Background and Purpose Transcranial color-coded duplex sonography (TCCS) enables visualization of the intracranial parenchymal structures and measurement of blood flow velocity in the basal cerebral arteries. The present study aims to evaluate prospectively the clinical usefulness of TCCS in patients with acute stroke.
Methods Eighty-four consecutive patients with central nervous symptoms suggesting acute stroke were investigated within the first 48 hours after clinical onset. TCCS was performed with a 2.5-MHz sector transducer through the temporal bone window. CT was available in all patients.
Results Forty-eight patients suffered from an infarction or a transient ischemic attack (TIA) in the territory of the middle cerebral artery (MCA). Fifteen of them showed an MCA occlusion, and 12 of the 15 developed recanalization during follow-up. Twelve revealed an increased, decreased, or oscillating flow pattern in the MCA main stem, and 21 patients had no ultrasonic abnormalities. The positive and negative predictive values of a pathological flow pattern in patients with MCA infarctions or TIA were .92 and .48, respectively. Fifteen patients suffered from an intracerebral hematoma, which could be diagnosed by TCCS in 14 cases. The positive and negative predictive values of a pathological parenchymal echo pattern were .88 and .96, respectively. Three patients suffered from an infarction and one from a TIA in the posterior cerebral artery territory. One female patient with an acute deterioration of a hemiparesis showed a glioma. The dropout rate due to an insufficient acoustic temporal bone window was 20% (17/84).
Conclusions TCCS is a noninvasive bedside method that provides rapid and reliable data regarding stroke subtype and mechanism immediately after onset. Window failure is a serious limitation of this method.
The development of TCD1 has opened a new noninvasive diagnostic window for the evaluation of intracranial hemodynamics. Among other applications, the clinical usefulness has been demonstrated in detecting intracranial artery stenosis, in monitoring vasospasm after subarachnoid hemorrhage,2 in follow-up studies of MCA occlusion,3 4 and in the diagnosis of cerebral circulatory arrest.5 However, soon some distinct problems of conventional hand-held TCD examinations were recognized, such as artery identification under pathological conditions, unknown insonation angle, documentation, and exploration of the “acoustic window.”
TCCS is a relatively new method that allows the simultaneous visualization of brain parenchyma in B-mode and the basal cerebral artery blood flow with the use of color flow imaging.6 7 It enables angle-corrected measurements of BFV8 and better documentation and discrimination of the spatial relationship of the basal cerebral arteries and provides additional information regarding parenchymal abnormalities.9 10 11 However, thus far there have been no prospective studies evaluating the usefulness of TCCS in a routine clinical setting.
This study aims to evaluate the applicability and reliability of TCCS to differentiate stroke subtypes and mechanisms immediately after admission of stroke patients to the hospital.
Subjects and Methods
A series of 84 consecutive patients with suspected cerebrovascular disease (mean age, 62.9±13.7 years; 48 male, 36 female) were enrolled within a period of 16 months. Each patient was examined by TCCS within 48 hours after onset of the neurological symptoms (mean time, 23.9±13.7 hours). In 41 patients the TCCS examination was the first diagnostic modality, and therefore the investigator was blinded to the results of other methods such as CT or angiography. In the remaining 43 patients the investigator knew that the patient under investigation suffered from an infarction or a hematoma, but data regarding the localization of the lesion or the hemodynamic or angiographic findings were not available. All patients were investigated with CT scan (Siemens Somatom HiQ) with sufficient latency to stroke onset to evaluate the presence and extent of ischemic lesions. The final neurological diagnosis (Table 1⇓) was based on patient history, clinical examination, and CT findings. In 8 patients arterial digital subtraction angiography of the extracranial and intracranial arteries was available.
Extracranial color Doppler examination (HP SONOS 1000, Hewlett-Packard Co; 5-MHz sector or 7.5-MHz linear scanner) of the brain-supplying arteries was performed before the transcranial study in all patients
TCCS was performed with a 2.5-MHz, 90° sector transducer transtemporally in projection to the orbitomeatal line. With a scanning depth of 16 cm, the butterfly-shaped brain stem was visualized first to obtain a landmark for orientation. Intraparenchymal hemorrhage was investigated from the contralateral temporal window, taking advantage of better image quality in the far field of the 2.5-MHz probe.
The angle-corrected BFV of the basal cerebral arteries was assessed from the ipsilateral perspective. With a scanning depth of 4 to 6 cm, the echo-dense structure of the M1 section of the MCA in the sylvian fissure was visualized in B-mode. Angle-corrected measurements of BFV could be obtained by switching to the Doppler mode. The anterior cerebral artery, the intracranial portion of the internal carotid artery (C1 and C2 sections), and the posterior cerebral artery were differentiated with reference to the anatomic course and blood flow direction. Compression tests were not necessary. The complete procedure was documented on videotape. The diameter of the sample volume was 12 mm, and the maximal intensity of ultrasound was 207 W/cm2 (ISPTA=86 mW/cm2). The angle-corrected BFV was displayed graphically (5 to 154 cm/s). During BFV measurements, the B-mode image was refreshed every 3 seconds to verify correct positioning of the probe. The spatial resolutions of the ultrasound beam in axial and lateral directions were 1.5 and 2.3 mm, respectively, at a focus depth of 6 cm (technical data from Hewlett-Packard Co).
An occlusion of MCA was suspected when (1) the M1 section of the artery was visible in the B-mode image and a Doppler signal was lacking in this structure (Fig 1⇓, Table 2⇓) and (2) all other ipsilateral basal cerebral arteries were detectable.4 12 Patients suffering from MCA occlusion were reexamined at short intervals (days 1, 2, 5, 8, 11, 14, and 21) to corroborate the diagnosis by monitoring the progress of recanalization. According to a previous study,13 an increase or decrease in BFV was considered pathological if the side-to-side difference of the angle-corrected peak systolic BFV in the MCA between the symptomatic and asymptomatic sides exceeded 20 cm/s.
The diagnosis of intracerebral hemorrhage was based on the typical sharply demarcated hyperechogenic B-mode pattern of the blood clot, as reported previously.9 10 Hemorrhagic transformation of a primarily isoechogenic infarction was diagnosed by TCCS during follow-up. If hyperechogenicity occurred within the ischemic area, a secondary hemorrhagic transformation of the infarction was diagnosed. This diagnosis was always confirmed by subsequent CT scan.
These initial TCCS findings were interpreted by the investigator at the bedside of the patient to determine the TCCS diagnosis shown in Table 1⇑. Afterward no change of this initial ultrasound diagnosis was made.
For statistical purposes, CT scan served as the “gold standard.” A lacunar lesion was radiologically defined as a small (<1.5 cm), low-density, and sharply marginated area on CT scan, usually located within or near the internal capsule, corona radiata, or pons.14 Sensitivity, specificity, and predictive values were calculated from cross tables.
Of 48 patients with ischemia related to the MCA territory and sufficient ultrasound penetration through the skull, we detected MCA occlusion in 15. In 12 patients with ischemia in the MCA territory, pathological increase, decrease, or oscillation of MCA BFV was registered (Table 1⇑). Four of 6 patients with decreased BFV had an infarction in the MCA territory compared with only 1 of 5 patients with increased BFV (Table 2⇑).
During follow-up, 12 patients with MCA occlusion underwent recanalization within a period of 3 weeks (Fig 2⇓). One patient who died revealed typical signs of a cerebral circulatory arrest with oscillating BFV in the contralateral primarily unaffected MCA.
In 3 patients with MCA occlusion due to cardiac embolism, TCCS disclosed hemorrhagic transformation (Fig 3⇓) during follow-up, which was confirmed by CT scan. One stroke was classified as a hematoma in the parietofrontal lobe by TCCS on admission but later was determined to be a hemorrhagic infarction. The hemorrhagic transformation of an infarction is a source of error and could be misdiagnosed because the ischemic brain tissue and the edema could not be imaged by TCCS. This diagnosis could only be confirmed by follow-up examinations, as shown in Fig 3⇓.
The positive and negative predictive values for an infarction in the MCA territory diagnosed by an abnormal blood flow pattern in the MCA main stem were .92 and .48, respectively (n=48) (Table 3⇓). The dropout rate due to an insufficient acoustic temporal bone window was 20% (17/84).
The diagnosis of intracerebral hemorrhage could be established by TCCS in 14 patients (Table 1⇑, Fig 4⇓) with focal hyperechogenicity in the parenchyma. The positive and negative predictive values of an intracerebral hemorrhage in 8 patients in whom TCCS was performed as the first diagnostic modality were .88 and .96, respectively (Table 4⇓).
One small (1×1 cm) putaminal hemorrhage was overlooked by the first examiner. A retrospective analysis of the videotape showed a very low increase of echogenicity in the putamen due to clotted blood.
One female patient with a glioblastoma was introduced into the study because of apoplectic onset of neurological symptoms. The echo pattern in TCCS with a hypoechogenic center of the lesion and a hyperechogenic margin allowed no differentiation from an acute hemorrhage (Fig 5⇓).
Stroke subtype and mechanism play a key role in patient management. The yield of pathological vascular findings depends on early and rapid evaluation. Noninvasive ultrasound examination of the brain-supplying arteries is a first-line diagnostic modality. TCCS represents a promising tool in this respect, but its specific impact on stroke management in a clinical setting has not yet been determined. The purpose of our study was to evaluate prospectively the diagnostic potential of TCCS in a nonselected population of stroke victims in the acute phase of the disease. In 41 of the 84 stroke patients under study, the investigator was totally blinded to the results of other diagnostic modalities such as CT scan or angiography. The TCCS diagnosis of stroke subtype was made at the bedside of the patient and was later compared with the final diagnosis based on patient history, clinical examination, and angiographic and CT findings.
An intracerebral hematoma can be diagnosed as a hyperechogenic area within the brain tissue, as described previously.9 10 The positive and negative predictive values of this diagnosis in our series are high (.88 and .96, respectively), enabling reliable differentiation between ischemic stroke and intracerebral hemorrhage.
An MCA occlusion can be diagnosed easily if the visualization of an artery is possible by B-mode and no Doppler signal is detectable in this structure. In our study we found recanalization in most of our patients (12 of 15). We also observed for the first time that TCCS depicts parenchymal images of hemorrhagic transformation during the follow-up of patients with brain infarctions. This information emphasizes the advantage of TCCS over conventional “blind” TCD. An abnormal blood flow pattern such as MCA occlusion or increased or decreased BFV can be regarded as an indirect sign of ischemic stroke within the MCA territory. The positive predictive value of a pathological flow pattern to establish the diagnosis of an ischemic stroke is high (.92), but the negative predictive value is very low (.48). Thus, pathological hemodynamic findings are very suggestive of the development of an infarction. Ischemia in the MCA territory might be suspected in patients with related clinical symptoms if a pathological hemodynamic flow pattern is present but no hyperechogenic parenchymal lesion representing a hemorrhage can be visualized. The fact that in our study 55% of the MCA stroke patients revealed MCA infarction on CT scan despite normal BFVs can be explained by the delay between stroke onset and TCCS examination or involvement of distal MCA branches.
Zanette and coworkers15 compared cerebral angiography and conventional TCD within the first 4 hours after stroke onset in patients with focal cerebral ischemia in the carotid territory. The most significant Doppler finding was the absence of an MCA signal in patients with angiographically occluded carotid siphon or MCA main stem. Reduced BFV in the MCA was found if the occlusion of the MCA was located in the terminal tract of the main stem or in numerous (>3) terminal branches. According to this study,15 we interpreted decreased BFV in our patients as a result of occlusion of several MCA branches. An abnormal accelerated BFV obtained from the symptomatic MCA is indicative of early recanalization followed by postischemic hyperperfusion.4 In our study only one of six patients with increased BFV suffered from a partial MCA infarction, indicating early recanalization without structural damage of brain tissue in most of those patients.
The diagnostic improvement of TCCS compared with conventional TCD is the visualization of the brain parenchyma. Intracerebral hematomas can be diagnosed by TCCS with high sensitivity and specificity. Furthermore, the identification of the basal cerebral arteries in relation to the sylvian fissure and the brain stem is unequivocally possible without a carotid compression test. In patients undergoing thrombolytic therapy, TCCS is a noninvasive method to diagnose potential hemorrhagic complications of the therapy by a close duplex monitoring.16 This might help to reduce the number of potential stressful and expensive imaging procedures such as CT and MRI. The addition of the time-saving, noninvasive transcranial duplex examination advances the diagnostic potential of the extracranial duplex examination.
Disadvantages of TCCS compared with CT or MRI are the limited spatial resolution of the ultrasound images and the dropout rate because of insufficient acoustic bone window in 20% of our stroke patients. The diagnosis of parenchymal bleeding is particularly limited in small (<1×1 cm) and cortically located lesions.10 It is also not possible to distinguish whether a hyperechogenic clot in the brain parenchyma is a hemorrhagically transformed infarction or a primary intracerebral hemorrhage. An ischemic lesion or brain edema cannot be visualized by TCCS. Brain tumors like apoplectic gliomas may be misdiagnosed as acute intracerebral hemorrhages.
In conclusion, TCCS is a noninvasive bedside method that differentiates stroke subtypes and provides immediate information regarding hemodynamic status in the basal cerebral arteries. In connection with extracranial color duplex examination, it helps to expeditiously select patients for specific therapeutic procedures and provides information on those patients not suitable for angiography. TCCS is not likely to replace high-resolution imaging systems such as CT or MRI, but it may detect hemorrhage that already exists on a patient’s admission to the hospital, thereby expediting further decisions on the management of stroke patients.
Selected Abbreviations and Acronyms
|BFV||=||blood flow velocity|
|MCA||=||middle cerebral artery|
|TCCS||=||transcranial color-coded duplex sonography|
|TCD||=||transcranial Doppler sonography<\/.>|
- Received April 10, 1995.
- Revision received July 31, 1995.
- Accepted July 31, 1995.
- Copyright © 1995 by American Heart Association
Ringelstein EB, Biniek R, Weiller C, Ammeling B, Nolte PN, Thron A. Type and extent of hemispheric brain infarctions and clinical outcome in early and delayed middle cerebral artery recanalization. Neurology.. 1992;42:289-298.
Kaps M, Damian MS, Teschendorf U, Dorndorf W. Transcranial Doppler ultrasound findings in middle cerebral artery occlusion. Stroke.. 1990;21:532-537.
Bogdahn U, Becker G, Winkler J, Greiner K, Perez J, Meurers B. Transcranial color-coded real-time sonography in adults. Stroke.. 1990;21:1680-1688.
Tsuchiya T, Yasaka M, Yamaguchi T, Hasegawa Y, Kimura K, Omae T. Transcranial real-time color-flow Doppler ultrasonography, part 1: imaging of basal cerebral arteries and measurement of blood velocity. Stroke. 1990;21(suppl I):I-49. Abstract.
Seidel G, Kaps M, Dorndorf W. Transcranial color-coded duplex sonography of intracerebral hematomas in adults. Stroke. 1993;24:1519-1527.
Seidel G, Schweizer J, Kaps M, Brandl HG. Transkranielle farbkodierte Duplexsonographie der A. cerebri media bei extra- und intrakraniellen Stenosen. In: Schimrigk K, Haas H, Hamann G, eds. Verhandlungen der Deutschen Gesellschaft für Neurologie. Homburg/Saar, FRG: 1992;7:341-342.
Millikan C, Futrell N. The fallacy of the lacune hypothesis. Stroke. 1990;21:1251-1257.
Zanette EM, Fieschi C, Bozzao L, Roberti C, Toni D, Argentino C, Lenzi GL. Comparison of cerebral angiography and transcranial Doppler sonography in acute stroke. Stroke. 1989;20:899-903.
Kaps M, Seidel G, Gerriets T, Traupe H. Transcranial duplex monitoring discloses hemorrhagic complications following rt-PA thrombolysis. Acta Neurol Scand. In press.