- Email: [email protected]
Transcranial Doppler ultrasonography in patients with Moyamoya disease
Clinical Neurology and Neurosurgery 99 Suppl. 2 (1997) S101 – S105
Transcranial Doppler ultrasonography in patients with Moyamoya disease Kensaku Takase *, Michiharu Kashihara, Tsuneyo Hashimoto Department of Neurosurgery, Tokushima Prefectural Central Hospital, Tokushima 770, Japan
Abstract This study included 17 patients (five men and 12 women) with clinical diagnosis of Moyamoya disease from conventional angiograms. Bilateral basal arteries were measured by the transtemporal approach with a 2 MHz pulsed Doppler instrument (TC-2 64B EME). In 28 out of the 34 basal arteries (82.4%), reliable recordings were obtained. These transcranial Doppler ultrasonography (TCD) findings were classified into three patterns: (1) High-high pattern; the mean cerebral blood flow velocity (CBFV) was increased throughout the basal arteries by over 70 cm/s (seven arteries, 25.0%); (2) high-low pattern; the mean CBFV was fastest (over 70 cm/s) at the ICA or proximal MCA and the mean CBFV decreased remarkably distally (15 arteries, 53.6%); and (3) low-low pattern; the mean CBFV was less than 40 cm/s throughout the basal arteries (arteries, 15.4%). These CBFV patterns as assessed by TCD are compared with the patients age, clinical symptoms and angiographical stagings (Suzuki’s criteria). The high-high pattern on TCD was predominantly seen in the younger patients and in the earlier stages of the disease. The high-low pattern was the most common pattern of CBFV as assessed by TCD in Moyamoya patients. The low-low pattern on TCD was more common in the later stages following angiographic evaluation by Suzuki’s criteria. The above patterns based on TCD findings show a good correlation with the age of the patient and the clinical diagnosis at the onset. TCD appears to be very useful in the evaluation of patients with Moyamoya disease. © 1997 Elsevier Science B.V. Keywords: Moyamoya disease; Cerebral blood flow velocity; Transcranial Doppler ultrasonography; Non-invasive monitoring; Cerebrovascular accident
1. Introduction
2. Materials and methods
In 1982, Aaslid introduced transcranial Doppler ultrasonography (TCD) [1]. The development of TCD has made possible the non-invasive evaluation of patients with cerebrovascular diseases. TCD provides a new advantage to study the human cerebral circulation and appears to be very useful in the evaluation of ongoing clinical conditions. Because of either limited cases or technical problem, there is little information regarding TCD findings on Moyamoya disease, the pathogenesis of which is unknown but regarded as a progressing disease [2,3]. The purpose of our present study is to show the usefulness of TCD in patients with Moyamoya disease.
We evaluated the cerebral blood flow velocities in 17 patients (12 women and five men, 6–58 years old, mean age 30.0 years old) using TCD. All patients were diagnosed as having Moyamoya disease from the conventional angiograms. None of the patients had been treated with surgical therapy for cerebral ischemia due to Moyamoya disease. In all patients, more than 3 months had passed since the onset to the evaluation by TCD. We used a 2 MHz range-gated pulsed Doppler instrument (TC-2 64B, EME, Germany). Bilateral basal arteries (34 arteries in 17 cases) were measured by the transtemporal approach. Serial TCD measurements were performed at a distance of every 5 mm from the internal carotid artery (ICA) to the distal part of the middle cerebral artery (MCA). At each insonation depth, peak systolic, end diastolic and mean flow veloc-
* Corresponding author. Fax: +81 886 318354. 0303-8467/97/$17.00 © 1997 Elsevier Science B.V. All rights reserved. PII S 0 3 0 3 - 8 4 6 7 ( 9 7 ) 0 0 0 6 6 - 8
K. Takase et al. / Clinical Neurology and Neurosurgery 99 Suppl. 2 (1997) S101–S105
S102
ities (cm/s) were measured. The patients were classified according to the cerebral blood flow velocity patterns of the TCD recordings. Additionally, the flow velocity patterns were compared with the patients age, clinical symptoms or diagnosis at the onset and angiographical stages by Suzuki’s criteria. In this study, 17 patients were divided into three groups according to their age; (1) less than 20 years old (four cases, eight arteries); (2) between 20 and 39 years old (nine cases, 18 arteries); and (3) over 40 years old (four cases, eight arteries). Their age and clinical diagnoses are summarized in Table 1.
termed this pattern as the TCD high-high flow velocity pattern. The second pattern, in which high flow velocities in the terminal portion of the ICA and/or proximal portion of the MCA decreased in the peripheral MCA was categorized as high-low pattern. The last group had low flow velocities less than 40 cm/s in mean value at all insonation points, so we termed this the low-low pattern. In this series based on TCD findings, seven of 28 arteries (25.0%) showed the high-high flow velocity pattern, 15 arteries (53.6%) were categorized as the high-low pattern and six arteries (21.4%) were the low-low pattern. Based on the TCD findings, the high-low flow velocity pattern was the most common pattern in patients with Moyamoya disease.
3. Results
3.1. Success rates for insonation on TCD A satisfactory readable record was obtained in 28 out of the 34 arteries (82.4%). In five arteries (14.7%), the Doppler spectra could be detected at a few insonation points. In one artery (2.9%), the Doppler spectra could not be detected at any insonation point. In this series, we evaluated the 28 arteries which were detected perfectly by TCD.
3.2. The cerebral blood flow 6elocity patterns as e6aluated by TCD (Fig. 1) The cerebral blood flow velocity (CBFV) patterns from TCD were divided into three groups. The first group was comprised of a TCD flow pattern with a continuous high flow velocity measuring more than 70 cm/s in mean value at all insonation points, so we
3.3. Correlations between TCD CBFV patterns and patient age (Table 2) In patients less than 20 years old, the flow velocities of five basal arteries were satisfactory recorded. Of the five basal arteries, two (40.0%) showed the high-high pattern and three (60.0%) were the high-low pattern. There were no patients with the low-low velocity pattern in the younger age group of patients. In patients who were 20–39 years of age, 15 basal arteries were examined completely on TCD. Of the 15 basal arteries, nine (60.0%) showed the high-low pattern, four (26.7%) were the high-high pattern and two (13.3%) were categorized as the low-low pattern. In the patients over 40 years old, eight basal arteries were detected totally. Of the eight basal arteries, four (50%) were classified into the low-low pattern, three (37.5%) were the high-low pattern and only one artery (12.5%) was the high-high flow TCD pattern.
3.4. Correlation between clinical symptoms at onset and TCD CBFV patterns (Table 3)
Table 1 Summary of patients Case No.
Age
Sex
Clinical diagnosis or symptoms
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17
29 6 35 33 30 12 49 45 30 27 44 30 15 21 33 58 13
F M F F F F F F F F M M F F F M M
R putaminal hematoma Headache Subarachnoid hemorrhage Intraventricular hemorrhage L cerebral infarction L cerebral infarction L cerebral infarction Subarachnoid hemorrhage R cerebral infarction Headache L cerebral infarction R cerebral infarction L cerebral infarction Subarachnoid hemorrhage Headache R cerebral infarction Headache
In patients with ischemic neurological symptoms at onset, 13 basal arteries were completely examined in the present TCD study. Of the 13 basal arteries, seven (53.8%) showed the high-low pattern, four (30.8%) were the low-low pattern and two (15.4%) were the high-high pattern. In the Moyamoya patients with hemorrhagic strokes, eight basal arteries were examined satisfactory. Five arteries (62.5%) had the highlow pattern, two (25.0%) were the low-low and one (12.5%) was the low-low pattern on TCD records. Whereas, in the patients without any episodes of strokes, four of seven (57.1%) basal arteries showed the high-high pattern and three (42.9%) were the highlow pattern. In the patients without strokes, there were no basal arteries with the low-low TCD flow velocity pattern.
K. Takase et al. / Clinical Neurology and Neurosurgery 99 Suppl. 2 (1997) S101–S105
S103
Fig. 1. (a) High-high pattern on TCD examination. In this case, the cerebral blood flow velocities (CBFV) were more than 70 cm/s in mean value at all points. (b) High-low pattern. CBFVs in the internal carotid artery and proximal middle cerebral artery (MCA) were elevated and were reduced in the peripheral MCA. (c) Low-low pattern. These patients showed low flow velocities throughout the whole of the basal artery (less than 40 cm/s in mean value).
3.5. Correlation between angiographical stages and TCD CBFV patterns (Table 4) According to the angiographic evaluations by Suzuki’s criteria, 19 arteries (67.8%) were classified as
the third stage, five (17.9%) were fourth stage, three (10.7%) were second stage and one (3.6%) was fifth stage. All of the three second stage arteries were detected as the high-high pattern on TCD monitoring. Of 19 Suzuki third stage arteries, the 12 (63.1%) showed
K. Takase et al. / Clinical Neurology and Neurosurgery 99 Suppl. 2 (1997) S100–S104
S104
Table 2 Correlation between CBFV patterns on TCD and patients’ age TCD patterns Age (years)
Total arteries (%)
B20 (%)
20–39 (%)
\40 (%)
High-high High-low Low-low
2 (40.0) 3 (60.0) 0 (0.0)
4 (26.7) 9 (60.0) 2 (13.3)
1 (12.5) 3 (37.5) 4 (50.0)
7 (25.0) 15 (53.6%) 6 (21.4)
Total
5 (17.8)
15 (53.6)
8 (28.6)
28 (100.0)
the high-low TCD pattern, four (21.1%) were the highhigh pattern and three arteries (15.8%) had the low-low flow velocity pattern. As to the fourth stage, three of the five arteries (60.0%) were the high-low and two (40.0%) were the low-low pattern. The one fifth Suzuki stage artery demonstrated the low-low pattern on CBFV study.
4. Discussion In this series, the Doppler spectra were detected perfectly in 28 out of 34 basal arteries (82.4%). According to recent reports, it is sometimes difficult to detect the Doppler waveforms on TCD in cases with Moyamoya disease [3]. In patients with Moyamoya disease, the terminal portion of the ICA and/or the proximal portion of the MCA show severe stenosis or occlusion. In this study, the clinical symptoms at the onset showed significant correlation with the TCD flow velocity patterns. In patients with episodes of strokes (ischemic and hemorrhagic strokes), the high-low TCD pattern was predominant, whereas in patients without strokes, the high-high pattern was the most common TCD-based pattern. Ischemic or hemorrhagic strokes may occur in patients with high-low or low-low TCD-based patterns. There was a clear-cut relationship between the angiographical stages and the TCD blood flow velocity patterns. Although blood supply to the affected hemisphere is very complicated and is derived from many different cerebral vessels, including embryonal, lenticulostriate and leptomeningeal branches, the blood flow velocity of the proximal MCA represents the progression of the disease and collateral flow.
Table 4 Correlation between angiographical staging and CBFV patterns on TCD Stage
CBFV patterns on TCD
Total arteries (%)
High-high (%)
High-low (%)
Low-low (%)
II III IV V
3 4 0 0
0 12 3 0
0 3 2 1
Total
7 (25.0)
(100.0) (21.1) (0.0) (0.0)
(0.0) (63.1) (60.0) (0.0)
15 (53.6)
(0.0) (15.8) (40.0) (100.0)
6 (21.4)
3 19 5 1
(10.7) (67.8) (17.9) (3.6)
28 (100.0)
In previous reports, Hashimoto and Nakamura reported that pulsatile flow velocities decreased in the MCAs of Moyamoya disease patients [4]. Fujimoto et al. found significantly reduced flow velocities in the MCA [2,5]. Weiller et al. and Steinke et al. observed low flow velocities in the distal M1 segment in vessels of Moyamoya disease patients [6,7]. On the contrary, Laborde et al. noted that vasoocclusion was identified by significant elevations of the contralateral blood flow velocities [8]. From the observation of these reports, there are apparent discrepancies concerning the blood flow velocities of the basal arteries in Moyamoya disease. In these reports, however, the authors have dealt with TCD tracings in only two or three insonated depths. Muttakin et al. reported in detail on eight cases with Moyamoya disease and noted that the flow velocities of the MCAs were lower than those of the ICAs [9]. In our study, serial Doppler tracing was carried out over the entire length of the basal arteries and the CBFV patterns based on the TCD examinations were classified into three types; the high-high, high-low and low-low flow velocity patterns. The high-high pattern, in which high blood flow velocities were recorded throughout the insonated basal arteries, was commonly seen in younger patients. The higher flow velocities in the ICA and MCA may correlate with the severity of the stenoses in these arteries. In early stage of Moyamoya disease, the stenotic rates in vessel diameter are relatively low and the collateralization is not considered of such great significance. In these cases, the flow velocities in the distal part of the MCAs would be
Table 3 Correlation between CBFV patterns on TCD and clinical diagnosis at onset TCD patterns High-high High-low Low-low Total
Ischemic strokes (%)
Hemorrhagic strokes (%)
Without strokes (%)
Total arteries (%)
2 (15.4) 7 (53.8) 4 (30.8)
1 (12.5) 5 (62.5) 2 (25.0)
4 (57.1) 3 (20.0) 0 (0.0)
7 (25.0) 15 (53.6) 6 (21.4)
13 (46.4)
8 (28.6)
7 (25.0)
28 (100.0)
K. Takase et al. / Clinical Neurology and Neurosurgery 99 Suppl. 2 (1997) S101–S105
maintained as high as those in the proximal MCAs and the MCA flow volume remains similar to that of normal individuals. Stenosis will cause an increase in velocity at the stenotic segment, if the same flow volume is preserved. In the later stages of Moyamoya disease, the hemodynamic resistance to blood flow through the severely stenotic lesions in the ICA and/or MCA becomes higher than in the early stage. In the patients with the high-low pattern of Doppler waveform, the velocity of blood flow through the MCA will increase in accordance with the increasing degree of narrowing until narrowing of the cortical artery occurs. The blood flow velocity increases at the site of stenosis and decreases on the distal side. It is also noted that a severe stenosis of the proximal MCA results in an abnormally low velocity with a dampened pulse wave in the distal MCA reflecting serious flow reduction. Severely stenotic lesions make the resistant vessels dilated, because the arteriolar and pial vessels in the affected area dilate due to the autoregulatory effort to maintain adequate blood flow. The blood flow velocities of basal arteries may reflect the proximal stenosis as well as the distal resistance of the vessels. Additionally, the collateral flow via the lenticulostriate, pial and/or meningeal arteries may cause a resistant stress to the blood flow in the peripheral MCAs. As a result, the blood flow velocity in the distal part of the MCAs become reduced. In the last stage of Moyamoya disease, the low-low TCD pattern shows severe reduction of the blood flow volume and collateralization becomes predominant. The flow pattern of the MCA in case of ICA occlusion is a decrease in velocity with a dampened pulse wave. In these cases, the blood flow velocity in the intracranial ICA become low on TCD records. Although the low-low TCD pattern will retain its low velocity all through the MCAs, in the final stage of Moyamoya disease, TCD shows no flow. The utility of TCD ultrasonography in the management of Moyamoya disease is not fully established, but this study demonstrates the possibility of providing a
.
.
S105
non-invasive detection of the disease, serial follow-up and staging. In the series of patients in this study, there are good correlations between patterns on TCD findings and the clinical symptoms at the onset. Additionally, the TCD patterns correspond well to the patient age distribution and angiographical stagings. It may be concluded that while cerebral angiograms are reflecting the morphological changes in the cerebral blood flow, the TCD tracing elicits flow in the affected hemispheres reflecting the stenosis of basal arteries and the blood supply from the collateral circulation. References [1] Aaslid R, Markwalder TM, Nornes H. Noninvasive transcranial Doppler ultrasound recording of flow velocity in basal cerebral arteries. J Neurosurg 1982;57:769 – 74. [2] Fujimoto T, Takahashi M, Mukouyama M, Fukushima Y. Criteria for diagnosis of Moyamoya disease by transcranial Doppler ultrasound. Neurosonology 1994;7:90 – 4. [3] Nishikawa M, Handa H. Doppler sonography. In: Kudo T, editor. The Spontaneous Occlusion of the Circle of Willis. Tokyo: Springer-Verlag, 1993:153 – 156. [4] Hashimoto T, Nakamura N. Transcranial assessment of intracranial aneurysm, arteriovenous malformation and Moyamoya disease. Jikei Med J 1991;38:369 – 75. [5] Fujimoto T, Fukushima Y. Evaluation of intracranial hemodynamics and vascular reactivity disease in Moyamoya by trnscranial Doppler ultrasound. In: Oka M, von Reutern GM, Furuhata H, Kodaira K, editors. Recent Advances in Neurosonology. International Congress Series, Excerpta Medica, 1992:83 – 86. [6] Weiller C, Muellges W, Leibold M, Rosenberger K, Biniek R, Ringelstein B. Infarctions and non-invasive diagnosis in Moyamoya disease: Two case reports. Neurosurg Rev 1991;14:75–77. [7] Steinke W, Tatemichi TK, Mohr JP, Massaro A, Prohovnik I, Solomon RA. Caudate hemorrhage with Moyamoya-like vasculopathy from atherosclerotic disease. Stroke 1992;23:1360–3. [8] Laborde G, Harders A, Klimek L, Hardenack M. Correlation between clinical, angiographic and transcranial Doppler sonographic findings in patients with Moyamoya disease. Neurol Res 1993;15:87 – 92. [9] Muttaqin Z, Ohba S, Arita K, Nakahara T, Pant B, Uozumi T, Kuwabara S, Oki S, Kurisu K, Yano T. Cerebral circulation in Moyamoya disease: A clinical study using transcranial Doppler sonography. Surg Neurol 1993;40:306 – 13.
Transcranial Doppler ultrasonography in patients with Moyamoya disease Kensaku Takase *, Michiharu Kashihara, Tsuneyo Hashimoto Department of Neurosurgery, Tokushima Prefectural Central Hospital, Tokushima 770, Japan
Abstract This study included 17 patients (five men and 12 women) with clinical diagnosis of Moyamoya disease from conventional angiograms. Bilateral basal arteries were measured by the transtemporal approach with a 2 MHz pulsed Doppler instrument (TC-2 64B EME). In 28 out of the 34 basal arteries (82.4%), reliable recordings were obtained. These transcranial Doppler ultrasonography (TCD) findings were classified into three patterns: (1) High-high pattern; the mean cerebral blood flow velocity (CBFV) was increased throughout the basal arteries by over 70 cm/s (seven arteries, 25.0%); (2) high-low pattern; the mean CBFV was fastest (over 70 cm/s) at the ICA or proximal MCA and the mean CBFV decreased remarkably distally (15 arteries, 53.6%); and (3) low-low pattern; the mean CBFV was less than 40 cm/s throughout the basal arteries (arteries, 15.4%). These CBFV patterns as assessed by TCD are compared with the patients age, clinical symptoms and angiographical stagings (Suzuki’s criteria). The high-high pattern on TCD was predominantly seen in the younger patients and in the earlier stages of the disease. The high-low pattern was the most common pattern of CBFV as assessed by TCD in Moyamoya patients. The low-low pattern on TCD was more common in the later stages following angiographic evaluation by Suzuki’s criteria. The above patterns based on TCD findings show a good correlation with the age of the patient and the clinical diagnosis at the onset. TCD appears to be very useful in the evaluation of patients with Moyamoya disease. © 1997 Elsevier Science B.V. Keywords: Moyamoya disease; Cerebral blood flow velocity; Transcranial Doppler ultrasonography; Non-invasive monitoring; Cerebrovascular accident
1. Introduction
2. Materials and methods
In 1982, Aaslid introduced transcranial Doppler ultrasonography (TCD) [1]. The development of TCD has made possible the non-invasive evaluation of patients with cerebrovascular diseases. TCD provides a new advantage to study the human cerebral circulation and appears to be very useful in the evaluation of ongoing clinical conditions. Because of either limited cases or technical problem, there is little information regarding TCD findings on Moyamoya disease, the pathogenesis of which is unknown but regarded as a progressing disease [2,3]. The purpose of our present study is to show the usefulness of TCD in patients with Moyamoya disease.
We evaluated the cerebral blood flow velocities in 17 patients (12 women and five men, 6–58 years old, mean age 30.0 years old) using TCD. All patients were diagnosed as having Moyamoya disease from the conventional angiograms. None of the patients had been treated with surgical therapy for cerebral ischemia due to Moyamoya disease. In all patients, more than 3 months had passed since the onset to the evaluation by TCD. We used a 2 MHz range-gated pulsed Doppler instrument (TC-2 64B, EME, Germany). Bilateral basal arteries (34 arteries in 17 cases) were measured by the transtemporal approach. Serial TCD measurements were performed at a distance of every 5 mm from the internal carotid artery (ICA) to the distal part of the middle cerebral artery (MCA). At each insonation depth, peak systolic, end diastolic and mean flow veloc-
* Corresponding author. Fax: +81 886 318354. 0303-8467/97/$17.00 © 1997 Elsevier Science B.V. All rights reserved. PII S 0 3 0 3 - 8 4 6 7 ( 9 7 ) 0 0 0 6 6 - 8
K. Takase et al. / Clinical Neurology and Neurosurgery 99 Suppl. 2 (1997) S101–S105
S102
ities (cm/s) were measured. The patients were classified according to the cerebral blood flow velocity patterns of the TCD recordings. Additionally, the flow velocity patterns were compared with the patients age, clinical symptoms or diagnosis at the onset and angiographical stages by Suzuki’s criteria. In this study, 17 patients were divided into three groups according to their age; (1) less than 20 years old (four cases, eight arteries); (2) between 20 and 39 years old (nine cases, 18 arteries); and (3) over 40 years old (four cases, eight arteries). Their age and clinical diagnoses are summarized in Table 1.
termed this pattern as the TCD high-high flow velocity pattern. The second pattern, in which high flow velocities in the terminal portion of the ICA and/or proximal portion of the MCA decreased in the peripheral MCA was categorized as high-low pattern. The last group had low flow velocities less than 40 cm/s in mean value at all insonation points, so we termed this the low-low pattern. In this series based on TCD findings, seven of 28 arteries (25.0%) showed the high-high flow velocity pattern, 15 arteries (53.6%) were categorized as the high-low pattern and six arteries (21.4%) were the low-low pattern. Based on the TCD findings, the high-low flow velocity pattern was the most common pattern in patients with Moyamoya disease.
3. Results
3.1. Success rates for insonation on TCD A satisfactory readable record was obtained in 28 out of the 34 arteries (82.4%). In five arteries (14.7%), the Doppler spectra could be detected at a few insonation points. In one artery (2.9%), the Doppler spectra could not be detected at any insonation point. In this series, we evaluated the 28 arteries which were detected perfectly by TCD.
3.2. The cerebral blood flow 6elocity patterns as e6aluated by TCD (Fig. 1) The cerebral blood flow velocity (CBFV) patterns from TCD were divided into three groups. The first group was comprised of a TCD flow pattern with a continuous high flow velocity measuring more than 70 cm/s in mean value at all insonation points, so we
3.3. Correlations between TCD CBFV patterns and patient age (Table 2) In patients less than 20 years old, the flow velocities of five basal arteries were satisfactory recorded. Of the five basal arteries, two (40.0%) showed the high-high pattern and three (60.0%) were the high-low pattern. There were no patients with the low-low velocity pattern in the younger age group of patients. In patients who were 20–39 years of age, 15 basal arteries were examined completely on TCD. Of the 15 basal arteries, nine (60.0%) showed the high-low pattern, four (26.7%) were the high-high pattern and two (13.3%) were categorized as the low-low pattern. In the patients over 40 years old, eight basal arteries were detected totally. Of the eight basal arteries, four (50%) were classified into the low-low pattern, three (37.5%) were the high-low pattern and only one artery (12.5%) was the high-high flow TCD pattern.
3.4. Correlation between clinical symptoms at onset and TCD CBFV patterns (Table 3)
Table 1 Summary of patients Case No.
Age
Sex
Clinical diagnosis or symptoms
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17
29 6 35 33 30 12 49 45 30 27 44 30 15 21 33 58 13
F M F F F F F F F F M M F F F M M
R putaminal hematoma Headache Subarachnoid hemorrhage Intraventricular hemorrhage L cerebral infarction L cerebral infarction L cerebral infarction Subarachnoid hemorrhage R cerebral infarction Headache L cerebral infarction R cerebral infarction L cerebral infarction Subarachnoid hemorrhage Headache R cerebral infarction Headache
In patients with ischemic neurological symptoms at onset, 13 basal arteries were completely examined in the present TCD study. Of the 13 basal arteries, seven (53.8%) showed the high-low pattern, four (30.8%) were the low-low pattern and two (15.4%) were the high-high pattern. In the Moyamoya patients with hemorrhagic strokes, eight basal arteries were examined satisfactory. Five arteries (62.5%) had the highlow pattern, two (25.0%) were the low-low and one (12.5%) was the low-low pattern on TCD records. Whereas, in the patients without any episodes of strokes, four of seven (57.1%) basal arteries showed the high-high pattern and three (42.9%) were the highlow pattern. In the patients without strokes, there were no basal arteries with the low-low TCD flow velocity pattern.
K. Takase et al. / Clinical Neurology and Neurosurgery 99 Suppl. 2 (1997) S101–S105
S103
Fig. 1. (a) High-high pattern on TCD examination. In this case, the cerebral blood flow velocities (CBFV) were more than 70 cm/s in mean value at all points. (b) High-low pattern. CBFVs in the internal carotid artery and proximal middle cerebral artery (MCA) were elevated and were reduced in the peripheral MCA. (c) Low-low pattern. These patients showed low flow velocities throughout the whole of the basal artery (less than 40 cm/s in mean value).
3.5. Correlation between angiographical stages and TCD CBFV patterns (Table 4) According to the angiographic evaluations by Suzuki’s criteria, 19 arteries (67.8%) were classified as
the third stage, five (17.9%) were fourth stage, three (10.7%) were second stage and one (3.6%) was fifth stage. All of the three second stage arteries were detected as the high-high pattern on TCD monitoring. Of 19 Suzuki third stage arteries, the 12 (63.1%) showed
K. Takase et al. / Clinical Neurology and Neurosurgery 99 Suppl. 2 (1997) S100–S104
S104
Table 2 Correlation between CBFV patterns on TCD and patients’ age TCD patterns Age (years)
Total arteries (%)
B20 (%)
20–39 (%)
\40 (%)
High-high High-low Low-low
2 (40.0) 3 (60.0) 0 (0.0)
4 (26.7) 9 (60.0) 2 (13.3)
1 (12.5) 3 (37.5) 4 (50.0)
7 (25.0) 15 (53.6%) 6 (21.4)
Total
5 (17.8)
15 (53.6)
8 (28.6)
28 (100.0)
the high-low TCD pattern, four (21.1%) were the highhigh pattern and three arteries (15.8%) had the low-low flow velocity pattern. As to the fourth stage, three of the five arteries (60.0%) were the high-low and two (40.0%) were the low-low pattern. The one fifth Suzuki stage artery demonstrated the low-low pattern on CBFV study.
4. Discussion In this series, the Doppler spectra were detected perfectly in 28 out of 34 basal arteries (82.4%). According to recent reports, it is sometimes difficult to detect the Doppler waveforms on TCD in cases with Moyamoya disease [3]. In patients with Moyamoya disease, the terminal portion of the ICA and/or the proximal portion of the MCA show severe stenosis or occlusion. In this study, the clinical symptoms at the onset showed significant correlation with the TCD flow velocity patterns. In patients with episodes of strokes (ischemic and hemorrhagic strokes), the high-low TCD pattern was predominant, whereas in patients without strokes, the high-high pattern was the most common TCD-based pattern. Ischemic or hemorrhagic strokes may occur in patients with high-low or low-low TCD-based patterns. There was a clear-cut relationship between the angiographical stages and the TCD blood flow velocity patterns. Although blood supply to the affected hemisphere is very complicated and is derived from many different cerebral vessels, including embryonal, lenticulostriate and leptomeningeal branches, the blood flow velocity of the proximal MCA represents the progression of the disease and collateral flow.
Table 4 Correlation between angiographical staging and CBFV patterns on TCD Stage
CBFV patterns on TCD
Total arteries (%)
High-high (%)
High-low (%)
Low-low (%)
II III IV V
3 4 0 0
0 12 3 0
0 3 2 1
Total
7 (25.0)
(100.0) (21.1) (0.0) (0.0)
(0.0) (63.1) (60.0) (0.0)
15 (53.6)
(0.0) (15.8) (40.0) (100.0)
6 (21.4)
3 19 5 1
(10.7) (67.8) (17.9) (3.6)
28 (100.0)
In previous reports, Hashimoto and Nakamura reported that pulsatile flow velocities decreased in the MCAs of Moyamoya disease patients [4]. Fujimoto et al. found significantly reduced flow velocities in the MCA [2,5]. Weiller et al. and Steinke et al. observed low flow velocities in the distal M1 segment in vessels of Moyamoya disease patients [6,7]. On the contrary, Laborde et al. noted that vasoocclusion was identified by significant elevations of the contralateral blood flow velocities [8]. From the observation of these reports, there are apparent discrepancies concerning the blood flow velocities of the basal arteries in Moyamoya disease. In these reports, however, the authors have dealt with TCD tracings in only two or three insonated depths. Muttakin et al. reported in detail on eight cases with Moyamoya disease and noted that the flow velocities of the MCAs were lower than those of the ICAs [9]. In our study, serial Doppler tracing was carried out over the entire length of the basal arteries and the CBFV patterns based on the TCD examinations were classified into three types; the high-high, high-low and low-low flow velocity patterns. The high-high pattern, in which high blood flow velocities were recorded throughout the insonated basal arteries, was commonly seen in younger patients. The higher flow velocities in the ICA and MCA may correlate with the severity of the stenoses in these arteries. In early stage of Moyamoya disease, the stenotic rates in vessel diameter are relatively low and the collateralization is not considered of such great significance. In these cases, the flow velocities in the distal part of the MCAs would be
Table 3 Correlation between CBFV patterns on TCD and clinical diagnosis at onset TCD patterns High-high High-low Low-low Total
Ischemic strokes (%)
Hemorrhagic strokes (%)
Without strokes (%)
Total arteries (%)
2 (15.4) 7 (53.8) 4 (30.8)
1 (12.5) 5 (62.5) 2 (25.0)
4 (57.1) 3 (20.0) 0 (0.0)
7 (25.0) 15 (53.6) 6 (21.4)
13 (46.4)
8 (28.6)
7 (25.0)
28 (100.0)
K. Takase et al. / Clinical Neurology and Neurosurgery 99 Suppl. 2 (1997) S101–S105
maintained as high as those in the proximal MCAs and the MCA flow volume remains similar to that of normal individuals. Stenosis will cause an increase in velocity at the stenotic segment, if the same flow volume is preserved. In the later stages of Moyamoya disease, the hemodynamic resistance to blood flow through the severely stenotic lesions in the ICA and/or MCA becomes higher than in the early stage. In the patients with the high-low pattern of Doppler waveform, the velocity of blood flow through the MCA will increase in accordance with the increasing degree of narrowing until narrowing of the cortical artery occurs. The blood flow velocity increases at the site of stenosis and decreases on the distal side. It is also noted that a severe stenosis of the proximal MCA results in an abnormally low velocity with a dampened pulse wave in the distal MCA reflecting serious flow reduction. Severely stenotic lesions make the resistant vessels dilated, because the arteriolar and pial vessels in the affected area dilate due to the autoregulatory effort to maintain adequate blood flow. The blood flow velocities of basal arteries may reflect the proximal stenosis as well as the distal resistance of the vessels. Additionally, the collateral flow via the lenticulostriate, pial and/or meningeal arteries may cause a resistant stress to the blood flow in the peripheral MCAs. As a result, the blood flow velocity in the distal part of the MCAs become reduced. In the last stage of Moyamoya disease, the low-low TCD pattern shows severe reduction of the blood flow volume and collateralization becomes predominant. The flow pattern of the MCA in case of ICA occlusion is a decrease in velocity with a dampened pulse wave. In these cases, the blood flow velocity in the intracranial ICA become low on TCD records. Although the low-low TCD pattern will retain its low velocity all through the MCAs, in the final stage of Moyamoya disease, TCD shows no flow. The utility of TCD ultrasonography in the management of Moyamoya disease is not fully established, but this study demonstrates the possibility of providing a
.
.
S105
non-invasive detection of the disease, serial follow-up and staging. In the series of patients in this study, there are good correlations between patterns on TCD findings and the clinical symptoms at the onset. Additionally, the TCD patterns correspond well to the patient age distribution and angiographical stagings. It may be concluded that while cerebral angiograms are reflecting the morphological changes in the cerebral blood flow, the TCD tracing elicits flow in the affected hemispheres reflecting the stenosis of basal arteries and the blood supply from the collateral circulation. References [1] Aaslid R, Markwalder TM, Nornes H. Noninvasive transcranial Doppler ultrasound recording of flow velocity in basal cerebral arteries. J Neurosurg 1982;57:769 – 74. [2] Fujimoto T, Takahashi M, Mukouyama M, Fukushima Y. Criteria for diagnosis of Moyamoya disease by transcranial Doppler ultrasound. Neurosonology 1994;7:90 – 4. [3] Nishikawa M, Handa H. Doppler sonography. In: Kudo T, editor. The Spontaneous Occlusion of the Circle of Willis. Tokyo: Springer-Verlag, 1993:153 – 156. [4] Hashimoto T, Nakamura N. Transcranial assessment of intracranial aneurysm, arteriovenous malformation and Moyamoya disease. Jikei Med J 1991;38:369 – 75. [5] Fujimoto T, Fukushima Y. Evaluation of intracranial hemodynamics and vascular reactivity disease in Moyamoya by trnscranial Doppler ultrasound. In: Oka M, von Reutern GM, Furuhata H, Kodaira K, editors. Recent Advances in Neurosonology. International Congress Series, Excerpta Medica, 1992:83 – 86. [6] Weiller C, Muellges W, Leibold M, Rosenberger K, Biniek R, Ringelstein B. Infarctions and non-invasive diagnosis in Moyamoya disease: Two case reports. Neurosurg Rev 1991;14:75–77. [7] Steinke W, Tatemichi TK, Mohr JP, Massaro A, Prohovnik I, Solomon RA. Caudate hemorrhage with Moyamoya-like vasculopathy from atherosclerotic disease. Stroke 1992;23:1360–3. [8] Laborde G, Harders A, Klimek L, Hardenack M. Correlation between clinical, angiographic and transcranial Doppler sonographic findings in patients with Moyamoya disease. Neurol Res 1993;15:87 – 92. [9] Muttaqin Z, Ohba S, Arita K, Nakahara T, Pant B, Uozumi T, Kuwabara S, Oki S, Kurisu K, Yano T. Cerebral circulation in Moyamoya disease: A clinical study using transcranial Doppler sonography. Surg Neurol 1993;40:306 – 13.