|
|
||||||||
J Thorac Cardiovasc Surg 2007;134:690-696
© 2007 The American Association for Thoracic Surgery
Surgery for Acquired Cardiovascular Disease |
a Department of Neurology, University of Magdeburg, Magdeburg, Germany
b Department of Cardiac Surgery, University of Magdeburg, Magdeburg, Germany
c Department of Anesthesia and Intensive Care, University of Magdeburg, Magdeburg, Germany.
Received for publication February 9, 2007; revisions received March 20, 2007; accepted for publication March 29, 2007. * Address for reprints: Michael Goertler, MD, Department of Neurology, Otto-von-Guericke-Universität Magdeburg, Leipziger Strasse 44, D-39120 Magdeburg, Germany. (Email: michael.goertler{at}medizin.uni-magdeburg.de).
| Abstract |
|---|
|
|
|---|
Methods: We prospectively assessed the incidence of stroke in relation to severity of carotid disease and corresponding autoregulatory reserve in 2797 patients who had coronary artery bypass graft and/or valve surgery with cardiopulmonary bypass. Patients underwent preoperative carotid sonography and, in case of severe extracranial disease, transcranial Doppler sonography with carbon dioxide stimulation to assess cerebrovascular reserve capacity.
Results: Sixty-seven (2.4%) patients had an ischemic stroke, which was fatal in 5. Anterior hemispheric stroke occurred in 42 (1.9%) patients with no/low-grade stenosis, 6 (1.8%) with medium-grade stenosis, 1 (0.6%) with high-grade stenosis/occlusion and normal autoregulation, and 3 (27.3%) with high-grade stenosis/occlusion and exhausted autoregulatory reserve. Increased risk was observed in patients with high-grade stenosis/occlusion and exhausted autoregulatory reserve also after adjustment for potential confounders (adjusted odds ratio [OR] 28.3, 95% confidence interval [CI] 5.8–139.1). Stroke risk was not increased in patients with stenosis/occlusion and normal autoregulation (1.5%, adjusted OR 0.6, 95% CI 0.2–1.6).
Conclusions: Cerebrovascular reserve capacity evaluated by preoperative transcranial Doppler carbon dioxide testing is a major determinant of stroke risk in patients with carotid artery stenosis/occlusion undergoing cardiac surgery with cardiopulmonary bypass. Its assessment facilitates identification of patients with an excess perioperative stroke risk.
| Introduction |
|---|
|
|
|---|
To investigate this hypothesis, we conducted a prospective cohort study, selecting consecutive patients who underwent CABG or valve surgery with CPB and had received preoperative carotid sonography including cerebral vasoreactivity testing.
| Materials and Methods |
|---|
|
|
|---|
|
Anesthesia and Cardiac Surgery
Standard anesthesia for cardiac surgery was induced by etomidate 0.2 mg/kg and sufentanil 1 µg/kg and was maintained with propofol 3 mg · kg–1 · h–1 and with, since 2003, sevoflurane 0.5 minimal alveolar concentration. Neuromuscular block was achieved by pancuronium 0.08 mg/kg. Standard CPB technique (HL20; Jostra, Hirrlingen, Germany) included nonpulsatile flow, membrane oxygenator, arterial line filter, moderate hypothermia at 29°C, and alpha-stat method for acid–base management. Heart surgery was performed after administration of cold crystalloid cardioplegic solution (Bretschneider) with a single period of aortic crossclamping. Maintenance of cerebral autoregulation under these conditions has been ascertained.16-22
Although the study aimed at evaluating the prognostic relevance of impaired cerebral autoregulation, we considered it unethical with respect to the hypothesized stroke risk not to attempt risk reduction in appropriate patients by combined carotid and cardiac surgery; that is, carotid endarterectomy precedes the cardiac procedure in the same operative setting. This strategy was also recommended in patients with severe bilateral stenosis or with contralateral occlusion (ie, high probability of an exhausted cerebrovascular reserve, in whom autoregulation could not be assessed because of insufficient acoustic temporal bone window). During the first study years, combined surgery was also considered in patients with recently symptomatic high-grade carotid stenosis and in patients in whom surgeons judged it adequate to prevent best perioperative stroke, even though transcranial Doppler had demonstrated normal autoregulation. The latter two indications were terminated after literature had demonstrated increased risk of a combined procedure, whereas, according to our own experience, isolated cardiac surgery showed low risk.
Neurologic Outcome Assessment
Diagnosis of a neurologic outcome event was made by a neurologist who was consulted for suspected signs or symptoms. Patients with focal deficits underwent computed tomographic (CT) scanning to differentiate between ischemic and hemorrhagic events. Ischemic stroke was diagnosed in case of a persisting focal neurologic deficit with compatible CT findings. Postoperative encephalopathy was diagnosed in patients who had prolonged unconsciousness, convulsion, and/or delirium. If the cause remained obscure or neurologic status could not be evaluated sufficiently, CT was performed to exclude or confirm cerebral infarction or bleeding as the underlying disease. An event was considered to be related to the operative procedure if it occurred within 30 days after surgery.
Statistical Analyses
Statistical analysis was made with the SPSS 11.0 software package (SPSS, Inc, Chicago, Ill). To identify differences in baseline and surgical characteristics between selected patients and those without preoperative carotid sonography, relevant variables were compared by
2 test and Mann–Whitney U test. The odds ratios (OR) and 95% confidence intervals (CI) were calculated to compare the incidence of stroke in patients with different degrees of carotid disease and impaired autoregulatory reserve. To evaluate independent contribution of carotid disease and autoregulatory reserve and to adjust for potential confounders, we estimated ORs and 95% CIs by multivariable logistic regression analysis.
| Results |
|---|
|
|
|---|
|
|
Carotid Stenosis and Outcome
Incidence of perioperative stroke in relation to carotid disease is presented in Figure 2. Forty-two anterior hemispheric strokes occurred in patients with no/low-grade stenosis, 21 of them more than 24 hours postoperatively in hemodynamically stable patients who had recovered from anesthesia without neurologic deficit. Thirteen of 21 patients whose deficit was diagnosed immediately after recovery from anesthesia revealed anterior hemispheric territorial infarction on subsequent CT scan, suggesting an embolic cause. Stroke was associated with an increased frequency of atrial fibrillation (23.8 vs 8.4%) and valve surgery (54.8 vs 27.5%). Six patients with medium-grade stenosis had an anterior hemispheric stroke distal to their stenosis, 2 of them more than 48 hours after the operation. One patient with normal cerebrovascular reserve capacity had a stroke distal to a unilateral occlusion on the third postoperative day after he had recovered postoperatively without any neurologic deficit. Three strokes occurred in hemispheres with exhausted cerebrovascular reserve distal to an occlusion (2 patients with bilateral occlusion, 1 with contralateral high-grade stenosis) and were diagnosed immediately after recovery from anesthesia. Three patients who did not undergo preoperative carbon dioxide testing had a stroke distal to unilateral occlusion (n = 2) and high-grade stenosis (n = 1). Both strokes distal to occlusion showed hemodynamic infarction pattern on CT scans and were diagnosed immediately after recovery from anesthesia, whereas the stroke distal to high-grade stenosis occurred 4 days after the operation.
|
Adjusted odds for categories of different degrees of carotid stenosis in relation to patients with no/low-grade stenosis are presented in Figure 2. The risk of an anterior hemispheric stroke in the flow area of a diseased carotid artery was increased in patients who exhibited an exhausted cerebrovascular reserve capacity distal to high-grade stenosis or occlusion (OR 19.9, 95% CI 5.1–77.1). This remained significant after adjustment for age, sex, risk factors, history of myocardial infarction, cardiogenic syncope/shock, cardiac surgery, stroke, and recently symptomatic carotid stenosis, concomitant peripheral arterial disease, atrial fibrillation, New York Heart Association classification, surgical urgency, type of surgery, number of diseased coronary arteries, and type of diseased valves (adjusted OR 28.3, 95% CI 5.8–139.1). Risk in the 482 patients with carotid disease, that is, at least medium-grade stenosis, and normal cerebrovascular reserve was low (7 strokes, 1.5%), not exceeding that in patients with no/low-grade stenosis (adjusted OR 0.6, 95% CI 0.2–1.6).
Regression analysis selected female sex (OR 2.7, 95% CI 1.5–4.8) and combined coronary and valve surgery (OR 2.6, 95% CI 1.3–5.3) as additional independent stroke predictors from the above listed covariables. However, when only patients with normal autoregulatory reserve were analyzed, multivariable regression selected female sex, combined coronary and valve surgery, and atrial fibrillation (OR 2.2, 95% CI 1.1–4.7) as independent predictors but not severity of carotid disease.
Postoperative encephalopathy was not associated with the severity of carotid disease and occurred in 51 (2.2%) patients with no/low-grade stenosis, 9 (2.7%) with medium-grade stenosis, and 1 (0.6%) with high-grade stenosis/occlusion and normal autoregulation. No patient with high-grade stenosis/occlusion and an exhausted autoregulatory reserve had encephalopathy.
Combined Carotid and Cardiac Surgery
Combined surgery was performed in 3 patients with exhausted reserve capacity distal to high-grade stenosis (of whom 1 had an ipsilateral stroke), 2 with bilateral severe disease and untestable autoregulation (1 ipsilateral stroke), 2 with recently symptomatic high-grade stenosis (1 contralateral stroke), and 5 selected on surgeons decision (1 contralateral stroke).
| Discussion |
|---|
|
|
|---|
Our data suggest an exhausted autoregulatory reserve distal to high-grade carotid stenosis/occlusion as the major determinant of perioperative stroke associated with carotid disease at cardiac surgery. Patients in whom stenosis/occlusion caused lowered cerebral perfusion pressure and induced subsequent cerebral vasodilatation (ie, who had impaired cerebral autoregulation to compensate an additional blood pressure decrease) had a 27% risk of a perioperative stroke. This represents a nearly 30-fold excess risk compared with patients with no/low-grade stenosis. In contrast, patients with carotid disease and normal autoregulation exhibited only a 1.5% incidence of a perioperative stroke, which was not increased compared with patients without carotid disease. However, with respect to the upper limit of the 95% CI of 1.6, a small risk increase cannot be excluded from our data. Hence, transcranial vasoreactivity testing in patients with high-grade carotid stenosis or occlusion before cardiac surgery may facilitate identification of patients with an excess risk of perioperative stroke by carotid disease.
Both variables found in former studies to be associated with an increased risk of a perioperative stroke—severity of stenosis and bilateralism—were strong predictors of an exhausted cerebrovascular reserve in our series. Risk increase found to be associated with these variables may be caused by patients with an impaired cerebral autoregulation. This also is corroborated by the 8.6% stroke risk determined in our patients with high-grade stenosis/occlusion who did not undergo hemodynamic testing, which matches the 8.4% risk determined for patients with comparable vascular findings in the cited review.1
Although cerebral embolism is considered the most frequent cause of a perioperative cerebrovascular event, which is corroborated by our study,7,8
carotid-originated embolism does not seem to be of relevance in stroke caused by carotid disease at cardiac surgery. Incidence of an ischemic stroke was lower in patients with severe stenosis than in the group with occlusion and was comparable with that in patients without relevant carotid disease when patients with hemodynamic impairment through their stenosis had been excluded.
Clinical Implications
The incidence of a perioperative stroke associated with carotid stenosis of 50% or more or occlusion at cardiac surgery with CPB was low in our series (2.5%, n = 13, Figure 2). Of these, 4 occurred in patients with isolated occlusion (unilateral/bilateral) not eligible for prophylactic endarterectomy. Seven strokes occurred in patients with medium-grade stenosis (1 distal to contralateral occlusion, 6 in isolated unilateral/bilateral stenosis) with questionable benefit from prophylactic endarterectomy.23
Only 2 patients with high-grade stenosis (1 with contralateral occlusion) might have benefited from prophylactic endarterectomy. However, because high-grade stenosis was diagnosed preoperatively in another 111 patients (Table 2), endarterectomy of 113 patients would have been necessary to potentially prevent 2 strokes. Even if carotid endarterectomy could be performed with a risk of perioperative stroke and death as low as in the Asymptomatic Carotid Surgery Trial (2.8%),24
no net benefit may be expected in this situation by the procedure. In contrast and with respect to the risks estimated from our study, only 4 patients (with proven impaired autoregulation) to 20 patients (without proven normal autoregulation) would have had to be operated on to reach this goal (Table 2). If an undetermined autoregulation can be limited to patients with an insufficient acoustic temporal bone window, the latter figure might be reduced to only 8 patients (Table 2).
Our data suggest that staged or combined endarterectomy of high-grade carotid stenosis before cardiac surgery with CPB is not necessary to prevent stenosis-associated perioperative stroke in patients with normal cerebral autoregulation and probably should not be recommended. Detection of an exhausted cerebrovascular reserve capacity identifies patients who are at high risk for hemodynamic stroke and might benefit from preceding carotid endarterectomy. However, even though we recommended combined surgery in these patients during our observation, this may be an adequate stroke prevention to be recommended only if stroke incidence at combined procedures remains below the 27% incidence during isolated cardiac surgery established in our study (Figure 2). This cannot be estimated up to date because hemodynamic testing has not been performed in published series and the number of patients with impaired autoregulation undergoing combined operations was far too low at our institution.
However, as an exhausted cerebrovascular reserve is predominantly found distal to occlusion rather than stenosis, nonsurgical precautions may be more effective in this group of patients to prevent perioperative stroke. Blood pressure increase may be the most obvious procedure to increase cerebral blood flow and subsequent oxygen supply in brain regions with impaired autoregulation, although association between blood pressure and incidence of perioperative stroke at cardiac surgery in general is controversially discussed.25
Therefore, we intended mean arterial pressure values during CPB nearby or even above the upper value of the standard 40 to 60 mm Hg pressure range in our patients by means of volume substitution and/or vasoconstrictive agents.
Study Limitations
Baseline characteristics of selected patients differed significantly from those of patients who did not undergo preoperative sonography. However, because we analyzed the effect of carotid disease on cardiac surgery and not its prevalence, pre-selection of patients for sonographic screening on the basis of clinical findings associated with carotid disease seemed reasonable and does not falsify our results. This also may hold true for the series of consecutive patients excluded because of data set loss. An insufficient acoustic temporal bone window, not enabling insonation of the middle cerebral artery in 20 (10.0%) of 200 patients with high-grade stenosis/occlusion, was the most important obstacle for preoperative transcranial carbon dioxide testing. Examination failure in a further 15 (7.5%) patients owing to organization deficiencies, unavailability of the device, and cardiopulmonary instability of patients should be avoidable, at least in part, but may not cause systematic bias. In contrast, bias is introduced by pre-selection of 12 patients with severe stenosis for combined carotid and cardiac surgery. This led to an underestimation of the prevalence of an exhausted autoregulatory reserve in patients with high-grade stenosis (3.5% vs 5.6%) and should be respected, particularly when the efficiency of surgical interventions to avoid perioperative stroke is estimated. As we have not investigated aortic arch atherosclerosis, probably the most frequent origin of cerebral embolism at cardiac surgery, and therefore the ratio of cardioembolic stroke may be underestimated in our cohort. However, this does not distort our finding that stroke caused by carotid stenosis is uncommon and may be restricted to patients with an impaired cerebral autoregulation.
In summary, our study suggests that cerebrovascular reserve capacity evaluated by preoperative transcranial carbon dioxide testing is a major determinant of stroke risk in patients with carotid artery stenosis/occlusion subjected to cardiac surgery with CPB. Identification of patients with an exhausted as well as normal reserve capacity may help to improve therapeutic strategies in the presence of combined carotid and cardiac disease.
| Acknowledgments |
|---|
| References |
|---|
|
|
|---|
This article has been cited by other articles:
![]() |
C. Weimar and S. C. Knipp Letter by Weimar and Knipp Regarding Article, "Patients With Severe Asymptomatic Carotid Artery Stenosis Do Not Have a Higher Risk of Stroke and Mortality After Coronary Artery Bypass Surgery" Stroke, January 1, 2012; 43(1): e12 - e12. [Full Text] [PDF] |
||||
![]() |
S. Venkatachalam, B. H. Gray, D. Mukherjee, and M. H. Shishehbor Contemporary management of concomitant carotid and coronary artery disease Heart, February 1, 2011; 97(3): 175 - 180. [Abstract] [Full Text] [PDF] |
||||
![]() |
T. Tovedal, O. Jonsson, V. Zemgulis, G. Myrdal, S. Thelin, and F. Lennmyr Venous obstruction and cerebral perfusion during experimental cardiopulmonary bypass Interact CardioVasc Thorac Surg, November 1, 2010; 11(5): 561 - 566. [Abstract] [Full Text] [PDF] |
||||
![]() |
K. Doi and H. Yaku Importance of cerebral artery risk evaluation before off-pump coronary artery bypass grafting to avoid perioperative stroke Eur J Cardiothorac Surg, November 1, 2010; 38(5): 568 - 572. [Abstract] [Full Text] [PDF] |
||||
![]() |
J. Fierstra, J. Poublanc, J. S. Han, F. Silver, M. Tymianski, A. P. Crawley, J. A. Fisher, and D. J. Mikulis Steal physiology is spatially associated with cortical thinning J. Neurol. Neurosurg. Psychiatry, March 1, 2010; 81(3): 290 - 293. [Abstract] [Full Text] [PDF] |
||||
![]() |
A.S. Vagal, J.L. Leach, M. Fernandez-Ulloa, and M. Zuccarello The Acetazolamide Challenge: Techniques and Applications in the Evaluation of Chronic Cerebral Ischemia AJNR Am. J. Neuroradiol., May 1, 2009; 30(5): 876 - 884. [Abstract] [Full Text] [PDF] |
||||
![]() |
S. Manabe, T. Shimokawa, T. Fukui, K.-u. Fumimoto, N. Ozawa, H. Seki, S. Ikenaga, and S. Takanashi Influence of carotid artery stenosis on stroke in patients undergoing off-pump coronary artery bypass grafting Eur J Cardiothorac Surg, November 1, 2008; 34(5): 1005 - 1008. [Abstract] [Full Text] [PDF] |
||||
![]() |
T. C. Lisle, K. M. Barrett, L. M. Gazoni, B. R. Swenson, C. D. Scott, A. Kazemi, J. A. Kern, B. B. Peeler, I. L. Kron, and K. C. Johnston Timing of Stroke After Cardiopulmonary Bypass Determines Mortality Ann. Thorac. Surg., May 1, 2008; 85(5): 1556 - 1563. [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 |