High Signal Intensity on T2-Weighted Magnetic Resonance Imaging and Cerebral Hemodynamic Reserve in Carotid Occlusive Disease
Background and Purpose The importance of MR imaging in carotid artery disease is unclear. We evaluated the sensitivity and specificity of the high signal intensity changes on MR images for diagnosis of hemodynamically compromised unilateral internal carotid artery disease.
Methods We evaluated the association of high signal intensities on T2-weighted MR images with changes in cerebral perfusion reserve measured using 99mTc-hexamethylpropyleneamine oxime single-photon emission CT and acetazolamide in 23 patients.
Results Eleven patients had a type I response (normal flow and normal perfusion reserve), 8 patients had a type II response (normal flow and decreased perfusion reserve), and 4 patients had a type III response (decreased flow and decreased perfusion reserve). High signal intensities in the centrum semiovale (11/12) and/or posterior periventricular white matter (6/12) were frequently seen in the hemodynamically compromised groups. Extensive high signal intensities were associated with severely impaired cerebral circulation. MR imaging had high sensitivity (0.92) and specificity (1.0) in predicting hemodynamically compromised patients when we used the presence of T2 high intensity in the centrum semiovale as a criterion.
Conclusions The centrum semiovale T2 hyperintensities lateralized to the side of carotid occlusion are specific and sensitive for the presence and severity of hemodynamic compromise from carotid occlusive disease.
In carotid occlusive disease, the most common cause of cerebral ischemia is the embolic event from severe carotid atherosclerosis.1 Other mechanisms include a hemodynamic stroke secondary to critical reduction of cerebral perfusion pressure and/or an inadequate intracranial collateral blood supply associated with ICAO.2 CO2- or ACZ-induced stimulation of the cerebral vasomotors provides an approach to examine the reduction in perfusion reserve due to extracranial cerebrovascular occlusive disease.3 4 Use of resting and post-ACZ 99mTc-labeled HMPAO SPECT allows an objective assessment of regional vasoreactivity to ACZ.5 6
Compared with the functional brain imaging approaches such as SPECT, MR imaging provides higher signal-to-noise ratios and is a suitable method for detection of structural alterations in the brain. However, the clinical relevance of MR images for diagnosis of hemodynamically compromised ICAO has not been fully appreciated.7 We compared the MR findings with the 99mTc-HMPAO SPECT findings with ACZ challenge in 23 patients with unilateral ICAO to address whether T2 high intensity on MR imaging is a sensitive and specific indicator of the hemodynamic status of the cerebral circulation.
Subjects and Methods
Between April 1992 and May 1996, MR imaging and 99mTc-HMPAO SPECT were done on 23 patients (11 men and 12 women aged 45 to 75 years [mean, 58.2 years]) with carotid occlusive disease (14 transient ischemic attacks and 9 minor strokes) who were selected according to the following criteria: hypodense area <1.5 cm diameter on a CT scan of the brain; and bilateral, biplane, common carotid arteriography (<70 years, n=20) or MR angiography (≥70 years, n=3) showing unilateral ICAO and no abnormality in the MCA territories. Eleven age-matched healthy control subjects underwent ACZ-SPECT imaging to establish the range of normal values for right-to-left asymmetry in 99mTc-HMPAO CBF. All subjects gave informed consent before the study.
MR images were obtained using a 1.5-T superconducting unit (Magnetom H15, Siemens). Multiple spin-echo sequences were done with a repetition time of 4500 milliseconds and an echo time of 90 milliseconds to produce T2-weighted images. Transaxial images of 8-mm-thick sections of the brain with a 2-mm gap were obtained. Two-dimensional Fourier transformation of images and a 256×256 data-acquisition matrix were used.
After MRI study, 296 MBq of 99mTc-HMPAO was injected intravenously, and 5 minutes later, 1 g ACZ was administered. After that, the baseline SPECT data were obtained. An additional dose of 592 MBq of 99mTc-HMPAO was injected immediately after stopping the baseline data acquisition. Five minutes later, the second SPECT data acquisition was started. After reconstruction,8 the baseline tomographic images were subtracted from the images in the second step to obtain ACZ images.
The MR images were individually visually interpreted by two observers blinded to the SPECT results but with clinical information of the occluded side. Focal high intensities on MR scans were qualitatively defined in the transaxial images for 10 brain regions from four tomographic slices (Fig 1⇓). The defined brain areas were delineated from the supraventricular (CS), high ventricular (CR and anterior and posterior white matters at the level of the body of the lateral ventricle and the CR), midventricular (anterior and posterior white matters at the level of the body of the lateral ventricle and the genu of corpus callosum), and low ventricular slices (putamen, thalamus, and anterior and posterior white matters at the level of the basal ganglia).9 They include internal watershed areas, basal ganglia, subcortical white matters, and periventricular areas.10 11 For each ventricular level, each white matter was assigned to anterior and posterior regions adjoining the frontal horn or the occipital horn of the lateral ventricle, respectively. For confidence of signal intensity lateralization (occluded versus contralateral hemisphere), measures of the distinct zone of high attenuation (namely, size [diameter], appearance [focal, multiple or confluent], and extent of visualization) were analyzed. Asymmetry was considered to be present if all of the above measures were larger in the occluded side than in the contralateral side.12 After each reader completed his interpretation, the two readers compared their results and arrived at a consensus reading for each patient's individual 10-brain region. The first agreement reached on the diagnostic interpretation on MR images was in 41 of 43 regions for regions with asymmetry, 157 of 187 for regions without asymmetry, and 198 of 230 for all regions (κ=0.64).
For semiquantitative measurements, a pair of circular ROIs (1 cm in diameter) was symmetrically drawn over the middle CS, and the difference of T2 relaxation time13 within the ROI was calculated as the AI14 with the following equation: AI=(T2 in Occluded Side−T2 in Contralateral Side)/Mean T2 of Both Sides. For SPECT evaluations, pairs of ROIs in the cortical territory of the MCA were defined in each hemisphere from four consecutive slices, and a hemispheric count density of the 99mTc was calculated as the average of the four slices weighted by ROI size. The hemispheric asymmetry of CBF was calculated in the same manner as the MR study.
The statistical significance of the relationship between the three SPECT categories of cerebral hemodynamic status and the frequency of side-to-side asymmetry in T2 high intensity were determined by χ2 analysis of a 3×10 contingency table. Subgroup analysis of these relationships was done with 2×2 contingency tables with two-tail probabilities from Fisher's exact test.
Based on AI values outside the mean±2 SD determined from the 11 healthy control subjects, we defined three types of SPECT responses to an ACZ challenge in the patients (Fig 2⇓): (1) normal perfusion before (%AI=0.8±1.0) and after (%AI=0.2±1.6) ACZ (11 of 23 [48%]); (2) normal perfusion before ACZ (%AI=−0.8±0.6) with hypoperfusion after ACZ (%AI=−12.5±3.3) (8 of 23 [35%]); and (3) hypoperfusion before ACZ (%AI=−4.8±1.4) and additional hypoperfusion after ACZ (%AI=−13.5±3.4) (4 of 23 [17%]).
There was a significant relationship among the three SPECT categories of ACZ responses and the presence of regional asymmetry of T2 high intensity in the occluded hemisphere (χ2=29.8, P=.04, df=18) (Fig 3⇓). Type II (Fig 4⇓) patients showed a higher frequency of T2 asymmetries in the CS (7/8 versus 0/11) and LP (4/8 versus 1/11) compared with type I patients; in type III patients (Fig 5⇓), the frequency of asymmetries was significantly higher than that in the type I in the CS (4/4 versus 0/11), LP (3/4 versus 1/11), CR (4/4 versus 2/11), and MA (2/4 versus 0/11). The criterion of presence of T2 asymmetry in the CS in the occluded hemisphere had a sensitivity of 0.92 (11/12) and specificity of 1.0 (11/11) for identifying patients with hemodynamically compromised ICAO (types II and III).
Post-ACZ CBF asymmetry and T2 asymmetry in the CS correlated excellently. There was also a significant inverse correlation between baseline CBF asymmetry and T2 asymmetry in the CS (Fig 6⇓).
We found a close association between the types of responses to an ACZ challenge and the distribution pattern of MR T2 high intensity in the hemisphere ipsilateral to the ICAO. There were significant differences in the frequencies of T2 high intensity in the CS and the LP between ICAO patients with normal baseline CBF and impaired cerebral perfusion reserve (type II) and ICAO patients without cerebral perfusion impairment (type I). As cerebral perfusion pressure falls further, the compensatory mechanisms of the brain will be exhausted and resting CBF will begin to fall (type III). T2 high intensities in type III extended over areas (ie, not only in the CS and LP but also in the CR and the MA) larger than those in type II.
In a postmortem study of the border zone from atherosclerotic carotid disease, Romanul and Abramowicz15 observed old infarcts in anterior border zones and recent larger infarcts within the territory of the major cerebral vessels. The most likely possibility is that lowered perfusion reserve initially produces cerebral ischemia in the most distal branches from ICAO such as the CS or the border-zone region between the MCA and PCA (LP), where perfusion pressure is lowest,16 and that with disease progression and decline in both resting perfusion and perfusion reserve, blood flow becomes insufficient more proximally, producing larger lesions.
Weiller et al5 reported that the infarcts were located in the most distal parts of territories of the lateral lenticulostriate group and of anterior choroidal artery in patients with hemodynamically compromised ICAO or MCA occlusion. Nakano et al17 reported that MCA occlusive disease frequently produces CR and/or striatocapsular infarct but was never associated with CS lesions, which usually were associated with ICAO. The question that now arises is whether a common mechanism can explain the MR high intensities in the CS and the CR in patients with ICAO. Supraventricular white matter of the CS corresponds to the border zone between the superficial territories of the anterior cerebral artery and MCA, and periventricular white matter of the CR corresponds to the border zone between the deep and superficial territories of the MCA.18 In patients with poor capacity for vasodilatation such as type III, CBF would be more sensitive to decreases in cerebral perfusion pressure. Critically reduced perfusion pressure and/or an insufficient collateral supply in the lenticulostriate arteries are presumably implicated in the processes of the striking increase of the high intensities in the CR in these patients. In individual cases of ICAO, the specific action of the carotid artery occlusion that causes small vessel alterations in the MCA territory is still unknown.
We did not find differences in T2 asymmetry in the Pu, Th, and the white matters of HA, HP, MP, and LA among the three hemodynamic types. The high intensities detected in these regions may be due to pathological processes that are independent of hemodynamic compromise. Cerebral arteries are subject to luminal narrowing from arteriosclerosis induced by age, hypertension, and other degenerative vascular conditions.10 11 It is important to determine whether patients with nonhemodynamic ICAO show a characteristic pattern on MR images.
In conclusion, our results show that the location and the degree of the high intensities in T2-weighted MR images accurately reflect the hemodynamic status of the cerebral circulation in patients with unilateral ICAO. We believe that the presence of high intensity in the CS on the occluded hemisphere provides the best criterion for identifying a hemodynamically compromised subgroup of individuals who may have transient ischemic attack or minor stroke due to a marginal basal level of perfusion that drops to a level inadequate to maintain function.
Selected Abbreviations and Acronyms
|CBF||=||cerebral blood flow|
|HA||=||high ventricular anterior white matter|
|HP||=||high ventricular posterior white matter|
|ICAO||=||internal carotid artery occlusion|
|LA||=||low ventricular anterior white matter|
|LP||=||low ventricular posterior white matter|
|MA||=||midventricular anterior white matter|
|MCA||=||middle cerebral artery|
|MP||=||midventricular posterior white matter|
|PCA||=||posterior cerebral artery|
|ROI||=||region of interest|
|SPECT||=||single-photon emission CT|
This work was supported by a grant for medical research from the Smoking Research Foundation in Japan. We wish to thank M. Harada for his invaluable support and technical help.
Reprint requests to Yoshinari Isaka, MD, Department of Nuclear Medicine, Osaka National Hospital, Hoenzaka 2-1-14, Chuo-ku, 540, Osaka, Japan. E-mail firstname.lastname@example.org.
- Received July 30, 1996.
- Revision received October 29, 1996.
- Accepted October 29, 1996.
- Copyright © 1997 by American Heart Association
Ringelstein EB, Zeumer H, Angelou D. The pathogenesis of strokes from internal carotid artery occlusion: diagnostic and therapeutic implications. Stroke.. 1983;14:867-875.
Powers WJ, Press GA, Grubb RL Jr, Gado M, Raichle ME. The effect of hemodynamically significant carotid artery disease on the hemodynamic status of the cerebral circulation. Ann Intern Med.. 1987;106:27-35.
Ringelstein EB, Sievers C, Ecker S, Schneider PA, Otis SM. Noninvasive assessment of CO2-induced cerebral vasomotor response in normal individuals and patients with internal carotid artery occlusion. Stroke.. 1988;19:963-969.
Weiller C, Ringelstein EB, Reiche W, Buell U. Clinical and hemodynamic aspects of low-flow infarcts. Stroke.. 1991;22:1117-1123.
Knop J, Thie A, Fuchs C, Siepmann G, Zeumer H. 99mTc-HMPAO SPECT with acetazolamide challenge to detect hemodynamic compromise in occlusive cerebrovascular disease. Stroke.. 1992;23:1733-1742.
Mills CM, De Groot J, Posin JP. Magnetic Resonance Imaging: Atlas of the Head, Neck, and Spine. Philadelphia, Pa: Lea & Febiger Inc; 1988.
Bladin CF, Chambers BR. Clinical features, pathogenesis, and computed tomographic characteristics of internal watershed infarction. Stroke.. 1993;24:1925-1932.
Moody DM, Bell MA, Challa VR. Features of the cerebral vascular pattern that predict vulnerability to perfusion or oxygenation deficiency: an anatomic study. Am J Neuroradiol.. 1990;11:431-439.
Mosmans PCM, Veering MM, Jonkman EJ. ISI values and interhemispheric differences in patients with ischemic cerebrovascular disease: correlation with clinical and angiographic findings. Stroke.. 1986;17:58-64.
Romanul FCA, Abramowicz A. Changes in brain and pial vessels in anterior borderzones: a study of 13 cases. Arch Neurol.. 1964;11:40-65.
Yamauchi H, Fukuyama H, Kimura J, Konishi J, Kameyama M. Hemodynamics in internal carotid artery occlusion examined by positron emission tomography. Stroke.. 1990;21:1400-1406.
Nakano S, Yokogami K, Ohta H, Goya T, Wakisaka S. CT-defined large subcortical infarcts: correlation of location with site of cerebrovascular occlusive disease. AJNR Am J Neuroradiol.. 1995;16:1581-1585.
Bogousslavsky J, Regli F. Centrum ovale infarcts: subcortical infarction in the superficial territory of the middle cerebral artery. Neurology.. 1992;42:1992-1998.