MRA Flow Quantification in Patients With a Symptomatic Internal Carotid Artery Occlusion
Background and Purpose Flow measurements in the collateral arteries of patients with internal carotid artery (ICA) occlusions may be important to estimate the risk of future stroke. Quantitative flow measurements in cerebropetal vessels can be reliably assessed by means of magnetic resonance angiography (MRA).
Methods Fifty-four patients with transient or minor ischemic neurological deficits and an angiographically proven ICA occlusion and 16 control subjects underwent two-dimensional phase-contrast MRA quantitative flow measurements through the common carotid arteries, basilar artery, ICAs, and middle cerebral arteries (MCA).
Results Patients with a unilateral ICA occlusion and a 0% to 69% stenosis of the contralateral ICA had increased flow in the contralateral ICA (P<.005) and in the basilar artery (P<.005) compared with control subjects. Even patients with a 70% to 99% stenosis contralateral to the ICA occlusion had increased flow in the ICA (P<.05) as well as increased flow in the basilar artery (P<.001). Total cerebropetal flow was not significantly different between these patients and control subjects. Patients with bilateral ICA occlusions had an increased flow in the basilar artery (P<.001), while the total cerebropetal flow was less than in control subjects (P<.001). In all patients, flow was decreased in the ipsilateral MCA (P<.001) and in the contralateral MCA (P<.05).
Conclusions The contralateral ICA is the main supplying artery in patients with an ICA occlusion. Total cerebropetal flow decreases only when both ICAs are occluded. In patients with symptomatic ICA occlusions, an open contralateral ICA is probably important to retain the cerebral blood flow within normal limits.
- carotid artery occlusion
- cerebral blood flow
- cerebral ischemia
- magnetic resonance angiography
Patients with transient or minor neurological deficits and an occlusion of the ICA have an annual stroke risk of approximately 6%.1 2 3 Both hemodynamic and thromboembolic origins have been postulated as the cause of stroke in these patients.4 5 6 7
In case of an ICA occlusion, cerebral perfusion pressure drops. Regional cerebral blood flow is, however, maintained as long as possible by vasodilatation to reduce cerebral vascular resistance (autoregulation) and by rerouting of the blood flow through compensatory pathways.5 Compensatory flow is possible through the contralateral ICA, BA, ophthalmic arteries, or leptomeningeal vessels. The latter two pathways seem to be particularly important if compensatory flow through the ICAs and the BA is inadequate.5 8
MR flow quantification is a fast, noninvasive, and widely available method to determine blood flow in the major cerebral arteries.9 10 11 12 13 14 15 In patients with carotid disease, it can be used to investigate the (re-) distribution of blood flow through various cerebral arteries. MR flow quantification techniques have been applied to study patients with a unilateral carotid stenosis.9 10 11 12 14 These studies showed that in patients with an ICA stenosis of more than 70%, flow is decreased compared with control subjects in the ipsilateral CCA11 12 and in the ipsilateral ICA.9 11 12 14 Only one study investigated flow in the contralateral CCA, the contralateral ICA, the BA, and the MCAs.12 This study showed that flow in both the contralateral CCA and the MCAs was decreased and that flow in the BA was increased. Flow in the contralateral ICA and total cerebropetal flow, calculated as cumulative flow through both ICAs and the BA, were unaltered in comparison with control subjects.12
Although arterial flow measurements in the cerebropetal arteries provide no information on the regional cerebral blood flow, they do indicate which arteries are important in the supply of blood to the brain. Changes in cerebropetal blood flow have, to the best of our knowledge, not been studied in patients with ICA occlusions. Particularly in patients with bilateral ICA occlusions, the flow distribution is likely to be altered.
The purpose of this study was to study the (re-) distribution of flow in the major cerebropetal arteries in patients with unilateral or bilateral occlusions of the ICA. We also investigated whether the total flow to the brain was altered in these patients and whether ICA occlusions affected the flow through the MCAs.
Subjects and Methods
Fifty-six consecutive patients with transient or minor retinal or hemispheric ischemia and an angiographically proven occlusion of the ipsilateral ICA were referred to the departments of neurology and vascular surgery between May 1995 and November 1996. Fifty-four of these patients were included in this study; two patients were not included because the MR flow quantification study could not be performed because of claustrophobia or metal artifact. Patients with a severe stroke in the past causing major disabilities were not included in this study.
Patients were divided into three groups: (1) patients with a unilateral occlusion of the ICA and a 0% to 69% contralateral ICA stenosis (n=27; mean age, 58 years; range, 42 to 82 years; 22 men, 5 women); (2) patients with a unilateral occlusion of the ICA and a severe (70% to 99%) contralateral ICA stenosis (n=17; mean age, 65 years; range, 37 to 82 years; 13 men, 4 women); and (3) patients with bilateral ICA occlusions (n=10; mean age, 62 years; range, 49 to 71 years; 9 men, 1 woman). The control group consisted of 16 subjects (mean age, 59 years; range, 40 to 73 years; 11 men, 5 women) who were treated in the Department of Neurology for other than intracranial diseases and who had no history of ischemic neurological deficits. These control subjects all had a normal MRI of the brain and a normal MRA of the cerebropetal arteries.
Study protocols were approved by the Human Research Committee of our hospital.
Cerebropetal vessels were investigated by iaDSA in all patients. The degree of stenosis was measured according to the North American Symptomatic Carotid Endarterectomy Trial criteria.16 Two patients had their angiography more than 1 year before the MR investigation. Reexamination of the ICAs in these two patients by means of color Doppler–assisted duplex, less than a week before the MR investigation, showed no alterations of the ICA lesions compared with the angiographic findings. In all other patients, contrast angiography was performed between 9 months before and 2 months after the MR investigation (median, 20 days before). Eight of these patients had their angiography more than 3 months before the MR investigation. One patient had an ICA occlusion in the C2 segment, distal of the ophthalmic artery branch. In two patients with ICA occlusions assessed by ultrasound examination, iaDSA showed a pseudo-occlusion of the ICA with no contribution to the cerebral perfusion. These ICAs were considered functionally occluded. In another patient, an ipsilateral ICA stenosis of 99% was diagnosed with iaDSA, 61 days before the MR investigation. However, the ultrasound examination, which was performed 5 days before the MR investigation, showed an occlusion of this ICA. MRI and MRA also did not show flow in this ICA; it therefore was considered occluded as well.
MRI and MRA
MRI and MRA studies were performed on a Philips Gyroscan ACS-NT 15 whole body system operating at 1.5 T (Philips Medical Systems). All patients and control subjects underwent the same MR protocol of the brain and cerebropetal arteries. The MRI examination consisted of 19 sagittal T1-weighted scout images (TR, 545 ms; TE, 15 ms; slice thickness, 4 mm; slice gap, 0.6 mm) and 15 transversal/oblique T2-weighted images (T2 turbo spin-echo; TR, 2462 ms; TE, 24 and 100 ms; slice thickness, 6 mm; slice gap, 1.5 mm). After MRI, quantitative flow measurements were performed in the CCAs, ICAs, BA, and MCAs. These measurements were performed with previously optimized scan protocols, showing the accuracy and precision with which time-averaged flow velocity can be measured by a nontriggered 2D phase-contrast sequence.17 These protocols feature a radiofrequency spoiled gradient-echo sequence with full echo sampling17 and were validated in vivo.18 2D phase-contrast flow measurements through the CCAs were performed perpendicular to the arteries, 20 to 30 mm below the carotid bifurcation (slice thickness, 5 mm; field of view, 250×250 mm; TR, 16 ms; TE, 9 ms; flip angle, 7.5°; velocity sensitivity, 1000 mm/s; and 8 averages) (Fig 1A⇓). Similar measurements were performed through the C3 segments of both ICAs and through the BA (Fig 1B⇓). These measurements were followed by two separate 2D phase- contrast quantitative flow measurements for the left and right MCA (slice thickness, 5 mm; field of view, 250×250 mm; TR, 17 ms; TE, 10 ms; flip angle, 8°; velocity sensitivity, 700 mm/s; and 24 averages) (Fig 1C⇓ and 1D⇓). All volume flow data were obtained by integrating across manually drawn regions of interest that enclosed the vessel lumen as closely as possible. Flow in the ECAs was calculated by subtracting flow through the ICA from flow through the CCA. Total cerebropetal blood flow was calculated by cumulating the flow through the contralateral ICA and the flow through the BA. Flow through the ophthalmic arteries was too small to be reliably assessed with our MR system.
The direction of blood flow in the A1 segment of both ACAs was measured with a 2D phase-contrast method, phase encoded in the anterior-posterior and left-right directions, with the following parameters: TR, 16 ms; TE:, 9.1 ms; flip angle, 7.5°; field of view, 250×250 mm; velocity sensitivity, 1000 mm/s; and 8 averages (Fig 1E⇑ and 1F⇑). Directional flow measurements through the posterior communicating arteries were not included in this study. Total study time per patient was approximately 30 minutes, of which 20 minutes were used for MRA.
For statistical analysis ANOVA was used to compare flow data by groups. Analysis of differences in flow in the contralateral ICA, BA, both ECAs, and MCAs was performed by the nonparametric Mann-Whitney U and Wilcoxon rank sum W tests. Statistical significance was corrected for repeated measures. Probability values were considered significant if P<.05. All flow data (Table⇓) are expressed as median and 25th and 75th percentiles. The median flow in patients is also expressed as a percentage of the median flow in control subjects.
Fig 1⇑ shows the images used for MR flow quantification and MR directional flow studies at different levels.
In control subjects, no significant differences in flow were found between the left and right ECA, ICA, or MCA. The mean of the flow measurements through the left and right corresponding arteries was therefore used to obtain normal values in control subjects (Table⇑).
The mean interval between the most recent ischemic event and the MR investigation was 2.6 months (range, 0 days to 7.2 months). In 4 patients, flow measurements failed in one or more arteries as a result of technical errors. In 3 patients, total flow could not be calculated. In 1 patient, iaDSA showed a 95% stenosis of the contralateral ICA, but MRA did not show flow (interval from iaDSA to MRA, 13 days).
In patients with a 0% to 69% stenosis of the contralateral ICA, the median flow in the contralateral ICA and in the BA was increased compared with control subjects (Table⇑). Flow in the MCAs was decreased on both the ipsilateral and contralateral sides, with the largest decrease on the ipsilateral side (ipsilateral MCA versus contralateral MCA, P<.05).
In patients with a 70% to 99% stenosis of the contralateral ICA, again, flow in the contralateral ICA and in the BA was increased in comparison with control subjects, whereas flow in both MCAs was decreased. In this group of patients, iaDSA showed a mean stenosis in the contralateral ICA of 81% (SD, 11%).
In patients with bilaterally occluded ICAs, flow in the BA was increased 2.5-fold compared with control subjects, while flow in both MCAs was decreased. Only in the group of patients with bilateral occlusions was total blood flow significantly decreased, at 55% of total flow in control subjects.
No significant differences in flow through the ECAs were found between patients and control subjects or between the ipsilateral and contralateral sides of patients.
Fig 2⇓ shows the flow through the contralateral ICA, the BA, and total cerebropetal flow in relation to the degree of stenosis of the contralateral ICA. Flow through the contralateral ICA and BA was not correlated to the degree of stenosis in the contralateral ICA. In patients with an open contralateral ICA, a critical degree of stenosis that would negatively affect the flow in the contralateral ICA was not found.
The differences in blood flow between the three groups of patients are presented in the Table⇑ by horizontal brackets. Fig 3⇓ shows the variations in flow through the contralateral ICA and BA in each group of patients and in control subjects. No significant differences in flow through any artery were found between patients with a 0% to 69% contralateral stenosis and patients with a 70% to 99% contralateral stenosis. Patients with a bilateral occlusion of the ICA had a significant increase in flow through the BA compared with patients with a 0% to 69% contralateral ICA stenosis. Compared with patients with a 70% to 99% contralateral ICA stenosis, this difference was not significant. The total cerebropetal flow and flow in the contralateral MCA was decreased in patients with bilateral occlusions compared with patients with a 0% to 69% contralateral stenosis and compared with patients with a 70% to 99% contralateral stenosis.
Collateral flow through the anterior communicating artery could be assessed in all but two patients by means of the MR directional flow investigation. No reversed flow in A1 segments was found in any of the control subjects. However, reversed flow in the ipsilateral A1 segment was found in 14 of the 27 patients with a 0% to 69% contralateral stenosis and in 9 of the 17 patients with a 70% to 99% contralateral stenosis. In the contralateral ICA and in both the ipsilateral and contralateral MCAs, flow was found to be similar in patients with and without reversed flow in the ipsilateral A1 segment.
Infarcts in the ipsilateral hemisphere were found in 20 of the 27 patients with a 0% to 69% contralateral stenosis, in 15 of the 17 patients with a 70% to 99% contralateral stenosis, and in 9 of the 10 patients with bilateral ICA occlusions. No differences in any flow were found between patients with and without visible cerebral infarcts in the ipsilateral hemisphere on MRI.
The most important finding in this study is that in patients with symptomatic unilateral ICA occlusion, flow in the contralateral ICA and in the BA was increased, but flow in both MCAs was decreased. A severe (70% to 99%) stenosis of the contralateral ICA did not decrease flow through this artery. When both ICAs were occluded, flow through the BA was particularly increased, but total cerebropetal blood flow and flow through both MCAs were decreased.
The contralateral ICA appears to be the main collateral artery. As long as this artery was not occluded, total cerebropetal blood flow did not significantly decrease in comparison with normal subjects. Apparently, flow compensation by the contralateral ICA and BA together is sufficient to keep total cerebropetal blood flow within normal limits, even when the contralateral ICA has a severe (70% to 99%) stenosis. These findings are supported by a transcranial Doppler sonography study by Müller and Schimrigk,8 in which the presence of an ICA stenosis contralateral to an ICA occlusion did not affect cerebral vasoreactivity, indicating that cerebral blood flow was not affected by the contralateral ICA stenosis.
There are several possibilities to explain the decrease in flow in both MCAs. First, flow in the ipsilateral MCA depends on collateral flow through the anterior and posterior communicating arteries. These pathways are longer and depend on smaller vessels than normal pathways. Second, if blood from the BA or the ECAs is redistributed via leptomeningeal vessels, this will bypass the normal pathways through the circle of Willis and the MCAs. Third, reduced total cerebropetal flow, as in patients with bilateral ICA occlusions, will cause a reduction in flow distally from the circle of Willis, including the MCAs. We do not have an explanation for the decrease in flow through the contralateral MCA in patients with an open contralateral ICA. This is probably not caused by a contralateral to ipsilateral steal, since flow through the contralateral MCA did not significantly differ between patients with and patients without retrograde flow in the A1 segment of the ipsilateral ACA. However, it still is possible that the contralateral ICA also supplies the ACA on the ipsilateral side through the anterior communicating artery, thereby lowering the flow in the contralateral MCA.
Under normal physiological circumstances, flow in the MCA is directly related to flow in the ICA, without contribution of the posterior circulation. In this study, the anterior circulation in control subjects accounted for 80% of the total flow, which is consistent with other MRA studies.12 13 15 In patients with unilateral ICA occlusions (with or without contralateral ICA stenoses), the anterior circulation provided only 60% of cerebropetal flow, whereas in patients with a bilateral occlusion of the ICA, the BA is the main supplying artery. In these patients, blood flowing through the BA will be redistributed via the posterior communicating arteries and via leptomeningeal vessels to compensate for reduced flow in the anterior circulation. However, compensatory flow through the posterior communicating arteries probably is limited in some patients. In this study, flow directions in the posterior communicating arteries were not investigated since the MRA directional flow assessments, which were different from the method used by Schomer et al,19 did not provide reliable results in these small arteries.
Our measurements of total cerebropetal flow might be incorrect since we did not include flow through the ophthalmic arteries.5 20 However, if retrograde ophthalmic flow is high, a significant increase in ECA flow is expected, which was not found in any patient group.
It is possible that the degree of stenosis in the contralateral ICA is increased in some patients during the interval between iaDSA and MRA. However, it is likely that this does not influence the flow characteristics in the group of patients with 0% to 69% contralateral ICA stenosis and the group with 70% to 99% contralateral ICA stenosis, since flow values are virtually the same in both groups.
MR spectroscopy studies have shown that the cerebral metabolism in patients with severe carotid disease is affected in noninfarcted areas, probably because of impaired cerebral blood flow.12 21 22 Cerebral blood flow studies using inhaled 133Xe, positron emission tomography, single-photon emission computed tomography, and transcranial Doppler sonography have demonstrated that cerebral blood flow is reduced, oxygen extraction fraction is increased, and vasomotor reactivity is impaired in patients with a symptomatic severe stenosis or occlusion of the ICA.5 6 7 8 20 23 24 25 26 27 An exhausted vascular reserve capacity has been shown to be a risk factor for stroke.24 26 When these results are combined, it appears that patients with low cerebropetal flow, like the patients with bilateral ICA occlusions in this study, are particularly vulnerable to cerebral ischemia, since autoregulation has probably reached its limits.
In conclusion, the results of this study show that in patients with carotid occlusions, it is most important to maintain an open contralateral ICA, since total cerebropetal blood flow does not decrease unless both ICAs are occluded. Clinical follow-up will be necessary to investigate whether low cerebropetal flow can be identified as a risk factor for stroke in these patients.
Selected Abbreviations and Acronyms
|ACA||=||anterior cerebral artery|
|CCA||=||common carotid artery|
|ECA||=||external carotid artery|
|iaDSA||=||intra-arterial digital subtraction angiography|
|ICA||=||internal carotid artery|
|MCA||=||middle cerebral artery|
|MRA||=||magnetic resonance angiography|
This study was supported by the Netherlands Heart Foundation (grants D94-012 and 94.085). We thank Professor B.C. Eikelboom from the Department of Vascular Surgery for enlisting additional patients for our study.
- Received February 27, 1997.
- Revision received April 16, 1997.
- Accepted May 2, 1997.
- Copyright © 1997 by American Heart Association
Kleiser B, Widder B. Course of carotid artery occlusions with impaired cerebrovascular reactivity. Stroke. 1992;23:171-174.
Müller M, Schimrigk K. Vasomotor reactivity and pattern of collateral blood flow in severe occlusive carotid artery disease. Stroke. 1996;27:296-299.
Vanninen RL, Koivisto K, Tulla H, Manninen H, Partanen K. Hemodynamic effects of carotid endarterectomy by magnetic resonance flow quantification. Stroke. 1995;26:84-89.
Van der Grond J, Eikelboom BC, Mali WTPM. Flow-related anaerobic metabolic changes in patients with severe stenosis of the internal carotid artery. Stroke. 1996;27:2026-2032.
Levine RL, Turski PA, Holmes KA, Grist TM. Comparison of magnetic resonance volume flow rates, angiography, and carotid Dopplers: preliminary results. Stroke. 1994;25:413-417.
Enzmann DR, Ross MR, Marks MP, Pelc NJ. Blood flow in major cerebral arteries measured by phase-contrast cine MR. AJNR Am J Neuroradiol. 1994;15:123-129.
Powers WJ, Press GA, Grubb RL Jr. The effect of hemodynamically significant carotid artery disease on the hemodynamic status of the cerebral circulation. Ann Intern Med. 1987;106:27-35.
Van der Grond J, Ramos LMP, Eikelboom BC, Mali WTPM. Cerebral metabolic differences between the severe and critical hypoperfused brain. Neurology. 1996;47:399-404.
Van der Grond J, Balm R, Kappelle LJ, Eikelboom BC, Mali WTPM. Cerebral metabolism of patients with stenosis or occlusion of the internal carotid artery: a 1H-MR spectroscopic imaging study. Stroke. 1995;25:822-828.
Kazumata K, Tanaka N, Ishikawa T, Kuroda S, Houkin K, Mitsumori K. Dissociation of vasoreactivity to acetazolamide and hypercapnia: comparative study in patients with chronic occlusive major cerebral artery disease. Stroke. 1996;27:2052-2058.
Yamauchi H, Fukuyama H, Nagahama Y, Nabatame H, Nakamura K, Yamamoto YL, Yonekura Y, Konishi J, Kimura J. Evidence for misery perfusion and risk for recurrent stroke in major cerebral arterial occlusive diseases from PET. J Neurol Neurosurg Psychiatry. 1996;61:18-25.