Embolization is an important cause of cardiac surgical

Transcription

1 Effect of Temperature and PaCO 2 on Cerebral Embolization During Cardiopulmonary Bypass in Swine David J. Cook, MD, Walter Plöchl, MD, and Thomas A. Orszulak, MD Department of Anesthesiology and the Division of Cardiovascular Surgery, Department of Surgery, Mayo Foundation and Mayo Clinic, Rochester, Minnesota Background. Patients experience cerebral embolization during cardiopulmonary bypass (CPB). This study determined if alterations in temperature and/or PaCO 2 can reduce cerebral and ocular embolization. Methods and Results. Forty-four pigs underwent CPB: 24 animals at 28 C, and 20 at 38 C. The two temperature groups were randomized to undergo embolization (67- m fluorescent microspheres) at either hypercarbia or hypocarbia. Before and after embolization, cerebral and ocular blood flow were determined at normocarbia. Reducing temperature or PaCO 2 reduced cerebral and ocular embolization. Hypocarbia reduced cerebral embolization by 60% and 45% in normothermic and hypothermic groups, respectively (p < and p < 0.05). Relative to normothermic animals, hypothermia reduced cerebral embolization by 37% under an elevated CO 2 condition (p < 0.05), but not under hypocarbic conditions. Similarly, regardless of temperature, fewer emboli were delivered to the eye in hypocarbic animals (p < 0.05), but hypothermia did not reduce ocular embolization. Conclusions. Cerebral embolization is determined by both temperature and PaCO 2 at the time of embolization. In CPB practice, reductions in temperature and/or PaCO 2 during periods of embolic risk may reduce brain injury. (Ann Thorac Surg 2000;69:415 20) 2000 by The Society of Thoracic Surgeons Embolization is an important cause of cardiac surgical neurologic morbidity. While control of embolic risk in the postoperative period is challenging, a variety of intraoperative interventions may reduce embolic risk. Increased surgical attention to the atherosclerotic aorta, echocardiographic assessment of de-airing, and technical innovations such as arterial line filters and membrane oxygenators should improve patient outcomes. Theoretically, simple physiologic interventions might also be used to reduce cerebral embolic risk in the operative period. Transcranial Doppler and echocardiographic data indicate that embolic risk during cardiopulmonary bypass (CPB) is clustered into fairly specific periods [1 3]. Manipulation of the aorta and the initial phases of ventricular ejection are primary periods of embolic risk, and during these times, exquisite control of physiologic variables is possible. In the nonbypass period, PaCO 2 is easily manipulated, and during CPB, PaCO 2, and temperature might both be rapidly altered to reduce delivery of emboli to the central nervous system (CNS). We recently demonstrated in an animal model that acute manipulations of CO 2 could reduce cerebral embolization by more than 50% during normothermic CPB [4]. Manipulation of temperature with or without alterations in PaCO 2 should also be effective. The aim of this study is Accepted for publication July 3, Address reprint requests to Dr Cook, Department of Anesthesiology, Mayo Clinic, 200 First St SW, Rochester, MN 55905; cook.david@mayo.edu. to determine the effect of hypothermia and PaCO 2 on cerebral embolization and compare those results with those obtained under normothermic conditions. Material and Methods After Institutional Animal Care and Use Committee approval, 44 pigs weighing kg were studied. Twenty-four pigs undergoing CPB at 28 C were compared with 20 previously reported pigs undergoing CPB at 38 C [4]. All 44 experiments were conducted during a 7-month period by the same personnel. Pigs were premedicated with IM telazol (4 mg/kg) and xylazine (2 mg/kg). Anesthesia was induced with 2% halothane and maintained with fentanyl (0.7 g/kg/min) and ketamine (28 g/kg/min). Muscle relaxation was obtained with pancuronium bolus (0.2 mg/kg IV) followed by an infusion (0.3 g/kg/min 1 ). Four-inch femoral artery catheters were placed for mean arterial blood pressure (MAP) measurements and blood sampling. A left thoracotomy was performed. Venous drainage to the CPB circuit was by a right atrial 36-Fr cannula. The blood was circulated by a centrifugal pump through a combined heat exchanger-oxygenator (Bentley Spiral Gold, Irvine, CA) and returned via a 4.5-mm ID aortic root cannula. CPB was initiated and nasopharyngeal temperature was maintained at either 28 C or 38 C depending on study group, Hgb at 7 to 8 g/dl, PaCO 2 at 35 to 40 mm Hg, and PaO 2 more than 150 mm Hg. MAP was maintained at 65 to 75 mm Hg by altering bypass pump 2000 by The Society of Thoracic Surgeons /00/$20.00 Published by Elsevier Science Inc PII S (99)

2 416 COOK ET AL Ann Thorac Surg TEMPERATURE, CO 2, AND EMBOLIZATION DURING CPB 2000;69: Table 1. Systemic Physiologic Values in Hypothermic (28 C) Animals During the Three Study Periods a Study Period Group Temperature ( C) MAP (mm Hg) Pump Flow (L/min/m 2 ) Hgb (g/dl) PaO 2 (mm Hg) PaCO 2 (mm Hg) ph Preembolization H L Embolization H b b L Postembolization H L a Values are mean SD; n 12 per group. b p 0.05 between groups by repeated-measures analysis of variance. H high PaCO 2 ; Hgb hemoglobin; L low PaCO 2 ; MAP mean arterial blood pressure. flow rate. When steady-state CPB conditions were reached, cerebral blood flow (CBF) was determined at normocarbia. Then, after equal random assignment, PaCO 2 was elevated or reduced to 50 to 55 mm Hg (group H,) or 25 to 30 mm Hg (group L), and an embolic load was administered into the aortic cannula. Approximately 30 minutes after embolization, when PaCO 2 was normalized, the postembolic CBF was measured to determine the effect of embolization on regional blood flow. CBF measurements were made with 15- m red (excitation/ emission wavelengths 580/605 nm) and yellow green (505/515 nm) fluorescent polystyrene microspheres (Molecular Probes, Eugene, OR), using the blood reference sample method [5, 6]. Four million microspheres were injected over 60 seconds into the aortic inflow line distal to the arterial filter. For the reference sample, blood was drawn from the femoral artery catheter by a Harvard withdrawal pump at 4.9 ml/min. The embolic load ( m orange [540/560 nm] fluorescent polystyrene microspheres; Molecular Probes) was diluted in 9 ml of 6% Dextran 70 with 0.025% Tween 80, sonicated, vortexed, and injected over 5 minutes [4]. At completion, the heart was fibrillated, and the brain and eyes were removed. Tissue samples ( 1 g) of left and right temporal and occipital lobes, thalamus, internal capsule, and cerebellar hemispheres were obtained. Microspheres were recovered from tissue by the sedimentation method [4, 5] and from blood using a commercial protocol (NuFlow Extraction Protocol ; Interactive Medical Technology, West Los Angeles, CA). The fluorescence in tissue and blood was determined by spectrofluorometry (SLM 8100; SLM-AMINCO, Rochester, NY). Cerebral and ocular blood flow and embolization were determined using previously described techniques [4 6]. Statistical Analysis Systemic physiologic data for preembolization, embolization, and postembolization periods were analyzed using two-factor repeated measures analysis of variance. For these models, the physiologic variable was the dependent variable, treatment group was the independent cross-classification factor, and time was the repeated factor. Linear regression was used to test for equal distribution of microspheres between left and right sides of the brain. To test if the regression line was different from the identity line, we performed individual t tests of intercept 0 and slope 1 along with the simultaneous F test for the identity line. All subsequent analyses for regional CBF and embolization were performed using the mean across left and right sides of the brain. Regional blood flow and embolization values for occipital and temporal lobes were combined and are presented as cortical grey. In addition, the mean number of emboli delivered to circumferential arterial territories (cortical grey and cerebellum) and penetrating arterial territories (internal capsule and thalamus) was determined. For each brain region, preembolic and postembolic blood flows were compared using the paired t test, and the two treatment groups were compared using the twosample t test. The two-sample rank sum test was used to compare embolization between group H and L at each temperature, as well as for between-temperature comparisons. In addition, with PaCO 2 treated as a continuous variable, linear regression was used to assess the association of embolization and PaCO 2. All data are presented as mean SD. A p value less than 0.05 was considered significant. Results In all 44 animals, there was no left to right lateralization of CBF or emboli. Paired preembolic (r 0.98) and postembolic (r 0.98) CBF values, as well as emboli counts (r 0.90), were well matched (y intercepts not significantly different from 0, slope not significantly different from 1), indicating adequate mixing of microspheres in the aortic root. Therefore, only the mean values for regional blood flows and embolization are presented. Hypothermic CPB (Groups 28H and 28L) Systemic physiologic data for the three study periods in the 24 hypothermic animals are presented in Table 1. Except for PaCO 2 (and ph), these did not differ between the groups 28H and 28L during any study period. At 28 C at normocarbia, the mean CBF over all brain regions did not differ between groups 28H and 28L (Table 2), and predicted differences in CBF between brain regions such as grey and white matter were demonstrated in both groups (cortical grey vs internal capsule: p for group H and L).

3 Ann Thorac Surg COOK ET AL 2000;69: TEMPERATURE, CO 2, AND EMBOLIZATION DURING CPB 417 Table 2. Cerebral and Ocular Blood Flow in Animals at 28 C Before and After Embolization at Normocarbia Group H Group L Region Preembolic flow Preembolic Flow (ml/100 g/min) Postembolic p Value a (ml/100 g/min) Postembolic Flow p Value a Cortical grey Internal capsule Thalamus Cerebellum Total brain Eye a Paired t test p value value comparing preembolic and postembolic flow within groups. Values are mean SD. Group H high PaCO 2 ; Group L low PaCO 2. n 12 per group. The difference in PaCO 2 between groups 28H and 28L (Table 1) resulted in a markedly reduced cerebral embolization in group 28L ( p 0.05) (Fig 1). A significant correlation existed between cerebral embolization and the PaCO 2 (r 0.53, p 0.01) (Fig 2). However, significant differences were not seen in all brain regions. Within groups, more emboli were found in circumferential (cortical grey and cerebellum) than in penetrating arterial territories (internal capsule and thalamus) (Fig 3). In group 28H, the mean number of emboli/g was in circumferential and in penetrating arterial territories ( p 0.01). In group 28L, the mean number of emboli per gram was in circumferential and 10 6 in penetrating arterial territories ( p 0.01). Embolization decreased total postembolic CBF by approximately 10% in the high-paco 2 animals ( p 0.049; Table 2), although only the cortical grey region showed a statistically significant ( p 0.026) decline. In the low- PaCO 2 group at 28 C, neither any region nor the total brain showed a decrease in CBF postembolization. At 28 C, 62% fewer emboli were delivered to the eyes of hypocarbic animals. As in the brain, a correlation existed between ocular embolization and PaCO 2 (r Fig 1. The effect of temperature on cerebral embolization at high and low PaCO 2.*p 0.05 by two-sample rank sum test between high and low CO 2 groups at a given temperature (n 12 per group at 28 C; n 10 per group at 38 C); # p 0.05 between animals at 38 C and 28 C at a given PaCO , p 0.013) (Fig 2). After embolization, OBF decreased by 16% and 18% in groups 28H and 28L, respectively, but this did not reveal statistical significance (Table 2). Normothermic (Groups 38H and 38L) vs Hypothermic CPB (Groups 28H and 28L) With the exception of pump flow ( p 0.001) and temperature ( p ), systemic physiologic variables at embolization did not differ between the two temperature groups. At embolization, mean nasopharyngeal temperatures were (group 38H) and C (group 38L) and (group 28H) and C (group 28L). The mean pump flow was (group 38H) and L/min/m 2 (group 38L), and in both 28 C groups. Pump flows and temperatures did not differ within the temperature groups. The other physiologic variables, MAP, Hgb, PaO 2, and ph, did not differ between normothermic and hypothermic animals under a given study condition. The temperature difference between groups resulted in a higher initial CBF in the normothermic animals in all regions. The mean CBF at 38 C and 28 C in group H animals was and ml/100 g/min ( p 0.001), respectively. In group L animals at 38 C and 28 C, the mean CBF was and 30 9 ml/100 g/min, respectively ( p 0.01). At both temperatures, there was a significant effect of PaCO 2 on cerebral embolization ( p 0.001) (Figs 1 and 2) and an independent effect of temperature ( p 0.03) on cerebral embolization. Under hypercarbic conditions, hypothermia reduced total cerebral embolization by a mean of 26% ( p 0.05). Total embolization was lowest in the hypothermic and hypocarbic group (28L), but the total embolization under this condition did not differ from the hypocarbic, normothermic group (38L). The number of emboli/g of brain was 19 7 (group 38L) and 16 8 (group 28L) ( p 0.40). This suggests that the response to changes in PaCO 2 is attenuated at lower bypass temperatures, but a statistically significant interaction between PaCO 2 and temperature was not demonstrated ( p 0.11). There was also a significant effect of PaCO 2 ( p 0.01) on ocular embolization, but in contrast to the brain, temperature did not effect ocular blood flow or emboli-

4 418 COOK ET AL Ann Thorac Surg TEMPERATURE, CO 2, AND EMBOLIZATION DURING CPB 2000;69: In hypothermic CPB practice, the principal debate over PaCO 2 has been regarding the advantages or disadvantages of -stat or ph-stat management. With the ph-stat strategy, PaCO 2 is elevated and the CBF is significantly higher [8, 9]. It has been speculated that this might increase cerebral embolization [10]. The same discussion is relevant to CPB temperature. Because 10 C of hypothermia may reduce CBF by 40% to 60%, cerebral embolization should be proportionately reduced during hypothermia. At normothermia, a 25-mm Hg reduction in PaCO 2 decreased cerebral embolization by approximately 60% [4]. Here, we demonstrate that at 28 C, a 28-mm Hg change in PaCO 2 reduced whole brain cerebral embolization by 45%. These data provide further evidence that hypocarbia can reduce cerebral embolization during periods of embolic risk. Furthermore, these findings favor -stat management during CPB and may help explain reports of better neurologic outcome with -stat management [9]. These data also provide an argument (although quantitatively and practically less compelling), in favor of moderate hypothermia during periods of embolic risk. Hypothermia reduces cerebral embolization, but during hypothermia, hypocarbia appears less effective. During hypothermia, hypocarbia significantly reduced cerebral embolization circumferential, but not penetrating, arterial territories (Fig 3). In contrast, during normothermia, embolization to both circumferential and penetrating arterial territories is affected by PaCO 2. Our data do not provide a clear explanation for this finding, although a variety are possible. First, this result may be a function of group size. Experimental variability may have its greatest statistical impact on tissues receiving the fewest absolute numbers of emboli. Although the changes were not statistically significant, in hypothermic animals, the mean reduction in embolization to internal capsule and thalamus was 55% and 29%, respectively. A slightly larger group size may have been necessary to assess the effect of PaCO 2 regions receiving the fewest Fig 2. The relationship between PaCO 2 and cerebral (squares) and ocular (diamonds) embolization in groups 28H and 28L. A significant correlation exists between the PaCO 2 and the number of emboli delivered per gram of brain and ocular bulb (n 12 per group). zation ( p 0.39). As for the brain, there was no significant interaction between PaCO 2 and temperature ( p 0.59). Comment Cerebral embolism is a primary cause of neurologic injury after cardiac surgery [3, 7]. Hypocarbia at the time of embolization markedly decreases cerebral embolization during normothermic CPB [4]. Most CPB is, however, conducted with mild to moderate hypothermia. The aim of this study was to determine the effect of hypothermia and PaCO 2 on cerebral embolization and compare those results with those under normothermic conditions. Fig 3. The effect of PaCO 2 on regional embolization in groups 28H (filled squares), and 28L (open squares). *p 0.05 by the twosample rank sum test (n 12 per group).

5 Ann Thorac Surg COOK ET AL 2000;69: TEMPERATURE, CO 2, AND EMBOLIZATION DURING CPB 419 emboli. Alternatively, a diminished effect of hypocarbia during hypothermia could be a function of regional differences in CO 2 or temperature responsiveness [11, 12]. Finally, under hypothermic conditions, there may be a greater trapping of emboli in circumferential territories [4, 13] such that the CO 2 effect in penetrating arterial territories is minimized. Studies using retinal angiography indicate that ocular embolization contributes to visual deficits after cardiac surgery [14]. As the eye is an outgrowth of the brain, PaCO 2 is a determinant of uveal and retinal blood flow [15], and one would hypothesize that ocular embolization could be reduced by hypocarbia. We found this to be true during normothermic CPB [4], and this study indicates the protective effect of hypocarbia during hypothermia as well. It is important to note that while hypothermia and/or hypocarbia reduced cerebral embolization, the decrease we identified was less than expected, particularly in the group that was both hypothermic and hypocarbic. While regional differences in blood flow response to CO 2 or temperature [11, 12] may account for some of this attenuation, there may be another reason indirectly related to CPB temperature. To maintain an equal MAP in normothermic and hypothermic animals, the total CPB flow was approximately 40% lower in the cold groups ( p 0.01). Lower pump flows at the time of embolization may increase cerebral embolization [16], because as pump flow is reduced a greater percentage of that flow will go to the cerebral circulation [17, 18]. Therefore, if a given embolic load enters the aortic root, a greater percentage of emboli may be delivered to the brain under lower CPB flow conditions [16]. This may have minimized some of the expected differences in embolization between normothermic and hypothermic animals. This study has a number of limitations. First, animals were not randomized to their temperature group and our results with normothermic animals were previously reported [4], so the criticism of using a historical control might be levied. The potential error with a historical control is that a change of technique may occur over time potentially altering results in the later group. While this is generally possible, all of these experiments were conducted over a 7-month period using the same techniques and personnel. Additionally, hypothermic studies were initiated before the normothermic studies were completed. Therefore, the criticism of having done the normothermic group first is essentially obviated. Finally, we believe repeating the normothermic animals for the purposes of this report would have been an unjustifiable consumption of animals. Second, our embolus model might be criticized. Clinically, cerebral emboli consist of a variety of compositions and sizes from which uniform polystyrene emboli differ. Nevertheless, a 67- m fluorescent microsphere is an excellent model. They approximate the size of emboli that occur during CPB [2, 19], and detection techniques allow us to detect as few as three emboli per gram of tissue. As such, the total embolization in this experiment approximates [2], or is less than [4], what may occur during clinical CPB. Additionally, the results of this study should not differ if emboli of different composition or sizes were used. Temperature and PaCO 2 should affect a wide range of emboli types and sizes similarly. While the etiology of postcardiac surgical neurologic dysfunction is multifactorial, cerebral embolization is a primary determinant. Importantly, most of cerebral embolization occurs during specific CPB surgical events [1 3], so attention to these periods should reduce embolic risk. Surgical management of the aorta [20] can reduce systemic embolization. This laboratory study demonstrates that physiologic interventions, hypothermia, and hypocarbia are also effective in reducing cerebral embolization. While clinical confirmation is indicated, these interventions can be rapidly achieved with modern CPB equipment and can be directed to periods of greatest risk. These interventions are simple, effective, and essentially without cost and may practically impact on patient outcomes. Dr Walter Plöchl is the recipient of an Erwin-Schrödinger research fellowship from the Austrian Science Foundation, Vienna, Austria. This work was supported by the American Heart Association Minnesota Affiliate and the Mayo Foundation. References 1. Van der Linden J, Casimir-Ahn H. When do cerebral emboli appear during open heart operations? A transcranial Doppler study. Ann Thorac Surg 1991;51: Barbut D, Yao FSF, Lo YW, et al. Determination of size of aortic emboli and embolic load during coronary artery bypass grafting. Ann Thorac Surg 1997;63: Clark RE, Brillman J, Davis DA, Lovell MR, Price TR, Magovern GJ. Microemboli during coronary artery bypass grafting. Genesis and effect on outcome. J Thorac Cardiovasc Surg 1995;109: Plöchl W, Cook DJ. Quantification and distribution of cerebral emboli during cardiopulmonary bypass in the swine: the impact of PaCO 2. Anesthesiology 1999;90: Van Oosterhout MF, Willigers HM, Reneman RS, Prinzen FW. Fluorescent microspheres to measure organ perfusion: validation of a simplified sample processing technique. Am J Physiol 1995;269:H Glenny RW, Bernard S, Brinkley M. Validation of fluorescent-labeled microspheres for measurement of regional organ perfusion. J Appl Physiol 1993;74: Pugsley W, Klinger L, Paschalis C, Treasure T, Harrison M, Newman S. The impact of microemboli during cardiopulmonary bypass on neuropsychological functioning. Stroke 1994;25: Prough DS, Stump DA, Roy RC, et al. Response of cerebral blood flow to changes in carbon dioxide tension during hypothermic cardiopulmonary bypass. Anesthesiology 1986; 64: Venn GE, Patel RL, Chambers DJ. Cardiopulmonary bypass: perioperative cerebral blood flow and postoperative cognitive deficit. Ann Thorac Surg 1995;59: Henriksen L, Hjelms E, Lindeburgh T. Brian hyperperfusion during cardiac operations. Cerebral blood flow measured in man by intra-arterial injection of xenon 133: evidence suggestive of intraoperative microembolism. J Thorac Cardiovasc Surg 1983;86: Palmer C, Vannucci RC, Christensen MA, Brucklacher RM. Regional cerebral blood flow and glucose utilization during hypothermia in newborn dogs. Anesthesiology 1989;71:730 7.

6 420 COOK ET AL Ann Thorac Surg TEMPERATURE, CO 2, AND EMBOLIZATION DURING CPB 2000;69: Busija DW, Leffler CW. Hypothermia reduces cerebral metabolic rate and cerebral blood flow in newborn pigs. Am J Physiol 1987;253:H Macdonald RL, Kowalczuk A, Johns L. Emboli enter penetrating arteries of monkey brain in relation to their size. Stroke 1995;26: Blauth CI, Arnold JV, Schulenberg WE, McCartney AC, Taylor KM, Loop FD. Cerebral microembolism during cardiopulmonary bypass. Retinal microvascular studies in vivo with fluorescein angiography. J Thorac Cardiovasc Surg 1988;95: Bill A. Blood circulation and fluid dynamics in the eye. Physiol Rev 1975;55: Sungurtekin H, Plöchl W, Cook DJ. Relationship between cardiopulmonary bypass flow rate and cerebral embolization in dogs. Anesthesiology 1999;91: Schwartz AE, Sandhu AA, Kaplon RJ, et al. Cerebral blood flow is determined by arterial pressure and not cardiopulmonary bypass flow rate. Ann Thorac Surg 1995; 60: Cook DJ, Proper JA, Orszulak TA, Daly RC, Oliver WC, Jr. Effect of pump flow rate on cerebral blood flow during hypothermic cardiopulmonary bypass in adults. J Cardiothorac Vasc Anesth 1997;11: Moody DM, Brown WR, Challa VR, Stump DA, Reboussin DM, Legault C. Brain microemboli associated with cardiopulmonary bypass: a histologic and magnetic resonance imaging study. Ann Thorac Surg 1995;59: Wareing TH, Davila-Roman VG, Daily BB, et al. Strategy for the reduction of stroke incidence in cardiac surgical patients. Ann Thorac Surg 1993;55: INVITED COMMENTARY In this study, Cook and coworkers explore the effects of core temperature and PaCO 2 on cerebral embolization of polystyrene microspheres during cardiopulmonary bypass. The authors performed this study to determine if simple physiologic interventions (essentially moderate hypothermia and -stat management instead of ph-stat management) can decrease the incidence of neurologic injury. If the results of this paper can be reproduced, and are applicable in humans, a significant improvement in morbidity might result from a small change in patient management during periods of embolic risk. The study does have some minor problems that will have to be resolved with additional research. The authors do not control cardiopulmonary bypass flow rates between normothermic and hypothermic groups. Differences in flow rates could possibly affect the amount of cerebral embolization that occurred during the study. The use of historical controls is another potential problem. Of course, the results in this study must be verified in both animal studies and controlled human trials before making changes in treatment. Despite these issues, however, Cook and coworkers describe a simple physiologic intervention that may decrease the incidence of potentially devastating neurologic injury. Keith J. Ruskin, MD Department of Anesthesiology Yale University School of Medicine 333 Cedar St, PO Box New Haven, CT by The Society of Thoracic Surgeons /00/$20.00 Published by Elsevier Science Inc PII S (99)01512-X