335 Three-Dimensional Transcranial Doppler Blood Flow Mapping in Patients With Cerebrovascular Disorders Kurt Niederkorn, MD, Lawrence G. Myers, BS, RVT, Catherine L. Nunn, RN, RVT, Marshall R. Ball, MD, and William M. McKinney, MD Downloaded from http://ahajournals.org by on October 8, 208 We investigated 60 patients with cerebrovascular disorders using a three-dimensional transcranial Doppler blood flow mapping system. A composite display of the circle of Willis is created with computer assistance, allowing accurate vessel identification and optimal data documentation of blood flow velocity and direction in the basal cerebral arteries. The basilar artery was insonated in every patient; the middle cerebral artery and the most distal internal carotid artery were found in 95% of the patients, the anterior cerebral artery in 85%, and the posterior cerebral artery in 84%. Insonation problems occurred predominantly in elderly women. Transcranial Doppler blood flow mapping showed an abnormal result in 23 of 60 patients (38%). An intracranial stenosis with >50% diameter reduction or occlusion was found in 0 of 3 patients (32%) with completed stroke, reversible ischemic neurologic deficit, or transient ischemic attack. Collateral blood flow mechanisms could be demonstrated in patients with extracranial carotid artery occlusions. Intra-arterial cerebral angiography performed in 2 patients confirmed the transcranial Doppler blood flow mapping diagnosis in 9 (90.5%). In one patient an arteriovenous malformation diagnosed by transcranial Doppler blood flow mapping was confirmed by magnetic resonance imaging. (Stroke 988;9:335-344) Doppler ultrasonography has been used successfully as a noninvasive method for the diagnosis of extracranial artery disease. Intracranial Doppler ultrasonography studies remained unsatisfactory due to attenuation of the skull 2 until transcranial Doppler ultrasonography (TCD) was introduced in 982. 3 TCD allows noninvasive, direct measurement of blood flow velocity and direction in the basal brain arteries; it is used for diagnosis and follow-up of vasospasm in the middle cerebral artery (MCA) after subarachnoid hemorrhage 45 and for monitoring the MCA during carotid endarterectomy 6 and cardiopulmonary bypass surgery. 7 These applications require relatively few investigative skills because insonation of the MCA is usually uncomplicated and clearly defined anatomically. 8 In patients with cerebrovascular disorders, in whom the entire circle of Willis (a variable 9 and topographically complex structure) needs to be examined, cor- From the Departments of Neurology (K.N., L.G.M., C.L.N., W.M.M.) and Radiology (M.R.B.), Bowman Gray School of Medicine, Wake Forest University, Winston-Salem, North Carolina. Address for correspondence: William M. McKinney, MD, Department of Neurology, Bowman Gray School of Medicine, 300 South Hawthorne Road, Winston-Salem, NC 2703. Received June 2, 987; accepted June 3, 988. rect identification of vessels becomes a problem. Repeated common carotid artery (CCA) compressions may be necessary to identify specific vessels. 30 Using hand-held transducers, a mental map of the circle of Willis is created subjectively by the investigator and cannot easily be reproduced. Threedimensional transcranial Doppler blood flow mapping (TDFM) allows for reproducible results and improved vessel identification. n We report the results of clinical use of TDFM for the noninvasive diagnosis of blood flow abnormalities within the circle of Willis in patients with cerebrovascular disorders, and we show that TDFM is a reliable and accurate tool. Subjects and Methods We investigated 60 consecutive patients (37 males, 23 females; mean age 56.8 ± 9., range 2/2-83 years). Patients were referred to the clinical neuroultrasound laboratory because of a variety of cerebrovascular disorders (Table ). Five patients were black, one was Indian, and the others were white. All patients had continuous-wave or pulsed Doppler and B-mode ultrasonographic examination of the carotid arteries, and 48 had an extracranial vertebral artery Doppler ultrasonographic study before TDFM was performed. The investigators
336 Stroke Vol 9, No, November 988 Downloaded from http://ahajournals.org by on October 8, 208 TABLE. Referral Diagnoses of 60 Patients With Cerebrovascular Disorders Diagnosis Stroke ICA territory VB territory RIND (ICA territory) Transient ischemic attack ICA territory VB territory Dizziness, vertigo (isolated) Syncope Hearing loss Asymptomatic carotid bruit Status post SAH Status post ICH Suspected vascular malformation Migraine Total No. 9 2 2 5 3 3 2 4 3 4 60 ICA, internal carotid artery; VB, vertebrobasilar artery; RIND, reversible ischemic neurologic deficit; SAH, subarachnoid hemorrhage; ICH, intracerebral hemorrhage. were aware of the clinical history of the patients and the extracranial Doppler result. We used a pulsed Doppler device (Trans-scan, EME, Uberlingen, FRG) operating at 2 MHz (Figure ). The maximal pulse repetition frequency (PRF) was 2.2 khz, allowing measurement of a maximum Doppler shift of 6. khz (PRF/2). With highresolution color display, the Doppler frequency signals were converted by fast Fourier transformation into velocity (v), which was calculated assuming an insonation angle of 0 as v = 39 x f in khz. This assumption was possible (at an error rate of <5%) because the insonation angle of intracranial arteries was 0-30, cosine -0.86. The maximum velocity measurable without aliasing was therefore 39 x 6. = 237.9 cm/sec. However, a special design of the system we used allowed the display of frequencies between -PRF/2 and PRF. A headpiece with two scanning arms connected to the Doppler probes by ball joints and rods was placed on the patient's head on two curved pads positioned on the glabella and bregma (joint of the coronal and sagittal sutures), which were used as reference points. The probe position was maintained constant over the temporal skull during each examination once an optimal ultrasonic window was detected by potentiometers. Using a computer readout of the angles shown in Figure 2, the intracranial position of the sample volume was defined by the insonation depth (x coordinate) and the calculated distances between the glabella and bregma reference points and the sample volume (y and z coordinates). The sample volume measured 6x4 mm, and it could be moved bidirectionally in.5- mm steps. The recorded spectra, symbolized by colored dots, were displayed simultaneously in lateral, coronal, and horizontal views. The threedimensionality of TDFM can be appreciated by the investigator or reader by combining the three twodimensional displays mentally. The proximal MCA, the Al segment of the anterior cerebral artery (ACA), the distal part of the supraclinoid internal carotid artery (ICA), and the PI segment of the posterior cerebral artery (PCA) were then insonated in a stepwise fashion through the temporal window. The investigator created a composite display of these basal cerebral arteries in the lateral, coronal, and horizontal views (Figure 3). FIGURE. Transcranial Doppler blood flow mapping instrument consists of headpiece with Doppler probes placed over temporal bone and personal computer with color monitor displaying Doppler spectra and blood flow map. Patient is examined in supine position.
Niederkorn et al Transcranial Doppler Flow Mapping 337 Downloaded from http://ahajournals.org by on October 8, 208 FIGURE 2. Depth of sample volume (small circle) functions as x coordinate; y and z coordinates are calculated by computer readout of angles C,, C 2, H i, and H 2 (distances between reference points glabella and bregma and sample volume) and, together with x coordinate, describe intracranial position of sample volume. The basilar artery (BA) was examined with a 2- MHz hand-held probe; no CCA compressions were necessary with this method. Time for the entire investigation varied from 30 to 90 minutes. The topographic and spectral data of each sonogram were stored on the hard disk of a personal computer. For reading and evaluation, individual spectra could be recalled from the memory. Mean and peak blood flow velocities and pulsatility could be calculated. Permanent copies were made using a color printer, and the complete data could be stored on floppy disk for long-term documentation. The Doppler criteria 23 for stenosis of at least 50% diameter reduction were a >50% increase in velocity compared with the contralateral side and/ or with the normal reference values, 304 turbulence phenomena, and the absence of collateral blood flow mechanisms to avoid misinterpreting functionally increased blood velocities. Twenty-one of the 60 patients underwent intra-arterial cerebral angiography. All TDFM studies were performed without knowledge of the angiographic diagnosis. Results Based on a maximum searching time of 0 minutes per side 92.5% of the MCAs and ICAs, 85% of the ACAs, and 84.2% of the PCAs were found and identified. The BA was found in all of the 59 patients in whom it was searched for (this was not done in one child due to cooperation problems). A temporal window could not be found bilaterally in three patients (one man, two women, 5%) and unilaterally in four (all women, 6.7%); transtemporal insonation problems of some degree were therefore present in 2.7% of the male and in 26.% of the female patients (mean age of the women with insonation problems was 72.8 vs. 54.5 years for the other females). TDFM was normal in 37 of the 60 patients (6.7%); it was abnormal in the remaining 23 (38.3%). The nature and frequency of the 35 abnormal findings in the 23 patients are shown in Table 2. In nine of the 2 patients who had cerebral angiography all arteries insonated and angiographed were normal. In the 2 remaining patients TDFM was abnormal; their personal data, clinical symptoms, extracranial and intracranial Doppler results, and angiographic diagnoses are listed in Table 3. The sonographic and angiographic appearances of the ICA stenosis in Patient 9 is shown in Figure 4, and those of the extracranial ICA occlusion in Patient 6, with cross flow via the ACAs, in Figure 5. There was good topographic correlation between TDFM and angiography. In patients in whom one ACA or PCA could not be clearly identified by TDFM, there was hypoplasia or an unusual anatomic course of the vessel, resulting in an unfavorable insonation angle. The only discrepancies between TDFM and angiography occurred in Patients and 5, with unilateral cervical ICA obstructions. TDFM suggested complete cross flow via the ACAs and the anterior communicating arteries (ACoAs) in both patients. However, angiography showed unilateral filling of both A2 segments and no cross flow. An arteriovenous malformation (AVM) found by TDFM was confirmed by magnetic resonance imaging (MRI) (Figure 6). Discussion Our results in a series of 60 patients with cerebrovascular disorders demonstrate that threedimensional TDFM is capable not only of providing real-time information on the physiologic parameters of blood flow velocity, blood flow direction, and pulsatility index but also of creating a topographically accurate composite display of the circle of Willis. The MCA, ACA, and BA were found at rates similar to those reported in the literature. 0 The PCA, however, could be identified in 85% of sides, which is well above the 70% reported by Arnolds and von Reutern 0 in normal subjects and using hand-held transducers. This improvement can be attributed to the clear separation between the anterior and posterior parts of the circle of Willis possible with TDFM. The ability to selectively investigate the supraclinoid ICA, an area of predilection for atherosclerotic disease, 5 is another major advantage of TDFM. The ACA can be distinguished easily from the ICA, which is important in patients with ACA cross flow due to ICA occlusion. CCA compression for
338 Stroke Vol 9, No, November 988 TV v- - 80 MiMCA R MCA C^ CA NORMAL R MAP 60 70 80 90 too 0 20 30 2O 30 40 50 60 70 80 90 X» Downloaded from http://ahajournals.org by on October 8, 208 FIGURE 3. Color: Transcranial Doppler bloodflowmapping. Top left: Normal Doppler signal of right Ml segment (R) of middle cerebral artery (MCA) of 22-year-old man; scale is cm/sec. Dot (arrow, circle) represents position of sample volume in which signal was obtained in lateral (bottom left), anteroposterior (top right), and horizontal (bottom right) views. Red, flow toward probe; blue, flow away from probe; brightness, mean velocity. Medial to MCA, Cl segment of internal carotid artery (ICA), andal segment of anterior cerebral artery (ACA), posterior and slightly inferior to most proximal PI segments of posterior cerebral arteries (PCA) are represented. Bottom: Right carotid angiogram of same 22- year-old man, anteroposterior view. Topographic similarity between image in box and anteroposterior view in flow map can be appreciated.
Niederkorn et al Transcranial Doppler Flow Mapping 339 TABLE 2. Intracranial Abnormalities Found in 23 of 60 Patients by Transcranial Doppler Blood Flow Mapping Abnormality MCA stenosis MCA occlusion MCA vasospasm ICA stenosis ICA vasospasm ACA stenosis ACA vasospasm PCA stenosis VA-BA stenosis MCA patent, reduced blood flow above cervical ICA occlusion Cross flow via the ACAs Compensatory increased PCA velocity Reversed VA blood flow direction (subclavian steal) Arteriovenous malformation No. Data are number of abnormalities, not number of patients. MCA, middle cerebral artery; ICA, internal carotid artery; ACA, anterior cerebral artery (Al segment); PCA, posterior cerebral artery (PI segment); VA, vertebral artery; BA, basilar artery. 3 4 4 8 7 vessel identification, which is contraindicated in patients with CCA plaques or hemodynamically significant ICA obstructions, can be avoided completely, thus making TDFM truly noninvasive. Other advantages of TDFM are computerized documentation and data processing, which allow optimal interpretation and reproducibility. Ten patients with of the 4 intracranial obstructions we observed belonged to the subgroup of 3 patients with stroke, reversible ischemic neurologic deficit (RIND), or transient ischemic attacks (TIAs). Eight of those 0 patients had no significant extracranial carotid artery disease. Thus, in 32% of the patients with stroke, RIND, or TIA, TDFM identified a significant intracranial artery lesion. Of the three black patients with stroke, RIND, or TIA, two had severe intracranial artery stenoses, underlining the higher incidence of intracranial artery disease in blacks 67 and the importance of TDFM in these patients. Clinical studies of MCA stenosis 8 and ICA siphon stenosis 9 report high annual ipsilateral stroke rates of 7.8% and 4.7% per patientyear. Based on these observations, early noninvasive diagnosis of these lesions is of vital importance for the patient. The EC/IC Bypass Study Group 20 Downloaded from http://ahajournals.org by on October 8, 208 TABLE 3. Clinical and Laboratory Findings in 2 Patients With Abnormal Transcranial Doppler Blood Flow Mapping Who Underwent Angiography Pt/age/ race/sex /53/W/M 2/46/I/M 3/64/W/F 4/6/W/M 5/74/W/M 6/52/B/M 7/22/W/F 8/73/W/F 9/42/B/F 0/74/W/M /39/B/F 2/66/W/F History and symptoms Asymptomatic R carotid bruit Bilateral MID, status post L TEA Status post R MCA stroke Status post R MCA stroke Status post SAH TIAs in BA territory Status post R MCA stroke Dizziness, numbness L arm Progressive L side weakness Status post R MCA stroke Status post R ICH, evacuated Asymptomatic carotid bruits Extracranial vertebral Doppler ultrasonography >75% R ICA stenosis R ICA occlusion; >75% L ICA stenosis Bilateral ICA occlusion at origin R ICA occlusion at origin R ICA occlusion at origin Reduced VA signals Reduced R ICA signal L SA occlusion, reversed L VA flow direction Reduced R ICA signal Minimally reduced R ICA signal Normal >75% bilateral ICA stenosis Transcranial Doppler blood flow mapping L to R cross flow via ACAs R MCA patent, L to R cross flow via ACAs MCA flow normal; BA, PCA velocity increased L to R cross flow via ACAs No spasm, L to R cross flow via ACAs Distal BA severe turbulence R distal MCA no flow, R proximal MCA + R distal ICA highly reduced flow L VA reversed flow direction; BA flow normal R ICA high velocity (>200 cm/sec systolic), turbulence R ICA increased (50 cm/sec systolic) velocity,. turbulence R ICA, MCA, ACA increased velocity R ICA increased (70 cm/sec systolic) velocity, turbulence Angiography No cross flow, L Al fills bilateral A2 Confirmed Doppler results Bilateral MCA, ACA filled byba Confirmed Doppler results No cross flow, L Al fills bilateral A2 BA severe (>90%) multiple stenoses R MCA occlusion (Ml segment) SA occlusion, L subclavian steal, BA forward flow Supraclinoid R ICA 90% stenosis Supraclinoid R ICA 75% stenosis R ICA, MCA, ACA vasospasm Supraclinoid R ICA at least 75% stenosis Pt, patient number; W, white; I, Indian; B, black; M, male; F, female; R, right; L, left; MCA, middle cerebral artery; ICA, internal carotid artery; ACA, anterior cerebral artery; PCA, posterior cerebral artery; VA, vertebral artery; BA, basilar artery; MID, multi-infarct disease; TEA, carotid thromboendarterectomy; SAH, subarachnoid hemorrhage; TIA, transient ischemic attack; ICH, intracerebral hemorrhage. Percent stenosis refers to area reduction. Extracranial vertebral artery Doppler results were all confirmed by angiography. Angiographic results refer to intracranial conditions only.
340 Stroke Vol 9, No, November 988 Downloaded from http://ahajournals.org by on October 8, 208 FIGURE 4. Color: Transcranial Doppler bloodflowmapping of Patient 9 (42-year-old black woman). Bright dots (arrow), area of increased blood flow velocity in most distal right (R) internal carotid artery (ICA). Top left: Doppler signal from area with increased velocity and severe turbulence. Anterior cerebral artery spectrum is displayed below zero line. Bottom left: Poststenotic middle cerebral artery (MCA) Doppler signal is damped and almost nonpulsatile. Bottom: Angiogram of Patient 9 confirms presence of severe (90%) stenosis of most distal R ICA (arrow).
Niederkorn et al Transcranial Doppler Flow Mapping 34 Record Mo -;-*& REVERSED R ACA (CIRCLE) L ACA,! liee I -70 -so T.L ICA (BRRON) 48 28 e tto \ V!OO 90 80 70 60 50 40 30-^ -90-00 -no ii - Downloaded from http://ahajournals.org by on October 8, 208 FIGURE 5. Color: Transcranial Doppler blood flow mapping. Cross flow from left to right hemisphere is documented in patient with right (R) cervical internal carotid artery (ICA) occlusion. Blood flow velocity in left Al segment of anterior cerebral artery (ACA) is increased, in R ACA reversed. Insonation from left results in string of blue dots across midline. Bottom: Angiogram of same patient confirms diagnosis and shows strong crossflowfrom left to right hemisphere. \
342 Stroke Vol 9, No, November 988 Downloaded from http://ahajournals.org by on October 8, 208 FIGURE 6. Color: TranscranialDoppler blood flow mapping revealed increased blood flow velocities in right proximal middle cerebral artery (MCA) and convolute of low pulsatility and high-velocity blood flow signals anterior and inferior from MCA mainstem in 36-year-old man with history of seizures since 970; arteriovenous malformation had been diagnosed at that time by angiography, which identified MCA and anterior cerebral artery as feeders. There was no reaction to hyperventilation in these vessels. Bottom: Magnetic resonance imaging confirms size and location of arteriovenous malformation.
Niederkorn et al Transcranial Doppler Flow Mapping 343 Downloaded from http://ahajournals.org by on October 8, 208 reported no significant benefit in these cases, but anticoagulation has been associated with benign outcome in patients with MCA stenosis. 2 An advantage of TDFM is its capability to rule out MCA-ICA stenoses in patients considered for carotid endarterectomy. Before antihypertensive treatment is started in elderly patients, significant obstructions of the major intracranial arteries should be excluded to prevent hemodynamically induced ischemic events. Intracranial vertebral artery and BA stenoses carry a high risk of brainstem infarction. 22 Stenoses can be detected and followed noninvasively using TDFM, and the effectiveness of anticoagulation and fibrinolysis can be assessed directly. 23 In patients with extracranial ICA occlusions, collateral blood flow pathways and patency of the MCA were clearly demonstrated by TDFM. This is especially important in the clinical assessment of these individuals because normal postocclusive MCA blood flow velocity may be a good prognostic sign. 24 However, in two of four patients with ACA collateral blood flow who underwent angiography, there was partial disagreement with TDFM because cross flow was not complete and one Al segment supported both A2 segments of the ACAs but not the postobstructive MCA. A precise evaluation of every individual with collateral blood flow patterns is necessary to avoid misinterpretation of high blood flow velocities caused by collateral flow through relatively narrow channels such as the ACA, PCA, and communicating arteries. 25 These functional stenoses occur when severe extracranial carotid artery obstructions are present and when circulation in the postocclusive hemisphere is supplied by the contralateral side via the ACA and the ACoA and/or from the ipsilateral side via the PCA and the posterior communicating artery (PCoA). A regular extracranial Doppler and duplex or B-scan examination should always be performed before TDFM. In certain instances, especially before surgery, CCA compression tests may be helpful in assessing collateral blood flow and the capacity of the circle of Willis and in distinguishing between functional and morphologic stenoses. The transcranial Doppler diagnosis of MCA occlusion is difficult. The examination must demonstrate an ultrasonic window on the affected side, and the other ipsilateral vessels must be clearly insonated and identified. 2 Further experience with angiographic correlations and very careful evaluation of the extracranial and intracranial blood flow data is necessary to establish the validity of TDFM in the diagnosis of MCA occlusion. In the follow-up of such patients TDFM may be the method of choice to learn more about spontaneous recanalization and its possible correlation with clinical recovery. 26 AVMs can be detected and monitored after treatment with the help of TCD. 2728 In our patient the approximate size and location of the AVM was determined by TDFM and confirmed by MRI. We included only one patient with subarachnoid hemorrhage and one with an intracranial hemorrhage causing vasospasm of the ICA, MCA, and ACA. However, the value of TCD in the noninvasive diagnosis and clinical management of vasospasm has been demonstrated extensively. 4528 The number of patients in our study is too small to calculate sensitivity and specificity, but previous publications report a similar high accuracy of TCD in the diagnosis of intracranial artery blood flow abnormalities. 2-3 - 30 Although larger correlative studies are necessary to definitely establish the accuracy of TDFM compared with angiography and pathology, TDFM may be a valuable and practicable tool to noninvasively provide hemodynamic information from the circle of Willis to the clinician. References. Von Reutern G-M: Functional and morphological evaluation of the cerebral circulation by ultrasound, in Poeck K, Freund H-J, Ganshirt H (eds): Neurology Proceedings of the Xlllth World Congress of Neurology, Hamburg, September -6, 985. Berlin/Heidelberg/New York/Tokyo, Springer-Verlag, 986, pp 44-452 2. White DN, Curry GR, Stevenson RJ: The acoustic characteristics of the skull. Ultrasound Med Biol 972;4:225-252 3. Aaslid R, Markwalder T-M, Nornes H: Noninvasive transcranial Doppler ultrasound recording of flow velocity in basal cerebral arteries. J Neurosurg 982;57:769-774 4. Aaslid R, Huber P, Nornes H: Evaluation of cerebrovascular spasm with transcranial Doppler ultrasound. J Neurosurg 984;60:37-4 5. Seiler RW, Grolimund P, Aaslid R, Huber P, Nornes H: Cerebral vasospasm evaluated by transcranial ultrasound correlated with clinical grade and CT-visualized subarachnoid hemorrhage. J Neurosurg 986:64:594-600 6. Padayachee TS, Gosling RG, Bishop CC, Burnand K, Browse NL: Monitoring middle cerebral artery blood velocity during carotid endarterectomy. Br J Surg 986;73:98-00 7. Lundar T, Lindegaard K-F, Froysaker T, Aaslid R, Wiberg J, Nornes H: Cerebral perfusion during nonpulsatile cardiopulmonary bypass. Ann Thorac Surg 985;40:44-50 8. Gillard JH, Kirkham FJ, Levin SD, Neville BGR, Gosling RG: Anatomical validation of middle cerebral artery position as identified by transcranial pulsed Doppler ultrasound. J Neurol Neurosurg Psychiatry 986;49:025-029 9. Riggs HE, Rupp C: Variation in form of circle of Willis. Neurology 963;8:24-30 0. Arnolds BJ, von Reutern G-M: Transcranial Dopplersonography. Examination technique and normal reference values. Ultrasound Med Biol 986;2:5-23. Aaslid R: The Doppler principle applied to measurement of blood flow velocity in cerebral arteries, in Aaslid R (ed): Transcranial Doppler Sonography. Wien/New York, Springer-Verlag, 986, pp 22-38 2. Lindegaard K-F, Bakke SJ, Aaslid R, Nornes H: Doppler diagnosis of intracranial artery occlusive disorders. J Neurol Neurosurg Psychiatry 986;49:50-58 3. Niederkorn K, Neumayer K: Transcranial Doppler sonography: A new approach in the non-invasive diagnosis of intracranial brain artery disease. Eur Neurol 987;26:65-68 4. Hennerici M, Rautenberg W, Sitzer G, Schwartz A: Transcranial Doppler ultrasound for the assessment of intracranial arterial flow velocity Part Examination technique and normal values. Surg Neurol 987;27:439-448 5. Baker AB, Iannone A: Cerebrovascular disease. I. The large arteries of the circle of Willis. Neurology 959,9:32-332 6. Heyman A, Fields WS, Keating D: Joint study of extracranial arterial occlusion. VI. Racial differences in hospitalized patients with ischemic stroke. JAMA 972;222:285-289
344 Stroke Vol 9, No, November 988 7. Nishimaru K, McHenry LC, Toole JF: Cerebral angiographic and clinical differences in carotid system transient ischemic attacks between American Caucasians and Japanese patients. Stroke 984;5:56-59 8. Bogousslavsky J, Bamett HJM, Fox AJ, Hachinski VC, Taylor W, for the EC/IC Bypass Study Group: Atherosclerotic disease of the middle cerebral artery. Stroke 986; 7:2-20 9. Wechsler LR, Kistler JP, Davis KP, Kaminski MJ: The prognosis of carotid siphon stenosis. Stroke 986;7:74-78 20. EC/IC Bypass Study Group: Failure of extracranial-intracranial arterial bypass to reduce the risk of ischemic stroke. NEnglJ Med 985;33:9-200 2. Hinton RC, Mohr JP, Ackermann RH, Adair LB, Fisher CM: Symptomatic middle cerebral artery stenosis. Ann Neurol 979;5:52-57 22. Moufarru NA, Little JR, Furlan AJ, Leatherman JR, Williams GW: Basilar and distal vertebral artery stenosis: Long-term follow-up. Stroke 986;7:938-942 23. Ringelstein EB: Ultraschalldiagnostik am vertebrobasilaren Kreislauf. Teil II: Transnuchale Diagnose intrakranieller vertebrobasilarer Stenosen mit Hilfe eines neuartigen Impulsschall-Doppler-Systems. Ultraschall Med 985;6:60-67 24. Hennerici M, Rautenberg W, Herzog H, Langen KJ, Rota E, Kunert T, Aulich A, Feindegen LE: Transcranial Doppler and positron emission tomography in asymptomatic patients with extracranial arterial disease, in Proceedings of Intracranial Hemodynamics: Transcranial Doppler and Cerebral Blood Flow, February 28-March 3, Clearwater Beach, Fla 25. Lindegaard K-F, Bakke SJ, Grolimund P, Aaslid R, Huber P, Nornes H: Assessment of intracranial hemodynamics in carotid artery disease by transcranial Doppler ultrasound. J Neurosurg 985;63:890-898 26. Sindermann F, Dichgans J, Bergleiter R: Occlusion of the middle cerebral artery and its branches: Angiographic and clinical correlates. Brain 969;92:607-620 27. Lindegaard K-F, Grolimund P, Aaslid R, Nornes H: Evaluation of cerebral AVM's using transcranial Doppler ultrasound. J Neurosurg 986;65:335-344 28. Schwartz A, Hennerici M: Noninvasive transcranial Doppler ultrasound in intracranial angiomas. Neurology 986; 36:626-635 29. Gilsbach JM, Harders A: Early aneurysm operation and vasospasm. Neurochirurgia 985;28:00-02 30. Wechsler LR, Ropper AH, Kistler JP: Transcranial Doppler in cerebrovascular disease. Stroke 986;7:905-92 KEY WORDS angiography cerebrovascular disorders ultrasonic diagnosis Downloaded from http://ahajournals.org by on October 8, 208