DIAGNOSTIC NEURORADIOLOGY. Yuki Shinohara & Toshibumi Kinoshita & Fumiko Kinoshita

Transcription

1 Neuroradiology (2012) 54: DOI /s DIAGNOSTIC NEURORADIOLOGY Changes in susceptibility signs on serial T2*-weighted single-shot echo-planar gradient-echo images in acute embolic infarction: comparison with status on 3D time-of-flight magnetic resonance angiography Yuki Shinohara & Toshibumi Kinoshita & Fumiko Kinoshita Received: 4 March 2011 /Accepted: 17 May 2011 /Published online: 28 May 2011 # Springer-Verlag 2011 Abstract Introduction The present study compares changes in susceptibility signs on follow-up single-shot echo-planar gradient-echo T2*-weighted images (GRE-EPI) with vascular status on follow-up magnetic resonance angiography (MRA) in acute embolic infarction. Methods Twenty consecutive patients with acute embolic infarction repeatedly underwent MR imaging including GRE-EPI and MRA using a 1.5-T MR superconducting system. All patients underwent initial MR examination within 24 h of onset and follow-up MR imaging within 1 month after onset. Results Changes in susceptibility signs on follow-up GRE- EPI were compatible with vascular status on follow-up MRA in 19 of the 20 patients. Susceptibility signs disappeared with complete in 13 patients, migrated with partial in 3, did not change together with the absence of in 2, and became extended together with the absence of in 1. Cerebral hemorrhage obscured susceptibility signs in the one remaining patient. Conclusion Susceptibility signs on follow-up GRE-EPI can reflect changes in an acute embolus, such as or migration, in this study. Serial GRE-EPI in acute embolism complements the diagnostic certainty of MRA by directly detecting an embolus as a susceptibility sign. Y. Shinohara (*) : T. Kinoshita : F. Kinoshita Department of Radiology, Research Institute of Brain and Blood Vessels Akita, 6-10 Senshu-kubota-machi, Akita , Japan shino-y@akita-noken.jp Keywords T2*-weighted single-shot echo-planar gradient-echo image. Susceptibility sign. Acute embolic infarction. Recanalization. 3D time-of-flight magnetic resonance angiography Introduction T2*-weighted imaging plays an important role in acute ischemic stroke because it can detect acute intracranial hemorrhage (ICH) with high sensitivity [1 4]. It can similarly detect the susceptibility of an acute embolus, which is referred to as a susceptibility sign [5 10]. Evaluation of the status of occluded vessels in acute embolic infarction is important to predict the effectiveness of thrombolytic therapy, including recombinant tissue-plasminogen activator (rt-pa), as well as to track prognosis, such as hemorrhagic transformation by or the progression of vasogenic edema due to persistent occlusion [11]. In general, follow-up magnetic resonance angiography (MRA) is used to assess the vascular status of acute embolism. To our knowledge, only a few investigations have compared changes in susceptibility signs on follow-up T2*-weighted images and status on followup MRA in patients with acute embolic infarction [1, 2]. Moreover, T2*-weighted single-shot echo-planar gradientecho imaging (GRE-EPI) has never been used as the follow-up T2* sequence. This modality can detect the susceptibility effect with sensitivity that is comparable to T2*-weighted gradient-echo imaging (GRE) and scan the entire brain much faster than GRE [3, 12]. Others have reported that GRE-EPI can detect microbleeding with

2 428 Neuroradiology (2012) 54: sensitivity that is almost equal or superior to that of GRE, although the image is more distorted on GRE-EPI than on GRE because of artifacts arising from the skull base [3, 12]. The purpose of the present study was to establish the validity of changes in susceptibility sign on follow-up GRE-EPI in comparison with the vascular status on MRA. Materials and methods Patients This retrospective study analyzed data obtained from all patients with stroke admitted to our institution between May 1, 2007 and February 28, Patients were eligible Table 1 Initial and follow-up MR imaging data of MRA and GRE-EPI in 20 patients Patient Age/sex rt-pa Interval between initial and follow-up MR studies (days) Initial MR findings Location of occluded site on initial MRA Location of susceptibility sign on initial GRE-EPI Follow-up MR findings Location of occluded site on follow-up MRA/evaluation Location of susceptibility sign on follow-up GRE- EPI/evaluation 1 79/M + 7 Rt proximal M1 Rt proximal M1 Absent/complete 2 71/M 12 Rt distal M1 Rt distal M1 Absent/complete 3 87/M + 1 Rt distal M1 Rt distal M1 Absent/complete 4 66/F 3 Lt distal M1 Lt distal M1 M2 Absent/complete 5 83/F 8 Rt M2 Rt M2 Absent/complete 6 82/F 9 Rt M2 Rt M2 Absent/complete 7 70/M 4 Rt M2 Rt M2 Absent/complete 8 82/M 1 Rt M2 Rt M2 Absent/complete 9 80/M 14 Rt M2 Rt M2 Absent/complete 10 67/M 7 Lt M2 Lt M2 Absent/complete 11 62/M + 1 Lt distal M1 Lt M2 Absent/complete 12 77/M + 1 Lt M3 Lt M3 Absent/complete 13 52/F 3 Lt terminal ICA Lt terminal ICA distal Absent/complete M /M 1 Lt proximal M1 Lt proximal distal M1 Lt distal M1/partial 15 61/M + 1 Lt distal M1 Lt distal M1 Lt M3/partial 16 57/M 2 Rt distal M1 Rt distal M1 Rt M2/partial 17 93/M 13 Rt proximal M1 Rt proximal M1 M2 Rt proximal M1/no 18 85/F 7 Rt A2 Rt A2 Rt A2/no 19 72/M + 1 Lt distal M1 Lt distal M1 Lt distal M1/no 20 73/M 13 Rt proximal M1 Rt proximal M1 M2 Absent, but faint in-flow effect/ partial Lt distal M1/migration Lt M3/migration Rt M2/migration Rt proximal M1 M2/no change Rt A2/no change Lt distal M1 M2/extension Unclear due to hemorrhage MR magnetic resonance, MRA magnetic resonance angiography, GRE-EPI T2*-weighted single-shot echo-planar gradient-echo imaging, rt-pa recombinant tissue-plasminogen activator, Rt right, Lt left, ICA internal carotid artery, M1 horizontal segment of the middle cerebral artery, M2 insular segment of the middle cerebral artery, M3 opercular segment of the middle cerebral artery, A2 infracallosal segment of the anterior cerebral artery

3 Neuroradiology (2012) 54: if they met the following criteria: MR imaging including GRE-EPI and 3D time-of-flight MRA performed within 24 h of symptom onset, clinical diagnosis of acute thromboembolism confirmed from clinical information such as pattern of symptom onset, resting electrocardiography, continuous electrocardiographic monitoring, transthoracic echocardiography, transesophageal echocardiography and carotid ultrasonography, follow-up GRE-EPI and MRA performed within 1 month, and a single occluded intracranial vessel on initial MRA. Twenty-nine patients met these inclusion criteria. We excluded nine patients in which petrous, precavernous, and cavernous internal carotid artery (ICA) occlusion or basilar-top to posterior cerebral artery (PCA) occlusion was identified by initial MRA because of difficulties associated with evaluating these ICA susceptibility signs due to artifacts from the skull base and with distinguishing PCA susceptibility signs from the basal vein of Rosenthal on initial GRE-EPI. We finally analyzed data from 20 patients (15 men and 5 women; mean age, 75.3 years; range, years). At our institution, we do not routinely perform follow-up MR imaging at the same interval for the patients with acute embolic infarction. Thus, the timing of follow-up MR imaging was not standardized. Informed consent was not required because of the retrospective study design. Imaging examinations The first MR examination proceeded at a mean of 3.3 (range, 1 12) hours from stroke onset using a 1.5-T superconducting unit with echo-planar capabilities (Signa CVNV, GE Medical Systems, Milwaukee, WI, USA or Magnetom Vision, Siemens, Erlangen, Germany). The following were included in the protocol of this study: diffusion-weighted imaging (DWI) (TR/TE/excitations, 5,800 ms/78.9 ms/1; flip angle, 90 ; slice thickness/ interslice gap, 5 mm/1 mm; field of view, 23 cm; matrix size, ; two b values of 0 and 1,000 s/mm 2 ; a b c d e f Fig. 1 Imaging findings from patient 4. a Initial MRA reveals left M1 occlusion (arrow). b, c Initial GRE-EPI shows hypointensity on the left M1 M2 segment (arrowheads). d Follow-up MRA shows complete of the left M1 segment. e, f Follow-up GRE-EPI shows disappearance of susceptibility sign at the left M1 M2 segment

4 430 Neuroradiology (2012) 54: slices; acquisition time, 24 s), GRE-EPI (TE/excitations, 28 ms/1; flip angle, 90 ; slice thickness/interslice gap, 5 mm/1 mm; field of view, 23 cm; matrix, ; 19 slices; acquisition time, 4 s), fast spin-echo (FSE) T2- weighted imaging (TR/TE/excitations, 4,000 ms/93.8 ms/2; slice thickness/interslice gap, 5 mm/1 mm; field of view, 23 cm; matrix size, ; 19 slices; acquisition time, 2 min 16 s), and 3D time-of-flight MRA (TR/TE/excitations, 26 ms/6.7 ms/1; flip angle, 20 ; slice thickness/ interslice gap, 1.4 mm/0.7 mm; field of view, cm; matrix size, ; acquisition time, 7 min 10 s). Follow-up MR including DWI, GRE-EPI, FSE T2- weighted imaging, and 3D time-of-flight MRA was performed 1 to 14 days (mean, 5.5 days) after the first examinations. Six patients with rt-pa therapy performed follow-up MR 1 to 7 days (mean, 2 days) after the first examinations, while the other 14 patients without rt-pa therapy performed follow-up MR 1 to 14 days (mean, 6.9 days) after the first examinations. Among the six patients with rt-pa therapy, MR examinations were first performed before intravenous rt-pa therapy in four patients. The other two patients had initial MR examinations immediately after rt-pa therapy. Image analysis Susceptibility signs on GRE-EPI were defined as hypointensity within the intracranial artery in which the diameter of the hypointense signal within the vessel exceeded the diameter of the contralateral and/or adjacent vessel. We determined the location of each susceptibility sign as the proximal horizontal segment of the middle cerebral artery (MCA) (proximal M1), distal horizontal segment of the MCA (distal M1), insular segment of the MCA (M2), opercular segment of the MCA (M3), infracallosal segment of the anterior cerebral artery (A2), and terminal ICA. Thereafter, recent ischemic territory and patency of the intracranial artery were investigated using DWI and MRA, respectively. The MRA findings served as the reference standard for establishing a diagnosis of occluded intracranial arteries. We compared the occlusion site at the initial MRA with the proximal portion of the susceptibility sign on initial GRE-EPI. In addition, we compared status on follow-up MRA with changes in susceptibility signs on follow-up GRE-EPI. Recanalization of an occluded artery was visually classified on follow-up MRA as complete (in-flow effect from the initial occluded site to a b c d e f Fig. 2 Imaging findings from patient 15. a Initial MRA before rt-pa therapy reveals left M1 occlusion (arrow). b, c Initial GRE-EPI before rt-pa therapy shows hypointensity on the left M1 segment (arrowhead). d Follow-up MRA after rt-pa therapy shows partial of the left MCA branch at the M3 segment (arrow). e, f Follow-up GRE- EPI reveals migratory embolus at the left M3 segment (arrowhead)

5 Neuroradiology (2012) 54: the distal segment with no laterality compared with the contralateral unaffected vessel), partial (in-flow effect from the initial occluded site to the distal segment shorter than that of contralateral unaffected vessel), or no (no new in-flow effect from the initial occluded site to the distal segment). We classified susceptibility signs on follow-up GRE-EPI as disappearance (not detected on follow-up GRE-EPI), migration (more distal on follow-up compared with initial GRE-EPI), extension (elongated distally on follow-up GRE-EPI), or no change (initial and follow-up GRE-EPI findings in the same location). The images of all selected patients were retrospectively reviewed by two neuroradiologists (Y.S. and T.K.) who reached consensus regarding image interpretation. Results Occluded arteries on initial MRA corresponded to acute ischemic lesions on DWI in all 20 included patients. Occluded sites on initial MRA were compatible with the proximal sites of susceptibility signs on initial GRE-EPI in 19 of 20 patients (Table 1). Among the 19 patients, susceptibility signs on GRE-EPI were found along multiple segments continuously in 4 patients (patients 4, 13, 14, and 20), while occlusion sites at MRA were consistent with the proximal sites of susceptibility signs. According to the classification in this study, the other one patient (patient 11) showed a different section between the occlusion site at initial MRA (Lt distal M1) and the proximal portion of the susceptibility sign at initial GRE-EPI (Lt M2). However, the difference was likely to be regarded as almost compatible because the proximal portion of the susceptibility sign at GRE-EPI case was very close to the distal M1 portion. Changes in susceptibility signs on follow-up GRE-EPI were consistent with vascular status on follow-up MRA in 19 of the 20 patients. The susceptibility signs disappeared with complete in 13 patients (including 4 with rt-pa therapy; Fig. 1), migrated with partial in 3 (including 1 with rt-pa therapy; Fig. 2), and remained unchanged with no in 2. The susceptibility sign had extended from the M1 to the M1 M2 segment on follow-up GRE-EPI 1 day after the initial study in one patient with rt-pa therapy, although simultaneous follow-up a b c d Fig. 3 Imaging findings from patient 19. a Initial MRA reveals left M1 occlusion (arrow). b, c Initial GRE-EPI shows strong hypointensity on the left M1 segment (arrowhead). d f Follow-up GRE-EPI e f shows distal growth of susceptibility sign towards the left M2 segment (arrowheads), although follow-up MRA shows no of occlusion site (arrow)

6 432 Neuroradiology (2012) 54: MRA showed no from the initial occlusion site (patient 19; Fig. 3). Findings from one patient between follow-up GRE-EPI and MRA were inconsistent as the susceptibility sign had become obscured by hemorrhage due to partial and decompression craniectomy (patient 20; Fig. 4). The duration between initial assessment to follow-up GRE-EPI was likely to be shorter among patients with migrated or extended susceptibility signs on follow-up GRE-EPI. The intervals for four patients with migrated or extended susceptibility signs ranged from 1 to 2 (mean, 1.3) days, including two with rt-pa therapy. Discussion Evaluating the status of an embolus during follow-up for acute ischemic stroke is important to assess therapeutic responses and estimate prognosis, such as risk for hemorrhagic transformation by or brain edema due to persistent occlusion [11]. The Diffusion and Perfusion Imaging Evaluation for Understanding Stroke Evaluation study found that early reperfusion is closely associated with symptomatic ICH [13]. Kimura et al. reported that the presence of between 1 and 24 h after rt-pa infusion is independently associated with ICH, whereas early within 1 h after rt-pa infusion improves neurological recovery and decreases the occurrence of ICH [14]. Other studies using follow-up GRE T2*-weighted imaging to evaluate susceptibility signs have shown that the changes are almost completely compatible with status on follow-up MRA [1, 2]. However, the diagnostic value of susceptibility signs on serial GRE-EPI assessments has not been reported. The present study found good agreement between susceptibility signs on follow-up GRE-EPI and status on follow-up MRA. GRE-EPI has some advantages over GRE for detecting susceptibility signs. One is the short acquisition time of GRE-EPI which minimizes the risk of artifacts arising due to patient movement compared with GRE or MRA [3, 5, 7, 12]. In our institution, the acquisition time of GRE-EPI for the entire brain is 4 s, while those of GRE and MRA are 1 min 46 s and 7 min 10 s, respectively. Another is good T2* sensitivity on GRE-EPI due to increased blooming of the embolus with susceptibility effect. The blooming effect is more obvious on GRE-EPI than on GRE [12]. The enlarged susceptibility effect on GRE-EPI associated with an embolus might have contributed to the ability to detect emboli in more distal or smaller vessels in our patients. On the other Fig. 4 Imaging findings from patient 20. a Initial MRA reveals right M1 occlusion (arrow). b Initial GRE-EPI shows strong hypointensity on the right M1 M2 segment (arrowhead). c Follow-up MRA shows of the right M1 segment, which has faint in-flow effect compared to contralateral unaffected right MCA. d Follow-up GRE-EPI shows hemorrhage due to the partial and decompressive craniectomy, which makes it difficult to evaluate changes in susceptibility sign a b c d

7 Neuroradiology (2012) 54: hand, there is a well-established drawback of GRE-EPI. The image of GRE-EPI near the skull base is more distorted than that of GRE because of the strong susceptibility artifact, which may decrease the detectability of susceptibility sign near the skull base, such as distal ICA, basilar-top, and proximal PCA segments [3, 12]. Although GRE-EPI is likely to have sensitivity comparable to GRE, further study of a side-by-side comparison of GRE-EPI with GRE in the same patients with acute embolic infarction will be warrantedinthisregard. The susceptibility sign had become extended on followup GRE-EPI in one patient without on follow-up MRA. This might indicate a migrated embolus or a secondary thrombus formed at the distal portion of occluded vessels without changing the proximal side of the embolus. In addition, the interval from initial to follow-up MR examination in this study for patients with migratory or extended susceptibility signs tended to be short. Although the number of patients with migratory or extended susceptibility sign was small in this study, a short interval before follow-up GRE-EPI might lead to the earlier detection of additional radiological findings in occluded vessels compared with follow-up MRA. Further studies are needed to clarify whether these early findings on follow-up GRE-EPI should influence a decision regarding the choice of therapeutic strategies such as intravenous rt-pa injection or mechanical thrombectomy. Although follow-up GRE-EPI was useful in this study, evaluating changes in an embolus using GRE-EPI alone requires further attention. Hypointense lesions on GRE-EPI are required to differentiate an embolus from other paramagnetic substances such as hematomas including microbleeds, intracranial large veins, and arterial wall calcification. The susceptibility sign was obscured on follow-up GRE-EPI due to hemorrhage in one patient in this study. Moreover, a cortical vessel sign on GRE-EPI, which reflects deoxygenation of the superficial cerebral veins in ischemic territory, might be confused with a peripheral susceptibility sign [15]. A susceptibility sign on GRE-EPI must be compared with an in-flow effect on MRA or a flow void on T2-weighted imaging to accurately evaluate changes in an embolus. Our study has several limitations. First, MRA is less accurate as a reference for vessel occlusion and than catheter angiography. This is because MRA overestimates the severity of arterial stenosis caused by saturation with slow flow and thus might produce a false diagnosis of vascular occlusion, although this modality has proven accurate in evaluating vessel patency [16, 17]. Second, we did not consider whether the emboli were platelet-rich white or fibrin-rich red types. Although the GRE-EPI sequence shows a powerful susceptibility effect, different features of emboli might have influenced the findings regarding susceptibility signs on GRE-EPI [6, 8]. Third, we did not include the patients with petrous, precavernous, and cavernous ICA occlusion or basilar-top to PCA occlusion, which will have strong susceptibility artifact arising from the skull base. It may introduce bias into the evaluation for detectability of susceptibility sign on GRE- EPI. These patients should be included when the detectability of susceptibility sign on GRE-EPI is investigated and compared to other T2* sequences on further study. In conclusion, susceptibility signs on follow-up GRE- EPI can reflect changes in an acute embolus, such as or migration. Serial GRE-EPI in acute embolic infarction complements the diagnostic certainty of follow-up MRA by directly detecting an embolus as a susceptibility sign. Acknowledgments We thank the technical staff of the Research Institute of Brain and Blood Vessels Akita. Conflict of interest References We declare that we have no conflict of interest. 1. Kim HS, Lee DH, Choi CG, Kim SJ, Suh DC (2006) Progression of middle cerebral artery susceptibility sign on T2*-weighted images: its effect on and clinical outcome after thrombolysis. AJR Am J Roentgenol 187: Assouline E, Benziane K, Reizine D, Guichard JP, Pico F, Merland JJ, Bousser MG, Chabriat H (2005) Intra-arterial thrombus visualized on T2* gradient echo imaging in acute ischemic stroke. Cerebrovasc Dis 20: Kinoshita T, Okudera T, Tamura H, Ogawa T, Hatazawa J (2000) Assessment of lacunar hemorrhage associated with hypertensive stroke by echo-planar gradient-echo T2*-weighted MRI. Stroke 31: Linfante I, Llinas RH, Caplan LR, Warach S (1999) MRI features of intracerebral hemorrhage within 2 hours from symptom onset. Stroke 30: Flacke S, Urbach H, Keller E, Träber F, Hartmann A, Textor J, Gieseke J, Block W, Folkers PJ, Schild HH (2000) Middle cerebral artery (MCA) susceptibility sign at susceptibility-based perfusion MR imaging: clinical importance and comparison with hyperdense MCA sign at CT. Radiology 215: Schellinger PD, Chalela JA, Kang DW, Latour LL, Warach S (2005) Diagnostic and prognostic value of early MR imaging vessel signs in hyperacute stroke patients imaged <3 hours and treated with recombinant tissue plasminogen activator. AJNR Am J Neuroradiol 26: Rovira A, Orellana P, Alvarez-Sabín J, Arenillas JF, Aymerich X, Grivé E, Molina C, Rovira-Gols A (2004) Hyperacute ischemic stroke: middle cerebral artery susceptibility sign at echo-planar gradient-echo MR imaging. Radiology 232: Cho KH, Kim JS, Kwon SU, Cho AH, Kang DW (2005) Significance of susceptibility vessel sign on T2*-weighted gradient echo imaging for identification of stroke subtypes. Stroke 36: Teitelbaum GP, Ortega HV, Vinitski S, Clark RA, Watanabe AT, Matsumoto AH, Rifkin MD, Barth KH (1990) Optimization of gradient-echo imaging parameters for intracaval filters and trapped thromboemboli. Radiology 174:

8 434 Neuroradiology (2012) 54: Atlas SW, Thulborn KR (1998) MR detection of hyperacute parenchymal hemorrhage of the brain. AJNR Am J Neuroradiol 19: del Zoppo GJ, Poeck K, Pessin MS, Wolpert SM, Furlan AJ, Ferbert A, Alberts MJ, Zivin JA, Wechsler L, Busse O, Greenlee R Jr, Brass L, Mohr JP, Feldmann E, Hacke W, Kase CS, Biller J, Gress D, Otis SM (1992) Recombinant tissue plasminogen activator in acute thrombotic and embolic stroke. Ann Neurol 32: Liang L, Korogi Y, Sugahara T, Shigematsu Y, Okuda T, Ikushima I, Takahashi M (1999) Detection of intracranial hemorrhage with susceptibility-weighted MR sequences. AJNR Am J Neuroradiol 20: Lansberg MG, Thijs VN, Bammer R, Kemp S, Wijman CA, Marks MP, Albers GW, DEFUSE Investigators (2007) Risk factors of symptomatic intracerebral hemorrhage after tpa therapy for acute stroke. Stroke 38: Kimura K, Iguchi Y, Shibazaki K, Kobayashi K, Uemura J, Aoki J, Yamashita S, Terasawa Y, Matsumoto N (2008) Recanalization between 1 and 24 hours after t-pa therapy is a strong predictor of cerebral hemorrhage in acute ischemic stroke patients. J Neurol Sci 270: Morita N, Harada M, Uno M, Matsubara S, Matsuda T, Nagahiro S, Nishitani H (2008) Ischemic findings of T2*-weighted 3-Tesla MRI in acute stroke patients. Cerebrovasc Dis 26: Hirai T, Korogi Y, Ono K, Nagano M, Maruoka K, Uemura S, Takahashi M (2002) Prospective evaluation of suspected stenoocclusive disease of the intracranial artery: combined MR angiography and CT angiography compared with digital subtraction angiography. AJNR Am J Neuroradiol 23: Stock KW, Wetzel S, Kirsch E, Bongartz G, Steinbrich W, Radue EW (1996) Anatomic evaluation of the circle of Willis: MR angiography versus intraarterial digital subtraction angiography. AJNR Am J Neuroradiol 17: