| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
J Thorac Cardiovasc Surg 2003;126:1829-1836
© 2003 The American Association for Thoracic Surgery
Cardiopulmonary support and physiology |
a Klinik und Poliklinik für Herzchirurgie, Universitätsklinikum Bonn, Germany
b Cleveland Clinic Foundation, Cleveland, Ohio,, USA
c Neurologische Klinik, Universitätsklinikum, Bonn, Germany.
Received for publication May 29, 2002; revisions received November 1, 2002; accepted for publication January 21, 2003.
* Address for reprints: Christoph Schmitz, MD, Klinik und Poliklinik für Herzchirurgie, Herzzentrum Universität Bonn, Sigmund-Freud-Strasse 25, 53105 Bonn, Germany
c.schmitz{at}uni-bonn.de
 |
Abstract |
|---|
 Top Abstract MethodsResultsDiscussionLimitationsConclusionsDiscussionAPPENDIX 1APPENDIX 2References |
|---|
METHODS: Patients (N = 582) were enrolled in a prospective, controlled study and alternately assigned to the therapy arm (n = 304; intra-aortic filtration) or control arm (n = 278). Preoperative, procedural, and postoperative data were collected. Neurologic examinations included the National Institutes of Health Stroke Scale, Glasgow Coma Scale, and memory tests. Investigators administering neurologic tests were blinded to the study arm. By the use of logistic regression and propensity matching, composite neurologic outcomes (transient ischemic attack, stroke, delirium, coma, and memory deficit) were evaluated.
RESULTS: Patients in the filter group experienced a lower incidence of adverse neurologic outcomes than patients in the control group (4.3% vs 11.9%) (P < .001). There were significantly less transient ischemic attacks (0% vs 1.4%), delirium (3.0% vs 6.5%), and memory deficit (1.3% vs 6.2%). There were fewer strokes in the filter group compared with the control group (0.7% vs 2.2%), although the sample size was too small for a significant finding. Both groups experienced 1 coma outcome. The use of a filter was associated with an adjusted odds ratio of 0.375, implying that a patient who does not receive a filter is 2.7 times more likely to experience an adverse neurologic event. Logistic modeling also demonstrated that there are increasing chances of poor neurologic outcome with increasing age. The model indicates that there may be an increasing protective benefit from the filter with increasing age, although the interaction was not significant.
CONCLUSIONS: The extraction of particulate emboli using intra-aortic filtration resulted in decreased neurologic outcomes.
Several studies have examined intra-aortic filtration and described its effects. In 2000, Reichenspurner and colleagues6 reported that the extraction of particulate emboli in patients undergoing cardiac surgery was feasible using intra-aortic filtration. In a series of 77 patients, this article highlighted the capture and removal of fibrous atheroma and other particulate debris. A larger study by Harringer and the International Council of Emboli Management (ICEM)7 showed that the filter extracted particulate material from 240 of 243 patients, and that 62% of the filters contained fibrous atheroma, which is of particular interest because the body does not lyse fibrous atheroma.
Whereas 2 studies established the filter's ability to capture and remove emboli, more recent studies examined the clinical relevance of emboli capture and removal. By using a published and validated stroke risk index, Schmitz and Blackstone8 found that patients undergoing coronary artery bypass graft (CABG) receiving intra-aortic filtration experienced approximately half the expected number of severe neurologic outcomes. In high-risk patients undergoing combined CABG and intracardiac procedures, even greater benefit was found.9 In these patients, Wimmer-Greinecker9 reported 74% fewer type I outcomes (stroke, transient ischemic attack [TIA], coma, and death as the result of neurologic causes) when compared with a previous study.10 The current investigation used a prospective, controlled study design to isolate the effect of the filter and examine whether the extraction of particulate emboli can reduce neurologic outcomes during cardiac surgery.
|
Methods |
|---|
|
TopAbstractMethodsResultsDiscussionLimitationsConclusionsDiscussionAPPENDIX 1APPENDIX 2References |
|---|
Preoperatively, patients were nonrandomly assigned to either the filter or control arm of the study at alternating intervals. Some flexibility in the time periods was used to keep enrollment numbers in both groups similar. Patients in both study arms received standard surgery per the normal protocols and procedures of our center. These procedures included cardiopulmonary bypass using aortic and 2-stage venous cannulas, a target flow rate of 2.4 L/m2 · min, and a core body temperature of 32°C to 34°C. Myocardial protection was achieved using Buckberg's cold blood cardioplegia or Bretschneider's cardioplegia. Central anastomoses were sewn using a partial clamp after removal of the aortic crossclamp.
Patients in the filter arm received intra-aortic filtration in addition to their standard procedures. The intra-aortic filtration device consisted of a 120-micron heparin-coated polyester mesh placed in the aorta through a side port in the aortic cannula (EMBOL-X System, Edwards Lifesciences, Irvine, Calif). Filter placement occurred immediately before crossclamp release, and the device remained indwelling until the patient was weaned off bypass. Full details of the methods of filter usage have been previously described.6-8 Filters were collected and sent to a central laboratory for analysis (Stanford University, Palo Alto, Calif).7 Final enrollment included 304 patients in the filter group and 278 patients in the control group.
Neurologic assessment was performed on all patients at several points: preoperatively, 48 hours postoperatively, and at discharge (typically on the seventh postoperative day). Examinations included the National Institutes of Health Stroke Scale, Glasgow Coma Scale, and memory deficit questions. Neurologic examiners received identical training in the Neurology Department of Bonn University Hospital and were blinded to the study arm of each patient.
Neurologic end points for the composite outcome included stroke, TIA, coma, delirium, and memory deficit. Strokes were defined as a new focal neurologic deficit, confirmed by cranial computed tomography (CT) scan showing a new ischemic lesion. CT scans were ordered for patients with neurologic problems equivalent to an National Institutes of Health stroke score change of 4 or greater from baseline that did not resolve within 24 hours. Changes were defined as TIA if they substantially resolved in 24 hours and presented a negative CT scan. Coma applied to patients not regaining consciousness in the 24 hours after termination of sedation. Delirious patients exhibited confusion or an altered state of consciousness during hospitalization. Coma and delirium were also assessed clinically. The determination of memory loss used 5 questions regarding the patient's operation and hospital stay. Memory deficit in nondelirious patients was defined as clinical retrograde amnesia and assessed through postoperative questioning. Patients giving erroneous answers to 2 or more memory assessment questions were judged as memory deficient. Patients assessed with memory deficit or delirium at any time were categorized as memory deficit or delirium end points even if symptoms resolved by discharge.
Statistical methodology
A statistical analysis investigated the effects of filter use on a composite neurologic outcome consisting of stroke, TIA, coma, delirium, or memory deficit. Each patient with at least 1 of these components was included as an adverse outcome. Because filters were not assigned in a prospective randomization scheme, a propensity score approach was taken to account for possible differences between treatment groups.11 Of the 582 patients enrolled in this study, 46 experienced a qualifying composite event.
In all tables of univariable results, P values associated with continuous variables were generated by Wilcoxon 2-sample tests because the factors were not normally distributed. For the categoric variables, patient demographics between groups were compared using 
2 tests or, if small expected cell frequencies, exact tests. Continuous variables are presented as means ± standard deviations, and categoric variables are presented as percentages.
Logistic regression modeling was used for both the propensity model and the model for composite outcome. Linearity assumptions for logistic models were checked, and appropriate data transformations were made. Lists of variables considered in the models are given in Appendix 1. The strategy for both models was to include only those factors that are known before procedure, including the type of procedure. Any variables with low frequency of occurrence were not used. Mean value imputation was used in the case of missing values for "missingness" of up to approximately 10%.
The strategy for obtaining a propensity model was to first develop a parsimonious model to describe patients who received filters. Stepwise selection techniques were used with a P value of .15 for entry. A P value of .10 was considered statistically significant. Model selection was validated using bootstrap techniques.12 After the model was finalized, additional nonsignificant factors were included to adjust for other possible differences. Selected demographic, preoperative, and procedural variables for the matched population are shown in Appendix 2. This final nonparsimonious model produced a propensity score for each patient that predicts the probability of being a patient who received a filter.
To assess the quality of this score, patients were ranked by increasing propensity for filter and grouped accordingly into 5 subgroups or quintiles. If the propensity score has been constructed well, there should not be significant differences in baseline characteristics between patients in the filter and non-filter groups. In addition, the propensity score was used to match patients from the filter and non-filter groups.13 In this way, a subset of the total population was obtained that behaved like a randomized group. The final model was verified on this subset.
Logistic modeling of the outcome variable included the filter group variable and then used stepwise selection techniques to determine other significant adjustments. A P value of .15 was used for entry and .10 was considered statistically significant. Model selection was validated using bootstrap techniques. After all baseline and procedural adjustments were made, the propensity score was added to the model.
|
Results |
|---|
|
TopAbstractMethodsResultsDiscussionLimitationsConclusionsDiscussionAPPENDIX 1APPENDIX 2References |
|---|
|
|
|
Selected demographic, clinical history, and procedural descriptors for composite outcomes are shown in Table 4. A summary of individual composite measures is shown in Table 5. There was a significant difference between the filter and no-filter groups for TIA (P = .051), delirium (P = .044), and memory deficit (P = .002). The number of strokes was decreased in the filter group compared with the no-filter group (0.7% vs 2.2%), although the difference was not significant. Comparing the composite outcomes between the filter (13/304, 4.3%) and no-filter (33/278, 11.9%) groups showed significantly (P = .001) less neurologic outcomes in the patients with filters.
|
|
|
|
Particulate debris was extracted from 98% of patients receiving intra-aortic filtration. Greater than 72% of the filters contained fibrous atheroma, including grumous and fibrocalcific atheroma. Platelets and fibrin were found in 69% of filters, thrombus clot was found in 8%, and medial tissue was found in 1.3%.
|
Discussion |
|---|
|
TopAbstractMethodsResultsDiscussionLimitationsConclusionsDiscussionAPPENDIX 1APPENDIX 2References |
|---|
Despite this intense effort, neurologic injury remains a devastating and tenacious surgical complication. Patients, families, surgeons, and the entire health care system bear a great burden each time a cardiac surgery ends in an adverse cerebral outcome. Roach and colleagues'5 landmark article looked at 2108 patients across 24 centers and demonstrated that 6.1% of patients undergoing CABG experience an adverse cerebral outcome, a far more common phenomenon than previously recognized. They further reported that these patients demonstrated 5 to 10 times the mortality and 2 to 4 times the length of stay in intensive care, and required 3 to 6 times the prolonged care when compared with patients without adverse cerebral events. Although a multifactorial cause of this problem is recognized, aortic atheromatosis has repeatedly been reported as the greatest risk factor, and recent articles have highlighted the central role of particulate emboli in neurologic injury.1,2,5
Early studies using transesophageal echocardiography and transcranial Doppler demonstrated the embolic activity associated with surgical manipulation of the heart and the aorta during surgery.3 In patients receiving intra-aortic filtration, the filter is deployed during the surgical period most associated with embolization, including partial clamping, sewing anastomoses, and removal of the aortic crossclamp. Intra-aortic filtration has emerged as the only technology to date capable of capturing and removing particulate emboli from the circulation. Previous reports by the ICEM Study Group focused analysis on a multicenter registry designed to gain statistical relevance by pooling data. These studies have been instrumental in demonstrating that intra-aortic filtration can be performed safely, that it successfully captures and removes particulate debris from the aorta, and that postoperative complications are reduced when compared with risk indices and historical controls.
Prior studies by the ICEM benefited from the ability to draw from 17 participating centers, allowing the aggregate patient group to reach more than 1600 patients in a consecutive enrollment registry. A study of this size has the advantage of providing meaningful background statistics, because category sizes remain large even when subgroups are examined. Because of this, important subset studies such as that reported by Schmitz and Blackstone8 on behalf of the ICEM Study Group could look at a subset of patients undergoing only CABG in enough detail to assess their outcomes with a published, validated stroke risk index. They reported approximately 50% fewer neurologic outcomes in patients receiving intra-aortic filtration when compared with expected events (1.5% vs 3.4%, P = .03).
Similarly, there were enough patients for Wimmer-Greinecker9 to examine a high-risk group of patients receiving concomitant CABG and intracardiac surgery. In his presentation at the 2001 European Association for Cardiothoracic Surgery meeting in Lisbon, Wimmer-Greinecker reported 74% fewer type I neurologic outcomes in this group when compared with a previous study.
This is the first study comparing a control group with a filter group in a single center. Although the study loses the power of the aggregate numbers, it gains the advantage of greater homogeneity through greater control over the procedural variables and patient population. All patients underwent operations by the same surgeons, at the same center, with the same tools and techniques. The procedure mix was very similar. The same outcome definitions and assessments were applied to all patients across both arms. The effect of the intra-aortic filter is thus more isolated than in previous investigations.
Exclusion criteria were applied in a further effort to ensure homogeneity across study arms. Patients with carotid stenosis greater than 70% were excluded because of the possibility of carotid effects on neurologic outcome confounding the effects of the surgery. In addition, patients with less than 20% left ventricular ejection fraction were also excluded. These patients were excluded to ensure that all patients could be assessed with the complete range of neuropsychologic tests that were part of the full protocol.
Of the 46 composite neurologic outcomes, the filter group had 28.3% (13/46) and the control group had 71.7% (33/46). As expected, individual subcategories of neurologic injury had less statistical power, although TIA (P = .05), delirium (P = .04), and memory deficit (P = .002) each achieved significance independently. The stroke subcategory trended toward reduced occurrence in the filter group. In parallel with the overall reduction of individual outcomes, fewer patients experienced a composite adverse neurologic outcome in the filter group (13/304, 4.3%) when compared with the control group (33/278, 11.9%) (P = .001). That some preoperative variables provided a protective benefit or were predictive of an outcome may be caused in part by the low incidence of outcomes in this population or other possible coexisting conditions.
Preoperative variables between the 2 groups showed some differences, but these were adjusted for in the logistic model. Variables with significant influence and sufficient frequency were included in the propensity score model. That some patients were more likely to receive a filter or not with the presence of certain preoperative variables was probably because a formal randomization scheme was not applied during patient selection. The propensity score was included in the model of outcome as an adjustment to compensate for bias. Although the propensity matching is a multivariate method that is the observational analog of randomization and controls for observed variables, it does not control for unobserved variables. Nonetheless, this model considered 37 variables and used all significant variables (n = 9) that provided good matching (P > .05 for variables) using a large percentage of the population. The model results remained consistent within the matched subset.
Important findings in this study include the benefit of filter use and the detriment of increasing age. Many studies have identified older patient age as a significant risk for adverse neurologic outcomes. The reason for this may in part be caused by increased preoperative risk conditions. The current results indicate that an 80-year-old patient with the filter might expect the equivalent outcome of a 55-year-old without the filter. Although the interaction of age and filter on outcomes was not significant, probably because of the low incidence of outcomes, there is a trend in this direction that warrants further investigation. It seems the older population may greatly benefit from intra-aortic filtration.
The mechanism for these results is most likely the extraction of particulate emboli from the aorta before they can travel distally and occlude important blood vessels. The chain of events—aortic atheroma, surgical manipulation, embolization, distal travel, and arterial occlusion—is apparently impeded when the emboli are captured and removed, making arterial occlusion less likely.
Two approaches have now been used to examine intra-aortic filtration. Previous ICEM studies have demonstrated the breadth of applicability of the filter in extracting particulate emboli safely in many patients. In addition, these studies have strongly suggested a clinical benefit. In our single-center controlled study, the evidence for a clinical benefit from filtration has become even more compelling. An even stronger study would combine breadth with control in a multicenter, randomized, controlled study.
|
Limitations |
|---|
|
TopAbstractMethodsResultsDiscussionLimitationsConclusionsDiscussionAPPENDIX 1APPENDIX 2References |
|---|
A model built on 46 events must be interpreted with caution and emphasizes the problem of evaluating outcomes when the incidence is very low. Most of the information available for analysis was clinical history (binary variables). This does not allow for the best predictive model construction. However, in the propensity-matching analysis, 81% of the total patients and 85% of the outcomes were included. This resulted in a well-matched evaluation between the filter and non-filter groups.
Patients in this study were not assigned per a randomization scheme. Although our analysis of preoperative risk factors, demographics, and procedure category using propensity score matching account for many of the differences between groups, an even stronger case could be made in a study with a formal randomization scheme.
This study only looks at outcomes occurring during hospitalization. More study is needed to look at long-term effects of filtration on neurologic outcomes.
In this investigation, outcomes were assessed on a dichotomous scale; a patient was in an outcome category or not. Using graduated assessments of neurologic condition would enable more nuanced and powerful analysis.
Finally, this study looked at damage to only 1 distal organ: the brain. However, many other distal organs may be damaged by particulate emboli. In their autopsy study, Blauth and colleagues14 reported emboli in the kidneys, gastrointestinal tract, and lower extremities. Further studies should examine intra-aortic filtration's effect on these organs. Studies are currently in progress to evaluate more subtle neurologic effects as may be detected by neuropsychologic testing.
|
Conclusions |
|---|
|
TopAbstractMethodsResultsDiscussionLimitationsConclusionsDiscussionAPPENDIX 1APPENDIX 2References |
|---|
|
Discussion |
|---|
|
TopAbstractMethodsResultsDiscussionLimitationsConclusionsDiscussionAPPENDIX 1APPENDIX 2References |
|---|
Dr Schmitz. Originally we had a strict randomization scheme, but it turned out that we were not able to enroll large numbers of patients. That is why we changed our enrollment scheme in that way. All patients were enrolled preoperatively, but the treatment assignment was based on alternate time periods, and we tried to balance the numbers in both groups.
Dr Vanermen. In conjunction with earlier studies on intra-aortic filtration, what do you believe is the current understanding of the indications for use of the device?
Dr Schmitz. We had a lot of discussion about this question in our ICEM group. Most centers try to find high-risk patients for adverse cerebral outcome, and different centers have different ideas. Some centers think that intracardiac repair, open heart surgery, might be the right way to look at it; others look at older patients. We know that the most important predictor for stroke is aortic calcification, and as we can deal very well with heart plaques, it is very difficult to look at soft plaques unless you do epiaortic scanning. However, as far as I know, this has not been done in many centers.
We included even low-risk patients, and our data showed that we found material in 98% of filters. So I think whenever we manipulate the ascending aorta, an intra-aortic filter is indicated.
Dr M. Mack (Dallas, Tex). Just to make that answer practical, coronary bypass surgery on pump, routinely, yes or no?
Dr Schmitz. Actually we are still collecting numbers, but if we look at the data we have so far, it seems to be indicated.
Dr Mack. Beating heart surgery?
Dr Schmitz. Even in beating heart surgery, if you perform a central anastomosis, if you use techniques not to touch the ascending aorta, there is no filter necessary.
Dr Mack. Valve surgery?
Dr Schmitz. Yes.
Dr K. Allen (Indianapolis, Ind). I enjoyed your presentation. We have been involved in the multicenter, randomized prospective trial in the United States and have experience in more than 100 patients with this device. As in your experience, we have been reassured by the ease of use of the device as well as the consistent finding of debris within the filter. I have 2 questions.
The first question involves safety. You did not comment on any potential injuries involving the filter. Could you comment specifically, for example, about the incidence of aortic dissection that may or may not have been seen between the 2 groups?
And second, I noticed that you did not perform a multivariable analysis of your data specifically looking at risk factors that might contribute to stroke. It would be interesting to see if things such as increased age, prior vascular surgery, cerebrovascular disease, or history of stroke were driving the results that you saw rather than the use of the filter. I enjoyed your presentation very much.
Dr Schmitz. Thank you very much for your questions. Regarding your first question about safety, in our series, we did not experience any problems with filters. I am aware that aortic dissection has been discussed in an article by the Hannover group just recently. They experienced this problem with the filter. So I feel that it is very important to emphasize that the cannula is properly seated and that the filter is de-aired, which can be seen at the tip of the filter before it is deployed. I think if you check that, you can minimize the risk of aortic dissection.
Regarding your second question about which preoperative variables may have driven the outcome, we looked at the variables that have been published that have an influence on negative neurologic outcome, and we did not find any difference. Only gender was a difference; the fact that female gender has a higher risk for neurologic outcome has been published in an article. That is why we looked at that, but we did not find any other differences.
Dr M. Song (Seoul, Korea). I enjoyed your article.
Obviously, intra-aortic filtration has some effect to prevent brain injury. However, I think the most important procedures that seem to be related to neurologic injury is the aortic cannulation process itself and then crossclamping techniques. In our experience, in 12% of patients there were some plaque or calcification on the ascending aorta on which we were going to cannulate.
So my question is, if there are patients with some plaque or calcification on the cannulation site, would you still cannulate with your new intra-aortic filter device or would you perform another technique such as off-pump coronary artery bypass or some different site of cannulation?
Dr Schmitz. Thank you very much for this question. No. I think the use of an intra-aortic filter is 1 possible therapy for these patients. If we have, for example, a porcelain aorta, for sure we would try to go for an off-pump coronary artery bypass without any manipulation of the ascending aorta.
Dr R. Shemin (Boston, Mass). This is a very nice article, and I think people will struggle with the appropriate application in some groups of patients to use it. Did you perform any epiaortic mapping and/or, say, transcranial Doppler to try to correlate what you would find with those techniques and the amount of debris that was obtained in individual patients?
Dr Schmitz. No, we did not do so, but that would be very interesting.
Discussion
Dr H. Vanermen (Aalst, Belgium). In view of the fact that this was a nonrandomized study, could you give us some more detail on how patients were chosen to go into 1 arm of the study or another, and could this potentially bias the results?
Dr Schmitz. Originally we had a strict randomization scheme, but it turned out that we were not able to enroll large numbers of patients. That is why we changed our enrollment scheme in that way. All patients were enrolled preoperatively, but the treatment assignment was based on alternate time periods, and we tried to balance the numbers in both groups.
Dr Vanermen. In conjunction with earlier studies on intra-aortic filtration, what do you believe is the current understanding of the indications for use of the device?
Dr Schmitz. We had a lot of discussion about this question in our ICEM group. Most centers try to find high-risk patients for adverse cerebral outcome, and different centers have different ideas. Some centers think that intracardiac repair, open heart surgery, might be the right way to look at it; others look at older patients. We know that the most important predictor for stroke is aortic calcification, and as we can deal very well with heart plaques, it is very difficult to look at soft plaques unless you do epiaortic scanning. However, as far as I know, this has not been done in many centers.
We included even low-risk patients, and our data showed that we found material in 98% of filters. So I think whenever we manipulate the ascending aorta, an intra-aortic filter is indicated.
Dr M. Mack (Dallas, Tex). Just to make that answer practical, coronary bypass surgery on pump, routinely, yes or no?
Dr Schmitz. Actually we are still collecting numbers, but if we look at the data we have so far, it seems to be indicated.
Dr Mack. Beating heart surgery?
Dr Schmitz. Even in beating heart surgery, if you perform a central anastomosis, if you use techniques not to touch the ascending aorta, there is no filter necessary.
Dr Mack. Valve surgery?
Dr Schmitz. Yes.
Dr K. Allen (Indianapolis, Ind). I enjoyed your presentation. We have been involved in the multicenter, randomized prospective trial in the United States and have experience in more than 100 patients with this device. As in your experience, we have been reassured by the ease of use of the device as well as the consistent finding of debris within the filter. I have 2 questions.
The first question involves safety. You did not comment on any potential injuries involving the filter. Could you comment specifically, for example, about the incidence of aortic dissection that may or may not have been seen between the 2 groups?
And second, I noticed that you did not perform a multivariable analysis of your data specifically looking at risk factors that might contribute to stroke. It would be interesting to see if things such as increased age, prior vascular surgery, cerebrovascular disease, or history of stroke were driving the results that you saw rather than the use of the filter. I enjoyed your presentation very much.
Dr Schmitz. Thank you very much for your questions. Regarding your first question about safety, in our series, we did not experience any problems with filters. I am aware that aortic dissection has been discussed in an article by the Hannover group just recently. They experienced this problem with the filter. So I feel that it is very important to emphasize that the cannula is properly seated and that the filter is de-aired, which can be seen at the tip of the filter before it is deployed. I think if you check that, you can minimize the risk of aortic dissection.
Regarding your second question about which preoperative variables may have driven the outcome, we looked at the variables that have been published that have an influence on negative neurologic outcome, and we did not find any difference. Only gender was a difference; the fact that female gender has a higher risk for neurologic outcome has been published in an article. That is why we looked at that, but we did not find any other differences.
Dr M. Song (Seoul, Korea). I enjoyed your article.
Obviously, intra-aortic filtration has some effect to prevent brain injury. However, I think the most important procedures that seem to be related to neurologic injury is the aortic cannulation process itself and then crossclamping techniques. In our experience, in 12% of patients there were some plaque or calcification on the ascending aorta on which we were going to cannulate.
So my question is, if there are patients with some plaque or calcification on the cannulation site, would you still cannulate with your new intra-aortic filter device or would you perform another technique such as off-pump coronary artery bypass or some different site of cannulation?
Dr Schmitz. Thank you very much for this question. No. I think the use of an intra-aortic filter is 1 possible therapy for these patients. If we have, for example, a porcelain aorta, for sure we would try to go for an off-pump coronary artery bypass without any manipulation of the ascending aorta.
Dr R. Shemin (Boston, Mass). This is a very nice article, and I think people will struggle with the appropriate application in some groups of patients to use it. Did you perform any epiaortic mapping and/or, say, transcranial Doppler to try to correlate what you would find with those techniques and the amount of debris that was obtained in individual patients?
Dr Schmitz. No, we did not do so, but that would be very interesting.
|
APPENDIX 1 |
|---|
|
TopAbstractMethodsResultsDiscussionLimitationsConclusionsDiscussionAPPENDIX 1APPENDIX 2References |
|---|
|
MI, Myocardial infarction; NYHA, New York Heart Association; TIA, transient ischemic attack; GI, gastrointestinal; CABG, coronary artery bypass graft.
|
APPENDIX 2 |
|---|
|
TopAbstractMethodsResultsDiscussionLimitationsConclusionsDiscussionAPPENDIX 1APPENDIX 2References |
|---|
|
P value: Wilcoxon 2-sample test, 2, and Fisher exact test.
|
References |
|---|
|
TopAbstractMethodsResultsDiscussionLimitationsConclusionsDiscussionAPPENDIX 1APPENDIX 2References |
|---|
This article has been cited by other articles:
|
|
 |
|
  D. J. Cook, J. Huston III, M. R. Trenerry, R. D. Brown Jr, K. J. Zehr, and T. M. Sundt III Postcardiac Surgical Cognitive Impairment in the Aged Using Diffusion-Weighted Magnetic Resonance Imaging Ann. Thorac. Surg., April 1, 2007; 83(4): 1389 - 1395. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 P. Boivie, M. Hansson, and K. G. Engstrom Intraluminal aortic manipulation by means of intra-aortic filter, cannulation, and external clamp maneuvers evaluated versus dislodged embolic material J. Thorac. Cardiovasc. Surg., February 1, 2006; 131(2): 283 - 289. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 G. M. McKhann, M. A. Grega, L. M. Borowicz Jr, W. A. Baumgartner, and O. A. Selnes Stroke and Encephalopathy After Cardiac Surgery: An Update Stroke, February 1, 2006; 37(2): 562 - 571. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 K. Prasongsukarn and M. A. Borger Reducing Cerebral Emboli During Cardiopulmonary Bypass Seminars in Cardiothoracic and Vascular Anesthesia, June 1, 2005; 9(2): 153 - 158. [Abstract] [PDF]
|
|
|
|
|
|
D. D. Doblar Intraoperative Transcranial Ultrasonic Monitoring for Cardiac and Vascular Surgery Seminars in Cardiothoracic and Vascular Anesthesia, June 1, 2004; 8(2): 127 - 145. [Abstract] [PDF]
|
|
|
|
 |
|
 S. Martens, M. Dietrich, C. Herzog, M. Doss, G. Schneider, A. Moritz, and G. Wimmer-Greinecker Automatic connector devices for proximal anastomoses do not decrease embolic debris compared with conventional anastomoses in CABG Eur. J. Cardiothorac. Surg., June 1, 2004; 25(6): 993 - 1000. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME |
