Cerebrovascular reactivity before and after carotid endarterectomy

Cerebrovascular Reactivity Before and After Carotid Endarterectomy V. D’Angelo, M.D.,* G. Catapano, M.D.,* V. Bozzini, M.D.,* D. Catapano, M.D.,* P. De Vivo, M.D.,† P. Ciritella,† L. Parlatore, Math. Consult. *Department of Neurosurgery, †Neurological Intensive Care Unit and Anaesthesiology, Casa Sollievo della Sofferenza Hospital, I.R.C.C.S., S. Giovanni Rotondo (FG) Italy

D’Angelo V, Catapano G, Bozzini V, Catapano D, De Vivo P, Ciritella P., Parlatore L. Cerebrovascular reactivity before and after carotid endarterectomy. Surg Neurol 1999;51:321– 6. BACKGROUND

The hemodynamic relevance of internal carotid artery (ICA) stenosis often does not correlate with anatomic features, as angiographically defined. The cerebrovascular reactivity (CVR) has been advocated as a means of defining the cerebral hemodynamic impairment. METHODS

We assessed the results of pre- and postoperative CVR evaluation, using the CO2 transcranial doppler method, in 25 patients with high-grade ICA stenosis. The patients with history of stroke, evidence of cerebral CT infarction or symptoms from the contralateral circulation or the brain stem were excluded to avoid the effects of cerebral infarction on the hemodynamic studies. Statistical analysis was used to evaluate the CVR changes after carotid endarterectomy. RESULTS

Preoperative evaluation showed that CVR was generally well correlated with the degree of ICA stenosis and concomitant contralateral carotid steno-occlusion. Before endarterectomy the mean CVR value was 66.5% (moderately reduced). After surgery the overall mean value of CVR was 84.1% (normal), with a statistically significant improvement. CONCLUSION

The results of this study suggest that the CVR evaluation allows one to obtain hemodynamic information of clinical interest in the patients with ICA stenosis and that carotid endarterectomy is effective to restore the CVR in patients with cerebral hemodynamic impairment. © 1999 by Elsevier Science Inc. KEY WORDS

Cerebrovascular reactivity, transcranial doppler, carotid artery stenosis, carotid endarterectomy.

Address reprint requests to: Dr. Vincenzo D’Angelo, Department of Neurosurgery, “Casa Sollievo della Sofferenza” Hospital, viale Cappuccini, 71013 San Giovanni Rotondo (FG) Italy. Received March 7, 1997; accepted August 4, 1997. © 1999 by Elsevier Science Inc. 655 Avenue of the Americas, New York, NY 10010

umerous techniques are available to evaluate regional cerebral hemodynamics in patients with occlusive disease of extracranial vessels. Studies have demonstrated that in the individual patient the degree of internal carotid artery (ICA) stenosis correlates poorly with the ipsilateral cerebral circulation. The predominant mechanism of ischemic damage is thromboembolic; however, in a small subgroup of patients the clinical correlates of cerebral ischemia are attributable to a critical reduction of cerebral perfusion pressure. The identification of patients with impairment of cerebral hemodynamics may have a great effect on their management. A parameter used to test the hemodynamic status of the cerebral circulation is the cerebral vasomotor reactivity (CVR). The CVR is the capacity of cerebral blood flow to agree with changes in regional metabolic demand. The CVR has been measured in stroke patients by various techniques to examine the changes of cerebral blood flow to exogenous vasodilatory stimuli, such as CO2 or acetazolamide [1,2,5,6,9,13,21,22]. In recent years transcranial doppler velocimetry (TCD), a simple and non-invasive method, has become available to assess the vasomotor response of the cerebral circulation by measuring the flow velocity changes. The flow velocity measurement corresponds to regional cerebral blood flow changes and may be used to assess the cerebral vasoreactivity [4,9,19,20,23,26]. In the present study we assessed the value of CVR in patients with high-grade ICA stenosis before and after endarterectomy. For this purpose we used TCD velocimetry and CO2 as vasodilatory stimuli. The aims of this study were to test the CVR in patients with high-grade ICA stenosis and its change after endarterectomy.

N

0090-3019/99/$–see front matter PII S0090-3019(97)00503-X

322 Surg Neurol 1999;51:321–6

1

D’Angelo et al

Clinical and Angiographic Findings

PT./AGE/SEX

CLINICAL DATA

ICA TREATED SURGICALLY (% STENOSIS)

1/66/M 2/75/M 3/57/M 4/64/M 5/59/M 6/61/M 7/62/M 8/65/M 9/64/M 10/62/M 11/71/M 12/66/M 13/65/M 14/74/F 15/72/M 16/59/F 17/65/M 18/56/F 19/71/M 20/48/M 21/65/M 22/67/F 23/66/M 24/67/F 25/69/F

TIA TIA TIA Amaur. fugax; Vertigo Amaur. fugax; Vertigo TIA TIA TIA; Amaur. fugax TIA TIA; Amaur. fugax TIA TIA Amaur. fugax; Vertigo Amaur. fugax; Vertigo TIA; Amaur. fugax TIA TIA TIA; Amaur. fugax Amaur. fugax TIA TIA TIA TIA TIA TIA; Amaur. fugax

.90 70–90 .90 70–90 70–90 70–90 .90 .90 70–90 70–90 .90 70–90 70–90 .90 70–90 70–90 70–90 70–90 70–90 .90 70–90 70–90 .90 70–90 70–90

Patients and Methods We studied 25 patients with high-grade ICA stenosis treated with endarterectomy. The mean age was 65.1 years (range, 48 –75 years); there were 20 men and 5 women. The patients had experienced TIAs or amaurosis fugax attributable to ICA stenosis. Patients with a history of stroke, evidence of cerebral CT infarction, or symptoms from the contralateral circulation or the brain stem were excluded. This selection was made to avoid the effects of cerebral infarction on the hemodynamic studies (Table 1). The ICA stenosis was assessed preoperatively by doppler ultrasound and four-vessel angiography studies. In 17 patients the degree of stenosis was between 70% and 90%. In eight patients the degree of stenosis was .90%. We divided the patients in two groups: a) unilateral ICA stenosis (18 pts); b) associated lesion of the contralateral ICA (high-grade stenosis or occlusion) (7 pts). We assessed the hemodynamic status of the cerebral circulation ipsilateral to the ICA treated surgically by measuring the CVR in the middle cerebral artery (MCA) territory. The study was done 2 days before endarterectomy and 10 days

CONTRALATERAL ICA No lesion No lesion No lesion No lesion No lesion No lesion No lesion No lesion No lesion No lesion No lesion No lesion No lesion No lesion No lesion No lesion No lesion No lesion Stenosis .70% Stenosis .70% Stenosis .70% Stenosis .70% Stenosis .70% Occlusion Occlusion

after surgery. The CVR was calculated from the blood flow velocity changes measured by transcranial doppler sonography in relation to endtidal CO2 concentrations (ETCO2) for normocapnic, hypercapnic, and hypocapnic conditions. The ETCO2 was measured from expired gas by a Nellcor N-100 gas-analyzer (Nellcor Inc., Hayward USA). In the present study, hypercapnia was achieved by a rebreathing method. Through a to-and-from circuit, a face mask and a 3-l balloon, the patient rebreathes his expirate air. The CO2 in the gas mixture increases with the ETCO2. Before starting the test the basal ETCO2 is recorded. Our protocol allows an increase of 50% with respect to the basal value of ETCO2 and the level achieved is kept stable for 2–3 minutes. Hypocapnia was achieved by hyperventilation, reaching a steadystate about 30% less than the basal value. During the breathing of room air the MCA mean velocity was set arbitrarily at 100%. The percentage changes of mean velocity during hypercapnia and hypocapnia were correlated with ETCO2 and were analyzed by curve fitting (Figures 1 and 2). The percentage changes of blood flow velocity reflect the responsiveness of the MCA vascular

CVR Changes After Carotid Endarterectomy

1

Patient N° 12. Preoperative blood flow velocity changes in the middle cerebral artery for normocapnic (a), hypercapnic (b), and hypocapnic (c) condition. CVR 5 63%.

territory to capnic stimulus, the so-called CO2induced cerebral vasomotor reactivity [19,20].

Results In our work we analyzed the results according to the grading of compromise of CVR measured with the TCD method proposed by Ringelstein et al [19,20]. Before endarterectomy the mean CVR value was 66.5%, i.e., in the range considered “moderately reduced.” In a few patients there was a great hemodynamic impairment. We analyzed the influence on CVR of the degree of ICA stenosis and of contralateral ICA pathology. In the 17 patients with stenosis between 70% and 90%, the mean CVR value was 70.4%. In the eight patients with

2

Surg Neurol 323 1999;51:321–6

stenosis greater than 90% the mean CVR value was 60.7%. In the patients of group a) with no lesion of the contralateral ICA the mean preoperative value of CVR was 71.8%, whereas in the group of patients with stenosis or occlusion of the contralateral carotid artery the mean CVR value was 54.0% (Tables 2 and 3). The CVR was measured with the same method in the early postoperative period (10 days after surgery). After endarterectomy the overall mean value of CVR was 84.1%, with a statistically significant increase. The CVR value was increased in all but one patient in whom it was slightly decreased after surgery, but both the values were in the normal range. The CVR was mainly increased in the patients with the worst preoperative vasomotor reactivity (Table 4).

Patient N° 12. Postoperative blood flow velocity changes in the middle cerebral artery for normocapnic (a), hypercapnic (b), and hypocapnic condition (c). CVR 5 97%.

324 Surg Neurol 1999;51:321–6

2

D’Angelo et al

Correlation between CVR and Degree of Stenosis

DEGREE OF STENOSIS

PREOPERATIVE CVR (%)

POSTOPERATIVE CVR (%)

70–90% .90%

70.4 60.7

87.5 78.6

STATISTICAL ANALYSIS Statistical analysis was used to evaluate the CVR changes after carotid endarterectomy. For this purpose the following tests of non-parametric analysis and data analysis were used: 1) Rank-order Test U 5 24.525 (p 5 0.00001); 2) Kruskal-Wallis one-way Test (ANOVA by ranks) Kw 5 13.573 (p 5 0.0002) 3) Wilcoxon signed-rank Test T1 5 268 T2 5 8 z 5 3.9539 (p 5 0.00004). 4) Cluster-analysis one-dimensional CVR threshold value 5 76 Before surgery 80% of pts were below the threshold value. After surgery 72% of pts were above the threshold value. 5) Cluster-analysis two-dimensional All pts were situated in a single cluster (postoperative CVR 5 or . preoperative CVR) 6) Increasing-rate test 5 100x (post. CVR-pre. CVR)/pre. CVR. Most points had high increasing rate. 7) ROC-Test: roc area (mean 5 0.8, SD 5 0.06) was .0.5. After surgery the number of points above the threshold was increased (roc threshold 5 76).

Discussion The weight of hemodynamic factors in the pathophysiology of cerebral ischemia associated with extracranial vessel occlusive diseases is still unclear. The predominant mechanism of stroke is thromboembolic; however, a small subgroup of patients may experience clinical symptoms because of a critical reduction of the cerebral perfusion pressure. These

3

Correlation between CVR and Contralateral ICA Pathology

CONTRALATERAL ICA No lesion Stenosis or occlusion

PREOPERATIVE POSTOPERATIVE CVR (%) CVR (%) 71.8 54.0

86.8 78.3

4

CVR Before and After Surgery

PT.

PREOPERATIVE CVR (%)

POSTOPERATIVE CVR (%)

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 Mean

62 61 80 107 76 71 76 62 76 70 58 63 62 68 86 80 100 75 65 34 62 60 46 50 72 66.5

70 94 88 97 94 80 100 74 88 84 89 97 80 74 94 80 100 80 70 58 86 76 76 78 90 84.1

patients suffer “true” cerebrovascular insufficiency and the cerebral blood flow is insufficient to fulfill the regional metabolic demand. The identification of patients with impairment of the hemodynamic status of the cerebral circulation may have a great effect on their management. Recent works have demonstrated a major risk of stroke in patients with extracranial carotid artery steno-occlusive diseases in whom a high degree of hemodynamic compromise is present [27]. According to Weiller et al, such patients may experience low-flow infarct, i.e., the result of a critically reduced perfusion pressure distal to an occlusive carotid lesion in the neck, whereas the artery supplying the ischemic brain territory is usually not affected. In the work of Weiller no patients with hemodynamically intact hemispheres have low-flow infarct [25]. It is speculative that in such patients with major impairment of cerebral hemodynamics a medical or surgical treatment that improves the cerebral blood flow may be beneficial, whereas we would not expect a benefit from antithrombotic treatment. This has to be proved by standardized clinical trials. Recent studies have made it possible to assess the hemodynamic status of the cerebral circulation

CVR Changes After Carotid Endarterectomy

in extracranial carotid artery diseases. A method frequently used is the measurement of CVR, i.e., the capacity of circulation to mantain the blood flow in accordance with regional metabolic demand. Various techniques are available to test the CVR, directly or indirectly, such as PET, SPECT, Xenon 133 inhalation, stable Xenon TC, and transcranial doppler velocimetry [1,2,5,6,9,13,21,22]. The study of CVR involves the use of two rCBF measurements, one performed at rest and the second after the application of a vasodilatory stimulus, such as CO2 or acetazolamide. The difference between the two studies reflects the responsiveness of the cerebral circulation and the degree of increase in cerebral perfusion will indicate the reserve vasodilating capacity. In the present study we used transcranial doppler sonography and the vascular territory of the MCA ipsilateral to the carotid lesion. TCD is a noninvasive and reproducible method and assesses indirectly the rCBF changes of a vascular territory by measurement of blood flow velocity in the supply artery [4,9,19,20,23,26]. This requires that both the diameter of the supply artery and the size of the perfusion territory remain constant. Dahl et al in a recent work have documented a good correlation between the TCD and SPECT study of rCBF in the assessment of CVR [3]. We used CO2 as a vasodilatory stimulus. In the literature no differences are found in the measurement of CVR between CO2 and acetazolamide. In our work we achieved the condition of hypercapnia using a method of rebreathing (see subjects and methods). Our results are in agreement with the literature. In our patients with high-grade ICA stenosis before endarterectomy the overall CVR value was lower than normal subjects and was in the range considered by Ringelstein et al “moderately reduced” [19, 20]. However in individual patients this value may be “severely reduced,” and this is a condition with a poor prognosis for ischemic stroke. The degree of ICA stenosis may have an influence on hemodynamic compromise. In 17 patients with ICA stenosis between 70% and 90% the preoperative CVR value was greater than in patients with ICA stenosis .90%. Moreover the CVR value was lower in those patients with associated contralateral ICA stenosis. This is in disagreement with the results of Kleiser et al, which found no influence on CVR from contralateral ICA pathology [7]. This is probably attributable to the effect that patency of the intracranial anastomoses (circle of Willis, leptomeningeal anastomoses) has on cerebral hemodynamics. In our study, after endarterectomy the CVR was increased in all but one patient. The increase of CVR was highly

Surg Neurol 325 1999;51:321–6

significant in the patients with the most hemodynamic compromise. In agreement with Ringelstein et al, the increase of the CVR value was documented in the early postoperative period. These findings suggest some final considerations about the pathophysiology and clinical management of extracranial occlusive diseases. First, the CO2-induced cerebral vasomotor reactivity measured by TCD represents a valid method to test the hemodynamic status of the intracranial circulation, as documented by Ringelstein et al. Actually, as stated by Powers [17,18], the concept of hemodynamically significant carotid stenosis and the relevance of hemodynamic factors in cerebral ischemia should be based on assessment of cerebral, not carotid or vertebral, hemodynamics. Moreover, in extracranial carotid artery stenosis in a few patients the impairment of cerebral hemodynamics can contribute to TIAs or stroke, although the predominant mechanism is thromboembolic. Such patients may experience a so-called low-flow infarct attributable to critically reduced perfusion pressure [25,27]. New clinical trials are necessary to analyze the role of treatments actually used in these patients. In conclusion, the primary aim of endarterectomy is to remove a source of emboli; however it is effective in restoring the CVR in patients with cerebral hemodynamic impairment. REFERENCES 1. Bushnell DL, Sudha G, Barnes WE, Litooy F, Niemiro M, Steffen G. Evaluation of cerebral perfusion reserve using 5% CO2 and SPECT neuroperfusion imaging. Clin Nuc Med 1991;16:263–7. 2. Challet F, Celsis P, Clanet M, et al. SPECT study of cerebral blood flow reactivity after acetazolamide in patients with transient ischemic attacks. Stroke 1989; 20:458 – 64. 3. Dahl A, Russell D, Nyberg-Hansen R, Rootwelt K, Soren JB. Cerebral vasoreactivity in unilateral carotid artery disease. Stroke 1994;25:621– 6. 4. Fujoka K, Kuhen K, Sola-Pierce N, Spencer MP. Transcranial pulsed doppler for evaluation of cerebral arterial hemodynamics. J Vasc Techn 1989;April:95–9. 5. Gibbs JM, Leenders KL, Wise RJS, Jones T. Evaluation of cerebral perfusion reserve in patients with carotid artery occlusion. The Lancet 1984;28:182– 6. 6. Herold S, Brown MM, Frackowiak RSJ, et al. Assessment of cerebral hemodynamic reserve: correlation between PET parameters and CO2 reactivity measured by intravenous 133 Xenon injection technique. J Neurol Neurosurg Psychiatry 1988;51:1045–50. 7. Kleiser B, Widder B. Course of carotid artery occlusion with impaired D’Angelo 13 cerebravascular reactivity. Stroke 1992;23:171– 4. 8. Kuroda S, Kamiyama H, Abe H, Houkin K, Isobe M, Mitsumori K. Acetazolamide test in detecting reduced cerebral perfusion reserve and predicting long-term

326 Surg Neurol 1999;51:321–6

9.

10. 11.

12. 13. 14.

15. 16.

17.

18. 19.

20. 21.

22.

23. 24.

25.

prognosis in patients with internal carotid artery occlusion. Neurosurgery 1993;32:912–9. Maeda H, Matsumoto M, Handa N, Hougaku H, Ogawa S, et al. Reactivity of cerebral blood flow to carbon dioxide in various types of ischemic cerebrovasculat disease: Evaluation by the transcranial doppler method. Stroke 1993;24:670 –5. Newell D, Aaslid R, Lam A, Mayberg T, Winn R. Comparison of flow and velocity during dynamic autoregulation testing in humans. Stroke 1994;25:793–97. Nighossian N, Touillas P, Philippon B, Itti R, Adeleine P. Cerebral blood flow reserve assessment in symptomatic versus asymptomatic high-grade internal carotid artery stenosis. Stroke 1994;25:1010 –13. Norrving B, Nilsson B, Risberg J. rCBF in patients with carotid occlusion. Resting and collateral flow related to collateral pattern. Stroke 1982;13:155– 62. Polcyn RE, Koeppe R, Hutchins G et al. Hypercapnia and cerebral blood flow: an approach to testing perfusion reserve. J Nucl Med 1984;25:110. Powers WJ, Press GA, Grubb RL, Gado M, Raiche ME. The effect of hemodynamically significant carotid artery disease on the hemodynamic status of cerebral circulation. Ann Intern Med 1987;106:27–35. Powers WJ, Fox PT, Raichle ME. The effect of carotid artery disease on the cerebrovascular response to physiologic stimulation. Neurology 1988;38:1475– 8. Powers WJ, Grubb RL, Raiche ME. Clinical results of extracranial-intracranial bypass surgery in patients with hemodynamic cerebravascular disease. J Neurosurg 1989;70:61–7. Powers WJ, Tempel LW, Grubb RL. Influence of cerebral hemodynamics on stroke risk; one year follow-up of 30 medically treated patients. Ann Neurol 1989;25: 325–30. Powers WJ. Cerebral hemodynamics in ischemic cerebrovascular disease. Ann Neurol 1991;29:231– 40. Ringelstein EB, Sievers C, Ecker S, Schneider PA, Otis SM. Non-invasive assessment of CO2-induced cerebral vasomotor reactivity in normal individuals and patients with internal carotid artery occlusion. Stroke 1988;19:963–9. Ringelstein EB, Otis SM. Physiological testing of vasomotor reserve. In: Newell DW, Aaslid R, eds. Transcranial doppler. New York: Raven Press, 1992;83–99. Rodriguez G, Nobili F, De Carli F, Francione S, Marenco S, Celestino MA, Hassan K, Rosadini G. Regional cerebral blood flow in chronic stroke patients. Stroke 1993;24:94 –9. Russell D, Dybevold S, Kjartansson O, Nyberg-Hansen R, Rootwelt K, Wiberg J. Cerebral vasoreactivity and blood flow before and 3 months after carotid endarterectomy. Stroke 1990;21:1029 –32. Schroeder T. Cerebrovascular reactivity to acetazolamide in carotid artery disease. Neurol Res 1986;8: 231– 6. Vorstrup S, Brun B, Lassen NA. Evaluation of cerebral vasodilating capacity by the acetazolamide test before EC-IC bypass surgery in patients with occlusion of the internal carotid artery. Stroke 1986;17:1291. Weiller C, Ringelstein EB, Reiche W, Buell U. Clinical and hemodynamic aspects of low-flow infarcts. Stroke 1991;22:1117–23.

D’Angelo et al

26. Widder B, Paulat K, Hackspacher J, Mayr E. Transcranial doppler CO2-test for the detection of hemodynamically critical carotid artery stenoses and occlusion. Eur Arch Psychiatry Neurol Sci 1986;236:162– 8. 27. Yonas H, Smith HA, Durham SR, Pentheny SL, Johnson WD. Increased stroke risk predicted by compromised cerebral blood flow reactivity. J Neurosurg 1993;79:483–9.

COMMENTARY

The authors have applied a reasonably well investigated technique of transcranial doppler assessment for cerebrovasomotor reactivity to the novel setting of pre- and post-carotid endarterectomy. The percent change of velocity in the middle cerebral artery from hypocapnia to hypercapnia suggests collateral reserve in the symptomatic hemisphere. Patients were excluded if there was CT evidence of infarction, or symptoms involving the contralateral anterior or posterior circulation. The study findings suggest what is intuitive, that cerebrovascular reactivity improves in the ipsilateral hemisphere after carotid endarterectomy. Patients with the lowest preoperative vasomotor reactivity tended to have the most significant postoperative increase in percentage change. The technique is of interest not for what it infers in these patients who suffered no postoperative neurologic sequelae, but for what may be predictive for patients at risk of neurologic sequelae during carotid endarterectomy. A noninvasive and inexpensive method of depicting preoperative cerebral reserve to the hemisphere in jeopardy that could potentially predict which patients will not fare well with any embolic events or prolonged carotid clamping during surgery, or those who may have a predisposition to post-endarterectomy hyperemia and possible hemorrhagic consequences, would be quite useful. In an age when the risk of invasive angiography is giving way to the less anatomic detail of MRA and carotid ultrasound, and when the costs of medicine are of ever-increasing importance, this technique may prove to be quite beneficial in the future as a predictor of patients with poor preoperative collateral supply and increased risk of hemorrhagic complications. John L. D. Atkinson, M.D., FACS Department of Neurologic Surgery Mayo Clinic Rochester, Minnesota