Home SVCC Area: English - Español - Português
Aortic Arch Repair Using the Hypothermic
Circulatory Arrest Technique
Takashi Hirotani, MD
Cardiovascular Surgery, Tokyo Saiseikai
Minato-ku, Tokyo, Japan
The hypothermic circulatory arrest technique is currently the standard method for aortic arch repair. However, some patients require an extended period of circulatory arrest beyond the safe threshold for the repair of aortic arch lesions due to either anatomical abnormalities or complex lesions. Therefore, the most important and urgent problem to be resolved is extension of the limited time allowed for repairing aortic arch lesions, incorporating such recently introduced methods as the cerebral protection method using brain protective agents, namely "cerebroplegia" and the retrograde cerebral perfusion technique. Since 1991, we have been using pharmacological mixtures of thiopental sodium, nicardipine and mannitol as prophylactic agents to protect the brain against ischemia in all patients requiring the circulatory arrest technique for aortic arch repairs. In Japan, the selective cerebral perfusion technique and the retrograde cerebral perfusion technique are usually used since limits of the time allowed to perform aortic arch repair are usually anticipated as not so severe. However, we believe that the hypothermic circulatory arrest technique associated with pharmacological brain protection is a more appropriate technique for aortic arch repairs than either the selective cerebral perfusion technique or the retrograde cerebral perfusion technique since it seems that our method is simpler and provides a better exposure of the operative fields. In addition, when the hypothermic circulatory arrest technique is used, the aortic arch and the brachiocephalic vessels do not necessarily have to be manipulated during surgery.
Herein, the details of our method and its results are described.
From November 1991 to February 2001, 88 consecutive patients underwent aortic arch repair using the hypothermic circulatory arrest technique associated with pharmacological brain protection. The clinical records of these patients were retrospectively reviewed.
The patients, 66 % of whom were male, ranged in age from 21 to 83 years. In 49 patients (56%), the operation was performed on an emergency basis; 43 for an acute type A aortic dissection and 6 for ruptured aortic aneurysms. The aortic arch lesions consisted of 61 aortic dissections, 21 atherosclerotic aneurysms, and 7 pseudoaneurysms.
For all patients, electroencephalogram (EEG) electrodes were attached as usual and a catheter for drawing blood was inserted into the bulbus jugularis through the internal jugular vein to enable monitoring of the cerebral metabolic status during cardiopulmonary bypass (CPB).
Surgery was performed through a median sternotomy in 66 patients, and through a posterolateral thoracotomy (the 4th or 5th intercostal space depending on the distal extent of the aneurysm) in 20 patients, and through a thoraco-retroperitoneal incision in 2 patients. When the patient had an aortic dissection, both the femoral artery, which had the stronger pulsation, and the right axillar artery were cannulated for arterial access to avoid malperfusion caused by retrograde perfusion through the femoral artery. A venous drainage cannula was placed through the right atrium when a median sternotomy was chosen for the approach or through the main pulmonary artery when the approach was via a left thoracotomy or was thoraco-retroperitoneal approach. In all patients, The left side of the heart was vented through the left atrium. All patients were placed on CPB and cooled to achieve profound hypothermia. The whole body was systemically cooled by using a heat exchanger in a pump-oxygenator. To regulate the blood pH during CPB, the alpha-stat strategy was used. Blood was sampled through the catheter in the bulbus jugularis of the internal jugular vein every 5 min to monitor the partial pressure of oxygen in the internal jugular vein (Pjo2) during CPB. In 56 patients who required repairs of both the ascending and transverse aorta, the ascending aorta was first cross-clamped just proximal to the origin of the innominate artery, and the ascending aortic repair, involving an aortic valve repair or replacement, was performed while the systemic cooling was progressed. Systemic cooling was continued until the nasopharyngeal temperature reached 2° to 3° below that at which attendant neurologists confirmed the total disappearance of EEG activity. After a constant temperature was attained and the Pjo2 had also stabilized, 15 (in 30 patients) or 30 mg/kg (in 58 patients) of thiopental sodium, 20 mg of nicardipine and 300 ml of mannitol were infused into the venous reservoir of the CPB circuit. Circulatory arrest was established at 5 min. after the infusion of these pharmacological agents.
After the patients had been placed in the Trendelenburg position, the aortic arch was opened. The aortic arch was repaired during circulatory arrest. Usually, a prosthetic graft with four limbs consisting of 3 limbs to reconstruct the brachiocephalic vessels and 1 limb to establish CPB arterial access after aortic arch repair, was used to reconstruct the aortic arch. Subsequently, the arch graft was cross-clamped at the preinnominate area and then CPB was resumed after aortic arch reconstruction was completed using the CPB arterial access limb to establish an antegrade flow. The patients were not rewarmed until the suppressed Pjo2 levels started to reverse and the circulatory arrest was deemed to have been sufficiently tolerated. The prosthesis was then secured to the proximal aorta during rewarming. CPB was terminated when normothermia had been achieved. The extent of the replacement is shown in . Thirteen patients underwent a resuspension of the aortic valve, 11 patients underwent concomitant aortic valve and ascending aorta replacement, and 5 patients underwent coronary artery bypass grafting.
Quantitative variables that approximated a normal distribution were presented as the means ± standard deviation. Nominal variables were analyzed nonparametrically using the chi-squared test. The correlation between the incidence of neurological complications and the duration of circulatory arrest was analyzed using the Spearman rank correlation test. The predictors of postoperative stroke were determined by the multivariate logistic regression analysis. The relationship between the duration of circulatory arrest and the decrease in Pjo2 during circulatory arrest was determined by a linear regression analysis. Significance was defined as p < 0.05.
The duration of circulatory arrest ranged from 16 to 84 min. (mean 43.0 ± 15.4 min.). There were 46 patients whose circulatory arrest duration exceeded 45 min. and among them, 12 patients had a circulatory arrest duration of over 60 min. due to the fact that they had either complicated lesions or anatomical abnormalities. The nasopharyngeal temperature at which circulatory arrest was established ranged from 10° to 22° (mean 14.1 ± 2.0).
Operative mortality was 11.4 % (10/88). The causes of death are listed in . Among these 10 patients, 4 patients were strictly evaluated as having no neurological complications postoperatively by attendant neurologists. Four patients died either during or soon after the operation and therefore were not able to undergo a neurological evaluation. Two patients died of stroke. The duration of circulatory arrest in these 2 patients was 22 and 46 min, respectively.
All patients except for the 4 patients who died either during or soon after the operation were able to undergo a neurological evaluation. All of them underwent computed tomography (CT) and careful postoperative neurological examinations conducted by attendant neurologists. Six patients (7.1%) had neurological complications of whom 3 suffered permanent neural deficits (3.6%) and 2 of these 3 patients died of stroke. The 3 remaining patients had transient neural deficits and recovered completely prior to discharge. There was no significant difference (p=0.09) in the rate of neurologic complications between low- (15 mg/kg) and high-dose (30 mg/kg) thiopental groups; 4/28 (14.3%) in the low-dose thiopental group and 2/56 (3.6%) in the high-dose group
The relationship between the incidence of neurological complications and the duration of circulatory arrest is shown in (). According to the Spearman rank correlation test, no correlation between the incidence of neurological complications and the duration of circulatory arrest was found (p=0.46).
During CPB, Pjo2 increased to 406±197 mmHg as core cooling progressed, then increased further to 481±189 mmHg after cerebroplegic infusion. There was a significant difference (p<0.0001) between the Pjo2 values before and after cerebroplegic infusion.
The cerebral metabolic rate during circulatory arrest is considered to correlate with the rate of decrease in Pjo2 since the brain mainly consumes the dissolved form of oxygen under profound hypothermia. To evaluate the ability of our cerebroplegia to reduce the cerebral metabolic rate, the differences in the rate of decrease in Pjo2 according to the dose of thiopental during circulatory arrest were analyzed. The correlation between the duration of circulatory arrest and the decrease in Pjo2 during circulatory arrest, as determined by a linear regression analysis, was significant. The slopes of the correlation lines were 7.29 for the high-dose thiopental group and 9.23 for the low-dose group ( ). This shows a tendency toward a decreasing slope of the correlation line with increasing thiopental dose, suggesting that during circulatory arrest the cerebral metabolic rate was lower in the high-dose than in the low-dose thiopental group, although the difference was not statistically significant according to the multiple regression analysis.
Multivariate logistic regression analysis
To identify the independent risk factors of postoperative neurological complications, a multivariate logistic regression analysis was used. The variables included age, sex, the duration of circulatory arrest, nasopharyngeal and rectal temperature at circulatory arrest, dose of thiopental (low or high), the extent of graft replacement, the method used to reconstruct the brachiocephalic vessels (individual or en bloc reconstruction), the etiology of the aortic disease, the surgical approach (median sternotomy or posterolateral thoracotomy), urgency of the operation, any history of prior stroke, and concomitant procedures (coronary artery bypass grafting, aortic valve replacement, aortic root replacement with coronary arteries re-implantation). As a result, each variable of the posterolateral thoracotomy approach and concomitant coronary artery bypass grafting was determined to be an independent risk factor.
Surgeons should naturally make all possible efforts to shorten the duration of circulatory arrest. However, complicated surgery frequently requires longer periods of circulatory arrest, thus necessitating increased protection against cerebral ischemia. We have used cerebroplegia consisting of thiopental sodium, nicardipine and mannitol to reduce the cerebral metabolism and to extend the limits of the time allowed for circulatory arrest before cerebral ischemic damages occur.
The safe period for cerebral ischemia is generally considered to be 45 min or less when profound hypothermia alone is used [1,2]. According to the study of a large series by Svensson and colleagues (1), an increased stroke rate was evident after 40 min. of circulatory arrest, and the stroke rate after 45 to 59 min. of circulatory arrest was 10.7 % while that after 60 to 120 min. of circulatory arrest was 14.6 %. On the other hand in our series, the incidence of postoperative stroke was 7.1 % overall and 6.5 % in patients in whom circulatory arrest duration was over 45 min. Many other authors have also reported the safe limit of circulatory arrest to be from 45 to 60 min. [3,4]. Nevertheless, in our study, the duration of circulatory arrest was not found to be related to the incidence of neurological complications, and according to a multivariate logistic regression analysis, the duration of circulatory arrest was not determined to be an independent predictor of the postoperative stroke. The only difference between these other reports and ours is the use of brain protecting agents. Our pharmacological combination of thiopental, nicardipine and mannitol therefore appears to extend the time allowable for aortic arch repairs.
Thiopental is believed to have protective effects against global cerebral ischemia, and some reports have been described, e.g., decreased neuropsychiatric complications in open ventricle operations (5). It has been also described that thiopental decreases the cerebral metabolic oxygen consumption (CMRO2) by 45% at normothermia and by 80% under hypothermia at 18° (6). In addition, thiopental is also indicated to decrease the accumulation of free radicals during ischemia (7). Furthermore, a high level of thiopental was demonstrated by Rung et al to reduce cerebral metabolic requirements to the lowest possible level during circulatory arrest in infants (8).
In addition to thiopental, we used nicardipine and mannitol to protect the brain against ischemic injury. Nicardipine is one of the well-known calcium antagonists. Cerebral ischemia causes a rapid shift of Ca++ from the extracellular spaces into cells. Nicardipine directly reduces Ca++ entry into ischemic cells (9)and interrupts the postischemic cerebral hypoperfusion phenomenon (10) which exacerbates cerebral damage. Mannitol is well known to reduce cerebral edema after ischemia. Mannitol can also scavenge free radicals (11) and thus reduce the degree of tissue damage caused by superoxide radicals. We have used nicardipine and mannitol as brain protective agents prior to circulatory arrest. However, these drugs have not been demonstrated to be effective under profound hypothermia by any comparative studies. Further studies are planned to prove these drugs are effective in protecting the brain against ischemic injury when these drugs are administered prior to circulatory arrest.
The partial pressure of oxygen in the internal jugular vein (Pjo2) and EEG were chosen for monitoring cerebral metabolism. The level of hypothermia used in this study ranged from 10° to 22°. At this level, the dissociation curve of hemoglobin is shifted toward the left. The oxygen delivered to the tissue is, therefore, mainly transported in the dissolved form, so the oxygen saturation of hemoglobin in the internal jugular vein cannot be used to monitor cerebral metabolism. Instead we used Pjo2, which correlates directly with dissolved oxygen as an indicator of cerebral metabolism, and controlled the speed of rewarming to keep the value above the control value before establishment of CPB. We also used these parameters to evaluate the protective effects of cerebroplegia against cerebral ischemia. According to the significant increase of Pjo2 after cerebroplegic infusion from 406±197 to 481±189 mmHg (p<0.0001), cerebral metabolism was considered to be reduced by the administration of pharmacological cerebroplegia. In addition, in a previous study we demonstrated that a higher dose (30 mg/kg) of thiopental reduced the cerebral metabolic rate to a greater extent than a lower dose (15 mg/kg) of thiopental even during profound hypothermic CPB (12), although the present study could only suggest that the cerebral metabolic rate tended to be lower in the high-dose compared with in the low-dose thiopental group.
The retrograde cerebral perfusion technique was developed by Dr. Yuichi Ueda and colleagues in 1988 (13). This technique appears to be an adjunct of the hypothermic circulatory arrest technique which is used to extend the safe limit of circulatory arrest. This technique has been shown to prevent the brain from being re-warmed during circulatory arrest. It remains controversial, however, as to whether or not the brain can be perfused by this technique.
Furthermore, it remains uncertain as to whether or not particulate debris or noxious metabolites can be washed out even though air is known to be washed out easily. There is also some concern about the existence of component valves and multiple venovenous anastomoses between the deep and superficial veins of the head. Our method of pharmacological brain protection appears to be a simpler way than the retrograde cerebral perfusion technique to extend the safe limits of hypothermic circulatory arrest.
The selective cerebral perfusion technique is believed to provide a longer safe period than the hypothermic circulatory arrest technique [14,15]. In addition, since profound hypothermia is not necessarily required, severe coagulopathy is rare and the cardiopulmonary bypass time is usually shorter than that needed for the hypothermic circulatory arrest technique. However, this technique is not simple to perform and therefore is not suitable for emergency operations and also does not provide as good an exposure of the operative fields as the hypothermic circulatory arrest technique. In addition, the optimal cerebral circulatory flow has yet to be elucidated. In particular, when the body temperature changes, the question remains as to what degree we should change the selective cerebral circulatory flow. Another important problem is that when using the selective cerebral perfusion technique, catheters have to be introduced into the brachiocephalic vessels. As a result, intimal injury and cerebral embolism occur more frequently than with the hypothermic circulatory arrest technique.
We believe that our hypothermic circulatory arrest technique associated with pharmacological brain protection is a more appropriate technique for aortic arch repairs than either the retrograde cerebral perfusion technique or the selective cerebral perfusion technique since it seems that our method is simpler and provides a better exposure of the operative fields. In addition, when the hypothermic circulatory arrest technique is used, the aortic arch and the brachiocephalic vessels do not necessarily have to be manipulated during surgery.
On the other hand, according to our multivariate logistic regression analysis, each variable of the posterolateral approach and concomitant coronary artery bypass grafting was determined to be an independent predictor of the postoperative stroke. When the posterolateral approach is chosen, the orifices of the brachiocephalic vessels are located at the bottom of the incised aorta and atherosclerotic debris may easily fall into these orifices. In addition, under the lateral decubitus position, de-airing the ascending and transverse aorta is difficult. We connect the vent to the arterial line and flush blood through the left ventricle and the proximal aorta to remove air. Nevertheless, air embolism may occur more frequently in a posterolateral approach than in an anterior approach. Concomitant coronary artery bypass grafting is required when severe atherosclerosis exists in the coronary arteries, and in such patients, atherosclerosis of the aorta also usually tends to be severe. As a result, it was considered that concomitant coronary artery bypass grafting was determined to be an independent risk factor of postoperative stroke.
Our experience is still insufficient to draw any definitive conclusions. However, about a half of all subjects (52%) in this study required over 45 min. of circulatory arrest and the majority (93.5%) of them did not have any neurological complications. This good result may be attributed to our pharmacologial combination of thiopental, nicardipine and mannitol. The ultimate objective of our study is the development of a cerebroplegia which can provide sufficient time for all cardiovascular surgeons even when repairing aortic arch lesions, no matter how complicated.
1. Svensson LG, Crawford ES, Hess KR, Coselli JS, Raskin S, Shenaq SA, Safi HJ.: Deep hypothermia with circulatory arrest. Determinants of stroke and early mortality in 656 patients. J Thorac Cardiovasc Surg 1993;106:19-31
2. Treasure T, Naftel OC, Conger KA, Garcia JH, Kirklin JW, Blackstone EH The effect of hypothermic circulatory arrest time on cerebral function, morphology, and biochemistry. J Thorac Cardiovasc Surg 1983;86:761-770
3. Ergin MA, Galla JD, Lansman SL, Quintana C, Bodian C, Griepp RB Hypothermic circulatory arrest in operations on thoracic aorta. Determinants of operative mortality and neurologic outcome. J Thorac Cardiovasc Surg 1994;107: 788-799
4. Griepp EB, Griepp RB Cerebral consequences of hypothermic circulatory arrest in adults. J Cardiac Surg 1992;7:134-155
5. Nussmeier NA, Arlund C, Slogoff S. Neuropsychiatric complications after cardiopulmonary bypass: cerebral protection by a barbiturate. Anesthesiology 1986; 64: 165-170
6. Tan PSK. The anesthetic management of circulatory arrest. Br. J. Hospital Med 1990; 43: 36-44
7. Majewska MD, Strosznajder J, Lazarewicz J. Effect of ischemic anoxia and barbiturate anesthesia on free radical oxidation of mitochondrial phospholipids. Brain Res 1978; 158: 423-434
8. Rung GW, Wickey GS, Myers JL,Salus JE, Hensley FA, Martin DE Thiopental as an adjunct to hypothermia for EEG suppression in infants prior to circulatory arrest. J Cardiothorac Vasc Anesth 1991;5:337-342
9. Hadani M, Young W, Flamm ES Nicardipine reduces calcium accumulation and electrolyte derangements in regional cerebral ischemia in rats. Stroke 1998;19:1125-1132
10. White BC, Winegar CD, Wilson RF, Krause GS Calcium blockers in cerebral resuscitation. J Trauma 1983;23:788-794
11. Mizoi K,Imaizumi S,Yoshimoto T Development of new cerebral protective agents: the free radical scavengers. Neurological Res 1986;8:75-80
12. Hirotani T, Kameda T, Kumamoto T, Shirota S, Yamano M Protective effect of thiopental against cerebral ischemia during circulatory arrest. Thorac Cardiovasc Surg 1999;47:223-228
13. Ueda Y, Miki S, Kusuhara K, Okita Y, Tahata T, Yamanaka K. Surgical treatment of aneurysm or dissection involving the ascending aorta and aortic arch, using circulatory arrest and retrograde colleagues cerebral perfusion. J.Cardiovasc.Surg. 1990; 31: 553-558
14. Soma Y, Kawada K, Kono N, Imamura H, Yotsu R, Odagiri S, Inoue T. Clinical results of cardiopulmonary bypass with selective cerebral perfusion for aneurysm of the ascending aorta and the aortic arch. Ann Thorac Surg 1982;34:659-663
15. Kazui T, Kimura N, Yamada O, Komatsu S. Surgical outcome of aortic arch aneurysms using selective cerebral perfusion. Ann Thorac Surg 1994;57:904-911
Your questions, contributions and commentaries
will be answered
by the lecturer or experts on the subject in the Cardiovascular Surgery list.
Please fill in the form (in Spanish, Portuguese or English) and press the "Send" button.
2nd Virtual Congress of Cardiology
Dr. Florencio Garófalo
Dr. Raúl Bretal
Dr. Armando Pacher
Technical Committee - CETIFAC
Copyright© 1999-2001 Argentine Federation of Cardiology
All rights reserved
This company contributed to the Congress: