Received 18 January 2004; received in revised form 2 April 2004; accepted 5 April 2004; Available online 18 May 2004

Transcription

1 European Journal of Cardio-thoracic Surgery 26 (2004) Hypothermic circulatory arrest with and without cold selective antegrade cerebral perfusion: impact on neurological recovery and tissue metabolism in an acute porcine model q Christian Hagl a, *, Nawid Khaladj a, Sven Peterss a, Klaus Hoeffler a, Michael Winterhalter b, Matthias Karck a, Axel Haverich a a Division of Thoracic and Cardiovascular Surgery, Hannover Medical School, Carl-Neuberg-Strasse, D Hannover, Germany b Department of Anaesthesiology, Hannover Medical School, Hannover, Germany Received 18 January 2004; received in revised form 2 April 2004; accepted 5 April 2004; Available online 18 May 2004 Abstract Objective: Clinically, selective antegrade cerebral perfusion (SACP) seems to be associated with a better neurological outcome compared to hypothermic circulatory arrest (HCA) alone, but the pathophysiological mechanisms are not well understood. Therefore, this study was undertaken to assess the effects of HCA with and without SACP on the cerebral integrity using multimodal neurophysiological monitoring. Methods: 12 pigs were randomly assigned to 100 min HCA at 20 8C brain temperature with ðn ¼ 6Þ and without ðn ¼ 6Þ SACP. Haemodynamics, metabolics and neurophysiology (EEG, SSEP, ICP, spectroscopy, cerebral tissue monitoring) were monitored. Animals were sacrified 4 h after reperfusion and the brains perfused for histopathological assessment. Results: There were no clinically relevant differences in hemodynamics between groups. During reperfusion, EEG and SSEP recovery was significantly faster in the SACP group ðp, 0:05Þ: The rise in ICP during reperfusion was markedly reduced in the SACP group (P, 0:01 for the trend). Three hours after reperfusion, median ICP was 130% compared to baseline in the SACP group and 225% in the HCA group ðp, 0:01Þ: Invasive as well as noninvasive cerebral monitoring indirectly indicates the occurrence of tissue acidosis in the HCA group even 4 h after HCA. Conclusions: Cold SACP is associated with better neurophysiological recovery and less cerebral edema, indicated by lower intracranial pressures during reperfusion. Neurophysiological recovery correlated well with the rise in ICP. HCA alone causes prolonged acidosis in the brain tissue during reperfusion. From these data, SACP appears to be superior to HCA alone, but further studies have to elucidate the optimal regimes for SACP. q 2004 Elsevier B.V. All rights reserved. Keywords: Aortic surgery; Hypothermic circulatory arrest; Selective antegrade cerebral perfusion; Cerebral protection; Pig model 1. Introduction Despite the widespread use of hypothermic circulatory arrest (HCA) in aortic arch surgery there remains continued concern about possible adverse cerebral sequelae. Therefore, different adjunctive perfusion techniques have been implemented to improve cerebral protection during these procedures [1]. q Presented at the joint 17th Annual Meeting of the European Association for the Cardio-thoracic Surgery and the 11th Annual Meeting of the European Society of Thoracic Surgeons, Vienna, Austria, October 12 15, * Corresponding author. Tel.: þ ; fax: þ address: hagl@thg.mh-hannover.de (C. Hagl). On the basis, that clinical as well as experimental data on the value of retrograde cerebral perfusion (RCP) techniques are still a matter of controversial discussion [2,3], selective antegrade cerebral perfusion (SACP) methods became increasingly popular. Recent reports from clinical studies clearly demonstrate a favourable outcome [4,5] in those patients, but prospective controlled studies are still missing. Reviewing data from the literature, there are a number of technical details how SACP can be used [3]. These techniques include different cannulation techniques, variable levels of general hypothermia and diverse protocols regarding antegrade flow and temperature. From the recently published data it is not possible to decide which technique may offer the best protection for the organ most sensitive to ischemia the brain. Furthermore, most /$ - see front matter q 2004 Elsevier B.V. All rights reserved. doi: /j.ejcts

2 74 C. Hagl et al. / European Journal of Cardio-thoracic Surgery 26 (2004) techniques are based on empiric findings, since experimental data relying on pathophysiological findings are limited. In the past, the Mount Sinai group affiliated with Randall Griepp established a clinically relevant porcine model which allowed a deeper insight into the cerebral consequences of hypothermic circulatory arrest, RCP as well as SACP. On the basis of these models, we investigated a modification to perform cold SACP in an acute porcine model of hypothermic circulatory arrest. The results were compared with animals which underwent HCA without adjunctive cerebral perfusion. 2. Material and methods 2.1. Study design Twelve female Landrace pigs, 3 4 months of age, weighing kg underwent 100 min of HCA at 20 8C brain temperature. All animals were randomly assigned to serve as control without SACP or with SACP for 90 min. Each animal underwent intra and postoperative hemodynamic and metabolic monitoring and recording of quantitative EEG as well as cortical somtosensory evoked potentials (SSEP). All animals were observed for 4 h after HCA and then electively sacrificed Anesthesia, perioperative management and neurophysiology All animals received human care in compliance with the guidelines Principles of Laboratory Animal Care formulated by the National Society for Medical Research and the Guide for the Care and Use of Laboratory Animals published by the National Institute of Health (NIH Publication No , revised 1996). The protocols for all experiments were approved by the Hannover Medical School Institutional Animal Care and Use Committee as well as the Land Niedersachsen. After pre-treatment with intramuscular azaperone (5 mg/kg) and atropine (0.5 mg), animals were anesthesized with intravenous thiopental sodium (15 mg/kg). After endotracheal intubation, the pigs were mechanically ventilated with a FiO 2 of 0.5 and isoflurane (1 2%). Continuous intravenous infusion of Fentanyl (1 mg kg 21 h 21 ) was administered to maintain adequate analgesia. Paralysis was achieved with intravenous pancuroniumbromide (0.1 mg/kg). The ventilator rate and the tidal volume were adjusted to maintain the arterial carbon dioxide tension between 35 and 45 mmhg. A positive endexpiratory pressure of 3 5 mmhg was frequently used. Before intervention, all animals received 1 g of Ceftriaxon intravenously. A transurethral Foley bladder catheter (8 10 F) was inserted for online measurement of urine output and temperatures probes were placed in the rectum and the esophageous. The Picco w (Pulsion Medical Systems AG, Munich, Germany) catheter was placed in the right femoral artery for arterial pressure monitoring, blood sampling and to allow detection of cardiac output by thermodilution. For venous infusion as well as detection of central venous pressure, a 3-luminal catheter was advanced via the right femoral vein towards the right atrium. After midline incision on the pigs head, a small burr hole (0.7 cm diameter) was drilled in the skull to allow introduction of the intraparenchymal microtip pressure catheter (Codman ICP Express w, Johnson and Johnson Prof. Inc., Raynham, MA, USA) and the Neurotrend w probe (Codman, Johnson and Johnson Professional, Inc. Raynham, MA, USA), a device which incorporates optical sensors for the measurement of ph, pco 2, and po 2, and a thermocouple for temperature measurement. The Neurotrend w sensor indicates the perfusion and metabolic acidosis/alkalosis status of cerebral tissue local to sensor placement. The sensor is contained within a polyethylene tube with an average outside diameter of less than 0.5 mm and an effective sensor length of 25 mm. The catheter was placed slightly posterior to the coronal suture approximately 0.5 cm from the midline and was pushed 2 cm vertical into the white matter of the brain. The proper location has been tested in pilot studies and standardization was tried to achieve as much as possible. Cervical and cortical somatosensory evoked potentials in response to stimulation of both median nerves as well as continuous EEG were monitored from needle electrodes which were placed in a standard fashion [6]. Analysis was performed by an investigator blinded to the protocol and are expressed as percent recovery compared to baseline measurements General surgical technique and cardiopulmonary bypass After preparation of the skull and the groin, the animals were approached via a left thoracotomy in the 4th intercostal space. The pericardium was entered and the thymus removed to allow better access towards the supra-aortic vessels. After systemic heparinization (400 IU/kg), nonpulsatile cardiopulmonary bypass (CPB) was instituted at a flow rate of ml/kg via a right-angled 8F single cannula (Polystan, Denmark) in the main pulmonary artery, with return to the distal ascending aorta (Medtronic, 12F). A 10F flexible cannula was passed from the left appendage of the atrium into the left ventricle to allow decompression. Surface cooling was used in all animals, without additional topical cooling of the head. Priming of the pump consisted of 1000 ml 0.9%NaCl, furosemide (0.5 mg/kg), heparine (5000 IU) and potassium chloride (1 mval/kg). For CPB, alpha-stat principles were used and the pressure was maintained at a minimum of 40 mmhg. Methylprednisolone (20 mg/kg body weight) was administered 15 min after initiation of CPB, but no barbiturates were given in

3 C. Hagl et al. / European Journal of Cardio-thoracic Surgery 26 (2004) the present model. Myocardial protection was achieved by repetitive administration of cold blood cardioplegia in the aortic root as well as topical cooling. After HCA, careful rewarming (avoiding a temperature gradient exceeding 8 8C between perfusate and core temperature) was continued to reach an esophageal temperature of approximately 36 8C. Internal defibrillation was performed after reaching a temperature of.27 8C and potassium,5.5 mmol/l. During weaning from CPB, catecholamines such as dobutamine or norepinehrine were frequently used for inotropic support or to treat vascular resistance vagaries Special considerations regarding selective antegrade cerebral perfusion In contrast to the human anatomy, the pig has only two main vessels arising from the aortic arch. The first branch (brachiocephalic artery) is the source of both carotid arteries with the bifurcation in the apex of the left chest cavity (compare Fig. 1a). Prior to the bifurcation into both carotid arteries the left subclavian artery arises from behind the brachiocephalic artery. In the present model, SACP can be performed via a left thoracotomy without the necessity of additional cannulas. After cross-clamping, the ascending aorta proximal to the cannula and the aortic arch between both supra-aortic vessels isolated perfusion of the bi-carotid trunk is possible (Fig. 1b). Perfusion was started after a 5-min interval of HCA, using a pressure regulating device keeping the pressure between 45 and 50 mmhg, which was usually associated with a flow of ml/min. Flow was indirectly controlled with a pair of fiberoptic optodes for Near-infrared spectroscopy (NIRS) (INVOS w, Somanetics, MA; USA) placed on the occipital part of the skull Study protocol Hemodynamics including ICP, blood gases, hematocrit as well as neurophysiological parameters (EEG, tissue O 2, CO 2 and ph) and temperatures were recorded online and analysed at the following time points. 1. Baseline before CPB at physiologic temperatures 2. During cooling on CPB after 15 min (C15 0 ) 3. During cooling on CPB after 30 min (C30 0 ) 4. During cooling on CPB after 45 min (C45 0 ) 5. After 45 min of HCA or SACP (45 0 HCA/ACP) 6. After 90 min of HCA or SACP (90 0 HCA/ACP) 7. During rewarming on CPB after 15 min (RW15 0 ) 8. During rewarming on CPB after 30 min (RW30 0 ) 9. During rewarming on CPB after 45 min (RW45 0 ) 10. During rewarming on CPB after 60 min (RW60 0 ) 11. Two hours after end of HCA or SACP (no CPB) (120 0 ) 12. Three hours after end of HCA or SACP (no CPB) (180 0 ) 13. Four hours after end of HCA or SACP (no CPB) (240 0 ) 3. Statistics The data were entered in an Excel spreadsheet and analysed using SPSS w software on a personal computer. Data are described as mean and standard deviation, median and range, or percent, as appropriate. The t-test or the Mann-Whitney test, as appropriate, has been used for comparisons at baseline. When the data were consistent with normality and equal variance assumptions, measurements during and after CPB were compared using repeated measures of ANOVA, with tests for average differences between groups. Otherwise the groups were compared separately at each time point using the Mann-Whitney or Fisher exact test. A difference of P, 0:05 was considered to be statistically significant. Fig. 1. Pig anatomy of the aortic arch and the supraaortic vessels. After cross-clamping, the ascending aorta proximal to the cannula and the aortic arch between both supra-aortic vessels, isolated perfusion of the bi-carotid trunk is possible. 4. Results One animal in the SACP-group and one animal of the HCA-group died before reaching the end of the experiment and were replaced for compensation.

4 76 C. Hagl et al. / European Journal of Cardio-thoracic Surgery 26 (2004) Table 1 Hemodynamics Variable Baseline C45 0 (CPB) HCA/SACP45 0 (CPB) HCA/SACP90 0 (CPB) RW60 0 (CPB) HR [b/min] SACP-group 88 ^ 9 36 ^ ^ ^ ^ ^ 20 HCA-group 80 ^ ^ ^ ^ ^ ^ 13 CVP [mmhg] SACP-group 1 (0 4) 2 (1 9) 2 (1 8) 2 (1 13) HCA-group 1 (0 5) 1 (0 7) 2 (0 8) 2 (2 10) MAP [mmhg] SACP-group 61 ^ 6 35 ^ 4 40 ^ 8 47 ^ 9 53 ^ ^ 15 HCA-group 66 ^ ^ 4 40 ^ ^ 5 49 ^ ^ 12 CO [l/min] SACP-group 4.9 ^ ^ ^ ^ ^ ^ ^ ^ 1.0 HCA-group 4.4 ^ ^ ^ ^ ^ ^ 0.8 HR, heart rate; MAP, mean arterial pressure; CO, cardiac output; CPB, cardiopulmonary bypass; HCA, hypothermic circulatory arrest; SACP, selective antegrade cerebral perfusion. Data are retrieved from all animals and are shown as mean ^ standard deviation or median and range; as appropriate. The values for cardiac output during CPB represent the actual pump flow or the SACP flow. A comparison of preoperative animal weights (SACP: 31.5 ^ 1.5 kg vs. HCA: 29.5 ^ 3.0 kg) and age (SACP: 13 ^ 3 weeks vs. HCA: 14 ^ 3weeks) showed no significant differences between the groups. Basic hemodynamic data showed some minor variations but no clinically relevant differences between groups in heart rate, central venous pressure, mean arterial pressure and cardiac output (Table 1). Slight but not significant differences were observed concerning the upward drift of the intracranial temperature in the HCA group. Acid base and blood gas parameters showed no significant differences among the groups Neurophysiology There was a marked increase in intracranial pressure in both groups two hours after reinstitution of the whole body reperfusion. After 4 h, the rise in ICP was significantly more pronounced in the HCA group (225% of baseline) than in the SACP group (130% of baseline, P, 0:01). Over the last 3 h of the experiments, the differences between groups increased steadily towards higher levels in the HCA group ðp. 0:01Þ (Fig. 2). The recovery of EEG (.8% of baseline) was observed at different time points in all animals after SACP, but none after HCA alone (Fig. 3). SSEP recovery was also earlier and more marked in the SACP group. Determination of tissue carbondioxide content revealed a significant increase of CO 2 in the HCA group during the period of circulatory arrest and remained elevated during the entire observation period ðp, 0:01Þ (Fig. 4). Tissue oxygen content showed a significant decrease during the no-flow period in the HCA group towards 0, followed by relatively high levels compared to the SACP group in the first hour during rewarming (Fig. 5). As expected, the tissue ph levels were lower in the HCA group during reperfusion but failed to reach statistical significance (Fig. 6). 5. Discussion With an increasing expertise in aortic arch operations, surgeons strive towards a more complete repair even in complex aortic pathologies. This is usually associated with prolonged periods of hypothermic circulatory arrest and a higher incidence of neurological morbidity and mortality. Therefore, the technique of SACP has been revived after it had been abandoned due to the disappointing results during the 60ties. The first clinical paper with favourable results by using a so called cold cerebroplegia has been published by Bachet and coworkers in 1991 [9]. In contrast to any no flow or retrograde technique it offers a more physiologic Fig. 2. Intracranial pressure (ICP) measurements (change from baseline) throughout the experiment, as described in the text. B, baseline before cardiopulmonary bypass. The measurements C were performed during cooling. Measurements at 45 and 90 0 HCA/SACP were performed during HCA or HCA þ SACP and RW15 60 indicated the measurement time points during rewarming. The last three measurements were performed 2 4 h ( ) after HCA or HCA þ SACP. *, P, 0:05 between groups.

5 C. Hagl et al. / European Journal of Cardio-thoracic Surgery 26 (2004) Fig. 3. Cumulative EEG recovery, throughout the experiment, as described in the text. B, baseline before cardiopulmonary bypass. The measurements C were performed during cooling. Measurements at 45 0 and 90 0 HCA/SACP were performed during HCA or HCA þ SACP and RW indicated the measurement time points during rewarming. The last three measurements were performed 2 4 h ( ) after HCA or HCA þ SACP. *, P, 0:01; **, P. 0:001 between groups. situation and it seems reasonable that it leads to a better protection of the brain. Recently, a number of different investigators reported in clinical settings excellent outcomes with this approach [4,5,7,8]. But besides the appealing concept to provide the cerebrum with oxygenated blood there are a number of open questions concerning the technique itself. From recent reports we learned that profound hypothermia can probably be avoided and we realized a marked tendency towards higher core and perfusion temperatures [7]. But what is the pathophysiological background of these techniques and what can be considered as safe enough for our patients? To answer these important questions we need prospective clinical trials but also clinically relevant animal models. As mentioned before, many of the basic work and contributions in animal models came from the Mount Sinai group in New York. They were able to establish a chronic pig model which allows besides intraoperative neurophysiological evaluation [10] sophisticated histopathology [11] and early [6] as well as late neurobehavioral testing [12]. Fig. 5. Brain tissue po 2 throughout the experiment. Data are given as change from baseline. The measurements C were performed during cooling. Measurements at 45 and 90 0 HCA/SACP were performed during HCA or HCA þ SACP and RW indicated the measurement time points during rewarming. The last three measurements were performed 2 4 h ( ) after HCA or HCA þ SACP. *, P, 0:05 between groups. In this context, Griepp and coworkers showed that at least min of hypothermic circulatory arrest at 20 8C is needed to produce some kind of cerebral damage in this model. Since this is a prerequisite to study different concepts of neuroprotection, these relatively long HCA times are needed, despite the fact, that clinically relevant HCA/SACP times are shorter. In the present study, we tested a new modification of an acute porcine model, which allows for SACP without the need of additional incisions or cannulas. In contrast to other studies, in which the whole upper body is perfused [13], the amount of blood reaching the brain can be defined by isolated perfusion of the bicarotid trunk. Furthermore chronic observations are possible since the number of incisions are limited and since a lateral thoracotomy is less traumatic than any median sternotomy in swine. Therefore a number of interesting questions concerning selective cerebral perfusion techniques could be answered in this model. In our first series we were interested to study the effects of cold cerebral perfusion in contrast to HCA alone. Besides the impressive early recovery of neurophysiological parameters in the SACP group, we realized Fig. 4. Brain tissue pco 2 throughout the experiment. Data are given as change from baseline. The measurements C were performed during cooling. Measurements at 45 and 90 0 HCA/SACP were performed during HCA or HCA þ SACP and RW indicated the measurement time points during rewarming. The last 3 measurements were performed 2 4 h ( ) after HCA or HCA þ SACP. *, P, 0:05; **, P, 0:01 between groups. Fig. 6. Brain tissue ph throughout the experiment. Data are given as change from baseline. The measurements C were performed during cooling. Measurements at 45 and 90 0 HCA/SACP were performed during HCA or HCA þ SACP and RW indicated the measurement time points during rewarming. The last three measurements were performed 2 4 h ( ) after HCA or HCA þ SACP. *, P, 0:05 between groups.

6 78 C. Hagl et al. / European Journal of Cardio-thoracic Surgery 26 (2004) once again the meaningful impact of the intracranial pressure during reperfusion. It seems that in contrast to HCA alone, continuous cold selective cerebral perfusion can reduce the increase of ICP during whole body reperfusion. Although ICP was rising in both groups the levels continued to be significantly lower after SACP during the whole observation period. In this context, Ehrlich and coworkers showed in 2001, that an interval of cold reperfusion after HCA can attenuate the rise in ICP usually seen after HCA, and can also reduce the amount of histopathological changes [14]. Furthermore studies of RCP have shown that RCP is often associated with a marked increase in ICP [12] and a worse neurological outcome. In a chronic porcine study published in 2002 we were able to show a close correlation between low ICPs during reperfusion and the occurance of EEG recovery [15]. Furthermore high ICP values during reperfusion correlated well with a worse neurobehavioral outcome. Unfortunately, we were not able to differentiate if high ICP values cause neurological damage or if it is just a reflection of incomplete cerebral protection. In the present study we used a monitoring device for local tissue ph and CO 2 with two modified optical fibers as well as detection of O 2 using a Clark electrode, allowing estimation of acidosis and its respiratory or metabolic components. Monitoring of the trend allows early detection of acidemia, which may indicate compromised tissue perfusion. Since there are marked interindividual differences regarding absolute values, continuous monitoring of these parameters and analysis as change from baseline have to be performed and may help to identify and distinguish hypoxic and ischemic episodes. The proper function of the device has been tested in pilot experiments in which modifications of cerebral perfusion have been performed (data not shown). In the present study, animals which were subjected to HCA without cerebral perfusion, had significant higher tissue CO 2 levels compared to those who had adjunctive ASCP. This was associated with a lower ph, indicating tissue acidosis. These phenomena could be observed even 4 hours after HCA. Interestingly, O 2 levels were significantly higher in the early reperfusion period after HCA, indicating luxury perfusion after HCA. Nevertheless, this better perfusion did not reverse tissue acidosis and seemed to cause tissue edema indicated by high ICP pressures. The reasons for these findings remain speculative, but capillary leakage or insufficient capillary perfusion due to arterio-venoes shunting may be involved in the pathophysiological pathways. 6. Limitations There are a number of limitations which have to be taken into account in the present study. Despite the use of SACP via the bicarotid trunk it is not known what percentage of blood flow is drained via the external carotid arteries. On the other hand, angiography indicated that the majority of flow reaches the brain via the internal carotid arteries. Since the vertebral arteries are not perfused in the present model, a patent Circle of Willis is a prerequisite for adequate cerebral perfusion. On the other hand Ye and coworkers were able to show in a swine model, that even unilateral antegrade cerebral perfusion provided uniform flow distribution to both hemispheres of the brain [16]. In this context, we feel that measurement of cerebral blood flow (e.g. by fluoroscopic techniques) may further validate the current model. Finally, sophisticated neuropathological methods have to be validated to assess neurological damage even four hours after HCA or SACP. 7. Conclusions Cold SACP is associated with earlier EEG recovery and lower intracranial pressure during reperfusion. With this technique, the amount of tissue acidosis can be significantly reduced. Further studies in a chronic model are warranted to evaluate the ideal temperature and perfusion mode for antegrade cerebral perfusion. Acknowledgements This study was supported by the German Research Foundation (HA 2971/2-1) and an Internal Grant from Hannover Medical School. The authors want to thank Helmut Preissler from Preissler Medizintechnik GmbH, Augsburg, Germany, for providing the neurophysiological devices for EEG and SSEP analysis. We would like to thank our research coordinators and technicians Astrid Diers-Ketterkatt, Petra Ziehme, Rosie Katt, Karin Peschel and Anja Giese for technical assistance and Kalle Napierski and Paul Zerbe for their care of the animals. References [1] Griepp RB. Cerebral protection during aortic arch surgery. J Thorac Cardiovasc Surg 2001;121: [2] Reich DL, Uysal S, Ergin MA, Griepp RB. Retrograde cerebral perfusion as a method of neuroprotection during thoracic aortic surgery. Ann Thorac Surg 2001;72: [3] Hagl C, Khaladj N, Karck M, Kallenbach K, Leyh R, Winterhalter M, Haverich A. Hypothermic circulatory arrest during ascending and aortic arch surgery: the theoretical impact of different cerebral perfusion techniques and other methods of cerebral protection. Eur J Cardiothorac Surg 2003;24: [4] Di Eusanio M, Wesselink RM, Morshuis WJ, Dossche KM, Schepens MA. Deep hypothermic circulatory arrest and antegrade selective cerebral perfusion during ascending aorta-hemiarch replacement: a retrospective comparative study. J Thorac Cardiovasc Surg 2003;125:

7 C. Hagl et al. / European Journal of Cardio-thoracic Surgery 26 (2004) [5] Hagl C, Ergin MA, Galla JD, Lansman SL, McCullough JN, Spielvogel D, Sfeir P, Bodian CA, Griepp RB. Neurologic outcome after ascending aorta-aortic arch operations: effect of brain protection technique in high-risk patients. J Thorac Cardiovasc Surg 2001;121: [6] Hagl C, Tatton NA, Weisz DJ, Zhang N, Spielvogel D, Shiang HH, Bodian CA, Griepp RB. Cyclosporine A as a potential neuroprotective agent: a study of prolonged hypothermic circulatory arrest in a chronic porcine model. Eur J Cardiothorac Surg 2001;19: [7] Dossche KM, Morshuis WJ, Schepens MA, Waanders FG. Bilateral antegrade selective cerebral perfusion during surgery on the proximal thoracic aorta. Eur J Cardiothorac Surg 2000;17: [8] Kazui T, Kimura N, Yamada O, Komatsu S. Surgical outcome of aortic arch aneurysms using selective cerebral perfusion. Ann Thorac Surg 1994;57: [9] Bachet J, Guilmet B, Goudot B, Termignon G, Dreyfus G, Teodori G. Cold cerebroplegia. A new technique of cerebral protection during surgery of the transverse aortic arch. J Thorac Cardiovasc Surg 1991; 102: [10] Mezrow CK, Midulla PS, Sadeghi AM, Gandsas A, Wang W, Dapunt OE, Zappulla R, Griepp RB. Evaluation of cerebral metabolism and quantitative electroencephalography after hypothermic circulatory arrest and low-flow cardiopulmonary bypass at different temperatures. J Thorac Cardiovasc Surg 1994;107: [11] Hagl C, Tatton NA, Khaladj N, Zhang N, Nandor S, Insolia S, Weisz DJ, Spielvogel D, Griepp RB. Involvement of apoptosis in neurological injury after hypothermic circulatory arrest: a new target for therapeutic intervention? Ann Thorac Surg 2001;72: [12] Juvonen T, Zhang N, Wolfe D, Weisz DJ, Bodian CA, Shiang HH, McCullough JN, Griepp RB. Retrograde cerebral perfusion enhances cerebral protection during prolonged hypothermic circulatory arrest: a study in a chronic porcine model. Ann Thorac Surg 1998;66: [13] Strauch JT, Spielvogel D, Haldenwang PL, Zhang N, Weisz D, Bodian C, Griepp RB. Hypothermic selective cerebral perfusion compared with hypothermic cardiopulmonary bypass on cerebral hemodynamics and metabolism. Eur J Cardio-Thoracic Surg 2003;24: [14] Ehrlich MP, McCullough J, Wolfe D, Zhang N, Shiang H, Weisz D, Bodian G, Griepp RB. Cerebral effects of cold reperfusion after hypothermic circulatory arrest. J Thorac Cardiovasc Surg 2001;121: [15] Hagl C, Khaladj N, Weisz DJ, Zhang N, Guo LJ, Bodian CA, Spielvogel D, Griepp RB. Impact of high intracranial pressure on neurophysiological recovery and behavior in a chronic porcine model of hypothermic circulatory arrest. Eur J Cardiothorac Surg 2002;22: [16] Ye J, Dai G, Ryner LN, Kozlowski P, Yang L, Summers R, Sun J, Salerno TA, Somorjai RL, Deslauriers R. Unilateral antegrade cerebral perfusion through the right axillary artery provides uniform flow distribution to both hemispheres of the brain: a magnetic resonance and histopathological study in pigs. Circulation 1999;100: II Appendix A. Conference Discussion Dr Y. Ueda (Nagoya, Japan): I would like to ask one question about the setting of a control model. Hypothermic circulatory arrest at 20 8C, 100 min, is quite an unusually long period. It produced almost a 100% of neurological damage according to the textbook of Kaplan and Barratt-Boyes. In their books, even at 18 8C, 60 min is almost dangerous for human surgery. So your model of 20 8C and 100 min is quite unfair to compare. You should compare the group of 18 8C, 40-min model compared to the antegrade cerebral perfusion group. How about your comments? Dr Hagl: It is true, that our model is a pretty tough one. It has been originally described by the Mount Sinai group affiliated with Dr Griepp. They gained quite a lot of experience over the years and they were able to demonstrate that you need at least min of hypothermic circulatory arrest at 20 8C to produce some kind of cerebral damage in a porcine model. And even to detect these neurological changes you need sophisticated neurophysiological and histopathological techniques. But on the other hand you need to produce significant damage to detect potential benefits of different cerebral protection methods. You are absolutely correct that the time frame is not really comparable with the clinical situation in human beings, which remains a limitation of our animal model. But we still believe that this model can give us further insight into the pathophysiolgy of hypothermic circulatory arrest and the effects of different potentially neuroprotective techniques. Dr A. Corno (Lausanne, Switzerland): First of all, you should be congratulated for your effort to move the surgeons away from deep hypothermic circulatory arrest towards a more physiologic perfusion. Two questions. First, why did you use in your model of circulatory arrest the alphastat instead of the ph-stat, since there are several recent reports about better cerebral protection given by ph-stat in both experimental and clinical studies? Second question, if you want to go far from deep hypothermic circulatory arrest, why do you use cold antegrade cerebral perfusion instead of normothermic? Dr Hagl: Concerning the first question you are right, there is some controversial discussion regarding the value of ph or alpha-stat in this setting. Actually, we have adjusted our model on our clinical situation in adults, where we exclusively use alpha-stat management. But there are a number of experimental and clinical reports in the literature, especially from the Boston group, who demonstrated positive effects of a ph-stat regime. This seems to be especially true in pediatric cardiac surgery. From the pathophysiological standpoint we know that ph-stat causes a breakdown of the autoregulation of the brain. This may be beneficial during cooling since you get more sustained and thorough cooling of the body and a more complete reduction of tissue metabolism. On the other hand during reperfusion you may increase intracranial pressure due to luxury perfusion and you also increase the embolic load. The solution may be using ph-stat during cooling and alpha-stat during reperfusion. But for this you need very experienced pump technicians and anesthesiologists to avoid confusion in the OR. Dr Corno: The second question is, if you want to move more towards better physiological perfusion, why are you still using cold perfusion instead of warm? Dr Hagl: That is a good question. Actually, we do not know what is the ideal temperature for cerebral perfusion and at what temperature can we stop the pump without doing harm to our patients. If you look in the literature, you will realize that there are a number of different concepts, but most of them are empirical. If you asked Dr Schepens, who is also in the audience, he will tell you that you can use relatively warm cerebral perfusion and the results of his clinical studies are really favourable and seem to support this thesis. We strongly feel, that our model may offer the opportunity to study exactly these important questions. May be, I can answer some of your questions in the next year. Did you plan to harvest the brain to measure the extent of neuronal apoptosis and necrosis and caspases activation in that model? Dr Hagl: Yes, actually we did that. But we are still fighting the problem that standard histopathology did not show any differences in the brains after four hours. Even with sophisticated apoptosis staining, and we did that in these brains, it is not that easy to find any differences. Due to these disappointing results we look now for something like heat shock protein, which has been shown to be involved in the pathophysiology of brain ischemia and which seems to be good marker even four hours after circulatory arrest. Mr Lang-Lazdunski: Actually, MAP-2 immunohistochemistry is an excellent technique because even after 3 h you can see the damage in the brain. Dr Hagl: Thanks for this comment. Actually, we have no experience with this marker, but I am eager to learn more about it. I would appreciate if we can discuss the issue further after the session.

8 80 C. Hagl et al. / European Journal of Cardio-thoracic Surgery 26 (2004) Dr J. Bachet (Paris, France): This beautiful scientific study demonstrates what we intuitively assumed when we published 15 years ago the technique of cold blood cerebral perfusion. If I remember well, all your pigs were put in general deep hypothermia. You are aware that the advantage of selective cerebral perfusion is, first, to implement the only way to correctly protect the brain during arch exclusion, but secondly, to get rid of all the drawbacks of deep hypothermia. My question is: do you think that you could do the same experiment with selectively perfusing the brain in moderate hypothermia of the whole body, and do you think that you could find exactly the same results, or even better results? Dr Hagl: Actually, I cannot tell you if we may find better results. I would speculate, that there is a difference in one or the other way, because we know that activation of the inflammatory system also depends on temperature. But there also may be another problem in this setting. As Dr Ueda mentioned before, 100 min of no flow perfusion in the lower body at 25 or even 28 8C, is quite a long time for no perfusion. You may get significant activation of acute phase reactants, causing hemodynamic instability, extravasation of protein and fluids, pulmonary edema and other problems. But as I mentioned before, we are open for all kind of suggestions and we may incorporate your idea in our next study.