. It seemed that $15^{circ}C$ of esophageal temperature was not safe for 20 minutes of TCA and continuous monitoring the EEG with CSA to identify the electrocerebral silence was useful.