Search

BACKGROUND AND PURPOSE Increased cerebral blood flow velocity of basilar artery (BA) is common but often neglected. By using digital subtraction angiography (DSA) to detect intracranial large artery stenosis, we performed transcranial Doppler (TCD) to evaluate cerebral hemodynamic changes of stroke patients. METHODS Acute stroke patients with TCD-detected raised peak systolic velocity (PSV) in BA (≥100 cm/second) were recruited. RESULTS Among 91 recruited patients, BA stenosis (>50%) was confirmed by using DSA in 29 patients (32%), among which 20 patients (70.0%) had coexistent internal carotid artery (ICA) and vertebral artery (VA) stenoses. Among patients without BA stenosis (n = 62, 68%), severe stenosis or occlusion of ICA was evaluated in 22 patients (22/62, 35.4%) and severe stenosis or occlusion of VA was detected in 22 patients (22/62, 35.4%). Among the values of PSV (120, 140, 160, and 180 cm/second) and the stenotic-to-prestenotic ratio (SPR) (1.5, 2.0, 2.5, and 3.0), 160 cm/second was found to show the highest predictive value (area under the receiver operator characteristic curve: .75, 95% CI: .65-.86), with a sensitivity of .70 and a specificity of .81. CONCLUSIONS Apart from an intrinsic BA stenosis, high-grade steno-occlusion of ICA or VA may contribute a lot to induce an elevated flow velocity in BA. A higher value of cutoff point may increase the accuracy of diagnosing BA stenosis.

BACKGROUND AND PURPOSE Microemboli signal (MES) detected by transcranial Doppler (TCD) may represent ongoing embolic phenomenon and is a predictor of recurrent stroke or transient ischemic attack. We sought to study the frequency of MES in stroke patients with large artery occlusive diseases treated with low molecular weight heparin (LMWH) or aspirin. METHODS Patients participating in the Fraxiparine in Ischemic Stroke (FISS)-tris study were recruited. MES detection was performed from middle cerebral artery on the 1st, 3rd, and 7th days after randomization. The correlation between the presence of MES and the treatment was determined by the χ2 test. RESULTS Among 47 patients, 26 were randomized to LMWH and 21 to aspirin. On day 1, MES were detected in 10 patients (7/26 on LMWH, 3/21 on aspirin; P= .475). On day 3, 12 patients had MES (7/25 on LMWH, 5/20 on aspirin; P= .821). On day 7, 11 patients had MES (6/25 on LMWH, 5/20 on aspirin; P= .938). The median of the number of MES on days 1, 3, and 7 was 4 (range 1-10), 5 (range 1-42), and 3 (range 1-33) for the LMWH group and 1 (range 1-15), 4 (range 1-10), and 2 (range 1-4) for the aspirin group. CONCLUSIONS There were no significant differences in the frequency of MES between patients with large artery occlusive disease treated with LMWH and aspirin.

BACKGROUND AND PURPOSE Current criteria for middle cerebral artery (MCA) stenosis by transcranial Doppler were mainly based on the flow velocity acceleration, ignoring other useful parameters. We tested the accuracy of new criteria using magnetic resonance angiography (MRA) as the confirmative method. METHODS A compositive score for each MCA was calculated according to following parameters: Velocity Scale (score 0-6 for peak systolic velocity

Background and Purpose. Small subcortical infarcts (SSI, maximum lesion diameter ≤2.0 cm) are usually considered as infarcts caused by small vessel disease. However, SSI can also be associated with large artery occlusive disease such as middle cerebral artery (MCA) stenosis. We performed a prospective study to investigate the relationship between MCA stenosis and SSI distribution and further to investigate the mechanism of SSI caused by MCA stenosis. Methods. Magnetic resonance angiography (MRA) and diffusion-weighed MRI (DWI) of consecutive acute ischemic stroke patients with recent SSI were studied. The distribution of acute infarcts on DWI was categorized as cortical infarct (CI), border zone infarct (BI), or perforating artery infarct (PAI). Results. Totally, 93 cases were recruited, among which 12 had single SSI with MCA stenosis (group 1) and 26 patients had multiple SSI with MCA stenosis (group 2), while 55 patients without MCA stenosis had single SSI (group 3). For patients with single SSI and MCA stenosis, 6 had BI and 6 had PAI; for patients with multiple SSI and MCA stenosis, 25 had BI, 4 had PAI and 9 had CI (compared with group1: P= .001); for patients with single SSI but without MCA stenosis, 20 had BI and 35 had PAI (compared with group1: P= .58). Conclusion. Multiple acute infarcts along the border zone are the commonest pattern in small infarcts with MCA stenosis, especially among those with multiple acute infarcts. Our data suggest that hemodynamic compromise and artery-to-artery embolism may be both important factors for infarcts in patients with MCA stenosis.

Background and Purpose. The relationships between cerebral vasomotor reactivity (CVR) and white matter lesions (WMLs) were investigated mainly in patients with carotid stenosis. We aimed to study the relationship in asymptomatic elderly subjects without large artery occlusive disease. Methods. A total of 33 elderly individuals (mean age was 76.2 years) who were free from neurological deficit or cognitive impairment were studied. Bilateral mean blood flow velocity was measured in the middle cerebral artery using a 2-MHz pulsed transcranial Doppler (TCD) system together with intravenous administration of acetazolamide as vasodilatory stimuli. WMLs on a fluid-attenuated inversion recovery (FLAIR) sequence of MRI were classified into two categories: subcortical deep white matter hyperintensity (SDWMH) and periventricular hyperintensity (PVH). The lesions in each category were then divided into three grades (grade 0-I, grade II, grade III) according to the Fazekas scale. Results. CVR was inversely associated with the extent of SDWMH and PVH. The differences in CVR were statistically significant among different severity of WMLs: for SDWMH (70%± 10% in grade 0-I, 60%± 10% in grade II, and 40%± 10% in grade III, P < .001); for PVH (80%± 10% in grade 0-I, 60%± 10% in grade II, and 40%± 10% in grade III, P < .001). Conclusions. Impaired CVR is related to the extent of WMLs in asymptomatic elderly individuals without large artery stenosis. The findings in our study suggest that dysfunction of cerebral vascular autoregulation might be an important factor in the development of WMLs in the asymptomatic elderly without large artery occlusive disease.

Background and Purpose. Controversies still exist as to the neuroimaging determinants of cognitive impairment in cerebral small vessel disease (SVD). The authors studied the neuroimaging correlates of cognitive performances among patients with stroke associated with SVD. Methods. The authors per formed cerebral computed tomography, magnetic resonance imaging, and diffusion-weighted imaging among 74 consecu tive patients admitted to the acute stroke unit because of stroke associated with SVD. They examined the association between cognitive performances and the following neuroimaging features: volume of white matter changes (WMC), multiplicity of lacunae, location of lacunae, total cerebral atrophy, and frontal and medial temporal lobe atrophy. Results. Apart from age and education, univariate linear regression analyses revealed that WMC volume, presence of thalamic lacunae, cerebral atrophy, and left frontal lobe atrophy predicted performance on the Mini-Mental State Examination while WMC volume, presence of thalamic infarcts, cerebral atrophy, and frontal lobe atrophy of both sides predicted performance on the Mattis Dementia Rat ing Scale-Initiation/Preservation subscale. In the multivariate analyses, education (R2= 0.22, P < .001), left frontal lobe atrophy (R2= 0.10, P= .004), and presence of thalamic lacunae (R2= 0.04, P= .049) were found to predict performance on the Mini-Mental State Examination while age (R2= 0.23, P < .001) and presence of thalamic lacunae (R2= 0.08, P= .011) were found to predict performance on the Mattis Dementia Rating Scale-Initiation/Preservation. Conclusions. Among patients with stroke associated with SVD, thalamic lacunae and frontal lobe atrophy are key determinants of cognitive performances.

Background and Purpose. Usually, microembolic signal (MES) monitoring is performed for MESs arising from downstream sources. The aim of this study is to describe the special characteristics of MESs originating from nearby sources in the middle cerebral artery (MCA). Methods. We analyzed 265 MESs detected in 5 patients (the study group) who had acute ischemic strokes within the territory of MCA stenoses and 48 MESs detected in 7 patients (the control group) from proximal sources: severe internal carotid artery stenoses and prosthetic heart valves. MESs were recorded with a 2-MHz, bigate transducer. We performed offline analyses of the recorded MESs using both post–fast Fourier transform (FFT) spectra and pre-FFT time domain signals. MESs were divided into 3 types–focused frequency signals (FFSs), bottom-frequency signals (BFSs), and multifrequency signals (MFSs)–according to the frequencies of the signals displayed on the post-FFT spectra. Results. In the study group, among 265 MESs, 245 (92.5%) were MFSs, 11 (4.2%) were FFSs, and 9 (3.4%) were BFSs. In the control group, among 48 MESs, 45 (93.5%) were FFSs, 2 (4.2%) were MFSs, and 1 (2.%) was a BFS. There was significant difference in MES type between MCA stenoses and other embolic sources (P < .05). On the post-FFT spectra, MFSs recorded from MCA stenoses occupied multiple frequencies along the vertical axis, and the low-frequency parts were usually bidirectional. On the pre-FFT time domain signals, the highest frequency parts presented distorted, amplitude-modulated sine waves, and the low-frequency parts presented totally irregular waves. The amplitudes of the low-frequency parts diminished with time in the proximal channel but enlarged with time in the distal channel over a short period. Conclusion. MESs recorded from MCA stenoses may have special characteristics of multiple frequencies on both post-FFT spectra and pre-FFT time domain signals. Our findings may represent rotating or vibrating emboli as they are just dislodged from the thrombus and are moving from the vessel wall to the center. Further clinical and laboratory studies are needed to confirm this hypothesis.

Background and Purpose. To evaluate the accuracy of B0 echo planar imaging (EPI) sequences for the detection of intracerebral bleeds. Methods. One hundred patients with acute strokes had magnetic resonance imaging and computed tomography (CT) examinations performed within 48 hours after the onset of symptoms. The detectability of intracerebral bleeds by the B0 EPI sequences was assessed. The results were compared to the gradient echo (GRE) sequence and CT brain examinations. The results of the GRE sequences were used as the gold standard. Results. The B0 EPI sequences detected 11 out of 11 acute, intracerebral hematomas; 6 out of 8 acute hemorrhagic strokes; 2 out of 2 acute, intraventricular hemorrhages; 8 out of 8 old hemorrhagic infarcts; 1 out of 1 subarachnoid hemorrhages; and 11 out of 22 patients with microbleeds. For the detection of acute, intracerebral hematomas and acute, hemorrhagic infarcts, B0 EPI sequences had a sensitivity of 89.5%, a specificity of 100%, and an accuracy of 98%. CT had a sensitivity of 57.9%, a specificity of 100%, and an accuracy of 92%. B0 EPI sequences did not miss any acute or chronic hemorrhages detected by CT examinations. Conclusions. B0 EPI sequences could not replace GRE images for the detection of both acute and chronic hemorrhages. Their sensitivity for the detection of acute and chronic blood products, however, was comparable, if not superior, to that of CT examinations.

To investigate the optimal values of flow velocity on transcranial Doppler (TCD) in grading the severity of middle cerebral artery (MCA) stenosis in comparison with magnetic resonance angiography (MRA).Both TCD and MRA examinations were performed on 148 asymptomatic patients. The peak flow velocities of each MCA were recorded. Severity of MCA stenosis on MRA was classified as normal-mild (50% lumen diameter reduction), moderate (50%-75%), and severe-void (75% and void of flow signal).Among 296 MCAs evaluated, normal-mild stenosis was found in 75 (25%), moderate stenosis in 112 (38%), and severe stenosis in 109 (37%). The mean of systolic velocity (Vs) of MCA differed significantly among these three groups: mean Vs = 121.83 +/- 22.52 cm/s in the normal-mild group; 155.96 +/- 21.62 cm/s for the moderate group; and 199.39 +/- 43.86 cm/s for the severe group (P.001). The optimal cutoff velocity for detection of MCA (50%) stenosis was found at Vs140 cm/s on TCD (area under the ROC curve is 0.87, P0.001). The best cutoff points for grading severity of on TCD were 140 cm/s and 180 cm/s.TCD enables grading of the severity of MCA stenosis according to the flow velocity. This method provides a noninvasive and reliable method for grading MCA stenosis and allows longitudinal monitoring of the relationship between clinical outcome and hemodynamic change.

BACKGROUND Age-related white matter changes (WMC) increase risk of cognitive and functional decline. They are considered as target for preventive trials. However, using magnetic resonance imaging (MRI) to screen for subclinical WMC is not cost-effective. We explored the clinical utility of transcranial Doppler ultrasound (TCD) in the evaluation of WMC. We hypothesized that the pulsatility index (PI) of TCD correlates with severity of WMC. METHODS Among 100 stroke patients with WMC, we obtained the mean PI of middle cerebral artery (MCA) and vertebral basilar artery (VB). We investigated the association between PI and WMC volume. We also obtained PI of 50 subjects who had no WMC on MRI and explored the ability of PI values in differentiating subjects with and without WMC. RESULTS Multivariate linear regressions found that mean MCA PI and mean VB PI were independent predictors for total WMC volume (beta = .43, P < .001 and beta = .37, P < .001, respectively). Mean MCA PI significantly differentiated patients with WMC from those without WMC (area under curve [AUC] .85 [95% confidence interval .78-.91]) with an optimal cut-off value at 1.15 that yielded sensitivity and specificity of 73.7% and 82%, respectively. AUC for mean VB PI was .81 (95% confidence interval .74-.88) with cut-off value at 1.06 that yielded sensitivity and specificity of 70% and 75%, respectively. CONCLUSIONS PI correlates with volume of WMC. It may help to differentiate those with and without WMC in stroke patients. Further study evaluating the clinical utility of TCD in screening for subclinical WMC among community elderly is warranted.