Cerebrovascular hemodynamic changes associated with carotid endarterectomy Clifford T. Araki, PhD, Viken L. Babildan, MD, Nancy L. Cantelmo, MD, and W'dlard C. Johnson, MD, Boston, Mass To evaluate the effect of carotid surgery on the cerebral circulation, transcranial Doppler sonography and ocular pneumoplethysmography were performed on 36 patients who underwent unilateral carotid endarterectomy. Ocular pneumoplethysmography and transcranial Doppler sonography tests were performed within a week before and after operation, and transcranial Doppler sonography was repeated ->30 days after operation. Middle and anterior cerebral arteries were insonated bilaterally, and flow velocities ipsilateral (iMCAFV or iACAFV) and contralateral (cMCAFV or cACAFV) to the side of surgery were recorded. The iMCAFV and iACAFV increased significantly in the immediate postoperative period, and the iMCAFV remained elevated on the second follow-up study. Patients with _>75% ipsilateral carotid stenosis (N = 23) had increased iMCAFV, iACAFV, and decreased cACAFV after operation, whereas those with < 75% stenosis (N = 13) had no significant transcranial Doppler sonography changes. Those with ->75% contralateral carotid stenosis (N = 17) had significant increases in iMCAFV, cMCAFV, and iACAFV after operation, whereas those with < 75% contrallateral carotid stenosis had no significant transcranial Doppler sonography changes. A subset of patients (N = 13) did not increase iMCAFV after surgery. The ocular pneumoplethysmography changes were significantly different in both groups when preoperative and postoperative values were compared. We conclude that carotid endarterectomy can cause lasting cerebral hemodynamic changes, but that its effects are not uniform among all patients (J VAsc S~G 1991;13:854-60.)
Transcranial Doppler sonography (TCD) provides a means ofnoninvasively evaluating intracranial arteries and collateral circulation around the circle of Willis by measuring blood flow velocity. In patients with significant extracranial carotid artery (ECA) stenosis, changes in the intracerebral circulation would be expected after carotid endarterectomy. The purpose of this study is to evaluate the immediate and delayed effects of carotid surgery on the intracerebral circulation by use of TCD, and ocular pneumoplethysmography (OPG). METHODS Fifty-four male patients with cerebrovascular disease admitted to the Boston Veterans AdminisFrom the Departments of Surgery and Neurology and the Vascular Laboratory, Boston Veterans Administration Medical Center, and the Vascular and Surgical Research Laboratories, Boston University School of Medicine. Presented at the Seventeenth Annual Meeting of the New England Society for Vascular Surgery, Newport, R. I., Sept. 13-14, 1990. Reprint requests: Nancy L Cantelmo, MD, Department of Surgery (112), Boston Veterans Administration Medical Center, 150 South Huntington Ave., Boston MA 02130. 24/6/28621 854
tration Medical Center were included in this study. They consisted of two groups. The first group (Surgery) included 36 patients who underwent cerebral angiography and unilateral carotid endarterectomy. Five of these patients were asymptomatic, and 31 were admitted with either transient ischemic attacks or cerebral infarction in the internal carotid artery (ICA) distribution. The group was evaluated with TCD and OPG within i week before operation (mean, 6 + 1.6 days). A second set of studies were performed within 1 week, but 24 hours after operation (mean, 2 + 1 days). Four patients were lost for follow-up, but the remaining 34 received TCD studies a third time, 30 days or more after operation (mean, 92 _+ 78 days). Patency of the carotid artery after operation was confirmed by duplex scanning, spectral analysis, or OPG in every case.
The second group (Control) was composed of 16 patients with transient or fixed ischemic neurologic deficits referable to the carotid circulation. Transcranial Doppler evaluation was performed in this group to assess technique reproducibility. These patients were selected retrospectively from the list of all patients with cerebrovascular disease tested in the
Volume 13 Number 6 June 1991
TCD laboratory who had received two TCD evaluations less than a week apart (mean, 6 -+ 2.5 days). Patients in the Control group were not subject to any surgical therapy and did not have clinically evident intervening cerebral ischemic events during the time of study. Angiography was performed on patients in the Surgery group by use of either conventional or digital arterial studies. Standard percutaneous technique was used, and biplane studies were obtained. Of the 36 preoperative angiograms, 32 provided intracranial views. Films were evaluated with regard to crosssectional area at the carotid bifurcation, the extent of intracranial disease, and hemispheric cross-filling through the arterior communicating artery. Percent stenosis was determined from biplanar films as follows: % area stenosis = (10011 - A2/B2]) + (100[1 - A12/BU])/2 where A and A ~ represent the smallest ICA diameter in each plane, and B and B ~represent the normal ICA distal to the stenosis in each plane. Transcranial Doppler studies were performed according to the technique described by Aaslid et al.,Z Arnolds and von Reutern, s and Hennerici et al.4 and others. A 2 M H z pulsed, range-gated Doppler was used (TC2-64, EME; Oberlingen, Federal Republic of Germany). Focal depth of the signal varied in 5 mm increments from 45 to 85 mm. Depth of insonation of the probe and direction of flow were used to identify the particular vessel studied. By use of the transtemporal window, Doppler signals were obtained for the middle cerebral artery (MCA) at a depth of 50 to 55 mm and from the anterior cerebral artery (ACA) at a depth of 70 to 75 mm and expressed as peak systolic flow velocity. The MCA and ACA were insonated in most patients. Peak systolic flow velocities were used in the analyses, and were noted as either ipsilateral (iMCAFV or iACAFV) or contralateral (cMCAFV or cACAFV) to the side of surgery. Ocular pneumoplethysmography (OPG-Gee Model OPG 4; Electrodiagnostic Instruments; Burbank, Calif.) was performed and interpreted in the method outlined by Gee. s The delta OPG was obtained by calculating the difference between ophthalmic artery systolic pressures on each side. Changes in MCA and ACA peak systolic velocity were analyzed through one way analysis of variance (ANOVA) with a posteriori testing by least significant difference. Patients in the Surgery group were further analyzed in subsets to better define the
Cerebrovascular changes with carotid endarterectomy 855
response of the sample population. Results of these subset groups are analyzed by obtaining the differences between the preoperative and first postoperative results of each individual patient and calculating the means of the value. A negative number indicates a decrease in the postoperative value compared with the preoperative study. Results are expressed as mean + standard error. Analysis of subsets were conducted by t test, paired t test, and correlation statistics. All analyses were calculated by use of the Statistical Analysis System (SAS Institute Inc., Cary, N.C.) on the IBM-XT (IBM Corp., Armonk, N.Y.) computer in the Surgical Research Laboratory at Boston University School of Medicine. RESULTS
The mean age of patients in the Control group was 67 _+ 7 years. The average MCA peak systolic velocity was 87.6 + 6.2 cm/sec with a mean difference of 2.4 +_ 3.1 cm/sec on a repeat study performed an average of 6 _+ 2.5 days between studies. The mean age of patients in the Surgery group was 64 _+ 4 years. The mean MCA peak systolic velocity in this group was 81.2 _+ 4.8 cm/sec, not significantly different from patients in the Control group. An analysis of percent stenosis and the ipsilateral velocity with both right and left hemispheres demonstrated a significant negative regression (p = 0.0106) with an r: of 0.164. No postoperative deaths occurred, and one stroke in the Surgery group occurred 3 weeks after operation. The pattern of cerebrovascular disease and hemispheric crossfiUing, as demonstrated by angiogram, varied among patients in the Surgery group. Twenty-six patients had a _ 75% stenosis on the side of surgery. Six patients were admitted with a - 75% stenosis on the contrallateral side and a lesser stenosis on the side of surgery. Bilateral lesions, of - 7 5 % were noted in 12 patients, and four patients had < 75% stenosis bilaterally. No significant intracranial disease was evident by angiography. Nineteen patients demonstrated hemispheric crosstilting through the anterior communicating artery, six of whom had _ 75% ICA stenotic disease. On the first postoperative evaluation after revascularization, significant increases in group mean iMCAFV and iACAFV were found (Table IA). A smaller increase in cMCAFV and a decrease in cACAFV were present on the contralateral side but not significant. Postoperative flow velocity changes became significant for all values when data were evaluated as mean differences between the preoperative and postoperative flow velocities computed for
Journal of VASCULAR SURGERY
856 Araki et al.
Table IA. Transcranial Doppler studies on patients in the Surgery group (n = 36) comparing preoperative to postoperative values Flow velocity cm/sec
Preoperative
Postoperative 1
Postoperative 2
iMCAFV cMCAFV iACAFV cACAFV
81.2 84.1 88.3 98.8
100.7 94.1 107.4 84.8
94.7 92.6 105.0 89.1
-+ 4.8 -+ 4.4 + 5.6 --- 6.3
+ 6.1" _+ 5.9 _+ 6.4* -+ 5.1
_+ 4.6* _+ 5.8 _+ 5.1 -+ 6.2
iMCAFV, Ipsilateral middle cerebral artery flow velocity; cMCAFV, contralateral middle cerebral flow velocity; iACAFV, ipsilateral anterior cerebral artery flow velocity;cACAFV, contralateral anterior cerebral artery flowvelocity. *p < 0.05.
each patient. Comparisons were performed by paired t test (Table IB). These same trends were maintained in the second postoperative evaluation with the iMCAFV and cACAFV remaining significantly elevated when compared with preoperative values (Table IA). Velocities in the Control group remained stable when the studies were compared, with a mean difference of 2.4 cm/sec. Velocities in the Surgery group increased 19.4 cm/sec between the preoperative and postoperative measurements. The difference between these two values is significant by t test kO < 0.05). Patients in the Surgery group were subgrouped according to the degree of ipsilateral carotid stenosis (Table II). Patients in group II, those with a stenosis --_75% on the side of operation (n = 26), showed a significant increase in iMCAFV and iACAFV after surgery. The cACAFV decreased significantly, whereas cMCAFV was not significantly affected. With the exception of cMCAFV, no significant changes were noted in group I, which consisted of patients with < 75% reduction on the side of surgery, (n = 10). Five of these patients had _>75% contralateral carotid stenosis, and five had 75% had significant increases in iMCAFV, cMCAFV, and iACAFV, whereas patients in group A with contralateral carotid stenosis of < 75% had no significant changes. A subset of patients in the Surgery group did not show an increase in iMCAFV. Additional analysis was performed to better define flow velocity changes in these patients. Patients who failed to show a
Table IB. Means of individual patient differences in those in the Surgery group (N = 36) comparing preoperative to postoperative values of transcranial Doppler studies Differences in flow velocities
Postoperative 1 to preoperative
Postoperative 2 to preoperative
iMCAFV cMCAFV iACAFV cACAFV
19.4 9.8 21.3 -14.1
18.3 10,3 20.4 -16,1
+ 5.8** -+ 4.0* + 7.5** _+ 5.8*
_+ 4.8** _+ 4.6* _+ 6.8 ~* _+ 5.5 *~
iMCAFV, Ipsilateral middle cerebral artery flow velocity; cMCAFV, contralateral middle cerebral flow velocity; iACAFV, ipsilateral anterior cerebral artery flow velocity; cACAFV, contralateral anterior cerebral artery flow velocity. *p < 0.05, **p < 0.01.
postoperative increase in iMCAFV of 10 cm/sec or more (n = 13) were compared with those that did (n = 23). An increase of 10 cm/sec represents a 95% confidence level. The 23 patients not only demonstrated an increase in iMCAFV after operation but also showed a significant increase in cMCAFV and iACAFV, which was accompanied by a decrease in cACAFV. For the 13 patients, not only was iMCAFV not elevated, no significant changes in iACAFV, cACAFV, nor cMCAFV were found. In addition, 12 of 13 patients demonstrated lower postoperative iMCAFV compared with preoperative values. Reductions ranged from - 5 to - 3 7 cm/sec. Changes persisted in the second postoperative measurement in all but two cases, in which increases to expected levels were noted. Patients in the subgroup that did improve iMCAFV were not homogeneous with respect to carotid disease. Six had only ipsilateral ---75% stenosis, three had bilateral ___75% stenosis, three had only contralateral _>75% stenosis, and one had no stenosis ___75% on either side. Patients in the two groups that did not improve iMCAFV did not differ significantly with respect to degree of stenosis on the side of operation or to differences between sides. Crossfilling from one hemisphere to the other through the anterior communicating artery also did not differ between groups. O f the patients in the Surgery group, 14 had preoperative angiographic and TCD evidence of collateral crossflow through the anterior communicating artery. The iACAFVs markedly decreased after operation in 8 of 14 patients suggesting a drop in collateral flow, and remained unchanged in 6. The latter had preoperative angiographic findings (such as ICA occlusion contralateral to the side of surgery),
Volume 13 Number 6 June 1991
Cerebrovascular changes with carotid endarterectomy
Table II. Means of individual patient differences comparing preoperative and first postoperative values of transcranial Doppler studies; group I (N = 10) with ipsilateral internal carotid artery stenosis of < 75% and group II (N = 26) with ipsilateral internal carotid artery stenosis of -> 75% Flow velocity (cm/sec) iMCAFV cMCAFV iACAFV cACAFV
Group I 1.3 19.4 3.0 -7.0
-+ 7.4 _+ 7.8* + 5.2 -+ 7.0
Group II 26.4 5.9 30.5 -17.2
_+ 7.1"* _+ 4.5 --- 10.5"* _+ 7.8*
857
TaMe III. Means of individual patient differences comparing preoperative and first postoperative transcranial Doppler values of group A (n = 19) with contralateral internal carotid artery stenosis of < 75% and group B (n = 17) with contralateral internal carotid artery stenosis of > 75% Flow velocity (cm/sec) iMCAFV cMCAFV iACAFV cACAFV
GroupA 13.4 10.0 18.7 -15.9
___ 7.4 _+ 6.7 -+ 11.1 _+ 9.0
Group B 26.2 9.6 24.0 -12.3
__+ 9.0** -+ 3.9* -+ 10.5" + 7.8
*p < 0.05, **p < 0.01.
*p < 0.05, **p < 0.01.
which explained the absence of postoperative change in ACAFV. Delta OPG represents the difference between the two carotid arteries of each patient. The mean of individual patient differences comparing preoperative to postoperative delta OPG was calculated. This difference was significant for the entire group at 12.3 _+ 1.7. Patients with 75% carotid stenosis, group II, had a mean delta OPG difference of 12.8 _.+ 2.0. Mean preoperative to postoperative difference in patients in group A with < 75% contralateral carotid stenosis was 9.9 _ 2.3 and patients in group B with _>75 % contralateral stenosis was 12.6 _+ 2.6. The OPG difference in the 13 patients who did not improve iMCAFV was 10.9 -+ 3.1. Those 23 patients who did improve iMCAFV had a mean preoperative to postoperative delta OPG difference of 11.0 + 2.0. All these preoperative to postoperative delta OPG changes were significant to p < 0.01.
Wechsler et al. 9'~° have demonstrated increases in MCA flow velocities on the side of endarterectomy. These studies did not have controls to confirm their finding, evaluated smaller study samples, and did not address the issue of subgroups. Schneider et al.s also showed that the increase in iMCAFV was more marked in patients with > 80% carotid stenosis. Our findings confirm and further expand these observations. The critical ICA stenosis causing distal hemodynamic change has been reported to range between 60% and 80% cross-sectional area reduction. 1~'~2 Severely stenotic ICA lesions can decrease cerebral blood flow and vasoreactivity, 13 and can reduce perfusion pressure. 1~ Our data indicate that the severity of hemodynamic change is proportional to the degree of ICA stenosis as demonstrated by the significant correlation between preoperative values of the latter and the MCAFV. In addition, patients with lesions causing more than 75% ipsilateral and bilateral stenoses have postoperative intracranial hemodynamic changes that involve flow in the MCA, ACA, and ophthalmic artery, all branches of the ICA. We also fotmd that the effect of carotid endarterectomy on iMCAFV is not limited to the perioperative period but extends to the chronic stage after operation, suggesting that the original changes are unlikely to be the effect of anesthetic agents on cerebral autoregulation or transient postoperative cerebral hyperemia. 14 Our data indicate that the intracranial hemodynamic effects of cervical flow limiting lesions can involve more than one vascular territory and induce bilateral responses. The postoperative drop in the ACAFV contralateral to the endarterectomized side indicates the presence of a dynamic equilibrium through the circle of Willis. Some of our findings are not readily explained.
DISCUSSION In this report we evaluate the cerebral hemodynamic effects of carotid revascularization procedures. Using TCD and OPG we show that several intracranial flow indexes change after carotid endarterectomy, and that this change persists in the chronic stage after operation. However, the change of the intracranial hemodynamic state is not uniform, as 13 of our 36 patients had small or even decreased postoperative changes in cerebral flow. The use of TCD sonography in monitoring intracranial blood flow during carotid endarterectomy has been described, 6-8 but little has been published on the effects of carotid endarterectomy on intracranial flow velocities. Schneider et al.8 and
lournal of VASCULAR SURGERY
858 Araki et al.
For example, why should patients with and without contralateral significant carotid stenosis improve their cMCAFV after endarterectomy of
Transcranial Doppler sonography (TCD) provides a means ofnoninvasively evaluating intracranial arteries and collateral circulation around the circle of Willis by measuring blood flow velocity. In patients with significant extracranial carotid artery (ECA) stenosis, changes in the intracerebral circulation would be expected after carotid endarterectomy. The purpose of this study is to evaluate the immediate and delayed effects of carotid surgery on the intracerebral circulation by use of TCD, and ocular pneumoplethysmography (OPG). METHODS Fifty-four male patients with cerebrovascular disease admitted to the Boston Veterans AdminisFrom the Departments of Surgery and Neurology and the Vascular Laboratory, Boston Veterans Administration Medical Center, and the Vascular and Surgical Research Laboratories, Boston University School of Medicine. Presented at the Seventeenth Annual Meeting of the New England Society for Vascular Surgery, Newport, R. I., Sept. 13-14, 1990. Reprint requests: Nancy L Cantelmo, MD, Department of Surgery (112), Boston Veterans Administration Medical Center, 150 South Huntington Ave., Boston MA 02130. 24/6/28621 854
tration Medical Center were included in this study. They consisted of two groups. The first group (Surgery) included 36 patients who underwent cerebral angiography and unilateral carotid endarterectomy. Five of these patients were asymptomatic, and 31 were admitted with either transient ischemic attacks or cerebral infarction in the internal carotid artery (ICA) distribution. The group was evaluated with TCD and OPG within i week before operation (mean, 6 + 1.6 days). A second set of studies were performed within 1 week, but 24 hours after operation (mean, 2 + 1 days). Four patients were lost for follow-up, but the remaining 34 received TCD studies a third time, 30 days or more after operation (mean, 92 _+ 78 days). Patency of the carotid artery after operation was confirmed by duplex scanning, spectral analysis, or OPG in every case.
The second group (Control) was composed of 16 patients with transient or fixed ischemic neurologic deficits referable to the carotid circulation. Transcranial Doppler evaluation was performed in this group to assess technique reproducibility. These patients were selected retrospectively from the list of all patients with cerebrovascular disease tested in the
Volume 13 Number 6 June 1991
TCD laboratory who had received two TCD evaluations less than a week apart (mean, 6 -+ 2.5 days). Patients in the Control group were not subject to any surgical therapy and did not have clinically evident intervening cerebral ischemic events during the time of study. Angiography was performed on patients in the Surgery group by use of either conventional or digital arterial studies. Standard percutaneous technique was used, and biplane studies were obtained. Of the 36 preoperative angiograms, 32 provided intracranial views. Films were evaluated with regard to crosssectional area at the carotid bifurcation, the extent of intracranial disease, and hemispheric cross-filling through the arterior communicating artery. Percent stenosis was determined from biplanar films as follows: % area stenosis = (10011 - A2/B2]) + (100[1 - A12/BU])/2 where A and A ~ represent the smallest ICA diameter in each plane, and B and B ~represent the normal ICA distal to the stenosis in each plane. Transcranial Doppler studies were performed according to the technique described by Aaslid et al.,Z Arnolds and von Reutern, s and Hennerici et al.4 and others. A 2 M H z pulsed, range-gated Doppler was used (TC2-64, EME; Oberlingen, Federal Republic of Germany). Focal depth of the signal varied in 5 mm increments from 45 to 85 mm. Depth of insonation of the probe and direction of flow were used to identify the particular vessel studied. By use of the transtemporal window, Doppler signals were obtained for the middle cerebral artery (MCA) at a depth of 50 to 55 mm and from the anterior cerebral artery (ACA) at a depth of 70 to 75 mm and expressed as peak systolic flow velocity. The MCA and ACA were insonated in most patients. Peak systolic flow velocities were used in the analyses, and were noted as either ipsilateral (iMCAFV or iACAFV) or contralateral (cMCAFV or cACAFV) to the side of surgery. Ocular pneumoplethysmography (OPG-Gee Model OPG 4; Electrodiagnostic Instruments; Burbank, Calif.) was performed and interpreted in the method outlined by Gee. s The delta OPG was obtained by calculating the difference between ophthalmic artery systolic pressures on each side. Changes in MCA and ACA peak systolic velocity were analyzed through one way analysis of variance (ANOVA) with a posteriori testing by least significant difference. Patients in the Surgery group were further analyzed in subsets to better define the
Cerebrovascular changes with carotid endarterectomy 855
response of the sample population. Results of these subset groups are analyzed by obtaining the differences between the preoperative and first postoperative results of each individual patient and calculating the means of the value. A negative number indicates a decrease in the postoperative value compared with the preoperative study. Results are expressed as mean + standard error. Analysis of subsets were conducted by t test, paired t test, and correlation statistics. All analyses were calculated by use of the Statistical Analysis System (SAS Institute Inc., Cary, N.C.) on the IBM-XT (IBM Corp., Armonk, N.Y.) computer in the Surgical Research Laboratory at Boston University School of Medicine. RESULTS
The mean age of patients in the Control group was 67 _+ 7 years. The average MCA peak systolic velocity was 87.6 + 6.2 cm/sec with a mean difference of 2.4 +_ 3.1 cm/sec on a repeat study performed an average of 6 _+ 2.5 days between studies. The mean age of patients in the Surgery group was 64 _+ 4 years. The mean MCA peak systolic velocity in this group was 81.2 _+ 4.8 cm/sec, not significantly different from patients in the Control group. An analysis of percent stenosis and the ipsilateral velocity with both right and left hemispheres demonstrated a significant negative regression (p = 0.0106) with an r: of 0.164. No postoperative deaths occurred, and one stroke in the Surgery group occurred 3 weeks after operation. The pattern of cerebrovascular disease and hemispheric crossfiUing, as demonstrated by angiogram, varied among patients in the Surgery group. Twenty-six patients had a _ 75% stenosis on the side of surgery. Six patients were admitted with a - 75% stenosis on the contrallateral side and a lesser stenosis on the side of surgery. Bilateral lesions, of - 7 5 % were noted in 12 patients, and four patients had < 75% stenosis bilaterally. No significant intracranial disease was evident by angiography. Nineteen patients demonstrated hemispheric crosstilting through the anterior communicating artery, six of whom had _ 75% ICA stenotic disease. On the first postoperative evaluation after revascularization, significant increases in group mean iMCAFV and iACAFV were found (Table IA). A smaller increase in cMCAFV and a decrease in cACAFV were present on the contralateral side but not significant. Postoperative flow velocity changes became significant for all values when data were evaluated as mean differences between the preoperative and postoperative flow velocities computed for
Journal of VASCULAR SURGERY
856 Araki et al.
Table IA. Transcranial Doppler studies on patients in the Surgery group (n = 36) comparing preoperative to postoperative values Flow velocity cm/sec
Preoperative
Postoperative 1
Postoperative 2
iMCAFV cMCAFV iACAFV cACAFV
81.2 84.1 88.3 98.8
100.7 94.1 107.4 84.8
94.7 92.6 105.0 89.1
-+ 4.8 -+ 4.4 + 5.6 --- 6.3
+ 6.1" _+ 5.9 _+ 6.4* -+ 5.1
_+ 4.6* _+ 5.8 _+ 5.1 -+ 6.2
iMCAFV, Ipsilateral middle cerebral artery flow velocity; cMCAFV, contralateral middle cerebral flow velocity; iACAFV, ipsilateral anterior cerebral artery flow velocity;cACAFV, contralateral anterior cerebral artery flowvelocity. *p < 0.05.
each patient. Comparisons were performed by paired t test (Table IB). These same trends were maintained in the second postoperative evaluation with the iMCAFV and cACAFV remaining significantly elevated when compared with preoperative values (Table IA). Velocities in the Control group remained stable when the studies were compared, with a mean difference of 2.4 cm/sec. Velocities in the Surgery group increased 19.4 cm/sec between the preoperative and postoperative measurements. The difference between these two values is significant by t test kO < 0.05). Patients in the Surgery group were subgrouped according to the degree of ipsilateral carotid stenosis (Table II). Patients in group II, those with a stenosis --_75% on the side of operation (n = 26), showed a significant increase in iMCAFV and iACAFV after surgery. The cACAFV decreased significantly, whereas cMCAFV was not significantly affected. With the exception of cMCAFV, no significant changes were noted in group I, which consisted of patients with < 75% reduction on the side of surgery, (n = 10). Five of these patients had _>75% contralateral carotid stenosis, and five had 75% had significant increases in iMCAFV, cMCAFV, and iACAFV, whereas patients in group A with contralateral carotid stenosis of < 75% had no significant changes. A subset of patients in the Surgery group did not show an increase in iMCAFV. Additional analysis was performed to better define flow velocity changes in these patients. Patients who failed to show a
Table IB. Means of individual patient differences in those in the Surgery group (N = 36) comparing preoperative to postoperative values of transcranial Doppler studies Differences in flow velocities
Postoperative 1 to preoperative
Postoperative 2 to preoperative
iMCAFV cMCAFV iACAFV cACAFV
19.4 9.8 21.3 -14.1
18.3 10,3 20.4 -16,1
+ 5.8** -+ 4.0* + 7.5** _+ 5.8*
_+ 4.8** _+ 4.6* _+ 6.8 ~* _+ 5.5 *~
iMCAFV, Ipsilateral middle cerebral artery flow velocity; cMCAFV, contralateral middle cerebral flow velocity; iACAFV, ipsilateral anterior cerebral artery flow velocity; cACAFV, contralateral anterior cerebral artery flow velocity. *p < 0.05, **p < 0.01.
postoperative increase in iMCAFV of 10 cm/sec or more (n = 13) were compared with those that did (n = 23). An increase of 10 cm/sec represents a 95% confidence level. The 23 patients not only demonstrated an increase in iMCAFV after operation but also showed a significant increase in cMCAFV and iACAFV, which was accompanied by a decrease in cACAFV. For the 13 patients, not only was iMCAFV not elevated, no significant changes in iACAFV, cACAFV, nor cMCAFV were found. In addition, 12 of 13 patients demonstrated lower postoperative iMCAFV compared with preoperative values. Reductions ranged from - 5 to - 3 7 cm/sec. Changes persisted in the second postoperative measurement in all but two cases, in which increases to expected levels were noted. Patients in the subgroup that did improve iMCAFV were not homogeneous with respect to carotid disease. Six had only ipsilateral ---75% stenosis, three had bilateral ___75% stenosis, three had only contralateral _>75% stenosis, and one had no stenosis ___75% on either side. Patients in the two groups that did not improve iMCAFV did not differ significantly with respect to degree of stenosis on the side of operation or to differences between sides. Crossfilling from one hemisphere to the other through the anterior communicating artery also did not differ between groups. O f the patients in the Surgery group, 14 had preoperative angiographic and TCD evidence of collateral crossflow through the anterior communicating artery. The iACAFVs markedly decreased after operation in 8 of 14 patients suggesting a drop in collateral flow, and remained unchanged in 6. The latter had preoperative angiographic findings (such as ICA occlusion contralateral to the side of surgery),
Volume 13 Number 6 June 1991
Cerebrovascular changes with carotid endarterectomy
Table II. Means of individual patient differences comparing preoperative and first postoperative values of transcranial Doppler studies; group I (N = 10) with ipsilateral internal carotid artery stenosis of < 75% and group II (N = 26) with ipsilateral internal carotid artery stenosis of -> 75% Flow velocity (cm/sec) iMCAFV cMCAFV iACAFV cACAFV
Group I 1.3 19.4 3.0 -7.0
-+ 7.4 _+ 7.8* + 5.2 -+ 7.0
Group II 26.4 5.9 30.5 -17.2
_+ 7.1"* _+ 4.5 --- 10.5"* _+ 7.8*
857
TaMe III. Means of individual patient differences comparing preoperative and first postoperative transcranial Doppler values of group A (n = 19) with contralateral internal carotid artery stenosis of < 75% and group B (n = 17) with contralateral internal carotid artery stenosis of > 75% Flow velocity (cm/sec) iMCAFV cMCAFV iACAFV cACAFV
GroupA 13.4 10.0 18.7 -15.9
___ 7.4 _+ 6.7 -+ 11.1 _+ 9.0
Group B 26.2 9.6 24.0 -12.3
__+ 9.0** -+ 3.9* -+ 10.5" + 7.8
*p < 0.05, **p < 0.01.
*p < 0.05, **p < 0.01.
which explained the absence of postoperative change in ACAFV. Delta OPG represents the difference between the two carotid arteries of each patient. The mean of individual patient differences comparing preoperative to postoperative delta OPG was calculated. This difference was significant for the entire group at 12.3 _+ 1.7. Patients with 75% carotid stenosis, group II, had a mean delta OPG difference of 12.8 _.+ 2.0. Mean preoperative to postoperative difference in patients in group A with < 75% contralateral carotid stenosis was 9.9 _ 2.3 and patients in group B with _>75 % contralateral stenosis was 12.6 _+ 2.6. The OPG difference in the 13 patients who did not improve iMCAFV was 10.9 -+ 3.1. Those 23 patients who did improve iMCAFV had a mean preoperative to postoperative delta OPG difference of 11.0 + 2.0. All these preoperative to postoperative delta OPG changes were significant to p < 0.01.
Wechsler et al. 9'~° have demonstrated increases in MCA flow velocities on the side of endarterectomy. These studies did not have controls to confirm their finding, evaluated smaller study samples, and did not address the issue of subgroups. Schneider et al.s also showed that the increase in iMCAFV was more marked in patients with > 80% carotid stenosis. Our findings confirm and further expand these observations. The critical ICA stenosis causing distal hemodynamic change has been reported to range between 60% and 80% cross-sectional area reduction. 1~'~2 Severely stenotic ICA lesions can decrease cerebral blood flow and vasoreactivity, 13 and can reduce perfusion pressure. 1~ Our data indicate that the severity of hemodynamic change is proportional to the degree of ICA stenosis as demonstrated by the significant correlation between preoperative values of the latter and the MCAFV. In addition, patients with lesions causing more than 75% ipsilateral and bilateral stenoses have postoperative intracranial hemodynamic changes that involve flow in the MCA, ACA, and ophthalmic artery, all branches of the ICA. We also fotmd that the effect of carotid endarterectomy on iMCAFV is not limited to the perioperative period but extends to the chronic stage after operation, suggesting that the original changes are unlikely to be the effect of anesthetic agents on cerebral autoregulation or transient postoperative cerebral hyperemia. 14 Our data indicate that the intracranial hemodynamic effects of cervical flow limiting lesions can involve more than one vascular territory and induce bilateral responses. The postoperative drop in the ACAFV contralateral to the endarterectomized side indicates the presence of a dynamic equilibrium through the circle of Willis. Some of our findings are not readily explained.
DISCUSSION In this report we evaluate the cerebral hemodynamic effects of carotid revascularization procedures. Using TCD and OPG we show that several intracranial flow indexes change after carotid endarterectomy, and that this change persists in the chronic stage after operation. However, the change of the intracranial hemodynamic state is not uniform, as 13 of our 36 patients had small or even decreased postoperative changes in cerebral flow. The use of TCD sonography in monitoring intracranial blood flow during carotid endarterectomy has been described, 6-8 but little has been published on the effects of carotid endarterectomy on intracranial flow velocities. Schneider et al.8 and
lournal of VASCULAR SURGERY
858 Araki et al.
For example, why should patients with and without contralateral significant carotid stenosis improve their cMCAFV after endarterectomy of
