| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Ann Thorac Surg 2005;79:1597-1605
© 2005 The Society of Thoracic Surgeons
a Department of Cardiovascular and Thoracic Surgery, University Hospital of Angers, Angers, France
b Clinical and Research Center for Memory Disorders and Degenerative Diseases and Laboratory of Psychology, University Hospital of Angers, Angers, France
c Laboratory of Immunology, University Hospital of Angers, Angers, France
d Department of Anaesthesiology, University Hospital of Angers, Angers, France
Accepted for publication August 23, 2004.
* Address reprint requests to Dr Baufreton, Dept of Cardiovascular and Thoracic Surgery, University Hospital of Angers, 4 rue Larrey, Angers, France; (E-mail: chbaufreton{at}chu-angers.fr).
 |
Abstract |
|---|
 Top Abstract IntroductionMaterial and MethodsResultsCommentAcknowledgmentsReferences |
|---|
METHODS: We randomly assigned 30 patients to undergo surgery using either standard noncoated or heparin-coated extracorporeal circuits. Closed cardiopulmonary bypass and controlled suctions of pericardial shed blood were standardized in both groups. Complement activation and cellular brain injury were assessed by measuring sC5b-9 and protein s100ß. Neuropsychometric tests were performed at least 2 weeks before operation and at discharge. They served to calculate z scores of cognitive domains and changes in neuropsychological functioning.
RESULTS: Peak value of sC5b-9 at the end of cardiopulmonary bypass in the noncoated group was significantly higher than in the heparin-coated group (p = 0.005). Changes in the heparin-coated group were not significant. Glial injury started after initiation of surgery and peaked at the end of cardiopulmonary bypass with significantly higher concentration of s100ß in the noncoated than in the heparin-coated group (p = 0.008). Values of s100ß and of sC5b-9 were significantly correlated (p = 0.03). Although no statistically significant between group difference was detected, z scores of attention and flexibility or executive functions were lowered postoperatively within the noncoated group (p = 0.033 and p = 0.028), whereas z scores were unchanged within the heparin-coated group.
CONCLUSIONS: Inhibition of complement activation by heparin-coated cardiopulmonary bypass reduced brain cell injury and was associated with preserved neuropsychological functioning after coronary artery bypass grafting.
|
Introduction |
|---|
|
TopAbstractIntroductionMaterial and MethodsResultsCommentAcknowledgmentsReferences |
|---|
Because inflammatory response was expected to contribute to neurologic injury after cardiac surgery, we postulated that minimizing the whole body systemic inflammatory response would contribute to the reduction of cerebral damage. Among the different components of the systemic inflammatory response, complement activation exerts its deleterious effects through biologic cascades that may finally result in poor clinical outcome after CPB [11]. C5a receptor expression on astrocytes plays an important role in controlling the inflammation in the brain and may be a central component of complement-mediated brain injury [12]. Heparin-coated circuits, used for CPB, reduce the complement activation as reflected by lower postoperative levels of sC5b-9 [13]. The goal of this prospective randomized study was to assess the influence of reduced complement activation on s100ß release by using heparin-coated CPB in patients undergoing coronary artery bypass grafting with a setting of controlled suction throughout the surgical procedure. In addition, we focused our attention on the postoperative CNS dysfunction of these patients using neuropsychometric tests.
|
Material and Methods |
|---|
|
TopAbstractIntroductionMaterial and MethodsResultsCommentAcknowledgmentsReferences |
|---|
Patients were randomly assigned by computerization (StatMate 1.01; GraphPad, San Diego, CA) to receive noncoated (NC) extracorporeal circuits or heparin-coated (HC) extracorporeal circuits for the CPB procedure. Investigators not present in the operating room but who were involved in the biochemical and the neuropsychological assessment were blinded to the randomization.
Anesthesia and Hemodynamics
Anesthesia was induced and maintained by weight-related doses of sufentanil, midazolam, and pancuronium bromide (total doses < 250 µg, 15 mg, and 16 mg respectively). Tranexamic acid (Exacyl; Sanofi Winthrop, Paris, France) was used routinely (100 mg/kg if creatine blood level was <100 µmol/L; 50 mg/kg otherwise). Monitoring included a radial arterial line, a pulmonary catheter for continuous cardiac output measurement (Swan-Ganz CCO/VIP; Baxter, Irvine, CA), and pressure monitoring kits (Baxter).
Surgical Procedure and Transesophageal Echocardiography
All the patients were operated on by the same surgeon. During the internal thoracic artery takedown procedure, an aortic ultrasound scanning was performed using a transesophageal echocardiography probe to determine the severity of the ascending aorta arteriosclerosis. Each patient was graded according to a standardized system as follows: grade 1, normal to mild intimal thickening; grade 2, severe intimal thickening; grade 3, atheroma protruding less than 5 mm into the lumen; grade 4, atheroma protruding 5 mm or more into the lumen; grade 5, atheroma with a mobile component [16]. Transesophageal echocardiography served also to determine the appropriate sites of cannulation and proximal anastomoses. The CPB was installed between an arterial catheter in the ascending aorta and a venous catheter using a two-stage cannula in the right atrium. The proximal anastomoses were performed first on the ascending aorta before initiation of CPB, and the distal anastomoses were completed during cardioplegic arrest. Myocardial protection was kept constant throughout the period of investigation and was achieved by cold-blood antegrade cardioplegia (2:1 blood cardioplegia with St. Thomas's Hospital solution, SLF 103; Aguettant, Lyon, France). Reperfusion with warm-blood cardioplegia was performed before release of the aortic cross-clamp.
Cardiopulmonary Bypass
Normothermic CPB (>36°C on the venous return) was performed using a centrifugal pump (Biopump; Medtronic Bio-Medicus Inc, Eden Prairie, MN), and a flow rate of 2.4 L · min–1 · m–2 was maintained. The circuit consisted of polyvinylchloride tubings, closed venous reservoir, hollow-fiber oxygenator (Maxima Carmeda; Medtronic, Anaheim, CA), cardiotomy reservoir (Intersept, Medtronic), and an arterial line filter (M 440, Medtronic). In the HC group, the circuit was treated with end point–attached covalently bonded heparin except for the cannulas (CBAS Carmeda; Medtronic, Kerkrade, the Netherlands).
The circuit was primed with 1,000 mL of Ringer's lactate solution and 500 mL gelatin (Gelofusine, B. Braun Medical SA, Paris, France). When necessary, Ringer's lactate solution (Fresenius Kabi, Sèvres, France) was added to the circuit to maintain filling volume. After the patient was disconnected from CPB, the residual blood remaining in the extracorporeal circuit was recirculated through a cell-saving device (Compact A; Dideco, Mirandola, Italy), and the autologous blood was thereafter transfused to the patient. Cardiotomy suction was never used, and the mediastinal shed blood was systematically processed using the cell-saving device before retransfusion throughout the surgical procedure.
A full anticoagulation protocol was applied through administration of a heparin dose based on the patient weight (250 IU/kg with 100 IU/kg in the prime). The level of anticoagulation was controlled by monitoring the activated clotting time (Hemocron 801; International Technidyne Corporation, Edison, NJ), and additional heparin was administered if the activated clotting time was less than 450 seconds. After cessation of CPB, 1 mg of protamine (Protamine Choay, Sanofi Winthrop) for every 100 IU of heparin (of the initial dose) was administered intravenously.
Blood Sampling and Biochemical Analyses
Serial arterial blood samples were collected for determination of blood levels of sC5b-9, the marker of terminal complement complex activation, and s100ß as a marker of glial damage.
For sC5b-9, blood samples were collected after induction of anesthesia before skin incision, after the end of CPB, and at 24 hours postoperatively. For s100ß protein, blood samples were collected after induction of anesthesia, before starting CPB, after the end of CPB (within the first minute), 4 hours after the initiation of CPB in intensive care unit, and at 24 hours postoperatively.
Blood samples were collected into sterile vacuum tubes for s100ß and with ethylenediaminetetraacetic acid for sC5b-9 measurements and immediately centrifuged. Aliquots of plasma were stored at –80°C and subsequently assayed in duplicate for each marker according to the manufacturer's procedure. sC5b-9 was measured by enzyme-linked immunosorbent assay from Quidel (InGen, Rungis, France) and s100ß by a commercially available, monoclonal, two-site immunoluminometric assay (LIA-mat S100; Nichols Institut Diagnostics, Paris, France).
Neuropsychometric Tests
Participants completed a battery of psychological and cognitive performance evaluations. Each patient completed the test battery at least 2 weeks before surgery, to avoid as much as possible anxiety influence near the operation, and again before discharge at postoperative day 7. Testing was carried out by 2 trained neuropsychologists blinded to the surgical technique used, in an office of the neuropsychological department.
A questionnaire was used to obtain demographic information (age, sex, education, and handedness). The short form of the Beck Depression Inventory [17] and the Goldberg Rating Anxiety Scale [18] were administered to determine the presence and intensity of depression and anxiety at the time of testing, before and after the surgery. Memory complaints and dysexecutive complaints were also quantified preoperatively and postoperatively by asking the participants to complete the Cognitive Difficulties Scale [18] and the dysexecutive questionnaire [19], respectively.
The battery of cognitive performance evaluations consisted of standardized neuropsychological tests. The tests chosen were consistent with the published recommendations of the consensus conference on assessment of neurobehavioral outcomes after cardiac surgery [20]. The tests were presented in the same order at both sessions. Eight cognitive domains were tested using specific neuropsychological tests [21] as follows: working memory, attention, flexibility or executive functions, immediate and delayed verbal memory, immediate and delayed visual memory, visuoconstruction, psychomotor processing speed, and fine motor speed. Details of neuropsychometric tests are provided in the Table 1.
|
Statistics
sC5b-9 has been demonstrated to be reduced by 50% after heparin-coated CPB [22]. Based on the hypothesis that neurologic injury after CPB could be of inflammatory origin through complement activation, we postulated that s100ß release would be decreased by the same extent. Previous data from a clinical report including the use of a cell-saving device [9] were considered to calculate the sample size. A power analysis determined that a sample size of 14 patients or more in each group would be adequate to detect a 50% difference in s100ß between the two groups with a power of 80% at the p level of significance of less than 0.05 (GraphPad; StatMate 1.01, San Diego, CA). Therefore, 30 patients were included in the study, including 2 additional patients (1 in each group) to compensate for possible further withdrawal.
Data were stored and analyzed using the SPSS software package (SPSS 11.5 for Windows; SPSS Inc, Chicago, IL). Data are presented as mean ± standard deviation unless otherwise indicated. A 
2 test with Fisher's exact test was used for categorical data if appropriate. The Mann-Whitney U test was used for quantitative variables as well as Wilcoxon's signed rank test for paired data. Increases in biochemical markers with time relative to baseline values were evaluated by the Friedman one-way analysis of variance to detect changes in each group. Wilcoxon's signed rank test was used post hoc to locate the changes. Changes among the groups at each data point were tested with the Mann-Whitney U test. Quantitative variables were also tested for correlations with Spearman's rank-correlation test.
All neuropsychometric test scores were converted to a z score by using the mean and standard deviation of the baseline scores of the patients with data at preoperative and postoperative points in a given cognitive domain. For example, for the entire group, mean baseline score for the Rey Complex Figure copy was 30.6 with a standard deviation of 3.5. For a patient with a score of 30 on this test, this would correspond to a z score of (30 – 30.6)/3.5 = –0.171. The z scores from multiple tests within a cognitive domain were averaged, so that each of the eight cognitive domains was summarized with a single z score. Z scores were computed in the same way postoperatively, again by using the mean and standard deviation of the baseline scores as previously published by Selnes and colleagues [23].
All the statistical significance thresholds were retained for a p value of less than 0.05.
|
Results |
|---|
|
TopAbstractIntroductionMaterial and MethodsResultsCommentAcknowledgmentsReferences |
|---|
|
|
|
|
Changes in composite z scores from baseline to discharge have been calculated for the different cognitive domains. The results have been reported separately for NC and HC groups with their within-group and between-group p values in Table 3. Attention and flexibility or executive functions were lowered postoperatively within the entire population of patients (p = 0.016 and p = 0.013, respectively; data not shown), as well as in the NC group (p = 0.033 and p = 0.028, respectively), whereas no significant changes were detected in the HC group (p = 0.12 and p = 0.22, respectively) nor for any of the six other cognitive domains. Moreover, in the NC group, postoperative value of z score for flexibility/executive functions was negatively correlated with peak levels of sC5b-9 (r = –0.67; r2 = 0.45; p = 0.017), indicating that higher complement activation during surgery implied lower flexibility or executive functions at discharge. Such a significant correlation was not detected between complement activation and z score for attention (r = –0.36; p = 0.24). There was also no correlation in the NC group between peak s100ß values and z score for attention (r = 0.06; p = 0.86) or z score for flexibility or executive functions (r = –0.32; p = 0.31).
|
|
Comment |
|---|
|
TopAbstractIntroductionMaterial and MethodsResultsCommentAcknowledgmentsReferences |
|---|
Neurologic injury after CPB may be caused by impaired cerebral blood flow as a result of hypoperfusion or cerebral embolization and may be assessed by s100ß release, a marker of glial injury that has even been reported to be a strong predictor of late mortality after cardiac surgery [24]. The serum s100ß elevation also parallels, to a lower extent, the cerebrospinal fluid s100ß elevation of patients showing spinal cord injury after thoracoabdominal aortic operations [25]. Owing to the lack of a specific marker of brain inflammation that could be measured after CPB, the role of the systemic inflammatory response in neurologic injury after cardiac surgery is still open for debate. Therefore we evaluated whether neurologic cellular damage could be related to inflammation in a surgical setting aimed at reducing complement activation while cardiotomy suction was avoided to ensure that s100ß measured would be released only by the brain.
Brain hypoperfusion was likely not involved in the higher glial injury observed in the NC group because the general condition of perfusion was completely controlled throughout the entire surgical procedure as reflected by similar hemodynamic measures in both groups. The severity of ascending aorta atheroma, and thus the risk of cerebral embolization, was also similar in both groups. However, the significant increase in s100ß that was observed similarly in both groups soon after initiation of surgery indicates that aortic manipulations to perform proximal anastomoses and to install CPB may have caused early glial injury before starting CPB.
Our results indicate also that systemic inflammation likely contributed to glial injury because higher sC5b-9 levels implied higher s100ß levels. The use of heparin-coated CPB was confirmed to decrease complement activation [13], and even to neutralize it when heparin-coated surfaces were combined with closed extracorporeal circuits and controlled suction. The fact that heparin-coated CPB was associated with significantly reduced s100ß elevation emphasizes the role of inflammation in postoperative neurologic injury. The presence of complement receptors in the brain [12] supports the hypothesis that complement activation during cardiac surgery may have a detrimental impact on the brain. Therefore, inflammation may be considered as an independent component of postoperative neurologic complications, and may be possibly combined in some circumstances with hypoperfusion or embolization processes. During cardiac surgery, efforts are mandatory to ensure that the brain should be subjected to the lowest inflammatory response and embolization process. Cardiopulmonary bypass technology offers at the present time some refinements to achieve this goal, such as heparin-coated circuits with closed reservoirs, cell-saving devices to avoid cardiotomy suction [26], and specific arterial cannulas with an intraaortic filter to capture particulate emboli [27]. Off-pump surgery may limit the systemic inflammatory response as well, but does not completely avoid the embolization process if proximal anastomoses are performed on the ascending aorta. In addition, the improved technique of CPB applied in this study may also be beneficial in a wide range of operations in cardiac surgery, particularly for open heart operations in which higher levels of s100ß than after coronary artery bypass grafting have been reported [28].
Because of the possible complement-mediated brain cell injury, systemic inflammation was suspected to potentially trigger CNS dysfunction. Indeed, complement acts as a trigger and oxygen-free radicals as effectors of the inflammatory cascade. Their subsequent harmful effects, reported either on the lung or on myocardial function after CPB [29, 30], belong to a similar pathophysiologic process that could also adversely affect CNS. Recently, brain oxidative damage has been identified as a cause of mild cognitive impairment, particularly in patients suffering from Alzheimer disease dementia before the onset of symptomatic dementia [31]. A CNS dysfunction rate of greater than 50% was expected to be detected in the time frame of the first postoperative week [32]. Even though significant differences were not detected between the two groups, attention and flexibility or executive functions were two cognitive domains significantly affected after surgery, at least in the NC group, without any noticeable effect of depression or anxiety in the measurements. However, the neuropsychological performances of the patients were considered lowered rather than severely altered at discharge by the neuropsychologists, and even normal in the HC group. The severity of the cognitive dysfunction after CPB is therefore questionable as previously discussed by Selnes and colleagues [33]. It might be argued that the lack of a nonsurgical control group could have masked the detection of such a clinical disturbance in surgical patients, particularly in candidates for cardiac surgery [34]. But preoperative values for the Mini-Mental State Examination, a test used to discriminate patients with global cognitive alteration, indicate that the patients included in this protocol did not have global CNS dysfunction and could be considered as a normal population. Furthermore, a testing location effect (possibly worse neuropsychometric test results owing to assessment in the surgical unit near the operation day) as previously advocated [33] is unlikely in our study because the neuropsychological testing was performed at least 2 weeks before surgery outside the surgical department and the measurements of depression and anxiety scales were found to be in the normal range. Rather, the weak postoperative neuropsychological dysfunction may be likely explained by the basic setting of CPB of this protocol, including, at least in both groups of patients, closed extracorporeal circuits and controlled suctions. When combined with heparin-coating, our CPB technique was aimed, successfully, at neutralizing complement activation, an adverse effect that was in this study correlated with altered flexibility or executive functions postoperatively in the NC group. Although no direct causal linkage may be established from this correlation, the altered flexibility or executive functions in the NC group seemed to be better explained by complement activation than by the brain cellular damage. Because complement activation is a part of the systemic inflammatory response, our findings appear to be consistent with the fact that cognitive dysfunction after coronary artery bypass grafting is transient and usually regresses with time [33]. However, the sample size calculation is a possible limitation in our study, and a larger population might be needed to likely provide a more evident relationship between cognitive dysfunction and brain cellular damage after coronary artery bypass grafting. Again, for the same reason (type 2 error), such a small population in this study may explain the discrepancy between intragroup and between-group analyses. Indeed, the sample size calculation has been designed primarily to demonstrate that cellular brain injury, as reflected by s100ß elevation, could be related to complement activation. Despite the statistical significance, the fact that complement activation only accounts for about 17% of the change in s100ß elevation, according to the regression coefficient, gives an idea of the complexity of the brain pathophysiology that occurs after cardiac surgery. Nevertheless, the correlation between one cognitive domain dysfunction and significant complement activation that has been detected in the NC group is in agreement with the link between CPB and brain inflammatory gene induction that has been demonstrated in animals [35]. Another evidence of the CPB-induced inflammatory cognitive dysfunction after cardiac surgery has been provided by the demonstration of the neuroprotective effect of lidocaine, a potent leukocyte inhibitor [36], in patients undergoing left heart valve procedures [37].
In conclusion, complement activation that occurs during coronary artery bypass grafting contributes to brain cell injury. The postoperative cognitive dysfunction detected in this study has occurred when complement was highly activated. Even if caution is required in attributing causality to this relationship, neuropsychological functioning at discharge was considered normal when this inflammatory pathway was controlled by surgical techniques. Therefore, procedures aimed at reducing complement activation during cardiac surgery, such as heparin-coated CPB, are recommended to reduce postoperative neurologic injury.
|
Acknowledgments |
|---|
|
TopAbstractIntroductionMaterial and MethodsResultsCommentAcknowledgmentsReferences |
|---|
|
References |
|---|
|
TopAbstractIntroductionMaterial and MethodsResultsCommentAcknowledgmentsReferences |
|---|
This article has been cited by other articles:
|
|
 |
|
  B. Ramlawi, H. Otu, J. L. Rudolph, S. Mieno, I. S. Kohane, H. Can, T. A. Libermann, E. R. Marcantonio, C. Bianchi, and F. W. Sellke Genomic expression pathways associated with brain injury after cardiopulmonary bypass. J. Thorac. Cardiovasc. Surg., October 1, 2007; 134(4): 996 - 1005.e4. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 R. A. Kumar, C. Cann, J. E. Hall, P. S. Sudheer, and A. R. Wilkes Predictive value of IL-18 and SC5b-9 for neurocognitive dysfunction after cardiopulmonary bypass Br. J. Anaesth., March 1, 2007; 98(3): 317 - 322. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 C. S. Rinder, M. J. Smith, H. M. Rinder, D. N. Cortright, R. M. Brodbeck, J. E. Krause, and B. R. Smith Leukocyte Effects of C5a-Receptor Blockade During Simulated Extracorporeal Circulation Ann. Thorac. Surg., January 1, 2007; 83(1): 146 - 152. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
K. G. Shann, D. S. Likosky, J. M. Murkin, R. A. Baker, Y. R. Baribeau, G. R. DeFoe, T. A. Dickinson, T. J. Gardner, H. P. Grocott, G. T. O'Connor, et al. An evidence-based review of the practice of cardiopulmonary bypass in adults: A focus on neurologic injury, glycemic control, hemodilution, and the inflammatory response. J. Thorac. Cardiovasc. Surg., August 1, 2006; 132(2): 283 - 290.e3. [Full Text] [PDF]
|
|
|
|
 |
|
 J. M Murkin Pathophysiological Basis of CNS Injury in Cardiac Surgical Patients: Detection and Prevention Perfusion, July 1, 2006; 21(4): 203 - 208. [Abstract] [PDF]
|
|
|
|
 |
|
 C. W. Hogue Jr, C. A. Palin, and J. E. Arrowsmith Cardiopulmonary bypass management and neurologic outcomes: an evidence-based appraisal of current practices. Anesth. Analg., July 1, 2006; 103(1): 21 - 37. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| ANN THORAC SURG | ASIAN CARDIOVASC THORAC ANN | EUR J CARDIOTHORAC SURG |
| J THORAC CARDIOVASC SURG | ICVTS | ALL CTSNet JOURNALS |
