(Stroke. 2005;36:32.)
© 2005 American Heart Association, Inc.
Original Contributions |
From the Stroke Treatment Team (OYC, ZG, JS, MSC, EAN, HS, C-IC, YI, JCG, AVA), the University of Texas-Houston, Texas; and the Department of Neurology (S.C.), Hospital Central Universitario de Asturias, Oviedo, Spain.
Correspondence to Dr Andrei V. Alexandrov, University of Texas-Houston Department of Neurology, 6431 Fannin Street MSB 7044, Houston, Texas 77005. E-mail avalexandrov{at}att.net
| Abstract |
|---|
|
|
|---|
Methods NVUE was performed with portable carotid duplex and TCD using standardized fast-track (<15 minutes) insonation protocols. Digital subtraction angiography (DSA) was the gold standard for identifying LAIT. These lesions were defined as proximal intra- or extracranial occlusions, near-occlusions,
50% stenoses or thrombus in the symptomatic artery.
Results One hundred and fifty patients (70 women, mean age 66±15 years) underwent NVUE at median 128 minutes after symptom onset. Fifty-four patients (36%) received intravenous or intra-arterial thrombolysis (median National Institutes of Health Stroke Scale (NIHSS) score 14, range 4 to 29; 81% had NIHSS
10 points). NVUE demonstrated LAITs in 98% of patients eligible for thrombolysis, 76% of acute stroke patients ineligible for thrombolysis (n=63), and 42% in patients with transient ischemic attack (n=33), P<0.001. Urgent DSA was performed in 30 patients on average 230 minutes after NVUE. Compared with DSA, NVUE predicted LAIT presence with 100% sensitivity and 100% specificity, although individual accuracy parameters for TCD and carotid duplex specific to occlusion location ranged 75% to 96% because of the presence of tandem lesions and 10% rate of no temporal windows.
Conclusions Bedside neurovascular ultrasound examination, combining carotid/vertebral duplex with TCD yields a substantial proportion of LAITs in excellent agreement with urgent DSA.
Key Words: carotid stenosis embolism stroke, acute thrombolysis ultrasonography, Doppler
| Introduction |
|---|
|
|
|---|
Considering previous publications2,3,5,7,13 and our own experience, we developed a fast-track neurovascular ultrasound examination (NVUE) protocol for an experienced sonographer to urgently evaluate acute stroke patients at bedside. We also developed detailed diagnostic criteria for normal, stenosed, and acutely occluded intracranial and extracranial vessels.1317
We routinely perform urgent bedside NVUEs with both standard fast-track TCD and carotid/vertebral duplex in patients with acute cerebral ischemia. We aimed to determine the yield and accuracy of these combined modalities to identify LAITs.
| Subjects and Methods |
|---|
|
|
|---|
An experienced sonographer and stroke neurologist arrived concurrently. The former performed and the latter evaluated urgent bedside NVUE results. To avoid delays in routine patient evaluation and thrombolytic therapy, bedside ultrasound examination was carried out simultaneously with clinical assessment, blood draws, computed tomography (CT), and chest x-ray, etc. Sonographers used standardized fast-track (<15 minutes) insonation protocols (Table 1) to identify suspected arterial obstruction. The neurological evaluation guided the sequence of vessels targeted by NVUE. Ultrasound results were interpreted at bedside without knowledge of angiographic results using previously published diagnostic criteria1319 (Table 2). In patients with no temporal windows, we performed a non-contractenhanced TCD examination of the orbital and posterior circulation vessels and carotid/vertebral duplex.
|
|
Patients received recombinant tissue plasminogen activator intravenously within 3 hours in a dose of 0.9 mg/kg as the standard of care.20 Patients were also screened for eligibility for experimental intra-arterial rescue using institutional review board (IRB)approved protocols. At our center, both ultrasound studies are initiated emergently without specific written informed consent in order not to delay institution of appropriate therapies or management strategies or select patients for experimental treatment. IRB approval was obtained in all cases when patients participated in our experimental protocols. All IRB approvals of our hyperacute experimental treatment protocols have mention of ultrasound tests as part of initial evaluation for symptoms of stroke. Patients presenting with sustained neurological deficits outside protocol time windows from symptom onset or hypoattenuation on CT scan did not receive interventional treatment. Patients with resolved neurological deficits at the time of examination were considered having transient ischemic attacks (TIAs).
Stroke neurologists decided to perform digital subtraction angiography (DSA) if patients either met clinical CT criteria for an IRB-approved experimental intra-arterial rescue protocol or ultrasound showed persisting arterial occlusion or reocclusion after full-dose intravenous recombinant tissue plasminogen activator therapy. DSA was the gold standard for the diagnosis of LAIT. LAIT was defined as an occlusion or near-occlusion, or
50% stenoses or thrombi in an artery (arteries) supplying brain area(s) affected by ischemia. Hence, current definition of LAIT may include chronic lesions, and caution should be exercised if these lesions are directly responsible for current patient symptoms. Patients with TIA, lacunar syndromes, and normal ultrasound findings were least likely to undergo urgent DSA.
After neuroradiologists dictated DSA reports, we compared NVUE results to DSA to determine true-positive, true-negative, false-positive, and false-negative ultrasound results. We calculated accuracy parameters, including sensitivity, specificity, and positive and negative predictive values. We used
2 analysis for dichotomous variables, Student t test for normally distributed continuous variables, and Wilcoxon 2-sample test for variables with nonparametric distribution. We determined statistical significance at a 2-tailed P<0.05.
| Results |
|---|
|
|
|---|
Fifty-four patients (36%) received either IV (n=24) or intra-arterial (IA, n=5) thrombolytic therapy or both (n=25). Their median National Institutes of Health Stroke Scale (NIHSS) score was 14 points (range 4 to 29); 81% had NIHSS
10 points. No delay in treatment resulted from ultrasound testing. Sixty-three (42%) patients were found ineligible for thrombolysis. Thirty-three patients (22%) had TIAs.
NVUE showed abnormal results consistent with our ultrasound criteria for LAIT in 98% of patients eligible for thrombolysis, 76% of acute stroke patients ineligible for thrombolysis, and 42% in patients with TIA, P<0.001 (Figure 1). In the thrombolysis group, 44% had internal carotid artery (ICA) lesions with or without middle cerebral artery (MCA) lesions, 52% had isolated MCA lesions, 2% had lesions in the vertebrobasilar (VB) system, and 2% had normal ultrasound findings. In patients ineligible for thrombolysis, 44% had ICA lesions; 30%, isolated MCA lesions; 2%, VB lesions; and 24% had normal findings. In the TIA group, 25% had isolated ICA lesions; none had tandem ICA+MCA lesions; 7%, isolated MCA lesions; 10%, VB lesions; and 58% had normal findings (Figure 1).
|
A total of 30 patients (20%) underwent urgent DSA at mean time of 230±207 minutes (maximum delay 308 minutes) after initial NVUE. DSA demonstrated LAIT presence in all patients with abnormal NVUE without false-positive ultrasound studies (sensitivity and specificity for LAIT presence 100%). DSA localized LAIT to the MCA in 10 patients, ICA in 4 patients, tandem ICA/MCA in 15 patients (Figure 2), and VB circulation in 1 patient (Figure 3).
|
|
Compared with DSA, the accuracy parameters for each component of NVUE were as follows: TCD 96% sensitivity, 75% specificity, 96% positive, and 75% negative predictive values (given the observed 10% rate of absent temporal windows). Carotid/vertebral duplex had 94% sensitivity, 90% specificity, 94% positive, and 90% negative predictive values (Table 3).
|
TCD and carotid/vertebral duplex ultrasound each showed 1 false-negative and 1 false-positive study (Figure 3). Tandem ICA/MCA lesions were responsible for these errors because of difficulties identifying the unilateral MCA occlusion with delayed collateral flow in the presence of bilateral carotid diseases and unilateral distal ICA stenosis beyond the reach of carotid duplex scanning area.
| Discussion |
|---|
|
|
|---|
The accuracy of NVUE demonstrated in this study is based on a relatively small sample of urgent invasive angiograms, and the selection bias introduced by abnormal ultrasound findings cannot be ruled out. Therefore, our findings should be interpreted with caution. However, TCD and CD taken separately have lesser accuracy parameters within the same patient sample. Knowledge of shortcomings of each test and the complimentary value of information derived from extra- and intracranial vessels helped to compensate for these shortcomings. Patients with posterior circulation symptoms, greater variability of posterior circulation vessels, and technical difficulties may further affect the performance of these tests.18,26 In our study, accuracy parameters were largely driven by findings in the anterior circulation. Individual accuracy parameters for TCD and CD were comparable to previous studies.2,3,5,7,27,28
Our study has limitations, including the need for considerable sonographer skill and expertise to complete testing promptly and efficiently. Other factors that can affect the yield and accuracy of NVUE include potential selection bias for diagnostic angiography and greater preponderance of patients with severe stroke symptoms or admitted very early after symptom onset. Also, our study had an average delay of over 3 hours between initial ultrasound and DSA that may provide ample time for thrombus propagation, dissolution, or reocclusion to occur. Our study also identified clinical situations when neither TCD nor CD could be reliably interpreted, including tandem lesions with poor collateral flow, limited arterial visualization, and absent temporal windows. Some of these shortcomings can be overcome by using ultrasound contrast agents.5,19,2931
Clinical implications of our findings affirm that rapid bedside ultrasound screening can be used in patient selection for interventional treatment.3234 With further developments in portable noninvasive technology, such examination can be feasibly taught to vascular neurologists, emergency physicians, and vascular technologists. The small devices required can become the new "stethoscope" for bedside evaluations.
In conclusion, urgent bedside NVUE yields a substantial proportion of LAITs in patients with acute cerebral ischemia. Transcranial and carotid ultrasound tests, if performed together by experienced sonographers, compensate for each others shortcomings and can achieve 100% accuracy in detection of lesions amenable for intervention.
Received June 20, 2004; revision received October 26, 2004; accepted October 27, 2004.
| References |
|---|
|
|
|---|
This article has been cited by other articles:
![]() |
V. K. Sharma, G. Tsivgoulis, A. Y. Lao, M. D. Malkoff, A. W. Alexandrov, and A. V. Alexandrov Quantification of Microspheres Appearance in Brain Vessels: Implications for Residual Flow Velocity Measurements, Dose Calculations, and Potential Drug Delivery Stroke, May 1, 2008; 39(5): 1476 - 1481. [Abstract] [Full Text] [PDF] |
||||
![]() |
G. Tsivgoulis, V. K. Sharma, S. L. Hoover, A. Y. Lao, A. A. Ardelt, M. D. Malkoff, and A. V. Alexandrov Applications and Advantages of Power Motion-Mode Doppler in Acute Posterior Circulation Cerebral Ischemia Stroke, April 1, 2008; 39(4): 1197 - 1204. [Abstract] [Full Text] [PDF] |
||||
![]() |
Y Iguchi, K Kimura, K Kobayashi, Y Ueno, K Shibazaki, and T Inoue Microembolic signals at 48 hours after stroke onset contribute to new ischaemia within a week J. Neurol. Neurosurg. Psychiatry, March 1, 2008; 79(3): 253 - 259. [Abstract] [Full Text] [PDF] |
||||
![]() |
G. Tsivgoulis, V. K. Sharma, A. Y. Lao, M. D. Malkoff, and A. V. Alexandrov Validation of Transcranial Doppler With Computed Tomography Angiography in Acute Cerebral Ischemia Stroke, April 1, 2007; 38(4): 1245 - 1249. [Abstract] [Full Text] [PDF] |
||||
![]() |
M. Saqqur, K. Uchino, A. M. Demchuk, C. A. Molina, Z. Garami, S. Calleja, N. Akhtar, F. O. Orouk, A. Salam, A. Shuaib, et al. Site of Arterial Occlusion Identified by Transcranial Doppler Predicts the Response to Intravenous Thrombolysis for Stroke Stroke, March 1, 2007; 38(3): 948 - 954. [Abstract] [Full Text] [PDF] |
||||
![]() |
Y Iguchi, K Kimura, K Kobayashi, Y Ueno, and T Inoue Ischaemic stroke with malignancy may often be caused by paradoxical embolism J. Neurol. Neurosurg. Psychiatry, December 1, 2006; 77(12): 1336 - 1339. [Abstract] [Full Text] [PDF] |
||||
![]() |
L. Sekoranja, J. Loulidi, H. Yilmaz, K. Lovblad, P. Temperli, M. Comelli, and R. F. Sztajzel Intravenous Versus Combined (Intravenous and Intra-Arterial) Thrombolysis in Acute Ischemic Stroke: A Transcranial Color-Coded Duplex Sonography-Guided Pilot Study Stroke, July 1, 2006; 37(7): 1805 - 1809. [Abstract] [Full Text] [PDF] |
||||
![]() |
A. V. Alexandrov and N. M. Bornstein Advances in Neurosonology 2005 Stroke, February 1, 2006; 37(2): 299 - 300. [Full Text] [PDF] |
||||
![]() |
R. Mikulik, A. V. Alexandrov, M. Ribo, Z. Garami, N. A. Porche, E. Fulep, J. C. Grotta, A. W. Wojner-Alexandrov, and J. Y. Choi Telemedicine-Guided Carotid and Transcranial Ultrasound: A Pilot Feasibility Study Stroke, January 1, 2006; 37(1): 229 - 230. [Abstract] [Full Text] [PDF] |
||||
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
|
Stroke Home | Subscriptions | Archives | Feedback | Authors | Help | AHA Journals Home | Search Copyright © 2005 American Heart Association, Inc. All rights reserved. Unauthorized use prohibited. |