Multiphase Arterial Spin Labeling Assessment of Cerebral Perfusion Changes Associated with Middle Cerebral Artery Stenosis Jun Chen, MD, Bin Zhao, MD, PhD, Min Bai, MD, Chunqing Bu, MD Rationale and Objectives: To assess the sensitivity and utility of multiphase pulsed arterial spin labeling (PASL) in detecting dynamic changes of cerebral perfusion in the presence of a middle cerebral artery (MCA) severe stenosis. Materials and Methods: Seventeen patients with severe stenoses in unilateral M1 segment of MCA were involved in this study. All patients underwent multiphase PASL imaging. Bilateral basal ganglia were drawn by hand as regions of interest (ROIs) on eight-phase images of each patient. The signal intensities of ROIs were measured and the time–intensity curves (TICs) were acquired through postprocessing on a magnetic resonance workstation. Whether there was a significant difference in the peak signal intensities of ROIs between the narrowed and normal sides was determined by the paired samples t test. Results: Three types of TICs were observed: eight cases with platform type, five cases with two-peak type, and four cases with singlepeak type. There was a significant difference in the peak signal intensities of ROIs between the narrowed and normal sides. Conclusions: Different types of TICs represent different cerebral hemodynamic changes. Multiphase PASL can sensitively detect the dynamic characteristics of cerebral perfusion and provide important dynamic perfusion information for clinical treatment of arterial stenosis. Key Words: Multiphase PASL; cerebral perfusion changes; middle cerebral artery stenosis. ªAUR, 2015
I
n recent years, cerebral perfusion imaging has gained much attention in ischemic cerebrovascular diseases. Perfusion imaging is highly sensitive in the detection of cerebral perfusion abnormalities. A mismatch between cerebral infarction volume and the volume of hypoperfusion may respond favorably to revascularization therapies (1). As compared with other perfusion techniques, such as positron emission tomography, single photon emission computed tomography (CT), and CT perfusion and dynamic susceptibility contrast (DSC) magnetic resonance imaging (MRI), arterial spin labeling (ASL) is a nonionizing and completely noninvasive technique for measuring cerebral perfusion. DSC perfusion imaging can provide more hemodynamic parameters than ASL imaging, such as cerebral blood volume, mean transit time, and time to peak, but ASL imaging
Acad Radiol 2015; 22:610–618 From the Shandong Medical Imaging Research Institute, Shandong University, Jingwu Rd No. 324, Jinan, Shandong Province 250021, China (J.C., B.Z.) and Department of Radiology, Liaocheng People’s Hospital, Liaocheng, Shandong Province, China (J.C., M.B., C.B.). Received July 22, 2014; accepted December 18, 2014. Address correspondence to: B.Z. e-mail: sdmuyxs@ sina.com ªAUR, 2015 http://dx.doi.org/10.1016/j.acra.2014.12.016
610
has its own unique advantages. A study (2) showed that ASL techniques could identify additional and complementary hemodynamic abnormalities in about half of the patients with normal DSC perfusion imaging findings. ASL has been proved to be a useful technique for evaluating cerebral perfusion. The role of ASL in clinical practice has increased steadily in recent years. Yun et al. (3) had quantitatively evaluated the hemodynamic changes after carotid artery stenting by measuring cerebral blood flow (CBF) using ASL. A study of Bivard et al. (4) showed that hyperperfusion of the initially ischemic area identified on ASL at 24 hours after stroke identifies patients with better tissue and clinical outcomes. Qiao et al. (5) found that ASL showed promise in the detection of perfusion abnormalities that correlated with clinically diagnosed transient ischemic attack in patients with otherwise normal neuroimaging results. ASL uses magnetically labeled water protons in arterial blood as the endogenous tracer. When the labeled tracer reaches the imaging slice, it gives raise to perfusion signal. Ischemic cerebrovascular diseases frequently manifest themselves first as disturbances in cerebral hemodynamics before cerebral infarction can be observed. We often found different cerebral ischemic strokes in the patients although the same degrees of stenosis in the same cerebral arteries
Academic Radiology, Vol 22, No 5, May 2015
MULTIPHASE ASL ASSESSMENT OF CEREBRAL PERFUSION CHANGES ASSOCIATED WITH MCA STENOSIS
Figure 1. Time-of-flight magnetic resonance angiographic (TOF MRA), T2weighted imaging, and diffusion-weighted imaging. TOF MRA and pulsed arterial spin labeling perfusion images from a 50year-old woman with a severe stenosis in the M1 segment of the left middle cerebral artery.
were detected in clinical practice, which demonstrated that there were different cerebral hemodynamic changes in the territories supplied by the narrowed arteries. In this study, multiphase pulsed ASL (PASL) technique was used to detect these hemodynamic changes. The goal of this study was to assess the sensitivity and utility of multiphase PASL in detecting dynamic changes of cerebral perfusion in the presence of a middle cerebral artery (MCA) severe stenosis.
MATERIALS AND METHODS This study was approved by the institutional review board at our institution. Written informed consent was obtained from all patients. Study Population
Seventeen patients (6 males and 11 females, aged 45–74 years with a mean age of 56 years) with cerebral ischemic symptoms in our institution were included in this study. Severe stenoses in unilateral M1 segment of MCA were confirmed in all patients by three-dimensional (3D) time-of-flight MR angiography (TOF MRA). The degrees of arterial stenosis were measured according to a study of Samuels et al. (6).
MRI Protocol
All MR examinations were performed on a 3.0-T MR system (Achieva; Philips Medical Systems, The Netherlands) using a 16-channel phased-array neurovascular coil. All patients underwent multisegment head and neck 3D TOF MRA with the following parameters. Intracranial segment: fast field echo (FFE), repetition time (TR)/echo time (TE) = 20 ms/3.5 ms, number of slices = 160, NSA = 2, field of view (FOV) = 220 220 mm2; cervical segment: FFE, TR/TE = 20 ms/3.5 ms, number of slices = 180, NSA = 2, FOV = 160 131 mm2; and thoracic segment: FFE, TR/TE = 22 ms/3.5 ms, number of slices = 100, NSA = 2, FOV = 250 238 mm2. Maximum intensity projection (MIP) and mobiview were processed on an MR workstation to obtain MRA images. Multiphase PASL images were acquired using a STAR (signal targeting by alternating radiofrequency) labeling technique with the following parameters: six slices positioned parallel to the anterior or posterior commissure line and comprising bilateral basal ganglia, single-shot FFE planar imaging, TR = 250 milliseconds, TE = 16 milliseconds, flip angle = 40 , matrix size = 64 64, voxel size = 3.48 3.48 mm2, FOV = 240 240 mm2, slice thickness = 6 mm, sensitivity encoding factor = 2.3, label thickness = 130 mm, staring delay time = 300 milliseconds, 611
CHEN ET AL
Academic Radiology, Vol 22, No 5, May 2015
Figure 2. Multiphase pulsed arterial spin labeling (PASL) images and time–intensity curve of right region of interest. Time-of-flight magnetic resonance angiographic and PASL perfusion images from a 50-year-old woman with a severe stenosis in the M1 segment of the left middle cerebral artery.
and interval of delay times = 250 milliseconds. Images were acquired at eight delay times, and the longest delay time was 2209 milliseconds. The labeling slab was chosen to be 20 mm below the imaging regions, and the labeling thickness was 130 mm. Two acquisitions were carried out: the labeled acquisition and control acquisition without arterial labeling. Thirty repetitions were acquired. The eight-phase PASL perfusion images were obtained by averaging the subtraction of the control and labeled acquisitions on a postprocessing MR workstation (Extended MRWorkstation; Philips Medical Systems). 612
Image Evaluation
On eight-phase PASL images, the signals of the territory supplied by normal MCA arose and then decreased quickly, and the signals of the area supplied by MCA with severe stenosis in M1 segment were significantly low. Bilateral basal ganglia were hand drawn as regions of interest (ROIs) on eight-phase images of each patient by an experienced radiologist. The signal intensities of ROIs were measured and the time–intensity curves (TICs) were acquired through postprocessing on an MR workstation (Extended MR
Academic Radiology, Vol 22, No 5, May 2015
MULTIPHASE ASL ASSESSMENT OF CEREBRAL PERFUSION CHANGES ASSOCIATED WITH MCA STENOSIS
Figure 3. Multiphase pulsed arterial spin labeling (PASL) images and time–intensity curve of left region of interest. Time-of-flight magnetic resonance angiographic and PASL perfusion images from a 50-year-old woman with a severe stenosis in the M1 segment of the left middle cerebral artery.
Workstation; Philips Medical Systems). X-axis of TIC represented the delay time, and y-axis represented the signal intensity of ROIs (7).
paired samples t test. A P value 95% (11). In recent years, the ease of implementation have made PASL a frequent tool for neuroimaging (12,13). PASL sequence was used for cerebral perfusion imaging in this study. Signal-to-noise ratio (SNR) of PASL is inherently low, which can be influenced by many factors. The duration
Academic Radiology, Vol 22, No 5, May 2015
MULTIPHASE ASL ASSESSMENT OF CEREBRAL PERFUSION CHANGES ASSOCIATED WITH MCA STENOSIS
Figure 5. Multiphase pulsed arterial spin labeling (PASL) images and time–intensity curve of right region of interest. Time-of-flight magnetic resonance angiographic and PASL perfusion images from a 45-year-old woman with a severe stenosis in the M1 segment of the right middle cerebral artery.
for the labeled tracer to travel from the labeling region to the imaging slices is known as arterial transit time (ATT). ASL is sensitive to ATT, which can either lead to overestimation of CBF because of bright intravascular labels or underestimation of CBF because of delayed arrival of labels (14). Longitudinal relaxation (T1) is another important factor. The labeled tracer decays with T1 (15). T1 of labeled tracer is long on 3-T MR, so labeled tracer can survive long enough to get into the imaging slices and high SNR can be obtained (16). It is advantageous to apply PASL at high-field MR system. The time
between labeling and image acquisition, called delay time, is a key point for minimizing the influence of ATT and T1. A short delay time results in incomplete delivery of the labeled tracer to the imaging slices, whereas a long delay time results in increased T1 decay and low SNR. In this study, the starting delay time of 300 milliseconds and eight delay times were used to detect the dynamic perfusion signal changes in basal ganglia. A limited number of patients were included in this study because the patients with severe stenoses in unilateral M1
615
CHEN ET AL
Academic Radiology, Vol 22, No 5, May 2015
Figure 6. Time-of-flight magnetic resonance angiographic (TOF MRA), T2-weighted imaging, and diffusionweighted imaging. TOF MRA and arterial spin labeling perfusion images from a 59year-old woman with a severe stenosis in the M1 segment of the right middle cerebral artery.
segment of MCA and normal other head and neck arteries were rare. There was another limitation that the perfusion were not quantified in absolute units in this study. However, the relative perfusion between the narrowed and normal sides in each patient was sufficient to identify hemodynamic changes associated with cerebral stenoses. High temporal resolution is needed to detect cerebral dynamic perfusion characteristics. Methods for improving temporal resolution include single-shot ASL (17) because the center of k space is acquired during each acquisition. Multiphase PASL imaging has the ability to improve temporal resolution using single-shot technique, which can accurately acquire cerebral perfusion signal changes. For the patients with cerebral arterial stenoses, PASL methods with a single delay time will obviously underestimate cerebral perfusion because of the delayed arrival of labels in the narrowed side. Multiphase PASL has the ability to detect the delayed perfusion signal changes regardless of the degree of stenosis. Three types of TIC were shown in this study. The most common type was platform type which meant the prolonged travel time of labeled tracer in ROI. The two-peak type meant partially delayed perfusion. The single-peak type with a reduced peak intensity was often associated with hypoperfusion in ROI. Perfusion deficit in the left basal ganglia could be seen on multiphase PASL images of case 1. This area will 616
derive little benefit from clinical treatment because no perfusion signal was seen. Delayed perfusion in the posterior part of basal ganglia could be seen on multiphase PASL images of cases 2 and 3, and subacute infarct was seen in the corresponding area. The delayed perfusion area will probably be salvaged by thrombolytic or endovascular recanalization therapy, so the patients with delayed perfusion may receive a benefit from the therapy of arterial stenosis. These cases demonstrated that the patients with the same degree of stenosis could show large variability in stroke risk and in the pattern of ischemic stroke. We also found that there was a significant difference in the peak signal intensities between the narrowed and normal sides. Hypoperfusion was detected in narrowed side although there was no obvious infarct. The quantitative measurement of the peak signal intensities on multiphase PASL images had high sensitivity in detecting cerebral abnormal perfusion. When no obvious infarct in the basal ganglia in the narrowed side was detected by T2WI and DWI, the dynamic signal changes on multiphase PASL images could provide complementary information to MRA, which was useful for identifying patients at high stroke risk or conversely those patients with adequate perfusion (eg, from collateral vessels which may not be visible on MRA) despite having a stenosis. In conclusion, different types of TIC represent different cerebral hemodynamic properties. Multiphase PASL can
Academic Radiology, Vol 22, No 5, May 2015
MULTIPHASE ASL ASSESSMENT OF CEREBRAL PERFUSION CHANGES ASSOCIATED WITH MCA STENOSIS
Figure 7. Multiphase pulsed arterial spin labeling images and time–intensity curve of right region of interest. Time-of-flight magnetic resonance angiographic and arterial spin labeling perfusion images from a 59-year-old woman with a severe stenosis in the M1 segment of the right middle cerebral artery.
sensitively identify important dynamic perfusion information of the patients with severe arterial stenoses. REFERENCES 1. Harris AD, Coutts SB, Frayne R. Diffusion and perfusion MR imaging of acute ischemic stroke. Magn Reson Imaging Clin N Am 2009; 17: 291–313. 2. Zaharchuk G, Bammer R, Straka M, et al. Arterial spin-label imaging in patients with normal bolus perfusion-weighted MR imaging findings: pilot identification of the borderzone sign. Radiology 2009; 252: 797–807.
3. Yun TJ, Sohn CH, Han MH, et al. Effect of carotid artery stenting on cerebral blood flow: evaluation of hemodynamic changes using arterial spin labeling. Neuroradiology 2013; 55:271–281. 4. Bivard A, Stanwell P, Levi C, et al. Arterial spin labeling identifies tissue salvage and good clinical recovery after acute ischemic stroke. Journal of Neuroimaging 2013; 23:391–396. 5. Qiao XJ, Salamon N, Wang DJ, et al. Perfusion deficits detected by arterial spin-labeling in patients with TIA with negative diffusion and vascular imaging. Am J Neuroradiol 2013; 34:2125–2130. 6. Samuels OB, Joseph GJ, Lynn MJ. A standardized method for measuring intracranial arterial stenosis. Am J Neuroradiol 2000; 21:643–646. 7. Fonda C, Ciccarone A, Mortilla M, et al. 3T arterial spin labeling (ASL) in pediatric patients: preliminary results. 6th congress and exhibition of the joint societies of paediatric radiology. 27–31 May 2011, London.
617
CHEN ET AL
8. Williams DS, Detre JA, Leigh JS, et al. Magnetic resonance imaging of perfusion using spin inversion of arterial water. Proc Natl Acad Sci USA 1992; 89:212–216. €nther M, Lichy MP, et al. Comparison of arterial spin-labeling 9. Weber MA, Gu techniques and dynamic susceptibility-weighted contrast-enhanced MRI in perfusion imaging of normal brain tissue. Invest Radiol 2003; 38:712–718. 10. Detre JA, Alsop DC. Perfusion magnetic resonance imaging with continuous arterial spin labeling: methods and clinical applications in the central nervous system. Eur J Radiol 1999; 30:115–124. 11. Wong EC. Quantifying CBF with pulsed ASL: technical and pulsed sequence factors. J Magn Reson Imaging 2005; 22:727–731. 12. Petcharunpaisan S, Ramalho J, Castillo M. Arterial spin labeling in neuroimaging. World J Radiol 2010; 10:384–398.
618
Academic Radiology, Vol 22, No 5, May 2015
13. Gregori J, Schuff N, Kern R, et al. T2-based arterial spin labeling measurements of blood to tissue water transfer in human brain. J Magn Reson Imaging 2013; 37:332–342. 14. Golay X, Hendrikse J, Lim TC. Perfusion imaging using arterial spin labeling. Top Magn Reson Imaging 2004; 15:10–27. 15. Detre JA, Wang J, Wang Z, et al. Arterial spin-labeled perfusion MRI in basic and clinical neuroscience. Curr Opin Neurol 2009; 22:348–355. 16. Wolf RL, Detre JA. Clinical neuroimaging using arterial spin labeled perfusion magnetic resonance imaging. Neurotherapeutics 2007; 4: 346–359. 17. Petcharunpaisan S, Ramalho J, Castillo M. Arterial spin labeling in neuroimaging. World J Radiol 2010; 2:384–398.
I
n recent years, cerebral perfusion imaging has gained much attention in ischemic cerebrovascular diseases. Perfusion imaging is highly sensitive in the detection of cerebral perfusion abnormalities. A mismatch between cerebral infarction volume and the volume of hypoperfusion may respond favorably to revascularization therapies (1). As compared with other perfusion techniques, such as positron emission tomography, single photon emission computed tomography (CT), and CT perfusion and dynamic susceptibility contrast (DSC) magnetic resonance imaging (MRI), arterial spin labeling (ASL) is a nonionizing and completely noninvasive technique for measuring cerebral perfusion. DSC perfusion imaging can provide more hemodynamic parameters than ASL imaging, such as cerebral blood volume, mean transit time, and time to peak, but ASL imaging
Acad Radiol 2015; 22:610–618 From the Shandong Medical Imaging Research Institute, Shandong University, Jingwu Rd No. 324, Jinan, Shandong Province 250021, China (J.C., B.Z.) and Department of Radiology, Liaocheng People’s Hospital, Liaocheng, Shandong Province, China (J.C., M.B., C.B.). Received July 22, 2014; accepted December 18, 2014. Address correspondence to: B.Z. e-mail: sdmuyxs@ sina.com ªAUR, 2015 http://dx.doi.org/10.1016/j.acra.2014.12.016
610
has its own unique advantages. A study (2) showed that ASL techniques could identify additional and complementary hemodynamic abnormalities in about half of the patients with normal DSC perfusion imaging findings. ASL has been proved to be a useful technique for evaluating cerebral perfusion. The role of ASL in clinical practice has increased steadily in recent years. Yun et al. (3) had quantitatively evaluated the hemodynamic changes after carotid artery stenting by measuring cerebral blood flow (CBF) using ASL. A study of Bivard et al. (4) showed that hyperperfusion of the initially ischemic area identified on ASL at 24 hours after stroke identifies patients with better tissue and clinical outcomes. Qiao et al. (5) found that ASL showed promise in the detection of perfusion abnormalities that correlated with clinically diagnosed transient ischemic attack in patients with otherwise normal neuroimaging results. ASL uses magnetically labeled water protons in arterial blood as the endogenous tracer. When the labeled tracer reaches the imaging slice, it gives raise to perfusion signal. Ischemic cerebrovascular diseases frequently manifest themselves first as disturbances in cerebral hemodynamics before cerebral infarction can be observed. We often found different cerebral ischemic strokes in the patients although the same degrees of stenosis in the same cerebral arteries
Academic Radiology, Vol 22, No 5, May 2015
MULTIPHASE ASL ASSESSMENT OF CEREBRAL PERFUSION CHANGES ASSOCIATED WITH MCA STENOSIS
Figure 1. Time-of-flight magnetic resonance angiographic (TOF MRA), T2weighted imaging, and diffusion-weighted imaging. TOF MRA and pulsed arterial spin labeling perfusion images from a 50year-old woman with a severe stenosis in the M1 segment of the left middle cerebral artery.
were detected in clinical practice, which demonstrated that there were different cerebral hemodynamic changes in the territories supplied by the narrowed arteries. In this study, multiphase pulsed ASL (PASL) technique was used to detect these hemodynamic changes. The goal of this study was to assess the sensitivity and utility of multiphase PASL in detecting dynamic changes of cerebral perfusion in the presence of a middle cerebral artery (MCA) severe stenosis.
MATERIALS AND METHODS This study was approved by the institutional review board at our institution. Written informed consent was obtained from all patients. Study Population
Seventeen patients (6 males and 11 females, aged 45–74 years with a mean age of 56 years) with cerebral ischemic symptoms in our institution were included in this study. Severe stenoses in unilateral M1 segment of MCA were confirmed in all patients by three-dimensional (3D) time-of-flight MR angiography (TOF MRA). The degrees of arterial stenosis were measured according to a study of Samuels et al. (6).
MRI Protocol
All MR examinations were performed on a 3.0-T MR system (Achieva; Philips Medical Systems, The Netherlands) using a 16-channel phased-array neurovascular coil. All patients underwent multisegment head and neck 3D TOF MRA with the following parameters. Intracranial segment: fast field echo (FFE), repetition time (TR)/echo time (TE) = 20 ms/3.5 ms, number of slices = 160, NSA = 2, field of view (FOV) = 220 220 mm2; cervical segment: FFE, TR/TE = 20 ms/3.5 ms, number of slices = 180, NSA = 2, FOV = 160 131 mm2; and thoracic segment: FFE, TR/TE = 22 ms/3.5 ms, number of slices = 100, NSA = 2, FOV = 250 238 mm2. Maximum intensity projection (MIP) and mobiview were processed on an MR workstation to obtain MRA images. Multiphase PASL images were acquired using a STAR (signal targeting by alternating radiofrequency) labeling technique with the following parameters: six slices positioned parallel to the anterior or posterior commissure line and comprising bilateral basal ganglia, single-shot FFE planar imaging, TR = 250 milliseconds, TE = 16 milliseconds, flip angle = 40 , matrix size = 64 64, voxel size = 3.48 3.48 mm2, FOV = 240 240 mm2, slice thickness = 6 mm, sensitivity encoding factor = 2.3, label thickness = 130 mm, staring delay time = 300 milliseconds, 611
CHEN ET AL
Academic Radiology, Vol 22, No 5, May 2015
Figure 2. Multiphase pulsed arterial spin labeling (PASL) images and time–intensity curve of right region of interest. Time-of-flight magnetic resonance angiographic and PASL perfusion images from a 50-year-old woman with a severe stenosis in the M1 segment of the left middle cerebral artery.
and interval of delay times = 250 milliseconds. Images were acquired at eight delay times, and the longest delay time was 2209 milliseconds. The labeling slab was chosen to be 20 mm below the imaging regions, and the labeling thickness was 130 mm. Two acquisitions were carried out: the labeled acquisition and control acquisition without arterial labeling. Thirty repetitions were acquired. The eight-phase PASL perfusion images were obtained by averaging the subtraction of the control and labeled acquisitions on a postprocessing MR workstation (Extended MRWorkstation; Philips Medical Systems). 612
Image Evaluation
On eight-phase PASL images, the signals of the territory supplied by normal MCA arose and then decreased quickly, and the signals of the area supplied by MCA with severe stenosis in M1 segment were significantly low. Bilateral basal ganglia were hand drawn as regions of interest (ROIs) on eight-phase images of each patient by an experienced radiologist. The signal intensities of ROIs were measured and the time–intensity curves (TICs) were acquired through postprocessing on an MR workstation (Extended MR
Academic Radiology, Vol 22, No 5, May 2015
MULTIPHASE ASL ASSESSMENT OF CEREBRAL PERFUSION CHANGES ASSOCIATED WITH MCA STENOSIS
Figure 3. Multiphase pulsed arterial spin labeling (PASL) images and time–intensity curve of left region of interest. Time-of-flight magnetic resonance angiographic and PASL perfusion images from a 50-year-old woman with a severe stenosis in the M1 segment of the left middle cerebral artery.
Workstation; Philips Medical Systems). X-axis of TIC represented the delay time, and y-axis represented the signal intensity of ROIs (7).
paired samples t test. A P value 95% (11). In recent years, the ease of implementation have made PASL a frequent tool for neuroimaging (12,13). PASL sequence was used for cerebral perfusion imaging in this study. Signal-to-noise ratio (SNR) of PASL is inherently low, which can be influenced by many factors. The duration
Academic Radiology, Vol 22, No 5, May 2015
MULTIPHASE ASL ASSESSMENT OF CEREBRAL PERFUSION CHANGES ASSOCIATED WITH MCA STENOSIS
Figure 5. Multiphase pulsed arterial spin labeling (PASL) images and time–intensity curve of right region of interest. Time-of-flight magnetic resonance angiographic and PASL perfusion images from a 45-year-old woman with a severe stenosis in the M1 segment of the right middle cerebral artery.
for the labeled tracer to travel from the labeling region to the imaging slices is known as arterial transit time (ATT). ASL is sensitive to ATT, which can either lead to overestimation of CBF because of bright intravascular labels or underestimation of CBF because of delayed arrival of labels (14). Longitudinal relaxation (T1) is another important factor. The labeled tracer decays with T1 (15). T1 of labeled tracer is long on 3-T MR, so labeled tracer can survive long enough to get into the imaging slices and high SNR can be obtained (16). It is advantageous to apply PASL at high-field MR system. The time
between labeling and image acquisition, called delay time, is a key point for minimizing the influence of ATT and T1. A short delay time results in incomplete delivery of the labeled tracer to the imaging slices, whereas a long delay time results in increased T1 decay and low SNR. In this study, the starting delay time of 300 milliseconds and eight delay times were used to detect the dynamic perfusion signal changes in basal ganglia. A limited number of patients were included in this study because the patients with severe stenoses in unilateral M1
615
CHEN ET AL
Academic Radiology, Vol 22, No 5, May 2015
Figure 6. Time-of-flight magnetic resonance angiographic (TOF MRA), T2-weighted imaging, and diffusionweighted imaging. TOF MRA and arterial spin labeling perfusion images from a 59year-old woman with a severe stenosis in the M1 segment of the right middle cerebral artery.
segment of MCA and normal other head and neck arteries were rare. There was another limitation that the perfusion were not quantified in absolute units in this study. However, the relative perfusion between the narrowed and normal sides in each patient was sufficient to identify hemodynamic changes associated with cerebral stenoses. High temporal resolution is needed to detect cerebral dynamic perfusion characteristics. Methods for improving temporal resolution include single-shot ASL (17) because the center of k space is acquired during each acquisition. Multiphase PASL imaging has the ability to improve temporal resolution using single-shot technique, which can accurately acquire cerebral perfusion signal changes. For the patients with cerebral arterial stenoses, PASL methods with a single delay time will obviously underestimate cerebral perfusion because of the delayed arrival of labels in the narrowed side. Multiphase PASL has the ability to detect the delayed perfusion signal changes regardless of the degree of stenosis. Three types of TIC were shown in this study. The most common type was platform type which meant the prolonged travel time of labeled tracer in ROI. The two-peak type meant partially delayed perfusion. The single-peak type with a reduced peak intensity was often associated with hypoperfusion in ROI. Perfusion deficit in the left basal ganglia could be seen on multiphase PASL images of case 1. This area will 616
derive little benefit from clinical treatment because no perfusion signal was seen. Delayed perfusion in the posterior part of basal ganglia could be seen on multiphase PASL images of cases 2 and 3, and subacute infarct was seen in the corresponding area. The delayed perfusion area will probably be salvaged by thrombolytic or endovascular recanalization therapy, so the patients with delayed perfusion may receive a benefit from the therapy of arterial stenosis. These cases demonstrated that the patients with the same degree of stenosis could show large variability in stroke risk and in the pattern of ischemic stroke. We also found that there was a significant difference in the peak signal intensities between the narrowed and normal sides. Hypoperfusion was detected in narrowed side although there was no obvious infarct. The quantitative measurement of the peak signal intensities on multiphase PASL images had high sensitivity in detecting cerebral abnormal perfusion. When no obvious infarct in the basal ganglia in the narrowed side was detected by T2WI and DWI, the dynamic signal changes on multiphase PASL images could provide complementary information to MRA, which was useful for identifying patients at high stroke risk or conversely those patients with adequate perfusion (eg, from collateral vessels which may not be visible on MRA) despite having a stenosis. In conclusion, different types of TIC represent different cerebral hemodynamic properties. Multiphase PASL can
Academic Radiology, Vol 22, No 5, May 2015
MULTIPHASE ASL ASSESSMENT OF CEREBRAL PERFUSION CHANGES ASSOCIATED WITH MCA STENOSIS
Figure 7. Multiphase pulsed arterial spin labeling images and time–intensity curve of right region of interest. Time-of-flight magnetic resonance angiographic and arterial spin labeling perfusion images from a 59-year-old woman with a severe stenosis in the M1 segment of the right middle cerebral artery.
sensitively identify important dynamic perfusion information of the patients with severe arterial stenoses. REFERENCES 1. Harris AD, Coutts SB, Frayne R. Diffusion and perfusion MR imaging of acute ischemic stroke. Magn Reson Imaging Clin N Am 2009; 17: 291–313. 2. Zaharchuk G, Bammer R, Straka M, et al. Arterial spin-label imaging in patients with normal bolus perfusion-weighted MR imaging findings: pilot identification of the borderzone sign. Radiology 2009; 252: 797–807.
3. Yun TJ, Sohn CH, Han MH, et al. Effect of carotid artery stenting on cerebral blood flow: evaluation of hemodynamic changes using arterial spin labeling. Neuroradiology 2013; 55:271–281. 4. Bivard A, Stanwell P, Levi C, et al. Arterial spin labeling identifies tissue salvage and good clinical recovery after acute ischemic stroke. Journal of Neuroimaging 2013; 23:391–396. 5. Qiao XJ, Salamon N, Wang DJ, et al. Perfusion deficits detected by arterial spin-labeling in patients with TIA with negative diffusion and vascular imaging. Am J Neuroradiol 2013; 34:2125–2130. 6. Samuels OB, Joseph GJ, Lynn MJ. A standardized method for measuring intracranial arterial stenosis. Am J Neuroradiol 2000; 21:643–646. 7. Fonda C, Ciccarone A, Mortilla M, et al. 3T arterial spin labeling (ASL) in pediatric patients: preliminary results. 6th congress and exhibition of the joint societies of paediatric radiology. 27–31 May 2011, London.
617
CHEN ET AL
8. Williams DS, Detre JA, Leigh JS, et al. Magnetic resonance imaging of perfusion using spin inversion of arterial water. Proc Natl Acad Sci USA 1992; 89:212–216. €nther M, Lichy MP, et al. Comparison of arterial spin-labeling 9. Weber MA, Gu techniques and dynamic susceptibility-weighted contrast-enhanced MRI in perfusion imaging of normal brain tissue. Invest Radiol 2003; 38:712–718. 10. Detre JA, Alsop DC. Perfusion magnetic resonance imaging with continuous arterial spin labeling: methods and clinical applications in the central nervous system. Eur J Radiol 1999; 30:115–124. 11. Wong EC. Quantifying CBF with pulsed ASL: technical and pulsed sequence factors. J Magn Reson Imaging 2005; 22:727–731. 12. Petcharunpaisan S, Ramalho J, Castillo M. Arterial spin labeling in neuroimaging. World J Radiol 2010; 10:384–398.
618
Academic Radiology, Vol 22, No 5, May 2015
13. Gregori J, Schuff N, Kern R, et al. T2-based arterial spin labeling measurements of blood to tissue water transfer in human brain. J Magn Reson Imaging 2013; 37:332–342. 14. Golay X, Hendrikse J, Lim TC. Perfusion imaging using arterial spin labeling. Top Magn Reson Imaging 2004; 15:10–27. 15. Detre JA, Wang J, Wang Z, et al. Arterial spin-labeled perfusion MRI in basic and clinical neuroscience. Curr Opin Neurol 2009; 22:348–355. 16. Wolf RL, Detre JA. Clinical neuroimaging using arterial spin labeled perfusion magnetic resonance imaging. Neurotherapeutics 2007; 4: 346–359. 17. Petcharunpaisan S, Ramalho J, Castillo M. Arterial spin labeling in neuroimaging. World J Radiol 2010; 2:384–398.
