Journal of Neurosurgical Anesthesiology Vol. 9, No. 2, p. 154-158 : © 1997 Lippincott-Raven Publishers, Philadelphia Case Report Cerebral Oximetry During Circulatory Arrest for Aneurysm Surgery Satwant K. Samra and *William F. Chandler Departments of Anesthesiology and *Neurosurgery, University of Michigan Medical Center, Ann Arbor, Michigan, U.S.A. Summary: A patient underwent surgical clipping of a complex giant intracra- nial carotid aneurysm with the aid of extracorporeal circulation and complete hypothermic circulatory arrest. During the entire procedure, cerebrovascular oxygen saturation (ScO,) was spectroscopically measured. The patient expe- rienced circulatory arrest for 34 min; for 15 of the 34 min ScO, was <34% (minimum 32%). The patient tolerated the procedure without new neurological deficit, thus demonstrating that the previously suggested ‘‘critical’’ level of 35% ScO, is not absolute. Key Words: Hypothermia—Hypoxia—Ischemia— Cerebral aneurysm—Near infrared spectroscopy. Deep hypothermia with induced circulatory ar- rest (DHCA) is sometimes a necessary adjunct for management of complex intracranial aneurysms (1,2). Although significantly reduced, cerebral me- tabolism with oxygen consumption continues even during deep hypothermia (3), while oxygen delivery is completely stopped during induced circulatory arrest. This combination creates a potential for neu- ral injury and deserves close monitoring. Near in- frared spectroscopy (NIRS) is a technique that can potentially monitor changes in cerebral oxygenation and tissue oxygen utilization at the mitochondrial level. At present, there is limited clinical experience with this monitoring technique, and a critical value of cerebrovascular oxygen saturation (ScO,), below which neuronal damage takes place, is not known. We report a case which, to the best of our knowl- Address correspondence and reprint requests to Dr. S. K. Samra at Department of Anesthesiology, 16323 UH, Box 0048, 1500 East Medical Center Drive, Ann Arbor, MI 48109-0048, U.S.A. 154 edge, documents the lowest level of ScO, possible during circulatory arrest without a new neurological deficit. CASE REPORT This 43-year-old woman presented to an outside hospital after a sudden syncopal episode that oc- curred while she was coming out of the shower. Found by her husband, she was unresponsive, frothing at the mouth, and having seizure-like movements. She had several episodes of vomiting with possible aspiration, and a left hemiparesis. Movement of her left arm returned within a few hours, but weakness of her left leg persisted. A CT scan showed subarachnoid hemorrhage, and an an- giogram revealed a giant right carotid-ophthalmic artery aneurysm. She was heavily sedated and pharmacologically paralyzed. The trachea was intu- bated and muscle paralysis maintained to facilitate mechanical ventilation before transfer to the Uni- versity hospital. CEREBRAL OXIMETRY DURING CIRCULATORY ARREST 155 Examination On admission, she was heavily sedated but her sensorium improved over the next 48 h. She be- came awake, alert, and oriented. A second angio- gram again demonstrated the giant right carotid- ophthalmic artery aneurysm that filled completely and a left carotid-ophthalmic artery aneurysm that was entirely thrombosed and calcified. She did not tolerate temporary balloon occlusion of the right carotid artery and developed ischemic EEG changes and a pronator drift following <30 s of ca- rotid occlusion. Based on these findings, it was clear that carotid occlusion to trap the aneurysm was not an option. The decision was made to use profound hypothermia with induced circulatory ar- rest as an adjunct for the surgical treatment of this complex carotid-ophthalmic aneurysm. Operation Anesthesia was induced with Pentothal and main- tained with isoflurane, oxygen, fentanyl, and vecu- tonium. Depth of anesthesia was maintained to achieve continuous burst suppression on EEG. In- traoperative monitoring included a continuous mon- itoring of ScO, using a tissue infrared spectrometer (INVOS model 3100; Somanetics, Troy, MI, U.S.A.) in addition to invasive hemodynamic mon- itoring and peripheral pulse oximeter. The cerebral oxygen sensor was placed on the forehead (nonop- erated, left side) lateral to the midline and protected from ambient light with an adhesive shield. This sensor consists of a near infrared light transmitter and two detectors (placed at a distance of 30 and 40 mm from it) housed within an adhesive strip. Near infrared light at 730 and 810 nm wavelengths, se- lected for maximum tissue penetration, is reflected by the tissues in a parabolic curve. When placed 30 mm from the transmitter, the detector receives light reflected predominantly through the scalp and skull; at 40 mm, the detector receives light reflected from the scalp, skull, and a small section of the brain tissue. The computer in the oximeter sub- tracts the reflected signal of superficial structures from that of the deeper tissues, thereby calculating the oxygen saturation of blood and brain tissue. Be- cause 75% of the blood volume in tissue beds is in the venous circulation, ScO, approximates venous blood oxygen saturation. We performed a right pterional craniotomy and exposed the aneurysm. It became clear that direct clipping of this thick-walled aneurysm was not fea- sible. We cannulated the left femoral artery and vein to initiate femoro-femoral extracorporeal cir- culation to induce deep hypothermia to 17°C. Ven- tricular fibrillation developed at 23°C. Circulatory arrest was accomplished by cessation of extracor- poreal circulation. We opened the aneurysm and performed an endarterectomy of the thick athero- matous wall of the aneurysm, which made feasible the application of the clip to the now well-defined neck. On gradual rewarming, there was spontane- ous conversion of ventricular fibrillation to sinus rhythm at 28°C, and the patient was weaned off the extracorporeal circulation without problems. Ade- quate hemostasis was achieved, and the craniotomy wound was closed. The patient was on extracorpo- real circulation for 172 min and experienced 34 min of complete circulatory arrest. The total duration of anesthesia was 10 h. There were no intraoperative problems or complications. Figure 1 shows the ScO, throughout the entire surgical procedure, and Fig. 2 demonstrates details of that same saturation during extracorporeal circu- lation (bypass). Numerical values of ScO, were re- corded on a floppy disc at 1-min intervals for later analysis. Mean ScO, was 69% prebypass and 77% postbypass. During extracorporeal circulation, ScO, was 63% pre- and 57% postcirculatory arrest. During circulatory arrest, mean ScO, reading over 34 min was 39%, with the range of 32 to 60%. A gradual decrease (from 60%) in saturation was first noticed 5 min after inducing circulatory arrest. It reached a 35% of 35% value over the next 14 min and remained <35% (lowest reading 32%) for the next 15 min, at which time circulation was restored. A second drop in saturation seen in Figs. 1 and 2 between 15:22 and 15:26 (clock time) was associ- ated with a brief period of mechanical hypoventila- tion during partial cardiac bypass with sinus thythm. This drop in ScO, was accompanied by a simultaneous decrease in peripheral oxygen satura- tion and an increase in end tidal CO). Postoperative Course The patient remained on controlled ventilation overnight. The next morning, the patient was easily arousable, followed verbal commands appropri- ately, and had no new neurological deficits. Over Journal of Neurosurgical Anesthesiology, Vol. 9, No. 2, 1997 156 aAUNAARHRYNSED SESaASHOHOK @— % Cerebrovascular Saturation —P> eon a maar roe oe Oo ° 238 TT OaH AH MO OT +o Ee 8ese Ze 5 & 3 TSSSSEERS SE SERS 5 82 22 8 6 « Clock Time S. K. SAMRA AND W. F. CHANDLER Extracorporeal circulation \" FIG. 1. Cerebrovascular satura- tion during surgery. 7 17:22:40 16:41:30 59 15:02:14 15:18:39 15:34:59 18:52:21 16:08:42 16:25:0 pas > the ensuring 5 days, she developed moderately se- vere vasospasm as indicated by increased velocities on transcranial Doppler studies but never devel- oped a neurological deficit. She was moved out of intensive care on the 12th postoperative day and discharged from the hospital 16 days after surgery. At the time of discharge, her mental status was at baseline according to her family. At the follow-up 3 months after leaving the hos- pital, she was bright, alert and oriented. Strength in her left leg had improved enough to walk without a cane. DISCUSSION The use of extracorporeal circulation to induced deep hypothermia and circulatory arrest is well rec- ognized as a useful adjunct in the repair of compli- cated intracranial aneurysms. In some patients, cir- culatory arrest is the only possible option. A ‘‘safe’’ 75 FIG. 2. Cerebrovascular satura- 70 tion during extracorporeal circu- lation. Circulatory arrest was in- duced for 34 min. During this time, pulse oximeter recorded he- moglobin saturation of 100%. A subsequent decrease in ScO, (ar- row marks hypoventilation) was associated with lack of mechani- cal ventilation (during partial by- pass) and was accompanied by pulse oximeter reading of 80% 65 60 55 50 45 7 <¢— % Cerebrovascutar saturation —2> duration of circulatory arrest is difficult to define and quite variable (2,4-7). Real time, clinically use- ful determinants of adequacy of cerebral oxygen- ation are poorly defined and difficult to monitor. None of the hemodynamic and neurophysiological monitors traditionally used for intraoperative mon- itoring provide information during circulatory ar- rest. NIRS, first described by Jobsis (8), has subse- quently been used by several investigators (9-13) to monitor cerebral oxygenation. Many of these investigators have suggested that NIRS may have clinical application for monitoring the brain during neurosurgical procedures and open heart surgery. Before this monitoring technique be- comes a clinical reality, a critical value of ScO,, below which neuronal damage is likely to occur, needs to be established. Ausman and co-workers (3) attempted to establish that value in a clinical study of seven patients undergoing intracranial aneurysm clipping during which deep hypothermia with circu- Hypoventilation saturation of hemoglobin. This graph emphasizes the response of the cerebral oximeter to cerebral ischemia as well as hypoxemia. Cire. Arrest a 4 4 4 = Zonnn 3 eraes 8 BYRSE eb Fk So essoG6 4G 2 reese sera 2 FEES Freese ee eeeveesses ie} 3 R ad 3 Journal of Neurosurgical Anesthesiology, Vol. 9, No. 2, 1997 CEREBRAL OXIMETRY DURING CIRCULATORY ARREST 157 latory arrest was used as an adjuvant. Duration of circulatory arrest in their study varied between 10 and 65 min and had a negative linear correlation with ScOQ,. In five patients, the ScO, reading re- mained >35%, four had excellent outcome with re- turn to premorbid levels, and one had a good out- come with mild new left hemiparesis. In the remain- ing two patients with a grade IV and grade 0 hemorrhage and circulatory arrest periods of 45 and 60 min, ScO, fell to 30 and 34%, respectively. Out- come was poor in these two patients, with changes of diffuse cerebral ischemia confirmed postmortem. These authors thus suggested a reading of 35% as the critical value of ScO, below which neurological damage may occur. The anesthetic management and technique of hy- pothermic circulatory arrest used for our patient was similar to those used by Ausman and co- workers except that continuous burst suppression was maintained for our patient. This is not men- tioned in their study. Our patient had ScO, <35% for 15 min, although the duration of circulatory ar- rest was 34 min (compared to 65 min). Our patient responded to verbal stimuli soon after discontinua- tion of anesthesia and left the hospital 2 weeks after surgery without a new neurological deficit. We re- port this case to emphasize that although ScO, has a potential to be a useful monitoring tool, there is a need for collection of clinical data to determine a “‘critical’’ reading below which cerebral injury oc- curs. In a clinical study of ScO, in patients under- going carotid endarterectomy under regional anes- thesia (14), it was noted that carotid occlusion re- sulted in an ipsilateral ScO, drop between 6 and 33% from baseline without a change in clinical neu- rological status. These findings suggest that while ScO, changes are capable of tracking changes in cerebral circulation, a critical value for ScO, change that is associated with neurological dysfunction in humans has not yet been defined. It is also possible that a “‘critical value’’ of ScO, or a ‘‘critical change in ScO,”’ is difficult to define because neurological injury may be a result of multiple factors. ScO, may be only one of these factors. A combination of de- gree (decrease in ScO,) and duration of desaturation may determine neurological outcome. Deep hypothermia is used because of the brain protection provided by hypothermia during a period of circulatory arrest. Cerebral metabolism, how- ever, has been shown to continue even at core tem- perature <15°C in primate models (15). At temper- atures of 15 to 18°C (clinical range), oxygen extrac- tion in a primate model was 30% of the baseline. It has been proposed that during DHCA, oxygenated blood in cerebral microvascular circulation should release oxygen to move along the partial pressure gradient to mitochondria (with PO, of 3-8 mm Hg). As partial pressure of oxygen in blood drops over time due to metabolic consumption, desaturation occurs. At 20°C hemoglobin saturation of 30% will be associated with capillary PO, of 7 to 10 mm Hg, and capillary to mitochondrial gradient will be abol- ished, leading to cerebral hypoxia (15). This hy- pothesis may explain why ScO, <30% may not be seen during DHCA. A ‘‘safe’’ duration of DHCA is difficult to define. Continuous ScO, monitoring may help guide the duration and management of DHCA and retrograde cerebral perfusion techniques. Our patient with good neurological outcome had a lower ScO, (32 vs. 35%) for a shorter duration (15 min vs. 65 min) compared to the mortality reported by Ausman and co-workers. It could be argued that maintenance of anesthetic depth at the level of burst suppression in our patient might have provided added ‘‘brain protection.”’ The greater decrease in ScO, during the period of circulatory arrest, how- ever, suggests that maintenance of burst suppres- sion did not totally suppress cerebral metabolic consumption of oxygen in our patient. This finding of continued oxygen consumption by the brain even during deep hypothermia (18°C) has been previ- ously reported (2). We conclude that ScO, monitoring has great po- tential for patients undergoing surgical procedures that necessitate circulatory arrest because all other monitors of neurological integrity and oxygenation (EEG, peripheral pulse oximetry) become nonfunc- tional during these circumstances. ScO, has been shown to have an excellent correlation with jugular venous saturation and middle cerebral blood flow velocity (13). Further clinical studies using this monitoring technique are needed to define the crit- ical value of ScO, that may be able to predict neu- rological outcome either alone or in combination with other factors. REFERENCES 1. Silverberg GD, Reitz BA, Ream AK, et al. Hypothermia and cardiac arrest in the treatment of giant aneurysms of the Journal of Neurosurgical Anesthesiology, Vol. 9, No. 2, 1997 158 fad S.K. SAMRA AND W. F. CHANDLER cerebral circulation and hemangioblastoma of the medulla. J Neurosurg 1981;55:337-46. . Spetzler RF, Medley MN, Rigamanti D, et al. Aneurysms of the basilar artery treated with circulatory arrest, hypother- mic and barbiturate cerebral protection. J Neurosurg 1988; 68:868-79. . Ausman J, McCormick P, Stewart M, et al. Cerebral oxygen metabolism during hypothermic circulatory arrest in hu- mans. J Neurosurg 1993;79:810-15. . Astrup J, Skorstedt P, Gerris F, et al. Increase in extracel- lular potassium in brain during circulatory arrest: effects of hypothermia, lidocaine and thiopental. Anesthesiology 1981; 55:256-62. . Bahmgartner WA, Silverberg GD, Ream AK, et al. Reprisal of cardiopulmonary bypass with deep hypothermia and cir- culatory arrest for complex neurosurgical operations. Sur- gery 1983;94:242-9, Drake CG, Barr HKW, Cotes JC, et al. The use of extracor- poreal circulation and profound hypothermia in the treat- ment of ruptured intracranial aneurysm. J Neurosurg 1964; 21:575-83. Tanaka J, Shiki K, Ason T, et al. Cerebral autoregulation during deep hypothermic non-pulsatile cardiopulmonary by- pass with selective cerebral perfusion in dogs. J Thorac Car- diovasc Surg 1988;95:124-32. Jobsis F. Non invasive, infrared monitoring of cerebral and Journal of Neurosurgical Anesthesiology, Vol. 9, No. 2, 1997 10. myocardial oxygen sufficiency and circulatory parameters. Science 1977;198:1264-7. . Brazy JE, Lewis DV, Mitnick MH, et al. Non invasive mon- M1, 12 13. 14, 15. itoring of cerebral oxygenation in preterm infants. Pediatrics 1985 ;75:217-25. McCormick PW, Stewart M, Goetting MC, et al. Non inva- sive cerebral optical spectroscopy for monitoring cerebral oxygen delivery and hemodynamics. Crit Care Med 1991; 19:89-97. McCormick PW, Stewart M, Goetting MG, et al. Regional cerebrovascular oxygen saturation measured by optical spectroscopy in humans. Stroke 1991;22:596-602. Shenaque S, Shankar P, Safi H, et al. Monitoring cerebral oxygenation during hypothermic circulatory arrest using near infrared spectroscopy [Abstract]. Anesth Analg 1994; 78:390. William IM, Picton A, Farrel A, et al. Light reflective cere- bral oximetry and jugular bulb venous oxygen saturation during carotid endarterectomy. Brit J Surg 1994;81:1291-5. Samra S, Dorje P, Zelenock GB, Stanley JC. Cerebral ox- imetry in patients undergoing carotid endarterectomy under regional anesthesia. Stroke 1996;27:49-55. McCormick PW, Stewart MC, Lewis GD, Zabramski JM. Measurement of human hypothermic cerebral oxygen me- tabolism by transmission spectroscopy; optical imaging of brain function and metabolism. Dimagl U, ed. New York: Plenum, 1993:33-42.