480 c 2008 Wiley Periodicals, Inc.

Transcription

1 480 c 2008 Wiley Periodicals, Inc. Antegrade Versus Retrograde Cerebral Perfusion in Relation to Postoperative Complications Following Aortic Arch Surgery for Acute Aortic Dissection Type A Efstratios Apostolakis, M.D., Ph.D., Efstratios N. Koletsis, M.D., Ph.D., Panagiotis Dedeilias, M.D., John N. Kokotsakis, M.D., Ph.D., George Sakellaropoulos, Ph.D., Argini Psevdi, M.D., Konstantin Bolos, M.D., and Dimitrios Dougenis, M.D., Ph.D. Cardiothoracic Surgery Department, School of Medicine, University of Patras, Greece; and Cardiac Surgery Department, Evangelismos General Hospital, Athens, Greece ABSTRACT Background: Aortic arch surgery is impossible without the temporary interruption of brain perfusion and therefore is associated with high incidence of neurologic injury. The deep hypothermic circulatory arrest (HCA), in combination with antegrade or retrograde cerebral perfusion (RCP), is a well-established method of brain protection in aortic arch surgery. In this retrospective study, we compare the two methods of brain perfusion. Materials and Methods: From 1998 to 2006, 48 consecutive patients were urgently operated for acute type A aortic dissection and underwent arch replacement under deep hypothermic circulatory arrest (DHCA). All distal anastomoses were performed with open aorta, and the arch was replaced totally in 15 cases and partially in the remaining 33 cases. Our patient cohort is divided into those protected with antegrade cerebral perfusion (ACP) (group A, n = 23) and those protected with RCP (group B, n = 25). Results: No significant difference was found between groups A and B with respect to cardiopulmonary bypass-time, brain-ischemia time, cerebral-perfusion time, permanent neurologic dysfunction, and mortality. The incidence of temporary neurologic dysfunction was 16.0% for group A and 43.50% for group B (p = 0.04). The mean extubation time was 3.39 ± 1.40 days for group A and 4.96 ± 1.83 days for group B (p = ). The mean ICU-stay was 4.4 ± 2.3 days for group A and 6.9 ± 2.84 days for group B (p = ). The hospital-stay was ± 4.06 days for group A and ± 6.91 days for group B (p = ). Conclusion: The antegrade perfusion seems to be related with significantly lower incidence of temporary neurological complications, earlier extubation, shorter ICU-stay, and hospitalization, and hence lower total cost. doi: /j x (J Card Surg ) Aortic arch surgery constitutes a significant challenge in cardiovascular surgery due to its demand for temporary interruption of brain perfusion. The partial or total replacement of aortic arch is thus associated with high mortality, as well as incidence of neurological injury. Mortality is widely ranged between 2.8% and 24%, depending on the extent of replaced ascending aorta and arch, age, emergency, comorbidity, etc. 1 5 The incidence of postoperative neurological dysfunction is expressed as permanent neurological dysfunction (PND) or temporary neurological dysfunction (TND). It varies between 5.5% and 33.3%, depending on the above mentioned factors, as well as the method of brain protection. 3,6 13 The deep hypothermic circulatory arrest (DHCA) alone or in combination with continuous cerebral perfusion (ante- or retrograde) are nowadays the established methods of brain protection in aor- Address for correspondence: Efstratios Koletsis, 31 Chlois Str, Voula, Athens, Greece. Fax: ekoletsis@hotmail.com tic arch surgery. The superiority of antegrade or retrograde cerebral perfusion (RCP) has been debated for many years. 6,7,14,15 Most of the experimental studies showed a clear superiority of antegrade cerebral perfusion (ACP) against RCP, on brain protection However, in clinical studies, the difference is not so clear, mainly because they are not comparable and not randomized. 3-5,12,23-25 This study concerns the comparison of the two methods of brain protection during the aortic arch replacement for acute aortic dissection in relation to the postoperative neurological events and its consequences. MATERIALS AND METHODS Between January 1998 and September 2006, in our hospitals, 212 patients were operated on emergency basis for acute type A dissection. Forty-eight of them underwent partial or total aortic arch replacement. The open aorta under circulatory arrest allowed intraoperative exploration of the arch, and the decision for

2 J CARD SURG APOSTOLAKIS, ET AL. ANTEGRADE VERSUS RETROGRADE CEREBRAL PERFUSION 481 TABLE 1 The Preoperative Clinical Characteristics and Findings Characteristics Number of Patients Group A Group B Gender Men 36 (75%) Women 12 (25%) 7 5 Age (years) Range (min-max) Mean ± SD 61 ± ± ± 17.1 History Hypertension 39 (81.1%) Diabetes mellitus 11 (22.9%) 5 6 Peripheral arteriopathy 7 (14.6%) 4 3 Acute renal failure 2 (4.2%) 1 1 Chronic renal failure 1 (2.1%) 1 Smoking or chronic obstructive pulmonary disease 25 (52.1%) Family history of dissection 3 (6.3%) 2 1 Substernal or back pain 46 (95.8%) Transient consciousness 3 (6.3%) 2 1 Hypotension 5 (10.4%) 3 2 Pulse abnormalities (absent, deficit, etc.) 13 (27.1%) 5 8 Aortic murmur (diastolic or systolo-diastolic) 12 (25.0%) 7 5 Diagnostic evidence (positive/performed) Chest X-ray 46/48 (95.8%) 21/23 25/25 Contrast-CT 32/35 (91.4%) 14/17 18/18 Angio-CT 11/11 (100%) 6/6 5/5 TEE 13/16 (81.3%) 6/7 7/9 Dissection type I 33 (68.7%) II 15 (31.3%) 5 10 Concomitant aortic valve insufficiency Regurgitation (>2+/4+) 19 (39.6%) 8 11 Aortic valve stenosis (chronic) 1 (2.1%) 1 Pericardial fluid (±tamponade) 3 (6.3%) 2 LV ejection fraction Normal (>50%) 40 (83.3%) Impaired (30% to 50%) 8 (16.7%) 4 4 TEE = transesophageal echocardiogram. replacement was based upon the existence of intima tear in the arch or extension of the ascending aortic tear in it. Thirty-six of them were men and 12 were women. Their age ranged from 29 to 78 years, with mean (±SD) age 61 ± 12.7 years (Table 1). Patients that were excluded from this study were those with: (a) preoperative neurological deficit, (b) unstable preoperative hemodynamic derangement (shock) or with severely affected left ventricular function (EF [ejection fraction] < 30%), or with signs of acute myocardial infarction, (c) preoperative consciousness disturbances, (d) history of a transient (or permanent) ischemic attack, or with known significant stenosis of carotid arteries, (e) symptomatic visceral ischemia, (f) uncontrolled postoperative acute renal or hepatic failure, and (g) preoperative disability of the upper or lower limbs. The type of dissection (by DeBakey) was type I in 33 cases (68.7%) and type II in the remaining 15 cases (31.3%). Concomitant significant (>2+/4+) aortic valve insufficiency was present in 19 cases (39.6%) (Table 1). Among the type I patients, 21 had aortic tear at the ascending aorta, 8 had the tear at the ascending aorta extending to the aortic arch, and in the remaining 4 ones the tear started from the aortic arch and extended for a short course to the descending aorta. From the history point of view, 39 patients had hypertension, 11 diabetes mellitus, and 25 were smokers or had chronic obstructive pulmonary disease (Table 1). The preoperative diagnosis was based on contrast-ct in 35 cases with sensitivity 91.4%, angio- CT in 11 cases (sensitivity 100%), and on transesophageal echocardiogram (TEE) in 16 cases (sensitivity 81.3%). The transthoracic echocardiogram was used in every case for the estimation of left ventricular and heart valve function (EF, aortic regurgitation [AR], mitral regurgitation [MR]) and the pericardial condition (fluid, tamponade). The left ventricular function was good in 40 cases (83.3%) and impaired in the remaining 8 cases (Table 1). The patients were divided in two groups depending on the cerebral perfusion method. In group A, 23 patients protected with ACP were enrolled and 25 patients with retrograde perfusion comprised group B. The two groups were similar concerning the clinical and intraoperative characteristics. During the initial period (1998 to 2001), we mainly used RCP, while after 2002 we mainly used ACP. The study was approved by the Hospital Ethical Committee. All patients, or in case of inability their relatives, gave their consent at the time of admission. At the same period of time in our institution, which is the largest refer-

3 482 APOSTOLAKIS, ET AL. ANTEGRADE VERSUS RETROGRADE CEREBRAL PERFUSION J CARD SURG ring hospital for emergency cardiac surgery in Greece, there were 128 admissions for acute type B dissections that were treated mainly conservatively and in some complicated cases surgically or with stent graft implantation. Operation s strategy Our policy was to reduce as much as possible the time between diagnosis and operation. Every patient was operated on after median sternotomy and using partial cardiopulmonary bypass through the common femoral artery and right atrium in 21 cases and through the common femoral artery and vein in 9 cases. In these cases, a long vein catheter was introduced through the femoral vein (Femoral Arterial Cannula, DLP, Medtronic, Minneapolis, MN, USA), and boosted up to the right atrium. In the remaining 18 cases, the bypass connection of the patient was performed through the right subclavian or axillary artery by using a polytetrafluoroethylene (PTFE) graft of 8 mm (Gore-Tex Stretch Vascular Graft, Thin Wall 8 mm, Creative Technologies Worldwide, Flagstaff, AZ, USA), and the venous drainage via the right atrium. Every patient was infused with Mannitol in the prime, Thiopental 5 mg/kg, 200 mg Lidocaine bolus, and 2 g magnesium bolus, 5 minutes before initiation of arrest. Circulatory arrest was implemented at a nasopharyngeal temperature from 16 to 20 C, and the head was packed in ice throughout hypothermic circulatory arrest (HCA) (Table 2). During the circulatory arrest, a continuous ACP or RCP was accomplished. ACP was applied either selectively through the aortic arch after direct cannulation of innominate and left common carotid artery (5 patients), or nonselectively through the (used for cardiopulmonary bypass [CPB]) right axillary artery (18 cases). The perfusion flow for ACP was 500 to 1000 ml/min so as to restore the right radial artery pressure to about 40 to 60 mmhg. RCP (25 cases) was obtained after insertion of a coronary sinus catheter into the superior vena cava distally and its occlusion proximally. The perfusion flow was 300 to 500 ml/min, and the recorded mean perfusion pressure was restored at a level between 20 and 25 mmhg. In five cases of RCP and for technical reasons (to allow bloodless field for the arch anastomosis, or in the presence of extensive calcification), the continuous flow was intermittently interrupted, for 3 to 8 minutes (Table 2). The anastomosis of the aortic arch was done first and just after it, the femoral arterial cannula was implanted into the distal neo-arch graft, and the cardiopulmonary bypass was reinitiated. The partial or total aortic arch replacement was performed either along with replacement of the ascending aorta, or of the aortic root as well (Table 3). The aortic valve was repaired in 13 cases (27.1%) and replaced in 12 cases (25%). The repair of the aortic valve consisted in resuspension of the commissures of the valve. In the majority of cases with aortic valve replacement we performed replacement of the aortic root using valvebearing graft, which is Bentall s operation (8 cases out of 12) (Table 3). The replacement of the aortic arch was total in 15 cases (31.25%) and partial in 33 cases (68.75%). The decision was based on the anatomy of the arch, the type of dissection, the quality of the dissected aortic wall, and the existence of atherosclerotic plaques. In addition to this, in the cases of total aortic arch replacement, the implantation of brachiocephalic vessels into the graft was performed, either through a button (in 12 cases), or separately through two (in two cases) or three (in one case) direct anastomoses (Table 3). Postoperative neurological estimation The neurological estimation was done preoperatively to exclude some patients (see exclusion criteria), and TABLE 2 The Surgical Methodology Regarding the Connection of Patients with Cardiopulmonary Bypass and the Cerebral Perfusion Methodology Number of Patients Group A Group B Surgical methodology Median sternotomy 48 (100%) Femoro-right atrium cannulation (initial) 21 (43.8%) Right subclavian-right atrium 18 (37.5%) 10 8 Femoro-femoral bypass 9 (18.8%) 2 7 Core temperature C 27 (56.3%) Core temperature C 21 (43.7%) 9 12 Cerebral perfusion methodology Antegrade through innominate and 5 (10.4%) 5 left common carotid artery Antegrade through right axillary artery 18 (37.5%) 18 ACP flow and pressure ml/min, right radial artery pressure mmhg Retrograde through superior vena cava 25 (52.1%) 25 RCP flow and pressure ml/min, perfusion pressure mmhg Continuous (ACP or RCP) flow 43 (89.6%) Intermittent RCP flow (3-5 minutes) 3 (6.3%) 3 Intermittent RCP flow (6-8 minutes) 2 (4.2%) 2

4 J CARD SURG APOSTOLAKIS, ET AL. ANTEGRADE VERSUS RETROGRADE CEREBRAL PERFUSION 483 TABLE 3 The Spectrum of Performed Operations Performed Operations Number of Patients Group A Group B Ascending + total arch 8 (16.7%) 5 3 Ascending + hemi-arch 15 (31.3%) 6 9 Ascending + total arch + AV repair (resuspension) 3 (6.3%) 2 1 Ascending + hemi-arch + AV repair (resuspension) 10 (20.8%) 4 6 Aortic valve + ascending (Bentall s) + total arch 3 (6.3%) 2 1 Aortic valve + ascending (Bentall s) + hemi-arch 5 (10.4) 2 3 Aortic valve + ascending (Cabrol s) + hemi-arch 1 (2.1%) 1 Aortic valve + ascending + total arch 1 (2.1%) 1 Aortic valve + ascending + hemi-arch 2 (4.2%) 2 Uniform branchiocephalic implantation 12 (25%) 7 5 Two separate anastomoses 2 (4.2%) 2 Three separate anastomoses 1 (2.1%) 1 every day postoperatively by neurologist. All patients with neurological symptoms underwent one or more brain CT-scans as required. For the classification of postoperative neurological findings, we used a scale (Table 4), which is a modification of the one proposed by Ergin et al. 11 Statistical analysis Continuous variables are presented as mean ±SD, unless otherwise specified. Comparisons between groups A and B were made using t-test or its nonparametric equivalent Mann-Whitney test, where appropriate. Comparisons of proportions, for the unpaired case, were performed using an appropriate standardized normal deviate. Chi-square test was used to test possible deviation from the independent assumption for discrete variables. The significance level was set to NS depicts statistically not-significant comparisons. RESULTS The time between the diagnosis and the operation ranged from 1 to 18 hours (mean time 4.4 hours) depending mainly on the distance of the primary diagnostic center and the availability of operation theater in our hospital. Bypass-time ranged from 151 to 221 minutes, with a mean time 176 ± minutes. The bypass- time was 179 ± minutes for group A and 184 ± minutes for group B (p = 0.58). Brain-ischemia time ranged from 24 to 56 minutes with a mean of 37 ± minutes for the entire sample of patients, 39 ± minutes for group A and 36 ± minutes for group B (p = 0.43). The distribution of brainischemia time between the two groups was similar, as it is shown in Table 5 (p = 0.68). Cerebral-perfusion time ranged from 21 to 51 minutes with a mean time 34 ± minutes. For group A, the cerebral perfusion time was 37 ± minutes, and for group B it was 34 ± minutes (p = 0.45) (Table 5). Mortality was 14.6% (7 patients) for the entire sample: 13.0% (3 patients) for group A and 16.6% (4 patients) for group B (NS) (Table 6). The causes of death were heart failure (2 patients), neurologic injury (1 patient), respiratory insufficiencyinfection (3 patients), and hepato-renal failure (1 patient). Respiratory complications like pneumonia, acute respiratory distress syndrome (ARDS), or simple atelectasis was present in about 20% of the patients (Table 6). Atrial fibrillation was present postoperatively in onefourth of the patients (Table 6). Overall, the incidence of neurologic injury was not significant between the two groups (Table 6). The incidence of PND was 4.3% for group A and 4.0% for group B. Of the two cases of PND were a case of hemiplegia in a patient of group A and a case of coma in a patient of group B, who died on the third TABLE 4 Our Classification of Neurological Dysfunctions after Thoracic Aortic Surgery (Modification from Ergin et al.) 21 Classification Grade Clinical Findings Duration Rehabilitation Permanent neurologic dysfunction (PND) I Hemiparesis <2 weeks Complete II Monoplegia <2-4 weeks Complete III Hemiplegia <3-6 months Complete or partial IV Paraplegia >6 months Partial or no V Coma Permanent No Temporary neurologic dysfunction (TND) I Simple confusion <48 hours Complete II Confusion + lethargy <72 hours Complete III Confusion + agitation <72 hours Complete IV Psychosis >72 hours Complete V Parkinsonism-choreoathetosis <4 weeks Complete or minor seizures

5 484 APOSTOLAKIS, ET AL. ANTEGRADE VERSUS RETROGRADE CEREBRAL PERFUSION J CARD SURG TABLE 5 The Conditions of Intraoperative Brain Protection Intraoperative Conditions Overall Group A Group B p CPB-time (minutes) 176 ± ± ± Cerebral perfusion-time (minutes) 34 ± ± ± Cerebral ischemia-time (minutes) 37 ± ± ± <30 minutes 6/14 (42.9%) 8/14 (57.1%) minutes 11/24 (45.8%) 13/24 (44.2%) >40 minutes 6/10 (60%) 4/10 (40%) Number of patients. postoperative day (Table 7). Both patients revealed stroke by CT scan. The patient with hemiplegia gradually improved and was discharged home. A majority of the neurologic dysfunctions (4 cases in group A and 10 in group B) were TND (Table 7). They were characterized by absence of stroke (confirmed by CT scan), diffuse neurologic dysfunction, which contributed to a reduced consciousness or agitation of the patient, absence of orientation and collaboration, and delay of mechanical respiratory weaning (Table 7). The extubation time ranged between 24 hours and 8 days with a mean time 4.12 ± 1.62 days. For group A, the extubation time was 3.39 ± 1.40 days while for group B it was 4.96 ± 1.83 days (p = ) (Table 8). The ICU-stay ranged from 2 to 10 days (mean 6.28 ± 2.11 days). The mean ICU-stay was 4.4 ± 2.30 days for group A and 6.9 ± 2.84 days for group B (p = ). The hospital-stay, that is the total postoperative stay in the hospital, ranged between 12 and 26 days (mean 17 ± 5.33 days). For group A, the mean hospital-stay was ± 4.06 days and for group B it was ± 6.91 days (p = ). The cost for group A was estimated to be 8700 euros, while for group B, 10,300 euros (Table 8). DISCUSSION The interruption of physiological brain perfusion during the aortic arch replacement is the major problem in aortic arch surgery. DHCA introduced by Griep et al. in 1975 offered safety for 30 to 40 minutes, but beyond this time brain protection was incompetent Ueda et al. in 1990 introduced the method of RCP through the superior vena cava as an adjunct to the brain protection. 29 At the same time Bachet et al. introduced cerebroplegia for the antegrade perfusion of the brain, with cold blood during the aortic arch replacement. 30 The advantages of RCP were the following: (a) it offers oxygen and substrates to the brain, (b) it obtains deairing of vessels of the arch, (c) it keeps off the toxic wastes of ischemia from the brain, (d) it removes the solid emboli from the arterial branches of the arch, (e) it avoids any maneuvers to the arch vessel which could be atheromatic, and (f) there is no catheter or cannula in the surgical field to impede the surgeon. However, some years later the above-mentioned advantages of RCP came in dispute. 3,22-24,31-35 After 1995 some experimental studies showed that during RCP, only a minimal amount of blood, not more than 3% to 10%, is TABLE 6 Complications and Mortality for the Two Groups Results and Complications Number of Patients (%) Group A (%) Group B (%) p Mortality 7 (14.6) 3/23 (13.0) 4/25 (16.0) 0.39 Overall ND 16 (33.3) 5/23 (21.7) 11/25 (44) 0.14 PND 2 (4.2) 1/23 (4.3) 1/25 (4.0) 0.48 TND 14 (29.2) 4/23 (17.39) 10/25 (40) 0.04 Bleeding 3 (6.2) 2/23 (8.7) 1/25 (4.0) 0.25 Arrhythmia-atrial fibrillation 10 (20.8) 4/23 (17.4) 6/25 (24.0) 0.29 Pericardial-pleural effusion 5 (10.4) 3/23 (13.0) 2/25 (8.0) 0.28 Pneumonia 4 (8.3) 1/23 (4.3) 3/25 (12.0) 0.17 ARDS 3 (6.2) 1/23 (4.3) 2/25 (8.0) 0.30 Atelectasis 2 (4.2) 1/23 (4.3) 1/25 (4.0) 0.48 Renal failure 3 (6.2) 2/23 (8.7) 1/25 (4.0) 0.51 Phrenic nerve palsy 2 (4.2) 1/23 (4.3) 1/25 (4.0) 0.48 Hepatic dysfunction 1 (2.1) 0/23 (0) 1/25 (4.0) 0.17 Trauma infection 2 (4.2) 1/23 (4.3) 1/25 (4.0) 0.48 Tracheostomy 4 (8.3) 1/23 (4.3) 3/25 (12.0) 0.17 PND = permanent neurologic dysfunction; TND = temporary neurologic dysfunction; ARDS.

6 J CARD SURG APOSTOLAKIS, ET AL. ANTEGRADE VERSUS RETROGRADE CEREBRAL PERFUSION 485 TABLE 7 The Kind of Neurological Dysfunction, According to Table 4, and Distribution between the Two Groups Grade Group A Group B PND III 1 V 1 TND II 2 4 III 2 4 IV 2 Sum 5 11 PND = permanent neurologic dysfunction; TND = temporary neurologic dysfunction. directed to the brain, whereas more than 90% was deviated through azygos to the vena cava (VC), or it was entrapped in the cerebral venous sinuses ,36 Therefore, the contribution of infused blood to the metabolic demands of the brain is very trivial. 36 From the eight reported clinical comparative studies between ACP and RCP, three showed a clear superiority of ACP. 10,37,38 Although three more studies present similar results between the two methods, they basically support ACP due to the prolonged ischemic and CPB time that accompanied their ACP method. 1,10,25 Unfortunately, all the above-mentioned studies as well as ours are nonrandomized. The only randomized comparative study but including a very small number of patients (10 patients) is that of Svensson et al., 2 which failed to show that the retrograde perfusion is equally able to protect the brain. There are two types of neurologic injuries during operations on aortic arch: strokes, which are due to localized embolic events from macro-material matter, and subtle, a diffuse injury without evidence of any embolic event. 3,9,29,39,40 Ergin et al. defined this latter neurological injury as TND, in contradistinction to the other type of neurological injury, which they called PND. There are three main characteristics of TND: 28,40 (a) there are no focal clinical neurologic findings, (b) the brain-ct is negative for any damage, and (c) there is a complete resolution after some days or weeks. 9 The clinical findings of TND are simple confusion, or confusion with agitation or lethargy, psychosis, or Parkinsonism. Ergin et al., by using neuropsychological tests, showed that the TND is related with deficits in memory and motor function, which may last up to 6 weeks postoperatively. 11,40 The incidence of PND ranges between 2.6% and 16% in several studies, 1,3,35,39 and is related to the age of the patient (>60 years), the severe atheromatous disease of aorta, and the additive surgery of descending aorta. 11 The incidence of TND ranges from 3.8% to 63% ,41 In contrast to PND, TND was directly related to the duration of HCA and the patient s age. 9,41 Our study, in accordance with others, 1,4,10,41 affirms that TND constitutes the majority of neurological dysfunctions during aortic arch surgery (Table 5 and IX). In addition, the incidence of TND in the retrograde group was found significantly greater (twofold) than in the antegrade group. The higher incidence of TND in the retrograde group explains the statistically significant longer extubation-time, ICU-stay, hospital-stay, as well as higher cost in the same group (Table 8). Fleck et al. observed that in their patients with TND, ICU-stay and hospital-stay were significantly longer, in comparison with the patients without TND. 41 This finding in our study may be related to three parameters: (a) reduced consciousness, or agitation, or lethargy, or psychosis of the patient with TND for a few days, which elongated its extubation (Table 10), (b) to a higher number of patients of RCP group who needed tracheostomy (Table 7), and (c) to a higher incidence of pneumonia in the same group (Table 7). The ACP through the axillary or subclavian artery is nowadays a safe method, that offers many advantages. 12,42-44 It offers the needed amount of blood to the brain through the same cannula of CPB, without additive maneuvers for selective perfusion of arch branches. 42 CONCLUSIONS Our study, despite its limitations (it is nonrandomized and with a relatively small number of patients) has shown, that: (a) ACP offers a better brain protection in comparison with that of RCP. It provides uniform brain perfusion and therefore better cooling; the surgeon is given more time to reconstruct the vessels of the arch because the perfusion time is not related to the TND, (b) the nonselective ACP through subclavian or axillary perfusion (18 of 23 cases, Table 3) is safe, feasible, and sufficient to meet the metabolic demands of brain, and (c) this perfusion offers the surgeon a bloodless field, and avoids the need for temporary disruption of flow, as is the case of RCP. TABLE 8 The Results till Discharge and Costs of Operations Overall Group A Group B p Extubation-time 4.12 ± ± ± ICU-stay 6.28 ± ± ± Hospital-stay 17 ± ± ± Calculated-cost (euros)

7 486 APOSTOLAKIS, ET AL. ANTEGRADE VERSUS RETROGRADE CEREBRAL PERFUSION J CARD SURG Acknowledgments: We wish to thank Dr. Antonios Tavernarakis for his valuable contribution in the neurological assessment of our patient cohort as well as to Mr. Efstratios Anagnostou for the collection of the data. REFERENCES 1. Usui A, Yasuura K, Watanabe T, et al: Comparative clinical study between retrograde cerebral perfusion and selective cerebral perfusion in surgery for acute type A aortic dissection. Eur J Cardiothorac Surg 1999;15: Svensson L, Nadolny E, Penney D, et al: Prospective randomized neurocognitive and S-100 study of hypothermic circulatory arrest, retrograde brain perfusion, and antegrade brain perfusion for aortic arch operations. Ann Thorac Surg 2001;71: Coselli J: Retrograde cerebral perfusion in surgery for aortic arch aneurysms. In: Ennker J, Coselli J, Hetzer R (eds): Cerebral Protection in Cerebrovascular and Aortic Surgery. Steinkopf Verlag, Darmstadt, Germany, 1997, pp Tan E, Dossche K, Morshuis W, et al: Is extended arch replacement for acute type A aortic dissection an additional risk factor for mortality? Ann Thorac Surg 2003;76: Wong C, Boncer R: Does retrograde cerebral perfusion affect risk factors for stroke and mortality after hypothermic circulatory arrest? Ann Thorac Surg 1999;67: Svensson L, Crawford E, Hess E, et al: Deep hypothermia with circulatory arrest: Determinants of stroke and early mortality in 656 patients. J Thorac Cardiovasc Surg 1993;106: Svensson L: Antegrade perfusion during suspended animation? J Thorac Cardiovasc Surg 2002;124: Mueller D, Fueguth G, Wimmer-Grenecker G, et al: Neurologic outcome after surgery of the aortic arch: Comparison of deep hypothermic arrest, antegrade and retrograde cerebral perfusion. Ind J Thorac Cardiovasc Surg 2004;20: Ergin M, Galla J, Lansman L, et al: Hypothermic circulatory arrest in operations on the thoracic aorta. Determinants of operative mortality and neurologic outcome. J Thorac Cardiovasc Surg 1994;107: Ueda T, Shimizu H, Ito T, et al: Cerebral complications associated with selective perfusion of the arch vessels. Ann Thorac Surg 2000;70: Ergin M, Uysal S, Reich D, et al: Temporary neurological dysfunction after deep hypothermic circulatory arrest: A clinical marker of long-term functional deficit. Ann Thorac Surg 1999;67: Dossche K, Schepens M, Morshuis W, et al: Antegrade selective cerebral perfusion in operations on the proximal thoracic aorta. Ann Thorac Surg 1999;67: Miyamoto Y, Onishi K, Mitsuno M: Aortic arch replacement with proximal first technique. Ann Thorac Cardiovasc Surg 2003;9: Engelman R, Pleet A, Rousou J, et al: Does cardiopulmonary bypass temperature correlate with postoperative central nervous system dysfunction? J Card Surg 1995;10: Svensson L: Surgery of the aortic arch, descending thoracic and thoracoabdominal surgery, and aortic dissection. In: Sellke F, et al (eds). Sabiston and Spencer Surgery of the Chest. 7th Ed, Amsterdam: Elsevier Saunders, 2005, p Usui A, Oohara K, Murakami F, et al: Body temperature influences regional tissue blood flow during retrograde cerebral perfusion. J Thorac Cardiovasc Surg 1997;114: Ye J, Ryner L, Kozlowski P, et al: Retrograde cerebral perfusion results in flow distribution abnormalities and neuronal damage. Circulation 1998;98: Bachtiary F, Dogan S, Dzemali O, et al: Mild hypothermic (32 C) and antegrade cerebral perfusion in aortic arch operations. J Thorac Cardiovasc Surg 2006;132: Duebener L, Hagino I, Schmitt K, et al: Direct visualization of minimal cerebral capillary flow during retrograde cerebral perfusion: An intravital fluorescence microcopy study in pigs. Ann Thorac Surg 2003;75: Katz M, Khazin V, Steinmetz A, et al: Distribution of cerebral flow using retrograde versus antegrade cerebral perfusion. Ann Thorac Surg 1999;67: Filgureiras C, Winsborrow B, Ye J, et al: A 31P-Magnetic Resonance study of anegrade and retrograde cerebral perfusion during aortic arch surgery in pigs. J Thorac Cardiovasc Surg 1995;110: Kawata M, Takamoto S, Kitahori K, et al: Intermittent pressure augmentation during retrograde cerebral perfusion under moderate hypothermia provides adequate neuroprotection: An experimental study. J Thorac Cardiovasc Surg 2006;132: Safi H, Letsou G, Iliopoulos D, et al: Impact of retrograde cerebral perfusion on ascending aortic and arch aneurysm repair. Ann Thorac Surg 1997;63: Coselli J, LeMaire S: Experience with retrograde cerebral perfusion during proximal aortic surgery in 290 patients. J Card Surg 1997;12: Matalanis G, Hata M, Buxton B: A retrospective comparative study of deep hypothermic circulatory arrest, retrograde, and antegrade cerebral perfusion in aortic arch surgery. Ann Thorac Cardiovasc Surg 2003;9: Griepp R, Stinson E, Hollingsworth, et al: Prosthetic replacement of the aortic arch. J Thorac Cardiovasc Surg 1975;70: Kouchoukos NT, Dougenis D. Surgery of the thoracic aorta. N Engl J Med 1997;336: Ergin M: Principles of cerebral protection during operations on the thoracic aorta. In: Franco K, Verrier E (eds). Advanced Therapy in Cardiac Surgery. BC Decker, Hamilton, Ontario, Canada, 1999, pp Ueda Y, Miki S, Kusuhara K, et al: Surgical treatment of aneurysm or dissection involving the ascending aorta and aortic arch, utilizing circulatory arrest and retrograde cerebral perfusion. J Cardiovasc Surg 1990;31: Bachet J, Guilmet D, Goudot B, et al: Cold cerebroplegia: A new technique of cerebral protection during operations on the transverse aortic arch. J Thorac Cardiovasc Surg 1991;102: Deeb G, Jenkins E, Bolling S, et al: Retrograde cerebral perfusion during hypothermic circulatory arrest reduces neurologic morbidity. J Thorac Cardiovasc Surg 1995;109: Usui A, Abe T, Murase M: Early clinical results of retrograde cerebral perfusion for aortic arch operations in Japan. Ann Thorac Surg 1996;62: Usui A, Hotta T, Himoura M, et al: Retrograde cerebral perfusion through a superior vena caval cannula protects the brain. Ann Thorac Surg 1992;53: Okita Y, Takamoto S, Ando M, et al: Mortality and cerebral outcome in patients who underwent aortic arch operations using deep hypothermic circulatory arrest with retrograde cerebral perfusion: No relation of early death,

8 J CARD SURG APOSTOLAKIS, ET AL. ANTEGRADE VERSUS RETROGRADE CEREBRAL PERFUSION 487 stroke, and delirium to the duration of circulatory arrest. J Thorac Cardiovasc Surg 1998;115: Coselli J: Retrograde cerebral perfusion is an effective means of neural support during deep hypothermic circulatory arrest. Ann Thorac Surg 1997;64: Oohara K, Usui A, Murase M, et al: Regional cerebral tissue blood flow measured by the coloured microsphere method during retrograde cerebral perfusion. J Thorac Cardiovasc Surg 1995;109: Neri E, Sassi C, Barabesi L, et al: Cerebral autoregulation after hypothermic circulatory arrest in operations on the aortic arch. Ann Thorac Surg 2004;77: Hagl C, Ergin M, Galla J, et al: Neurological outcome after ascending aorta-aortic arch operations: Effect of brain protection technique in high risk patients. J Thorac Cardiovasc Surg 2001;121: Ueda T, Shimizu H, Ito T, et al: Cerebral complications associated with selective perfusion of the arch vessels. Ann Thorac Surg 2000;70: Ergin A, Griepp E, Lansman S, et al: Hypothermic circulatory arrest and other methods of cerebral protection during operations on the thoracic aorta. J Card Surg 1994;9: Fleck T, Czerny M, Hutschala D, et al: The incidence of transient neurologic dysfunction after ascending aortic replacement with circulatory arrest. Ann Thorac Surg 2003;76: Budde J, Serna D, Osborne S, et al: Axillary cannulation for proximal aortic surgery is a safe in the emergent setting as in elective cases. Ann Thorac Surg 2006;82: Dougenis D, Daily BB, Kouchoukos NT: Reoperations on the aortic root and ascending aorta. Ann Thorac Surg. 1997;64: Numata S, Ogino H, Sasaki H, et al: Total arch replacement using antegrade selective cerebral perfusion with right axillary artery perfusion. Eur J Cardiothorac Surg 2003;23: