- Email: [email protected]
Contrast-Enhanced Color-Coded Doppler Sonography in Moyamoya Disease: A Retrospective Study
ARTICLE IN PRESS Ultrasound in Med. & Biol., Vol. ■■, No. ■■, pp. ■■–■■, 2018 Copyright © 2018 Published by Elsevier Inc. on behalf of World Federation for Ultrasound in Medicine & Biology. Printed in the USA. All rights reserved 0301-5629/$ - see front matter
https://doi.org/10.1016/j.ultrasmedbio.2018.01.002
●
Clinical Note
CONTRAST-ENHANCED COLOR-CODED DOPPLER SONOGRAPHY IN MOYAMOYA DISEASE: A RETROSPECTIVE STUDY Woo-Keun Seo,*,† Chang-Woon Choi,† Chi Kyung Kim,† and Kyungmi Oh† * Department of Neurology and Stroke Center, Samsung Medical Center, Sungkyunkwan University, Seoul, South Korea; and † Department of Neurology, College of Medicine, Korea University Guro Hospital, Korea University, Seoul, South Korea (Received 19 June 2017; revised 26 December 2017; in final form 8 January 2018)
Abstract—The purpose of this study was to validate the feasibility of contrast-enhanced transcranial Doppler sonography (CE-TCCD) in the diagnosis of Moyamoya disease (MMD). CE-TCCD data on patients with MMD were analyzed. The CE-TCCD data were classified qualitatively into four patterns by two independent investigators: normal vascular color Doppler signal (pattern 1), augmented color Doppler signal with identifiable vascular structure (pattern 2), confluent color Doppler signal filling more than two-thirds of the display frame without identifiable vascular structure (pattern 3) and confluent color Doppler signal filling full display (pattern 4). To investigate the validity, we compared the CE-TCCD data with traditional transcranial Doppler data and Suzuki grades on cerebral angiography. A total of 32 CE-TCCD studies from 16 MMD patients (male 37.5%, median age 48) were included in this study. The CE-TCCD findings were distributed across patterns 1 (n = 3), 2 (n = 12), 3 (n = 10) and 4 (n = 7) and were correlated with the Suzuki grades (p < 0.005) and hemodynamic parameters. Inter-rater reliability was promising (Cronbach α = 0.883). The CE-TCCD test provides distinctive patterns in MMD, according to their stage of progression. CE-TCCD patterns seem to be a reliable and valid means for the evaluation of MMD. (E-mail: [email protected]) © 2018 Published by Elsevier Inc. on behalf of World Federation for Ultrasound in Medicine & Biology. Key Words: Moyamoya disease, Transcranial doppler, Ultrasound contrast agent, Diagnosis.
and Stolz 2002). Ultrasound contrast agents (UCAs) composed of microbubbles have become an available tool for evaluating vascular systems (Cosgrove 2006; Dietrich et al. 2012). UCAs can intensify weak vascular signals and improve the visualization of intracranial arteries in patients with a poor temporal window through which conventional TCD cannot detect cerebral blood flow. Conventional TCD or TCCD can detect major basal cerebral arteries including the circle of Willis. However, in MMD patients, the middle cerebral artery or intracranial internal carotid artery disappears and is replaced by aberrant basal collateral vessels. We hypothesized that the signal of weak basal collateral vessels could be augmented by the use of UCAs in MMD patients. Therefore, based on the hypothesis that CE-TCCD can visualize basal collaterals in MMD, we studied CE-TCCD findings for patients with MMD.
INTRODUCTION Moyamoya disease (MMD) is a cerebrovascular condition characterized by progressive stenosis or occlusion of the terminal internal carotid or middle cerebral arteries with abnormal basal collaterals (Suzuki and Takaku 1969). The diagnosis of MMD is challenging because the only diagnostic gold standard test is invasive transfemoral catheterbased cerebral angiography (TFCA), which is not suitable for a screening test (Research Committee on the Pathology and Treatment of Spontaneous Occlusion of the Circle of Willis and Health Labour Sciences Research Grant for Research on Measures for Infractable Diseases 2012). Transcranial Doppler (TCD) or transcranial colorcoded Doppler sonography (TCCD) is a non-invasive test for the evaluation of intracranial cerebral vasculature (Zipper
Address correspondence to: Kyungmi Oh, Department of Neurology, Korea University College of Medicine, Korea University Guro Hospital, 80 Guro-dong, Guro-gu, Seoul, 08308, Republic of Korea. E-mail: okyunmi @korea.ac.kr Conflict of interest disclosure: W.K.S. received study funds (not related to this study) from Myung in Pharm. Co. LTD, Korea United Pharm. Inc. and Korea University grants (K1423491 and K1518381).
METHODS We identified patients who were diagnosed with MMD and underwent CE-TCCD at the Korea University Guro Hospital between January 2015 and January 2016. The 1
ARTICLE IN PRESS 2
Ultrasound in Medicine & Biology
diagnosis of MMD was made when TFCA revealed characteristics of steno-occlusive disease of the anterior circulation with basal collateral flow, after excluding other causes of steno-occlusive disease (Research Committee 2012). The study protocol was approved by the institutional review board of the Korea University Guro Hospital (KUGH 16195). Given the retrospective design of this study, the board allowed us to proceed with the analysis without obtaining informed consent from the patients. We rated the angiographic findings according to the Suzuki criteria, which classify the angiographic stage of MMD as narrowing of the carotid fork (stage I); initiation of the MMD (dilated major cerebral artery and a slight moyamoya vessel network, stage II); intensification of MMD (disappearance of the middle and anterior cerebral artery, and thick and distinct moyamoya vessels, stage III); minimization of the moyamoya (disappearance of the posterior cerebral artery and narrowing of individual moyamoya vessels, stage IV); reduction of the moyamoya (disappearance of all main cerebral arteries from the internal carotid artery system, further minimization of the moyamoya vessels and an increase in the collateral pathways from the external carotid artery system, stage V); and disappearance of the moyamoya (disappearance of the moyamoya vessels, with cerebral blood flow derived only from the external carotid artery and the vertebrobasilar artery systems, stage IV), and compared them with the CETCCD findings (Suzuki and Takaku 1969). Contrast-enhanced TCCD was performed on two machines (12 patients with ProSound Alpha7 premier [HitachiAloka Medical, Tokyo, Japan; using UST-52105, 1.25– 5 MHz phase array sector probe, display rate 13 Hz] and 4 patients with Affiniti 70 (Philips Ultrasound, Bothell, WA, USA; using S5-1 phased array sector probe]) using contrast harmonic imaging at a low mechanical index (1.1). Before starting the CE-TCCD examination, intravenous access was obtained through the right medial cubital vein with a 22-gauge intravenous catheter. After at least 5 min of rest in the supine position, transcranial Doppler sonographic examination was performed for the middle cerebral, anterior cerebral, posterior cerebral, vertebral and basilar arteries. For comparison with CE-TCCD patterns, we determined the following hemodynamic parameters of the middle cerebral arteries on both sides: mean flow velocity (MFV), peak systolic velocity (PSV), end-diastolic velocity (EDV), pulsatility index (PI) and resistance index (RI). As hemodynamic parameter values varied greatly in patients with MMD, the parameters on the side with the highest MFV were obtained as representative. The CE-TCCD images were obtained using a lowfrequency sector probe on a temporal window. Before the start of UCA (SonoVue, Bracco, Milan, Italy) infusion, color Doppler was turned on, and the color gain was set at minimum to avoid exaggerated vascular signals. Before
Volume ■■, Number ■■, 2018
UCA infusion, we checked the best visible temporal window. While the probe was manually maintained over the temporal window, 2 mL of activated UCA was infused in a bolus, followed by saline flushing. All these procedures were recorded using a movie recorder with which the machines were equipped. CE-TCCD findings were classified into four groups: normal vascular pattern (pattern 1: normal), augmented color Doppler signal with identifiable vascular structures (pattern 2: augmented), confluent color Doppler signal filling more than two-thirds of the display frame without identifiable vascular structures (pattern 3: confluent), confluent color Doppler signal filling the complete display (pattern 4: full) (Fig. 1). The CETCCD findings were classified by two independent vascular neurologists who were blind to the clinical information (W.K.S. and C.W.C.). Inter-rater reliability was tested using Cronbach’s α. To test the validity of CE-TCCD, the correlation with Suzuki grade was computed using Kendall’s rank correlation coefficient (τ). The association of CE-TCCD patterns with hemodynamic parameters, including PSV, EDV, MV, PI and RI, was tested using linear regression analyses. RESULTS During the study period, a total of 32 CE-TCCD studies (patients, 16; males, 37.5%; median age, 48 y; range, 19–64 y) were included (Table 1). The patients presented with several cerebrovascular diseases including ischemic stroke (n = 8), intracerebral hemorrhage (n = 1), transient ischemic attack (n = 4) and headache without focal neurologic symptoms (n = 2). Of the 16 patients, 6 (37.5%) were classified as “probable MMD” given the unilateral involvement. The distribution of Suzuki grades was as Table 1. Baseline characteristics of the patients (n = 16) Age Sex, male Hypertension Diabetes mellitus Smoking Dyslipidemia Total cholesterol, mg/dL Low-density lipoprotein cholesterol, mg/dL High-density lipoprotein cholesterol, mg/dL Triglycerides, mg/dL Presenting symptoms Ischemic stroke/transient ischemic attack Transient ischemic attack Border zone infarction Single subcortical infarction Diffuse multifocal infarction Posterior circulatory territorial infarction Hemorrhagic stroke (thalamic intracerebral hemorrhage) Headache No symptom/sign
46.06 ± 12.40 6 (37.5%) 6 (37.5%) 6 (37.5%) 0 (0%) 3 (18.9%) 174.64 ± 51.70 112.31 ± 51.23 47.13 ± 13.59 158.31 ± 98.92 4 (25%) 3 (18.8%) 1 (6.3%) 2 (12. 5) 1 (6.3%) 1 (6.3%) 3 (18.8%) 1 (6.3%)
ARTICLE IN PRESS CE-TCCD in diagnosis of moyomoya disease ● W.-K. Seo et al.
3
Fig. 1. Four patterns of contrast-enhanced color-coded Doppler sonographic images in Moyamoya disease. (A) Pattern 1: Normal pattern. (B) Pattern 2: Augmented color Doppler signal with identifiable vascular structure. (C) Pattern 3: Confluent color Doppler signal filling more than two-thirds of the display without identifiable vascular structures. (D) Pattern 4: Confluent color Doppler signal filling the complete display.
follows: grade 0 (n = 7), grade 1 (n = 4), grade 2 (n = 8), grade 3 (n = 4), grade 4 (n = 5) and grade 5 (n = 4) (counted separately for the right and left sides). MFV, PSV, EDV, PI and RI values were 67.69 ± 39.47 cm/s, 97.56 ± 58.59 cm/s, 47.91 ± 27.42 cm/s, 0.71 ± 0.16 and 0.49 ± 0.08, respectively. Among the 32 hemispheres from 16 patients, one artery exhibited an aliasing pattern before infusion of the UCA; therefore, the hemodynamic parameters for that artery were not obtained. Contrast-enhanced TCCD revealed the change from a small scattered signal into a complex flow signal filling the entire scan area in a mosaic pattern after infusion of the UCA (Supplementary Video S1, online only, available at https://doi.org/10.1016/j.ultrasmedbio.2018.01.002). CE-TCCD pattern 1 was observed in 3 hemispheres, pattern 2 in 12 hemispheres, pattern 3 in 10 hemispheres and pattern 4 in 7 hemispheres. PSV, MFV and EDV were decreased in CE-TCCD patterns 3 and 4, compared with pattern 1 (Table 2, Fig. 2). However, no significant associations were found for PI or RI. Inter-rater reliability ranged from acceptable to good (Cronbach’s α = 0.883). The CE-TCCD findings were significantly correlated with Suzuki grades (τ = −0.365,
p < 0.005). Figure 3 illustrates that pattern 1 is not associated with MMD stage, pattern 2 is associated with stages 0 and 1 and pattern 3 and 4 are associated with stages 2–5. DISCUSSION In this study, we determined the feasibility and reliability of a novel imaging modality called CE-TCCD for the diagnosis of MMD. CE-TCCD revealed four different patterns, which correlated well with the stages of MMD in the gold standard diagnostic test, in addition to consistency among independent reviewers. Considering that ultrasound is superior to magnetic resonance imaging or transfemoral catheter angiography (TFCA) in practicality, the discriminative value of CE-TCCD for tracking the progression of MMD adds to its clinical utility. The diagnosis of MMD continues to depend on cerebral angiography for the first description and early conceptualization of the disease (Research Committee on the Pathology and Treatment of Spontaneous Occlusion of the Circle of Willis and Health Labour Sciences Research Grant for Research on Measures for Infractable Diseases 2012). Recently, new neurovascular imaging modalities
ARTICLE IN PRESS 4
Ultrasound in Medicine & Biology
Volume ■■, Number ■■, 2018
Table 2. Association of contrast-enhanced transcranial doppler sonography patterns and hemodynamic parameters Pattern
Mean ± SD (cm/s)
Peak systolic velocity 1 144.5 ± 60.3 2 126.7 ± 69.1 3 74.0 ± 45.0 4 65.6 ± 21.2 End-diastolic velocity 1 72.3 ± 18.5 2 60.8 ± 33.2 3 36.5 ± 22.9 4 33.6 ± 6.0 Mean flow velocity 1 96.7 ± 32.1 2 87.7 ± 48.8 3 51.8 ± 30.4 4 46.6 ± 13.2 Pulsatility index 1 0.67 ± 0.15 2 0.72 ± 0.17 3 0.74 ± 0.16 4 0.66 ± 0.19 Resistance index 1 0.49 ± 0.13 2 0.50 ± 0.09 3 0.51 ± 0.06 4 0.47 ± 0.10
β
SE
p*
1 −74.287 −145.686 −140.383
52.94 53.35 53.91
0.178 0.014 0.018
1.000 −22.500 −56.471 −53.383
24.91 25.11 25.37
0.378 0.037 0.050
1.000 −35.250 −83.543 −80.433
37.34 37.64 38.03
0.358 0.040 0.049
1.000 −0.151 −0.114 −0.182
0.18 0.18 0.18
0.408 0.534 0.331
1.000 −0.096 −0.076 −0.115
0.09 0.09 0.09
0.279 0.395 0.207
SD = standard deviation; SE = standard error. * The tests were performed using linear regression analyses and results were compared with those for pattern 1.
Fig. 3. Distribution of the contrast-enhanced transcranial Doppler sonography patterns according to the Suzuki criteria. The number of tests of pattern 3 and pattern 4 was merged in Figure 3.
such as high-resolution vessel wall imaging (HR-VWI) have led to advancements in the diagnosis of MMD. Features such as small and concentric narrowing or occlusion of intracranial arteries constitute the distinguishing features of MMD (Kim et al. 2013; Ryoo et al. 2014). Ultrasonography also has been suggested as an alternative vascular imaging approach for MMD (Lee et al. 2004, 2016; Ruan et al. 2006). Symmetric high MFV and low PI on TCD without color Doppler image are characteristic features of MMD. However, these features are not
Fig. 2. Peak systolic velocity, mean velocity, pulsatility index and resistance index according to contrast-enhanced transcranial Doppler sonography patterns. *p < 0.05, **p < 0.01.
ARTICLE IN PRESS CE-TCCD in diagnosis of moyomoya disease ● W.-K. Seo et al.
specific to MMD and are observed in other conditions such as migraine, anemia and diffuse hyperperfusion (Lee et al. 2004). In terms of TCCD, basal collaterals in MMD may be displayed as scattered color Doppler signals with low velocity and resistance index (Ruan et al. 2006). The scattered color signals reflect the basal collateral because the signal is relatively fixed and the blood flow signal is detected by spectral Doppler analysis. CE-TCCD revealed a unique color Doppler signal pattern filling the display monitor. The blooming signal of the UCA filling in basal collateral vessels seems to produce a unique imaging pattern for MMD (Supplementary Video S1). The limitations of this study include small sample size and retrospective design. Therefore, cautious interpretation of the results is needed. Another limitation is that the classification of the patterns was based on visual inspection by two independent investigators and was not based on qualitative analyses using standard imaging technique. In addition, the power to discriminate MMD from other steno-occlusive diseases such as atherosclerosis should be validated by future studies. Despite these limitations, this study suggests that CE-TCCD could be considered an alternative for the screening and monitoring of cerebral vasculature in patients with MMD. SUPPLEMENTARY DATA Supplementary data related to this article can be found at https://doi.org/10.1016/j.ultrasmedbio.2018.01.002.
5
REFERENCES Cosgrove D. Ultrasound contrast agents: An overview. Eur J Radiol 2006; 60:324–330. Dietrich CF, Schreiber-Dietrich D, Hocke M. [Comments on the EFSUMB non-liver Guidelines 2011]. Praxis (Bern 1994) 2012;101:1175– 1181. Kim YJ, Lee DH, Kwon JY, Kang DW, Suh DC, Kim JS, Kwon SU. High resolution MRI difference between moyamoya disease and intracranial atherosclerosis. Eur J Neurol 2013;20:1311– 1318. Lee WJ, Jung KH, Lee KJ, Kim JM, Lee ST, Chu K, Lee SK, Roh JK. Sonographic findings associated with stenosis progression and vascular complications in moyamoya disease. J Neurosurg 2016;125: 689–697. Lee YS, Jung KH, Roh JK. Diagnosis of moyamoya disease with transcranial Doppler sonography: Correlation study with magnetic resonance angiography. J Neuroimaging 2004;14:319– 323. Research Committee on the Pathology and Treatment of Spontaneous Occlusion of the Circle of Willis and Health Labour Sciences Research Grant for Research on Measures for Infractable Diseases. Guidelines for diagnosis and treatment of moyamoya disease (spontaneous occlusion of the circle of Willis). Neurol Med Chir (Tokyo) 2012;52:245–266. Ruan LT, Duan YY, Cao TS, Zhuang L, Huang L. Color and power Doppler sonography of extracranial and intracranial arteries in Moyamoya disease. J Clin Ultrasound 2006;34:60– 69. Ryoo S, Cha J, Kim SJ, Choi JW, Ki CS, Kim KH, Jeon P, Kim JS, Hong SC, Bang OY. High-resolution magnetic resonance wall imaging findings of Moyamoya disease. Stroke 2014;45:2457– 2460. Suzuki J, Takaku A. Cerebrovascular “moyamoya” disease: Disease showing abnormal net-like vessels in base of brain. Arch Neurol 1969; 20:288–299. Zipper SG, Stolz E. Clinical application of transcranial colour-coded duplex sonography—A review. Eur J Neurol 2002;9:1–8.
https://doi.org/10.1016/j.ultrasmedbio.2018.01.002
●
Clinical Note
CONTRAST-ENHANCED COLOR-CODED DOPPLER SONOGRAPHY IN MOYAMOYA DISEASE: A RETROSPECTIVE STUDY Woo-Keun Seo,*,† Chang-Woon Choi,† Chi Kyung Kim,† and Kyungmi Oh† * Department of Neurology and Stroke Center, Samsung Medical Center, Sungkyunkwan University, Seoul, South Korea; and † Department of Neurology, College of Medicine, Korea University Guro Hospital, Korea University, Seoul, South Korea (Received 19 June 2017; revised 26 December 2017; in final form 8 January 2018)
Abstract—The purpose of this study was to validate the feasibility of contrast-enhanced transcranial Doppler sonography (CE-TCCD) in the diagnosis of Moyamoya disease (MMD). CE-TCCD data on patients with MMD were analyzed. The CE-TCCD data were classified qualitatively into four patterns by two independent investigators: normal vascular color Doppler signal (pattern 1), augmented color Doppler signal with identifiable vascular structure (pattern 2), confluent color Doppler signal filling more than two-thirds of the display frame without identifiable vascular structure (pattern 3) and confluent color Doppler signal filling full display (pattern 4). To investigate the validity, we compared the CE-TCCD data with traditional transcranial Doppler data and Suzuki grades on cerebral angiography. A total of 32 CE-TCCD studies from 16 MMD patients (male 37.5%, median age 48) were included in this study. The CE-TCCD findings were distributed across patterns 1 (n = 3), 2 (n = 12), 3 (n = 10) and 4 (n = 7) and were correlated with the Suzuki grades (p < 0.005) and hemodynamic parameters. Inter-rater reliability was promising (Cronbach α = 0.883). The CE-TCCD test provides distinctive patterns in MMD, according to their stage of progression. CE-TCCD patterns seem to be a reliable and valid means for the evaluation of MMD. (E-mail: [email protected]) © 2018 Published by Elsevier Inc. on behalf of World Federation for Ultrasound in Medicine & Biology. Key Words: Moyamoya disease, Transcranial doppler, Ultrasound contrast agent, Diagnosis.
and Stolz 2002). Ultrasound contrast agents (UCAs) composed of microbubbles have become an available tool for evaluating vascular systems (Cosgrove 2006; Dietrich et al. 2012). UCAs can intensify weak vascular signals and improve the visualization of intracranial arteries in patients with a poor temporal window through which conventional TCD cannot detect cerebral blood flow. Conventional TCD or TCCD can detect major basal cerebral arteries including the circle of Willis. However, in MMD patients, the middle cerebral artery or intracranial internal carotid artery disappears and is replaced by aberrant basal collateral vessels. We hypothesized that the signal of weak basal collateral vessels could be augmented by the use of UCAs in MMD patients. Therefore, based on the hypothesis that CE-TCCD can visualize basal collaterals in MMD, we studied CE-TCCD findings for patients with MMD.
INTRODUCTION Moyamoya disease (MMD) is a cerebrovascular condition characterized by progressive stenosis or occlusion of the terminal internal carotid or middle cerebral arteries with abnormal basal collaterals (Suzuki and Takaku 1969). The diagnosis of MMD is challenging because the only diagnostic gold standard test is invasive transfemoral catheterbased cerebral angiography (TFCA), which is not suitable for a screening test (Research Committee on the Pathology and Treatment of Spontaneous Occlusion of the Circle of Willis and Health Labour Sciences Research Grant for Research on Measures for Infractable Diseases 2012). Transcranial Doppler (TCD) or transcranial colorcoded Doppler sonography (TCCD) is a non-invasive test for the evaluation of intracranial cerebral vasculature (Zipper
Address correspondence to: Kyungmi Oh, Department of Neurology, Korea University College of Medicine, Korea University Guro Hospital, 80 Guro-dong, Guro-gu, Seoul, 08308, Republic of Korea. E-mail: okyunmi @korea.ac.kr Conflict of interest disclosure: W.K.S. received study funds (not related to this study) from Myung in Pharm. Co. LTD, Korea United Pharm. Inc. and Korea University grants (K1423491 and K1518381).
METHODS We identified patients who were diagnosed with MMD and underwent CE-TCCD at the Korea University Guro Hospital between January 2015 and January 2016. The 1
ARTICLE IN PRESS 2
Ultrasound in Medicine & Biology
diagnosis of MMD was made when TFCA revealed characteristics of steno-occlusive disease of the anterior circulation with basal collateral flow, after excluding other causes of steno-occlusive disease (Research Committee 2012). The study protocol was approved by the institutional review board of the Korea University Guro Hospital (KUGH 16195). Given the retrospective design of this study, the board allowed us to proceed with the analysis without obtaining informed consent from the patients. We rated the angiographic findings according to the Suzuki criteria, which classify the angiographic stage of MMD as narrowing of the carotid fork (stage I); initiation of the MMD (dilated major cerebral artery and a slight moyamoya vessel network, stage II); intensification of MMD (disappearance of the middle and anterior cerebral artery, and thick and distinct moyamoya vessels, stage III); minimization of the moyamoya (disappearance of the posterior cerebral artery and narrowing of individual moyamoya vessels, stage IV); reduction of the moyamoya (disappearance of all main cerebral arteries from the internal carotid artery system, further minimization of the moyamoya vessels and an increase in the collateral pathways from the external carotid artery system, stage V); and disappearance of the moyamoya (disappearance of the moyamoya vessels, with cerebral blood flow derived only from the external carotid artery and the vertebrobasilar artery systems, stage IV), and compared them with the CETCCD findings (Suzuki and Takaku 1969). Contrast-enhanced TCCD was performed on two machines (12 patients with ProSound Alpha7 premier [HitachiAloka Medical, Tokyo, Japan; using UST-52105, 1.25– 5 MHz phase array sector probe, display rate 13 Hz] and 4 patients with Affiniti 70 (Philips Ultrasound, Bothell, WA, USA; using S5-1 phased array sector probe]) using contrast harmonic imaging at a low mechanical index (1.1). Before starting the CE-TCCD examination, intravenous access was obtained through the right medial cubital vein with a 22-gauge intravenous catheter. After at least 5 min of rest in the supine position, transcranial Doppler sonographic examination was performed for the middle cerebral, anterior cerebral, posterior cerebral, vertebral and basilar arteries. For comparison with CE-TCCD patterns, we determined the following hemodynamic parameters of the middle cerebral arteries on both sides: mean flow velocity (MFV), peak systolic velocity (PSV), end-diastolic velocity (EDV), pulsatility index (PI) and resistance index (RI). As hemodynamic parameter values varied greatly in patients with MMD, the parameters on the side with the highest MFV were obtained as representative. The CE-TCCD images were obtained using a lowfrequency sector probe on a temporal window. Before the start of UCA (SonoVue, Bracco, Milan, Italy) infusion, color Doppler was turned on, and the color gain was set at minimum to avoid exaggerated vascular signals. Before
Volume ■■, Number ■■, 2018
UCA infusion, we checked the best visible temporal window. While the probe was manually maintained over the temporal window, 2 mL of activated UCA was infused in a bolus, followed by saline flushing. All these procedures were recorded using a movie recorder with which the machines were equipped. CE-TCCD findings were classified into four groups: normal vascular pattern (pattern 1: normal), augmented color Doppler signal with identifiable vascular structures (pattern 2: augmented), confluent color Doppler signal filling more than two-thirds of the display frame without identifiable vascular structures (pattern 3: confluent), confluent color Doppler signal filling the complete display (pattern 4: full) (Fig. 1). The CETCCD findings were classified by two independent vascular neurologists who were blind to the clinical information (W.K.S. and C.W.C.). Inter-rater reliability was tested using Cronbach’s α. To test the validity of CE-TCCD, the correlation with Suzuki grade was computed using Kendall’s rank correlation coefficient (τ). The association of CE-TCCD patterns with hemodynamic parameters, including PSV, EDV, MV, PI and RI, was tested using linear regression analyses. RESULTS During the study period, a total of 32 CE-TCCD studies (patients, 16; males, 37.5%; median age, 48 y; range, 19–64 y) were included (Table 1). The patients presented with several cerebrovascular diseases including ischemic stroke (n = 8), intracerebral hemorrhage (n = 1), transient ischemic attack (n = 4) and headache without focal neurologic symptoms (n = 2). Of the 16 patients, 6 (37.5%) were classified as “probable MMD” given the unilateral involvement. The distribution of Suzuki grades was as Table 1. Baseline characteristics of the patients (n = 16) Age Sex, male Hypertension Diabetes mellitus Smoking Dyslipidemia Total cholesterol, mg/dL Low-density lipoprotein cholesterol, mg/dL High-density lipoprotein cholesterol, mg/dL Triglycerides, mg/dL Presenting symptoms Ischemic stroke/transient ischemic attack Transient ischemic attack Border zone infarction Single subcortical infarction Diffuse multifocal infarction Posterior circulatory territorial infarction Hemorrhagic stroke (thalamic intracerebral hemorrhage) Headache No symptom/sign
46.06 ± 12.40 6 (37.5%) 6 (37.5%) 6 (37.5%) 0 (0%) 3 (18.9%) 174.64 ± 51.70 112.31 ± 51.23 47.13 ± 13.59 158.31 ± 98.92 4 (25%) 3 (18.8%) 1 (6.3%) 2 (12. 5) 1 (6.3%) 1 (6.3%) 3 (18.8%) 1 (6.3%)
ARTICLE IN PRESS CE-TCCD in diagnosis of moyomoya disease ● W.-K. Seo et al.
3
Fig. 1. Four patterns of contrast-enhanced color-coded Doppler sonographic images in Moyamoya disease. (A) Pattern 1: Normal pattern. (B) Pattern 2: Augmented color Doppler signal with identifiable vascular structure. (C) Pattern 3: Confluent color Doppler signal filling more than two-thirds of the display without identifiable vascular structures. (D) Pattern 4: Confluent color Doppler signal filling the complete display.
follows: grade 0 (n = 7), grade 1 (n = 4), grade 2 (n = 8), grade 3 (n = 4), grade 4 (n = 5) and grade 5 (n = 4) (counted separately for the right and left sides). MFV, PSV, EDV, PI and RI values were 67.69 ± 39.47 cm/s, 97.56 ± 58.59 cm/s, 47.91 ± 27.42 cm/s, 0.71 ± 0.16 and 0.49 ± 0.08, respectively. Among the 32 hemispheres from 16 patients, one artery exhibited an aliasing pattern before infusion of the UCA; therefore, the hemodynamic parameters for that artery were not obtained. Contrast-enhanced TCCD revealed the change from a small scattered signal into a complex flow signal filling the entire scan area in a mosaic pattern after infusion of the UCA (Supplementary Video S1, online only, available at https://doi.org/10.1016/j.ultrasmedbio.2018.01.002). CE-TCCD pattern 1 was observed in 3 hemispheres, pattern 2 in 12 hemispheres, pattern 3 in 10 hemispheres and pattern 4 in 7 hemispheres. PSV, MFV and EDV were decreased in CE-TCCD patterns 3 and 4, compared with pattern 1 (Table 2, Fig. 2). However, no significant associations were found for PI or RI. Inter-rater reliability ranged from acceptable to good (Cronbach’s α = 0.883). The CE-TCCD findings were significantly correlated with Suzuki grades (τ = −0.365,
p < 0.005). Figure 3 illustrates that pattern 1 is not associated with MMD stage, pattern 2 is associated with stages 0 and 1 and pattern 3 and 4 are associated with stages 2–5. DISCUSSION In this study, we determined the feasibility and reliability of a novel imaging modality called CE-TCCD for the diagnosis of MMD. CE-TCCD revealed four different patterns, which correlated well with the stages of MMD in the gold standard diagnostic test, in addition to consistency among independent reviewers. Considering that ultrasound is superior to magnetic resonance imaging or transfemoral catheter angiography (TFCA) in practicality, the discriminative value of CE-TCCD for tracking the progression of MMD adds to its clinical utility. The diagnosis of MMD continues to depend on cerebral angiography for the first description and early conceptualization of the disease (Research Committee on the Pathology and Treatment of Spontaneous Occlusion of the Circle of Willis and Health Labour Sciences Research Grant for Research on Measures for Infractable Diseases 2012). Recently, new neurovascular imaging modalities
ARTICLE IN PRESS 4
Ultrasound in Medicine & Biology
Volume ■■, Number ■■, 2018
Table 2. Association of contrast-enhanced transcranial doppler sonography patterns and hemodynamic parameters Pattern
Mean ± SD (cm/s)
Peak systolic velocity 1 144.5 ± 60.3 2 126.7 ± 69.1 3 74.0 ± 45.0 4 65.6 ± 21.2 End-diastolic velocity 1 72.3 ± 18.5 2 60.8 ± 33.2 3 36.5 ± 22.9 4 33.6 ± 6.0 Mean flow velocity 1 96.7 ± 32.1 2 87.7 ± 48.8 3 51.8 ± 30.4 4 46.6 ± 13.2 Pulsatility index 1 0.67 ± 0.15 2 0.72 ± 0.17 3 0.74 ± 0.16 4 0.66 ± 0.19 Resistance index 1 0.49 ± 0.13 2 0.50 ± 0.09 3 0.51 ± 0.06 4 0.47 ± 0.10
β
SE
p*
1 −74.287 −145.686 −140.383
52.94 53.35 53.91
0.178 0.014 0.018
1.000 −22.500 −56.471 −53.383
24.91 25.11 25.37
0.378 0.037 0.050
1.000 −35.250 −83.543 −80.433
37.34 37.64 38.03
0.358 0.040 0.049
1.000 −0.151 −0.114 −0.182
0.18 0.18 0.18
0.408 0.534 0.331
1.000 −0.096 −0.076 −0.115
0.09 0.09 0.09
0.279 0.395 0.207
SD = standard deviation; SE = standard error. * The tests were performed using linear regression analyses and results were compared with those for pattern 1.
Fig. 3. Distribution of the contrast-enhanced transcranial Doppler sonography patterns according to the Suzuki criteria. The number of tests of pattern 3 and pattern 4 was merged in Figure 3.
such as high-resolution vessel wall imaging (HR-VWI) have led to advancements in the diagnosis of MMD. Features such as small and concentric narrowing or occlusion of intracranial arteries constitute the distinguishing features of MMD (Kim et al. 2013; Ryoo et al. 2014). Ultrasonography also has been suggested as an alternative vascular imaging approach for MMD (Lee et al. 2004, 2016; Ruan et al. 2006). Symmetric high MFV and low PI on TCD without color Doppler image are characteristic features of MMD. However, these features are not
Fig. 2. Peak systolic velocity, mean velocity, pulsatility index and resistance index according to contrast-enhanced transcranial Doppler sonography patterns. *p < 0.05, **p < 0.01.
ARTICLE IN PRESS CE-TCCD in diagnosis of moyomoya disease ● W.-K. Seo et al.
specific to MMD and are observed in other conditions such as migraine, anemia and diffuse hyperperfusion (Lee et al. 2004). In terms of TCCD, basal collaterals in MMD may be displayed as scattered color Doppler signals with low velocity and resistance index (Ruan et al. 2006). The scattered color signals reflect the basal collateral because the signal is relatively fixed and the blood flow signal is detected by spectral Doppler analysis. CE-TCCD revealed a unique color Doppler signal pattern filling the display monitor. The blooming signal of the UCA filling in basal collateral vessels seems to produce a unique imaging pattern for MMD (Supplementary Video S1). The limitations of this study include small sample size and retrospective design. Therefore, cautious interpretation of the results is needed. Another limitation is that the classification of the patterns was based on visual inspection by two independent investigators and was not based on qualitative analyses using standard imaging technique. In addition, the power to discriminate MMD from other steno-occlusive diseases such as atherosclerosis should be validated by future studies. Despite these limitations, this study suggests that CE-TCCD could be considered an alternative for the screening and monitoring of cerebral vasculature in patients with MMD. SUPPLEMENTARY DATA Supplementary data related to this article can be found at https://doi.org/10.1016/j.ultrasmedbio.2018.01.002.
5
REFERENCES Cosgrove D. Ultrasound contrast agents: An overview. Eur J Radiol 2006; 60:324–330. Dietrich CF, Schreiber-Dietrich D, Hocke M. [Comments on the EFSUMB non-liver Guidelines 2011]. Praxis (Bern 1994) 2012;101:1175– 1181. Kim YJ, Lee DH, Kwon JY, Kang DW, Suh DC, Kim JS, Kwon SU. High resolution MRI difference between moyamoya disease and intracranial atherosclerosis. Eur J Neurol 2013;20:1311– 1318. Lee WJ, Jung KH, Lee KJ, Kim JM, Lee ST, Chu K, Lee SK, Roh JK. Sonographic findings associated with stenosis progression and vascular complications in moyamoya disease. J Neurosurg 2016;125: 689–697. Lee YS, Jung KH, Roh JK. Diagnosis of moyamoya disease with transcranial Doppler sonography: Correlation study with magnetic resonance angiography. J Neuroimaging 2004;14:319– 323. Research Committee on the Pathology and Treatment of Spontaneous Occlusion of the Circle of Willis and Health Labour Sciences Research Grant for Research on Measures for Infractable Diseases. Guidelines for diagnosis and treatment of moyamoya disease (spontaneous occlusion of the circle of Willis). Neurol Med Chir (Tokyo) 2012;52:245–266. Ruan LT, Duan YY, Cao TS, Zhuang L, Huang L. Color and power Doppler sonography of extracranial and intracranial arteries in Moyamoya disease. J Clin Ultrasound 2006;34:60– 69. Ryoo S, Cha J, Kim SJ, Choi JW, Ki CS, Kim KH, Jeon P, Kim JS, Hong SC, Bang OY. High-resolution magnetic resonance wall imaging findings of Moyamoya disease. Stroke 2014;45:2457– 2460. Suzuki J, Takaku A. Cerebrovascular “moyamoya” disease: Disease showing abnormal net-like vessels in base of brain. Arch Neurol 1969; 20:288–299. Zipper SG, Stolz E. Clinical application of transcranial colour-coded duplex sonography—A review. Eur J Neurol 2002;9:1–8.