- Email: [email protected]
Loss of labelling efficiency caused by carotid stent in pseudocontinuous arterial spin labelling perfusion study
Clinical Radiology 71 (2016) e21ee27
Contents lists available at ScienceDirect
Clinical Radiology journal homepage: www.clinicalradiologyonline.net
Loss of labelling efficiency caused by carotid stent in pseudocontinuous arterial spin labelling perfusion study D.Y.-T. Chen a, Y.-S. Kuo a, H.-L. Hsu a, F.-X. Yan a, H.-L. Liu b, C.-J. Chen a, Y.-C. Tseng a, * a
Department of Radiology and Brain and Consciousness Research Center, Taipei Medical University Shuang-Ho Hospital, New Taipei City, Taiwan b Department of Imaging Physics, The University of Texas M. D. Anderson Cancer Center, Houston, TX, USA
art icl e i nformat ion Article history: Received 21 December 2014 Received in revised form 8 September 2015 Accepted 1 October 2015
AIM: To elucidate the cause of cerebral hypoperfusion on the stent placement side after carotid artery stent placement (CAS) measured by pseudocontinuous arterial spin labelling (PCASL) perfusion imaging. MATERIALS AND METHODS: Consecutive patients with symptomatic internal carotid artery stenosis receiving CAS were included in the study. Cerebral blood flow (CBF) was measured by PCASL perfusion imaging at 3 T magnetic resonance imaging (MRI) the day before and 3 days after the procedure. Changes in cerebral haemodynamics after CAS were assessed. RESULTS: Twenty-two patients were included; 17 patients had increased or stationary CBF after CAS and five patients had significantly reduced CBF on the stenting side after CAS whereas CBF increased on the contralateral side. High stent position was noticed in the five patients. After labelling plane adjustment to avoid labelling on the stent, no more cerebral hypoperfusion was noticed. CONCLUSION: When using PCASL perfusion imaging to monitor post-stenting CBF, the stent may cause an artefact that leads to a low CBF in the territory of the stented vessel. Routinely adding a fast T2 star gradient-echo echo-planar-imaging covering the upper neck region before PCASL perfusion imaging to identify the stent position and avoid the stent-related artefact is recommended. Ó 2015 The Royal College of Radiologists. Published by Elsevier Ltd. All rights reserved.
Introduction Carotid artery stenting (CAS) is being increasingly used as an alternative treatment to carotid endarterectomy for * Guarantor and correspondent: Y.-C. Tseng, Department of Radiology, Shuang-Ho Hospital, No. 291, Zhongzheng Road, Zhonghe District, New Taipei City 23561, Taiwan. Tel.: þ886 2 22490088 1320; fax: þ886 2 22490088 1322. E-mail address: [email protected] (Y.-C. Tseng).
carotid artery stenosis.1 Alteration of cerebral perfusion after CAS has been investigated in many perfusion studies.2e7 Most of them reported improvement in cerebral perfusion on the stenting side after CAS, especially in patients with high-grade stenosis, with greater perfusion deficit prior to CAS, or with symptoms.4e6 Pseudocontinuous arterial spin labelling (PCASL) is a non-invasive technique that detects absolute cerebral blood flow (CBF) without the use of an exogenous contrast agent.8,9 These properties permit repeated measurement of
http://dx.doi.org/10.1016/j.crad.2015.10.004 0009-9260/Ó 2015 The Royal College of Radiologists. Published by Elsevier Ltd. All rights reserved.
e22
D.Y.-T. Chen et al. / Clinical Radiology 71 (2016) e21ee27
CBF in a short interval, which is useful in patients receiving CAS.3,7 In Shuang-Ho Hospital PCASL perfusion imaging was used to monitor the cerebral haemodynamic changes before and after CAS. Most patients had increased or stationary CBF after CAS; however, occasionally, patients had obviously decreased CBF on the stenting side after CAS. The aim of the present study was to elucidate the causes of the post-stenting hypoperfusion measured by ASL.
Materials and methods Patients This study was approved by the local ethics committee. From July 2012 to October 2013, consecutive patients with symptomatic internal carotid artery (ICA) stenosis (ICA stenosis >70% by North American Symptomatic Carotid Endarterectomy Trial method on conventional cerebral angiogram10) receiving CAS in Shuang-Ho Hospital were included in the study. To evaluate the haemodynamic changes after CAS, whole-brain CBF was measured using the PCASL technique before and after the procedure. Patients were excluded when severe dental metal artefact was noted on the PCASL perfusion imaging.
CAS Patients were pretreated with aspirin (325 mg/d) and clopidogrel (75 mg/d, Plavix; Bristol-Myers Squibb/Sanofi Pharmaceuticals, New York, NY, USA) for at least 3 days before CAS. Systemic anticoagulation was then given by intravenous administration of a bolus of heparin (70 U/kg) and then continuous heparin infusion (15 U/kg/h) to maintain an activated clotting time of more than 250 seconds. The self-expanding stent (Precise; Cordis, Miami Lakes, FL, USA) was deployed across the lesion, and a semicompliant angioplasty balloon was used to post-dilate the stent to achieve >90% luminal diameter. The postprocedural angiography was performed 20 minutes after stent deployment to evaluate both the stented site and the distal cerebral vasculature. After the procedure, antihypertensive agents were used to maintain systolic pressures of <130 mmHg.
Imaging acquisition All patients received brain magnetic resonance imaging (MRI) using a 3 T MRI system (Discovery MR750, General Electric Medical Systems, Milwaukee, WI, USA) with diffusion-weighted imaging (DWI), time-of-flight (TOF)magnetic resonance angiography (MRA) of the brain, and PCASL perfusion imaging on the day before and 3 days after CAS. The protocol applied on each scan was as follows1: DWI: single-shot spin-echo echo-planar imaging with repetition time (TR)/echo time (TE)¼8000 ms/68 ms, flip angle¼90 , b¼1000 s/mm2, diffusion direction¼all, field of view (FOV)¼230230 mm2, in-plane matrix¼512512, section thickness¼5 mm, intersection gap¼2 mm.2 TOFMRA: three-slab three-dimensional (3D) sequence with
TR/TE¼30 ms/2.9 ms, flip angle¼20o, FOV¼200200 mm2, in-plane matrix¼416256, section thickness¼1.4 mm, superior to inferior acquisition.3 PCASL perfusion imaging: 3D background suppressed fast-spin-echo stack-of-spiral readout module with eight in-plane spiral interleaves, TR/ TE¼5327 ms/10.5 ms, labelling duration¼1.5 seconds, postlabelling delay¼2525 ms, no flow-crushing gradients, inplane matrix¼128128, number of excitations (NEX)¼4, section thickness¼4 mm, echo train length¼36 to obtain 36 consecutive axial sections, labelling plane¼10-mm thick and placed 2 cm inferior to the lower edge of the cerebellum, total scan time¼336 seconds.
Data analysis The CBF maps were generated on an Advantage Windows workstation using Functool software (version 9.4, GE Medical Systems). Quantification of CBF was calculated with the following equation9:
PLD ST 1 e T1t eT1b PW CBF ¼ 6000,l LT SFPW PD 2εT1b 1 e T1b where T1 of the blood (T1b) was assumed to be 1.6 seconds at 3 T, T1 of the tissue (T1t) 1.2 seconds, partition coefficient (l) 0.9, labelling efficiency (ε) 0.6, saturation time of PD (ST) 2 seconds, labelling duration (LT) 1.5 seconds, and postlabelling delay (PLD) 2525 ms. PW was the perfusionweighted or the raw-difference image; PD was the signal intensity of the proton density image and SFPW was an empirical scaling factor (¼ 32) used to increase the dynamic range of the PW. To compare changes in CBF before and after CAS, two radiologists with 4 and 15 years of experience chose the regions of interest (ROIs) on two sections, the level of the basal ganglia and the level of the body of the lateral ventricle. On each section, the raters manually outlined the ROIs corresponding to the cortical flow territory of the middle cerebral artery (MCA) of both hemispheres according to the maps by Damasio,11 taking care to exclude regions of prior cerebral infarction. The absolute CBF (expressed in millilitres per 100 g of tissue per minute) of the MCA territories in each hemisphere were obtained by averaging the mean values within the ROIs of the two sections. Relative CBF was defined as stenting side CBF/contralateral side CBF.
Callback study In patients showing post-stenting cerebral hypoperfusion, callback studies were performed to confirm whether cerebral hypoperfusion persisted in PCASL perfusion imaging. In the callback study, the CBF was measured twice using two different labelling planes. The CBF was first measured by PCASL perfusion imaging using the routine labelling plane. On the second measurement, T2* echoplanar imaging (gradient-echo echo-planar-imaging sequence, TR/TE¼925 ms/20 ms, flip angle¼70o, FOV¼250250 mm2, in-plane matrix¼132128, section thickness¼5 mm, intersection gap¼1 mm, NEX¼4, scanning
D.Y.-T. Chen et al. / Clinical Radiology 71 (2016) e21ee27
time¼26 seconds) was performed that covered the cervical region to identify the extent of the stent-related susceptibility artefact. The labelling plane was then placed 1 cm above the upper margin of the area with stent-related susceptibility artefact to measure the post-adjustment CBF.
Results Patient characteristics Twenty-three patients received CAS in Shuang-Ho Hospital during the study period. One patient was excluded due to severe dental metal artefact in the PCASL perfusion imaging. Seventeen patients had increased or stationary CBF after CAS and five patients had obviously reduced CBF after CAS on the stenting side. The mean age of the patients with increased or stationary post-stenting CBF was 70.7 years (9.9 years; range, 56e89 years) and the mean preprocedural ICA stenosis rate on the stenting side was 84% (10%; range, 70e99%). The mean age of the patients with post-stenting hypoperfusion was 73.6 years (5.72 years; range, 64e78 years) and the mean pre-procedural ICA stenosis rate on the stenting side was 84.4% (12.9%; range, 70e99%). No significant difference was noted on both groups regarding age, gender, stenting side and stenosis grade (p¼0.21, 0.64, 0.52, 0.49 respectively). The five patients with post-stent hypoperfusion were included in the following analysis.
CBF measurement The pre-stenting and post-stenting (Day 3) CBF values of the five patients are listed in Table 1. The five patients with decreased perfusion on the stenting side after CAS had increased CBF on the contralateral side after CAS. On the stenting side, there were decreases of 20.2%, 20.1%, 43.2%, 27.2%, and 34% in the post-stenting CBF compared to the pre-stenting CBF, respectively. On the contralateral side, there were increases of 22.8%, 26.6%, 6.5%, 11.8%, and 2.1% in the post-stenting CBF, respectively. The relative CBF was 0.90, 1.08, 1.14, 0.83, and 0.81 in the pre-stenting scan and 0.58, 0.68, 0.61, 0.54, and 0.53 in the post-stenting scan for the five patients, respectively.
e23
Acute carotid stent thrombosis The post-procedural angiography 20 minutes after stent deployment showed good recanalization of the lumen without acute thrombosis in these five patients. TOF-MRA on Day 3 post-stent placement also showed normal or increased flow signal in the ICA and MCA on the stenting side. There were few punctate infarcts at the posterior watershed area on stenting side in Cases 1, 3, and 5, and no new infarction in Cases 2 and 4 in the post-stenting DWI.
Carotid stent position The carotid stent position was high in the five patients in the post-procedural angiography. The upper margins of the stents were all above the level of the C2 body. Conversely, the upper margins of the stents were below the C2 body in the other 17 patients who had increased or stationary CBF after CAS.
Callback study The duration between the callback examination and the CAS procedures were 6 months, 4 months, 3 months, 3 days and 3 days for Cases 1, 2, 3, 4, and 5, respectively. The preadjustment and post-adjustment CBF of the five patients are listed in Table 1. The post-adjustment CBF increased significantly from the pre-adjustment CBF on the stenting side. There were 21.2%, 45.9%, 48.8%, 88.3%, and 88.7% increases in the post-adjustment CBF compared to the preadjustment CBF in the five patients, respectively. The CBF remained relatively stationary (<10% changes) on the contralateral side before and after labelling plane adjustment in the five patients. The post-adjustment relative CBF was 1.13, 0.92, 1.17, 1.30, and 0.94 in the five patients, respectively. When we compared the post-adjustment CBF with the pre-stenting CBF, the CBF actually increased on both sides. There were 50.9%, 1.0%, 13.1%, 77.7%, and 25.8% increases on the stenting side and 20.7%, 18.1%, 9.5%, 13.6%, and 8.5% increases on the contralateral side in the five patients, respectively. The serial CBF maps (Fig 1), the pre- and post-stenting angiography images (Fig 2), and images of the labelling
Table 1 Cerebral blood flow (CBF) of the patients with post-stent placement hypoperfusion. Case (no.)
1 2 3 4 5
Pre-stenting CBF
Post-stent D3 CBF
Stent
Stent
35.2 47.8 59.5 37.2 42.3
Contra 39.1 44.3 52.4 44.8 52.1
28.1 38.2 33.8 27.1 27.9
Contra a
(e20.2%) (e20.1%)a (e43.2%)a (e27.2%)a (e34.0%)a
48.0 56.1 55.8 50.1 53.2
a
(þ22.8%) (þ26.6%)a (þ6.5%)a (11.8%)a (þ2.1%)a
Callback pre-adjust CBF
Callback post-adjust CBF
Stent
Stent
43.8 33.1 45.2 35.1 28.2
Contra 52.1 54.3 57.4 50.3 53.8
53.1 48.3 67.3 66.1 53.2
Contra b
(þ21.2%) (þ45.9%)b (þ48.8%)b (þ88.3%)b (þ88.7%)b
47.2 52.3 57.4 50.9 56.5
(e9.4%)b (e3.7%)b (0%)b (þ1.2%)b (þ5.0%)b
Data are mean values of CBF (in ml/100 g of tissue/min). Stent, stent placement side; Contra, the side contralateral to stent placement. a Numbers in parentheses represent Day 3 post-stent placement to pre-stent placement CBF change, defined as [(post-stent placement Day 3 CBF e pre-stent placement CBF)/pre-stent placement CBF]100%. b Numbers in the parentheses represent post-adjust to pre-adjust CBF change, defined as [(post-adjust CBF e pre-adjust CBF)/pre-adjust CBF]100%.
e24
D.Y.-T. Chen et al. / Clinical Radiology 71 (2016) e21ee27
Figure 1 CBF changes in Case 1. A 64-year-old man presented with intermittent left side weakness. (a) The pre-stent CBF map showed mild decreased perfusion in the right hemisphere especially at the right posterior watershed area. (b) The CBF map Day 3 post-stent placement, showed decreased perfusion on the stenting side and increased perfusion on the contralateral side. (c) The callback pre-adjustment CBF showed persistent hypoperfusion on the stenting side. (d) The callback post-adjustment CBF showed symmetrically increased perfusion on both sides.
plane and the stent position (Fig 3) of a representative case are presented to illustrate the condition.
Discussion Cerebral hyperperfusion syndrome, although rare, is a potentially devastating event that could complicate carotid artery stenting.12,13 The phenomenon was most common in patients with CBF increases of more than 100% compared with baseline values after carotid revascularisation.2,13 In the present cases, no patient had a CBF increase >100% compared with the baseline value after CAS; however, five patients had CBF decreases of >20% compared with the baseline value on the stenting side after CAS.
Acute carotid stent thrombosis was one of the major clinical concerns when hypoperfusion on the stenting side was detected after CAS. It is a rare but potentially devastating complication, occurring in 0.5e2% of cases.14,15 Stent thrombosis can be detected immediately after stent deployment when a post-stenting angiography shows newly formed blood clots in the stent or complete occlusion of the stent.15,16 Patients with acute carotid stent thrombosis presented with new-onset neurological deficits such as crescendo transient ischaemic attacks, contralateral hemiplegia, aphasia, change of consciousness or coma immediately or several days after CAS.15 None of the five patients with post-stenting hypoperfusion had the above findings.
Figure 2 Carotid angiography before and after CAS in Case 1. (aeb) Pre- and post-stent placement right common carotid angiography of the neck region. (a) Before CAS, the right proximal cervical ICA showed severe (95%) luminal stenosis but, (b) 20 minutes after CAS, successful stenting without acute carotid stent thrombosis was shown. (ced) Right common carotid angiography of the brain. Relative improvement of the blood flow in right anterior cerebral artery and MCA was noted after CAS (d) than before CAS (c).
D.Y.-T. Chen et al. / Clinical Radiology 71 (2016) e21ee27
e25
Figure 3 Stent position and labelling plane in Case 1. (a) The routine labelling plane was placed 2 cm below the FOV, at the level of the C2 dens (cross); dashed lines, C2 body and dens. (b) Post-stent placement angiography showed the upper margin of the stent (arrow) was at the level of the C2 dens, which superimposed on the labelling plane; dashed lines, C2 body and dens. (c) 3D reformation of post-stent placement TOF-MRA showed the upper margin of the stent (arrow); however, using a labelling plane 1 cm above this level still yielded hypoperfusion on the stenting side. (d) T2* gradient-echo echo-planar imaging better depicted the stent (arrows) and the susceptibility artefact surrounding the stent, which extended above the upper margin of the stent (open arrow). Using a labelling plane above the area with susceptibility artefact yielded increased cerebral perfusion on the stenting side.
Another potential cause of cerebral hypoperfusion after CAS is systemic hypotension due to carotid artery baroreceptor damage during CAS.17 Seventy-five percent of CAS patients had significant episodes of hypotension (>30 mmHg decrease in systolic blood pressure) within the first 24 hours of carotid revascularisation.17 Systemic hypotension could lead to decreased cerebral perfusion when it exceeds the cerebral autoregulation capacity18; however, systemic hypotension caused a more generalised decrease in the cerebral perfusion, especially in the watershed areas in bilateral hemispheres.18 In the present cases, CBF decreased on the stenting side whereas it increased on the contralateral side. Furthermore, no severe hypotension episode was recorded during the post-stenting course in the five patients. Extracranial carotid artery disease occurs most commonly in the carotid artery bifurcation, which usually appears at the level of the thyroid cartilage (approximately C4)19; however, the bifurcation might occur as high as C1 or as low as T2 in some patients.19 In the present routine PCASL perfusion imaging, the labelling plane was placed 2 cm inferior to the lower edge of the cerebellum, roughly at the level of the C2 dens (Fig 2). Therefore, at first, the cause of cerebral hypoperfusion was not considered to be an artefact of carotid stenting but pathophysiological. High stent position with the upper margin of the stent above the level of the C2 body was noticed in all five patients. The high stent position would make the uppermost part of the stent superimposed on the labelling plane. The susceptibility effect around the stent would cause resonance offset in the labelling plane and lead to a low labelling efficiency of the arterial spins.20,21 When the labelling efficiency was low in a feeding artery, the CBF measured in the territory of that artery would be low, which could mimic areas of hypoperfusion. To avoid labelling the stent, an MRA examination of the neck might help identify the position of the stent. In
the present series, TOF-MRA was performed during the first callback study, which covered the neck to identify the position of the stent and put the labelling plane 1 cm above the level (Fig 3); however, hypoperfusion was still noted in the post-adjustment PCASL perfusion imaging. Consequently, a T2* (gradient-echo) echo-planar imaging was performed, which better identified the extent of the susceptibility effect around the stent (Fig 3) and adjusted the labelling level to avoid the susceptibility affected region. After labelling plane adjustment, the CBF on the stenting side increased significantly, while remaining stationary on the contralateral side in the patient. The post-stenting hypoperfusion measured using PCASL perfusion imaging in the present cases was actually due to the carotid stent artefact. Of note, in Case 4, the patient presented with acute headache and limb weakness contralateral to the stenting side after CAS. On Day 3 post-stent placement, PCASL perfusion imaging revealed 27% decreased cerebral hypoperfusion on the stenting side. Emergent cerebral angiography was requested by the clinician due to concern regarding acute stent thrombosis; however, the TOF-MRA in the same study showed patent ICA flow distal to the stent. Repeated PCASL perfusion imaging on the same day showed persistent cerebral hypoperfusion on the stenting side before labelling plane adjustment. After labelling plane adjustment, it showed that there was actually a 78% increase of the CBF in the stenting side compared to the prestenting condition. The symptoms of the patient were actually caused by cerebral hyperperfusion syndrome, rather than hypoperfusion. The emergent cerebral angiography was not needed. After antihypertensive treatment, the symptoms resolved. Ideally, the labelling plane of the PCASL perfusion imaging should be placed in a region where the feeding arteries are straight and perpendicular to the labelling
e26
D.Y.-T. Chen et al. / Clinical Radiology 71 (2016) e21ee27
plane.22 An additional MRA could visualise this but would extend the examination time. Methods using the anatomical landmark to locate the labelling plane have also been proposed. Aslan and colleagues suggested choosing a plane that is 8.4cm inferior to the anterior commissureeposterior commissure line to yield the highest signal intensity.23 Dai and colleagues suggested placing the labelling plane 1.8 cm below the inferior border of the cerebellum.24 Labelling failure should be considered when the CBF is low in a vascular territory around the PCASL perfusion imaging; however, true hypoperfusion could not be excluded. The TOF-MRA may be helpful in differentiating between these two conditions. The flow signals would decrease in the regions with true hypoperfusion while they are preserved in labelling failure. Moreover, gradient-echo-based imaging covering the neck, which is sensitive to local magnetic field inhomogeneity, may help demonstrate the cause of labelling failure, such as tortuous vessels or susceptibility artefact (e.g., dental implant, blood, calcification, surgical materials) in the labelling plane.22 A carotid stent was also a potential cause of susceptibility artefact in the neck region. To the authors’ knowledge, no PCASL perfusion study measuring the CBF in the post-CAS patients has mentioned this artefact or discussed how not to label a stent to get reliable perfusion results. This might be because the location of the carotid stent is usually lower than the routine labelling plane. Seventeen of 22 patients (77%) were not affected by the artefact in the present PCASL perfusion study. Moreover, the CBF might increase substantially in the stenting side after CAS, which might compensate the effect of loss of labelling efficiency caused by the stent and lead to a pseudonormalised CBF measurement in some cases. Ignorance of the artefact may lead to misinterpretation of a patient with cerebral hyperperfusion syndrome as normal cerebral perfusion or cerebral hypoperfusion after CAS (as in the present Case 4). The present study has several limitations. First, the results are limited to a small sample size. More cases may help solidify the stent artefact in PCASL as well as the method proposed to avoid such an artefact. Second, the extent of the susceptibility artefact surrounding the stent was not quantified in the current study in which T2* gradient-echo echo-planar imaging was used to visualise the susceptibility artefact. Minor resonance offset might still exist when the labelling plane is placed 1 cm away from the stent where the susceptibility artefact was not be visualised. This minor phase shift could still interfere with the labelling efficiency, and hence, affect the quantification of CBF in PCASL perfusion imaging. Several methods have been proposed to overcome the labelling efficiency loss in PCASL, such as the multiple-phase offset method or phase-tracking error adjustment20,25 or using phase-contrast MRA to generate a global CBF calibration factor.23 These methods could be also considered when stent artefact is found in PCASL imaging in future studies. In conclusion, when using the PCASL technique to measure post-carotid stenting CBF, a stent-related
hypoperfusion artefact that leads to a low calculated CBF in the territory of the stented vessel might occur if the labelling plane was superimposed on the stent. This should not be mistaken for acute carotid stent thrombosis, which is a potential devastating complication after CAS. This artefact occurred, in particular, in patients with high carotid stent position. Routinely performing a quick T2* gradient-echo echo-planar imaging covering the cervical region in these patients is recommended to identify the extent of the susceptibility effect around the stent and place the labelling plane over the upper margin of the area to avoid the stentrelated artefact.
References 1. Brott TG, Hobson 2nd RW, Howard G, et al. Stenting versus endarterectomy for treatment of carotid-artery stenosis. N Engl J Med 2010;363(1):11e23. 2. Ogasawara K, Yukawa H, Kobayashi M, et al. Prediction and monitoring of cerebral hyperperfusion after carotid endarterectomy by using singlephoton emission computerized tomography scanning. J Neurosurg 2003;99(3):504e10. 3. Van Laar PJ, Hendrikse J, Mali WPTM, et al. Altered flow territories after carotid stenting and carotid endarterectomy. J Vasc Surg 2007;45(6):1155e61. 4. Waaijer A, van Leeuwen MS, van Osch MJ, et al. Changes in cerebral perfusion after revascularization of symptomatic carotid artery stenosis: CT measurement. Radiology 2007;245(2):541e8. 5. Gaudiello F, Colangelo V, Bolacchi F, et al. Sixty-four-section CT cerebral perfusion evaluation in patients with carotid artery stenosis before and after stenting with a cerebral protection device. AJNR Am J Neuroradiol 2008;29(5):919e23. 6. Yang B, Chen W, Yang Y, et al. Short- and long-term hemodynamic and clinical effects of carotid artery stenting. AJNR Am J Neuroradiol 2012;33(6):1170e6. 7. Yun TJ, Sohn CH, Han MH, et al. Effect of carotid artery stenting on cerebral blood flow: evaluation of hemodynamic changes using arterial spin labelling. Neuroradiology 2013;55(3):271e81. 8. Detre JA, Leigh JS, Williams DS, et al. Perfusion imaging. Magn Reson Med 1992;23(1):37e45. 9. Wang JJ, Alsop DC, Li L, et al. Comparison of quantitative perfusion imaging using arterial spin labelling at 1.5 and 4.0 tesla. Magn Reson Med 2002;48(2):242e54. 10. Ferguson GG, Eliasziw M, Barr HWK, et al. The North American Symptomatic Carotid Endarterectomy Trial d surgical results in 1415 patients. Stroke 1999;30(9):1751e8. 11. Damasio H. A computed tomographic guide to the identification of cerebral vascular territories. Arch Neurol 1983;40(3):138e42. 12. Coutts SB, Hill MD, Hu WY, et al. Hyperperfusion syndrome: toward a stricter definition. Neurosurgery 2003;53(5):1053e8. 13. van Mook WN, Rennenberg RJ, Schurink GW, et al. Cerebral hyperperfusion syndrome. Lancet Neurol 2005;4(12):877e88. 14. Diethrich EB, Ndiaye M, Reid DB. Stenting in the carotid artery: initial experience in 110 patients. J Endovasc Surg 1996;3(1):42e62. 15. Xiromeritis K, Dalainas I, Stamatakos M, et al. Acute carotid stent thrombosis after carotid artery stenting. Eur Rev Med Pharmacol Sci 2012;16(3):355e62. 16. Iancu A, Grosz C, Lazar A. Acute carotid stent thrombosis: review of the literature and long-term follow-up. Cardiovasc Revasc Med 2010;11(2):110e3. 17. McKevitt FM, Sivaguru A, Venables GS, et al. Effect of treatment of carotid artery stenosis on blood pressure: a comparison of hemodynamic disturbances after carotid endarterectomy and endovascular treatment. Stroke 2003;34(11):2576e81. 18. Bladin CF, Chambers BR. Frequency and pathogenesis of hemodynamic stroke. Stroke 1994;25(11):2179e82. 19. Osborn AG. Diagnostic cerebral angiography. Philadelphia: Lippincott Williams & Wilkins; 1999.
D.Y.-T. Chen et al. / Clinical Radiology 71 (2016) e21ee27 20. Shin DD, Liu TT, Wong EC, et al. Pseudocontinuous arterial spin labelling with optimized tagging efficiency. Magn Reson Med 2012;68(4):1135e44. 21. Wu WC, Fernandez-Seara M, Detre JA, et al. A theoretical and experimental investigation of the tagging efficiency of pseudocontinuous arterial spin labelling. Magn Reson Med 2007;58(5):1020e7. 22. Alsop DC, Detre JA, Golay X, et al. Recommended implementation of arterial spin-labeled perfusion MRI for clinical applications: a consensus of the ISMRM perfusion study group and the European consortium for ASL in dementia. Magn Reson Med 2014 Apr 8. http://dx.doi.org/10.1002/ mrm.25197.
e27
23. Aslan S, Xu F, Wang PL, et al. Estimation of labelling efficiency in pseudocontinuous arterial spin labelling. Magn Reson Med 2010;63(3):765e71. 24. Dai W, Robson PM, Shankaranarayanan A, et al. Reduced resolution transit delay prescan for quantitative continuous arterial spin labelling perfusion imaging. Magn Reson Med 2012;67(5):1252e65. 25. Jung Y, Woing EC, Liu TT. Multiphase pseudocontinuous arterial spin labelling (MP-PCASL) for robust quantification of cerebral blood flow. Magn Reson Med 2010;64(3):799e810.
Contents lists available at ScienceDirect
Clinical Radiology journal homepage: www.clinicalradiologyonline.net
Loss of labelling efficiency caused by carotid stent in pseudocontinuous arterial spin labelling perfusion study D.Y.-T. Chen a, Y.-S. Kuo a, H.-L. Hsu a, F.-X. Yan a, H.-L. Liu b, C.-J. Chen a, Y.-C. Tseng a, * a
Department of Radiology and Brain and Consciousness Research Center, Taipei Medical University Shuang-Ho Hospital, New Taipei City, Taiwan b Department of Imaging Physics, The University of Texas M. D. Anderson Cancer Center, Houston, TX, USA
art icl e i nformat ion Article history: Received 21 December 2014 Received in revised form 8 September 2015 Accepted 1 October 2015
AIM: To elucidate the cause of cerebral hypoperfusion on the stent placement side after carotid artery stent placement (CAS) measured by pseudocontinuous arterial spin labelling (PCASL) perfusion imaging. MATERIALS AND METHODS: Consecutive patients with symptomatic internal carotid artery stenosis receiving CAS were included in the study. Cerebral blood flow (CBF) was measured by PCASL perfusion imaging at 3 T magnetic resonance imaging (MRI) the day before and 3 days after the procedure. Changes in cerebral haemodynamics after CAS were assessed. RESULTS: Twenty-two patients were included; 17 patients had increased or stationary CBF after CAS and five patients had significantly reduced CBF on the stenting side after CAS whereas CBF increased on the contralateral side. High stent position was noticed in the five patients. After labelling plane adjustment to avoid labelling on the stent, no more cerebral hypoperfusion was noticed. CONCLUSION: When using PCASL perfusion imaging to monitor post-stenting CBF, the stent may cause an artefact that leads to a low CBF in the territory of the stented vessel. Routinely adding a fast T2 star gradient-echo echo-planar-imaging covering the upper neck region before PCASL perfusion imaging to identify the stent position and avoid the stent-related artefact is recommended. Ó 2015 The Royal College of Radiologists. Published by Elsevier Ltd. All rights reserved.
Introduction Carotid artery stenting (CAS) is being increasingly used as an alternative treatment to carotid endarterectomy for * Guarantor and correspondent: Y.-C. Tseng, Department of Radiology, Shuang-Ho Hospital, No. 291, Zhongzheng Road, Zhonghe District, New Taipei City 23561, Taiwan. Tel.: þ886 2 22490088 1320; fax: þ886 2 22490088 1322. E-mail address: [email protected] (Y.-C. Tseng).
carotid artery stenosis.1 Alteration of cerebral perfusion after CAS has been investigated in many perfusion studies.2e7 Most of them reported improvement in cerebral perfusion on the stenting side after CAS, especially in patients with high-grade stenosis, with greater perfusion deficit prior to CAS, or with symptoms.4e6 Pseudocontinuous arterial spin labelling (PCASL) is a non-invasive technique that detects absolute cerebral blood flow (CBF) without the use of an exogenous contrast agent.8,9 These properties permit repeated measurement of
http://dx.doi.org/10.1016/j.crad.2015.10.004 0009-9260/Ó 2015 The Royal College of Radiologists. Published by Elsevier Ltd. All rights reserved.
e22
D.Y.-T. Chen et al. / Clinical Radiology 71 (2016) e21ee27
CBF in a short interval, which is useful in patients receiving CAS.3,7 In Shuang-Ho Hospital PCASL perfusion imaging was used to monitor the cerebral haemodynamic changes before and after CAS. Most patients had increased or stationary CBF after CAS; however, occasionally, patients had obviously decreased CBF on the stenting side after CAS. The aim of the present study was to elucidate the causes of the post-stenting hypoperfusion measured by ASL.
Materials and methods Patients This study was approved by the local ethics committee. From July 2012 to October 2013, consecutive patients with symptomatic internal carotid artery (ICA) stenosis (ICA stenosis >70% by North American Symptomatic Carotid Endarterectomy Trial method on conventional cerebral angiogram10) receiving CAS in Shuang-Ho Hospital were included in the study. To evaluate the haemodynamic changes after CAS, whole-brain CBF was measured using the PCASL technique before and after the procedure. Patients were excluded when severe dental metal artefact was noted on the PCASL perfusion imaging.
CAS Patients were pretreated with aspirin (325 mg/d) and clopidogrel (75 mg/d, Plavix; Bristol-Myers Squibb/Sanofi Pharmaceuticals, New York, NY, USA) for at least 3 days before CAS. Systemic anticoagulation was then given by intravenous administration of a bolus of heparin (70 U/kg) and then continuous heparin infusion (15 U/kg/h) to maintain an activated clotting time of more than 250 seconds. The self-expanding stent (Precise; Cordis, Miami Lakes, FL, USA) was deployed across the lesion, and a semicompliant angioplasty balloon was used to post-dilate the stent to achieve >90% luminal diameter. The postprocedural angiography was performed 20 minutes after stent deployment to evaluate both the stented site and the distal cerebral vasculature. After the procedure, antihypertensive agents were used to maintain systolic pressures of <130 mmHg.
Imaging acquisition All patients received brain magnetic resonance imaging (MRI) using a 3 T MRI system (Discovery MR750, General Electric Medical Systems, Milwaukee, WI, USA) with diffusion-weighted imaging (DWI), time-of-flight (TOF)magnetic resonance angiography (MRA) of the brain, and PCASL perfusion imaging on the day before and 3 days after CAS. The protocol applied on each scan was as follows1: DWI: single-shot spin-echo echo-planar imaging with repetition time (TR)/echo time (TE)¼8000 ms/68 ms, flip angle¼90 , b¼1000 s/mm2, diffusion direction¼all, field of view (FOV)¼230230 mm2, in-plane matrix¼512512, section thickness¼5 mm, intersection gap¼2 mm.2 TOFMRA: three-slab three-dimensional (3D) sequence with
TR/TE¼30 ms/2.9 ms, flip angle¼20o, FOV¼200200 mm2, in-plane matrix¼416256, section thickness¼1.4 mm, superior to inferior acquisition.3 PCASL perfusion imaging: 3D background suppressed fast-spin-echo stack-of-spiral readout module with eight in-plane spiral interleaves, TR/ TE¼5327 ms/10.5 ms, labelling duration¼1.5 seconds, postlabelling delay¼2525 ms, no flow-crushing gradients, inplane matrix¼128128, number of excitations (NEX)¼4, section thickness¼4 mm, echo train length¼36 to obtain 36 consecutive axial sections, labelling plane¼10-mm thick and placed 2 cm inferior to the lower edge of the cerebellum, total scan time¼336 seconds.
Data analysis The CBF maps were generated on an Advantage Windows workstation using Functool software (version 9.4, GE Medical Systems). Quantification of CBF was calculated with the following equation9:
PLD ST 1 e T1t eT1b PW CBF ¼ 6000,l LT SFPW PD 2εT1b 1 e T1b where T1 of the blood (T1b) was assumed to be 1.6 seconds at 3 T, T1 of the tissue (T1t) 1.2 seconds, partition coefficient (l) 0.9, labelling efficiency (ε) 0.6, saturation time of PD (ST) 2 seconds, labelling duration (LT) 1.5 seconds, and postlabelling delay (PLD) 2525 ms. PW was the perfusionweighted or the raw-difference image; PD was the signal intensity of the proton density image and SFPW was an empirical scaling factor (¼ 32) used to increase the dynamic range of the PW. To compare changes in CBF before and after CAS, two radiologists with 4 and 15 years of experience chose the regions of interest (ROIs) on two sections, the level of the basal ganglia and the level of the body of the lateral ventricle. On each section, the raters manually outlined the ROIs corresponding to the cortical flow territory of the middle cerebral artery (MCA) of both hemispheres according to the maps by Damasio,11 taking care to exclude regions of prior cerebral infarction. The absolute CBF (expressed in millilitres per 100 g of tissue per minute) of the MCA territories in each hemisphere were obtained by averaging the mean values within the ROIs of the two sections. Relative CBF was defined as stenting side CBF/contralateral side CBF.
Callback study In patients showing post-stenting cerebral hypoperfusion, callback studies were performed to confirm whether cerebral hypoperfusion persisted in PCASL perfusion imaging. In the callback study, the CBF was measured twice using two different labelling planes. The CBF was first measured by PCASL perfusion imaging using the routine labelling plane. On the second measurement, T2* echoplanar imaging (gradient-echo echo-planar-imaging sequence, TR/TE¼925 ms/20 ms, flip angle¼70o, FOV¼250250 mm2, in-plane matrix¼132128, section thickness¼5 mm, intersection gap¼1 mm, NEX¼4, scanning
D.Y.-T. Chen et al. / Clinical Radiology 71 (2016) e21ee27
time¼26 seconds) was performed that covered the cervical region to identify the extent of the stent-related susceptibility artefact. The labelling plane was then placed 1 cm above the upper margin of the area with stent-related susceptibility artefact to measure the post-adjustment CBF.
Results Patient characteristics Twenty-three patients received CAS in Shuang-Ho Hospital during the study period. One patient was excluded due to severe dental metal artefact in the PCASL perfusion imaging. Seventeen patients had increased or stationary CBF after CAS and five patients had obviously reduced CBF after CAS on the stenting side. The mean age of the patients with increased or stationary post-stenting CBF was 70.7 years (9.9 years; range, 56e89 years) and the mean preprocedural ICA stenosis rate on the stenting side was 84% (10%; range, 70e99%). The mean age of the patients with post-stenting hypoperfusion was 73.6 years (5.72 years; range, 64e78 years) and the mean pre-procedural ICA stenosis rate on the stenting side was 84.4% (12.9%; range, 70e99%). No significant difference was noted on both groups regarding age, gender, stenting side and stenosis grade (p¼0.21, 0.64, 0.52, 0.49 respectively). The five patients with post-stent hypoperfusion were included in the following analysis.
CBF measurement The pre-stenting and post-stenting (Day 3) CBF values of the five patients are listed in Table 1. The five patients with decreased perfusion on the stenting side after CAS had increased CBF on the contralateral side after CAS. On the stenting side, there were decreases of 20.2%, 20.1%, 43.2%, 27.2%, and 34% in the post-stenting CBF compared to the pre-stenting CBF, respectively. On the contralateral side, there were increases of 22.8%, 26.6%, 6.5%, 11.8%, and 2.1% in the post-stenting CBF, respectively. The relative CBF was 0.90, 1.08, 1.14, 0.83, and 0.81 in the pre-stenting scan and 0.58, 0.68, 0.61, 0.54, and 0.53 in the post-stenting scan for the five patients, respectively.
e23
Acute carotid stent thrombosis The post-procedural angiography 20 minutes after stent deployment showed good recanalization of the lumen without acute thrombosis in these five patients. TOF-MRA on Day 3 post-stent placement also showed normal or increased flow signal in the ICA and MCA on the stenting side. There were few punctate infarcts at the posterior watershed area on stenting side in Cases 1, 3, and 5, and no new infarction in Cases 2 and 4 in the post-stenting DWI.
Carotid stent position The carotid stent position was high in the five patients in the post-procedural angiography. The upper margins of the stents were all above the level of the C2 body. Conversely, the upper margins of the stents were below the C2 body in the other 17 patients who had increased or stationary CBF after CAS.
Callback study The duration between the callback examination and the CAS procedures were 6 months, 4 months, 3 months, 3 days and 3 days for Cases 1, 2, 3, 4, and 5, respectively. The preadjustment and post-adjustment CBF of the five patients are listed in Table 1. The post-adjustment CBF increased significantly from the pre-adjustment CBF on the stenting side. There were 21.2%, 45.9%, 48.8%, 88.3%, and 88.7% increases in the post-adjustment CBF compared to the preadjustment CBF in the five patients, respectively. The CBF remained relatively stationary (<10% changes) on the contralateral side before and after labelling plane adjustment in the five patients. The post-adjustment relative CBF was 1.13, 0.92, 1.17, 1.30, and 0.94 in the five patients, respectively. When we compared the post-adjustment CBF with the pre-stenting CBF, the CBF actually increased on both sides. There were 50.9%, 1.0%, 13.1%, 77.7%, and 25.8% increases on the stenting side and 20.7%, 18.1%, 9.5%, 13.6%, and 8.5% increases on the contralateral side in the five patients, respectively. The serial CBF maps (Fig 1), the pre- and post-stenting angiography images (Fig 2), and images of the labelling
Table 1 Cerebral blood flow (CBF) of the patients with post-stent placement hypoperfusion. Case (no.)
1 2 3 4 5
Pre-stenting CBF
Post-stent D3 CBF
Stent
Stent
35.2 47.8 59.5 37.2 42.3
Contra 39.1 44.3 52.4 44.8 52.1
28.1 38.2 33.8 27.1 27.9
Contra a
(e20.2%) (e20.1%)a (e43.2%)a (e27.2%)a (e34.0%)a
48.0 56.1 55.8 50.1 53.2
a
(þ22.8%) (þ26.6%)a (þ6.5%)a (11.8%)a (þ2.1%)a
Callback pre-adjust CBF
Callback post-adjust CBF
Stent
Stent
43.8 33.1 45.2 35.1 28.2
Contra 52.1 54.3 57.4 50.3 53.8
53.1 48.3 67.3 66.1 53.2
Contra b
(þ21.2%) (þ45.9%)b (þ48.8%)b (þ88.3%)b (þ88.7%)b
47.2 52.3 57.4 50.9 56.5
(e9.4%)b (e3.7%)b (0%)b (þ1.2%)b (þ5.0%)b
Data are mean values of CBF (in ml/100 g of tissue/min). Stent, stent placement side; Contra, the side contralateral to stent placement. a Numbers in parentheses represent Day 3 post-stent placement to pre-stent placement CBF change, defined as [(post-stent placement Day 3 CBF e pre-stent placement CBF)/pre-stent placement CBF]100%. b Numbers in the parentheses represent post-adjust to pre-adjust CBF change, defined as [(post-adjust CBF e pre-adjust CBF)/pre-adjust CBF]100%.
e24
D.Y.-T. Chen et al. / Clinical Radiology 71 (2016) e21ee27
Figure 1 CBF changes in Case 1. A 64-year-old man presented with intermittent left side weakness. (a) The pre-stent CBF map showed mild decreased perfusion in the right hemisphere especially at the right posterior watershed area. (b) The CBF map Day 3 post-stent placement, showed decreased perfusion on the stenting side and increased perfusion on the contralateral side. (c) The callback pre-adjustment CBF showed persistent hypoperfusion on the stenting side. (d) The callback post-adjustment CBF showed symmetrically increased perfusion on both sides.
plane and the stent position (Fig 3) of a representative case are presented to illustrate the condition.
Discussion Cerebral hyperperfusion syndrome, although rare, is a potentially devastating event that could complicate carotid artery stenting.12,13 The phenomenon was most common in patients with CBF increases of more than 100% compared with baseline values after carotid revascularisation.2,13 In the present cases, no patient had a CBF increase >100% compared with the baseline value after CAS; however, five patients had CBF decreases of >20% compared with the baseline value on the stenting side after CAS.
Acute carotid stent thrombosis was one of the major clinical concerns when hypoperfusion on the stenting side was detected after CAS. It is a rare but potentially devastating complication, occurring in 0.5e2% of cases.14,15 Stent thrombosis can be detected immediately after stent deployment when a post-stenting angiography shows newly formed blood clots in the stent or complete occlusion of the stent.15,16 Patients with acute carotid stent thrombosis presented with new-onset neurological deficits such as crescendo transient ischaemic attacks, contralateral hemiplegia, aphasia, change of consciousness or coma immediately or several days after CAS.15 None of the five patients with post-stenting hypoperfusion had the above findings.
Figure 2 Carotid angiography before and after CAS in Case 1. (aeb) Pre- and post-stent placement right common carotid angiography of the neck region. (a) Before CAS, the right proximal cervical ICA showed severe (95%) luminal stenosis but, (b) 20 minutes after CAS, successful stenting without acute carotid stent thrombosis was shown. (ced) Right common carotid angiography of the brain. Relative improvement of the blood flow in right anterior cerebral artery and MCA was noted after CAS (d) than before CAS (c).
D.Y.-T. Chen et al. / Clinical Radiology 71 (2016) e21ee27
e25
Figure 3 Stent position and labelling plane in Case 1. (a) The routine labelling plane was placed 2 cm below the FOV, at the level of the C2 dens (cross); dashed lines, C2 body and dens. (b) Post-stent placement angiography showed the upper margin of the stent (arrow) was at the level of the C2 dens, which superimposed on the labelling plane; dashed lines, C2 body and dens. (c) 3D reformation of post-stent placement TOF-MRA showed the upper margin of the stent (arrow); however, using a labelling plane 1 cm above this level still yielded hypoperfusion on the stenting side. (d) T2* gradient-echo echo-planar imaging better depicted the stent (arrows) and the susceptibility artefact surrounding the stent, which extended above the upper margin of the stent (open arrow). Using a labelling plane above the area with susceptibility artefact yielded increased cerebral perfusion on the stenting side.
Another potential cause of cerebral hypoperfusion after CAS is systemic hypotension due to carotid artery baroreceptor damage during CAS.17 Seventy-five percent of CAS patients had significant episodes of hypotension (>30 mmHg decrease in systolic blood pressure) within the first 24 hours of carotid revascularisation.17 Systemic hypotension could lead to decreased cerebral perfusion when it exceeds the cerebral autoregulation capacity18; however, systemic hypotension caused a more generalised decrease in the cerebral perfusion, especially in the watershed areas in bilateral hemispheres.18 In the present cases, CBF decreased on the stenting side whereas it increased on the contralateral side. Furthermore, no severe hypotension episode was recorded during the post-stenting course in the five patients. Extracranial carotid artery disease occurs most commonly in the carotid artery bifurcation, which usually appears at the level of the thyroid cartilage (approximately C4)19; however, the bifurcation might occur as high as C1 or as low as T2 in some patients.19 In the present routine PCASL perfusion imaging, the labelling plane was placed 2 cm inferior to the lower edge of the cerebellum, roughly at the level of the C2 dens (Fig 2). Therefore, at first, the cause of cerebral hypoperfusion was not considered to be an artefact of carotid stenting but pathophysiological. High stent position with the upper margin of the stent above the level of the C2 body was noticed in all five patients. The high stent position would make the uppermost part of the stent superimposed on the labelling plane. The susceptibility effect around the stent would cause resonance offset in the labelling plane and lead to a low labelling efficiency of the arterial spins.20,21 When the labelling efficiency was low in a feeding artery, the CBF measured in the territory of that artery would be low, which could mimic areas of hypoperfusion. To avoid labelling the stent, an MRA examination of the neck might help identify the position of the stent. In
the present series, TOF-MRA was performed during the first callback study, which covered the neck to identify the position of the stent and put the labelling plane 1 cm above the level (Fig 3); however, hypoperfusion was still noted in the post-adjustment PCASL perfusion imaging. Consequently, a T2* (gradient-echo) echo-planar imaging was performed, which better identified the extent of the susceptibility effect around the stent (Fig 3) and adjusted the labelling level to avoid the susceptibility affected region. After labelling plane adjustment, the CBF on the stenting side increased significantly, while remaining stationary on the contralateral side in the patient. The post-stenting hypoperfusion measured using PCASL perfusion imaging in the present cases was actually due to the carotid stent artefact. Of note, in Case 4, the patient presented with acute headache and limb weakness contralateral to the stenting side after CAS. On Day 3 post-stent placement, PCASL perfusion imaging revealed 27% decreased cerebral hypoperfusion on the stenting side. Emergent cerebral angiography was requested by the clinician due to concern regarding acute stent thrombosis; however, the TOF-MRA in the same study showed patent ICA flow distal to the stent. Repeated PCASL perfusion imaging on the same day showed persistent cerebral hypoperfusion on the stenting side before labelling plane adjustment. After labelling plane adjustment, it showed that there was actually a 78% increase of the CBF in the stenting side compared to the prestenting condition. The symptoms of the patient were actually caused by cerebral hyperperfusion syndrome, rather than hypoperfusion. The emergent cerebral angiography was not needed. After antihypertensive treatment, the symptoms resolved. Ideally, the labelling plane of the PCASL perfusion imaging should be placed in a region where the feeding arteries are straight and perpendicular to the labelling
e26
D.Y.-T. Chen et al. / Clinical Radiology 71 (2016) e21ee27
plane.22 An additional MRA could visualise this but would extend the examination time. Methods using the anatomical landmark to locate the labelling plane have also been proposed. Aslan and colleagues suggested choosing a plane that is 8.4cm inferior to the anterior commissureeposterior commissure line to yield the highest signal intensity.23 Dai and colleagues suggested placing the labelling plane 1.8 cm below the inferior border of the cerebellum.24 Labelling failure should be considered when the CBF is low in a vascular territory around the PCASL perfusion imaging; however, true hypoperfusion could not be excluded. The TOF-MRA may be helpful in differentiating between these two conditions. The flow signals would decrease in the regions with true hypoperfusion while they are preserved in labelling failure. Moreover, gradient-echo-based imaging covering the neck, which is sensitive to local magnetic field inhomogeneity, may help demonstrate the cause of labelling failure, such as tortuous vessels or susceptibility artefact (e.g., dental implant, blood, calcification, surgical materials) in the labelling plane.22 A carotid stent was also a potential cause of susceptibility artefact in the neck region. To the authors’ knowledge, no PCASL perfusion study measuring the CBF in the post-CAS patients has mentioned this artefact or discussed how not to label a stent to get reliable perfusion results. This might be because the location of the carotid stent is usually lower than the routine labelling plane. Seventeen of 22 patients (77%) were not affected by the artefact in the present PCASL perfusion study. Moreover, the CBF might increase substantially in the stenting side after CAS, which might compensate the effect of loss of labelling efficiency caused by the stent and lead to a pseudonormalised CBF measurement in some cases. Ignorance of the artefact may lead to misinterpretation of a patient with cerebral hyperperfusion syndrome as normal cerebral perfusion or cerebral hypoperfusion after CAS (as in the present Case 4). The present study has several limitations. First, the results are limited to a small sample size. More cases may help solidify the stent artefact in PCASL as well as the method proposed to avoid such an artefact. Second, the extent of the susceptibility artefact surrounding the stent was not quantified in the current study in which T2* gradient-echo echo-planar imaging was used to visualise the susceptibility artefact. Minor resonance offset might still exist when the labelling plane is placed 1 cm away from the stent where the susceptibility artefact was not be visualised. This minor phase shift could still interfere with the labelling efficiency, and hence, affect the quantification of CBF in PCASL perfusion imaging. Several methods have been proposed to overcome the labelling efficiency loss in PCASL, such as the multiple-phase offset method or phase-tracking error adjustment20,25 or using phase-contrast MRA to generate a global CBF calibration factor.23 These methods could be also considered when stent artefact is found in PCASL imaging in future studies. In conclusion, when using the PCASL technique to measure post-carotid stenting CBF, a stent-related
hypoperfusion artefact that leads to a low calculated CBF in the territory of the stented vessel might occur if the labelling plane was superimposed on the stent. This should not be mistaken for acute carotid stent thrombosis, which is a potential devastating complication after CAS. This artefact occurred, in particular, in patients with high carotid stent position. Routinely performing a quick T2* gradient-echo echo-planar imaging covering the cervical region in these patients is recommended to identify the extent of the susceptibility effect around the stent and place the labelling plane over the upper margin of the area to avoid the stentrelated artefact.
References 1. Brott TG, Hobson 2nd RW, Howard G, et al. Stenting versus endarterectomy for treatment of carotid-artery stenosis. N Engl J Med 2010;363(1):11e23. 2. Ogasawara K, Yukawa H, Kobayashi M, et al. Prediction and monitoring of cerebral hyperperfusion after carotid endarterectomy by using singlephoton emission computerized tomography scanning. J Neurosurg 2003;99(3):504e10. 3. Van Laar PJ, Hendrikse J, Mali WPTM, et al. Altered flow territories after carotid stenting and carotid endarterectomy. J Vasc Surg 2007;45(6):1155e61. 4. Waaijer A, van Leeuwen MS, van Osch MJ, et al. Changes in cerebral perfusion after revascularization of symptomatic carotid artery stenosis: CT measurement. Radiology 2007;245(2):541e8. 5. Gaudiello F, Colangelo V, Bolacchi F, et al. Sixty-four-section CT cerebral perfusion evaluation in patients with carotid artery stenosis before and after stenting with a cerebral protection device. AJNR Am J Neuroradiol 2008;29(5):919e23. 6. Yang B, Chen W, Yang Y, et al. Short- and long-term hemodynamic and clinical effects of carotid artery stenting. AJNR Am J Neuroradiol 2012;33(6):1170e6. 7. Yun TJ, Sohn CH, Han MH, et al. Effect of carotid artery stenting on cerebral blood flow: evaluation of hemodynamic changes using arterial spin labelling. Neuroradiology 2013;55(3):271e81. 8. Detre JA, Leigh JS, Williams DS, et al. Perfusion imaging. Magn Reson Med 1992;23(1):37e45. 9. Wang JJ, Alsop DC, Li L, et al. Comparison of quantitative perfusion imaging using arterial spin labelling at 1.5 and 4.0 tesla. Magn Reson Med 2002;48(2):242e54. 10. Ferguson GG, Eliasziw M, Barr HWK, et al. The North American Symptomatic Carotid Endarterectomy Trial d surgical results in 1415 patients. Stroke 1999;30(9):1751e8. 11. Damasio H. A computed tomographic guide to the identification of cerebral vascular territories. Arch Neurol 1983;40(3):138e42. 12. Coutts SB, Hill MD, Hu WY, et al. Hyperperfusion syndrome: toward a stricter definition. Neurosurgery 2003;53(5):1053e8. 13. van Mook WN, Rennenberg RJ, Schurink GW, et al. Cerebral hyperperfusion syndrome. Lancet Neurol 2005;4(12):877e88. 14. Diethrich EB, Ndiaye M, Reid DB. Stenting in the carotid artery: initial experience in 110 patients. J Endovasc Surg 1996;3(1):42e62. 15. Xiromeritis K, Dalainas I, Stamatakos M, et al. Acute carotid stent thrombosis after carotid artery stenting. Eur Rev Med Pharmacol Sci 2012;16(3):355e62. 16. Iancu A, Grosz C, Lazar A. Acute carotid stent thrombosis: review of the literature and long-term follow-up. Cardiovasc Revasc Med 2010;11(2):110e3. 17. McKevitt FM, Sivaguru A, Venables GS, et al. Effect of treatment of carotid artery stenosis on blood pressure: a comparison of hemodynamic disturbances after carotid endarterectomy and endovascular treatment. Stroke 2003;34(11):2576e81. 18. Bladin CF, Chambers BR. Frequency and pathogenesis of hemodynamic stroke. Stroke 1994;25(11):2179e82. 19. Osborn AG. Diagnostic cerebral angiography. Philadelphia: Lippincott Williams & Wilkins; 1999.
D.Y.-T. Chen et al. / Clinical Radiology 71 (2016) e21ee27 20. Shin DD, Liu TT, Wong EC, et al. Pseudocontinuous arterial spin labelling with optimized tagging efficiency. Magn Reson Med 2012;68(4):1135e44. 21. Wu WC, Fernandez-Seara M, Detre JA, et al. A theoretical and experimental investigation of the tagging efficiency of pseudocontinuous arterial spin labelling. Magn Reson Med 2007;58(5):1020e7. 22. Alsop DC, Detre JA, Golay X, et al. Recommended implementation of arterial spin-labeled perfusion MRI for clinical applications: a consensus of the ISMRM perfusion study group and the European consortium for ASL in dementia. Magn Reson Med 2014 Apr 8. http://dx.doi.org/10.1002/ mrm.25197.
e27
23. Aslan S, Xu F, Wang PL, et al. Estimation of labelling efficiency in pseudocontinuous arterial spin labelling. Magn Reson Med 2010;63(3):765e71. 24. Dai W, Robson PM, Shankaranarayanan A, et al. Reduced resolution transit delay prescan for quantitative continuous arterial spin labelling perfusion imaging. Magn Reson Med 2012;67(5):1252e65. 25. Jung Y, Woing EC, Liu TT. Multiphase pseudocontinuous arterial spin labelling (MP-PCASL) for robust quantification of cerebral blood flow. Magn Reson Med 2010;64(3):799e810.