| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Neuroradiology |
1 From the Departments of Radiology (J.L.S., R.W.T., C.F.L., J.S.L.), Neurology (D.M.D.L.), and Neurosurgery (W.R.S.), University Hospitals of Cleveland and Case Western Reserve University, Lakeside Basement, 11000 Euclid Ave, Cleveland, OH 44106. Received August 14, 1998; revision requested October 15; revision received November 12; accepted January 25, 1999. Address reprint requests to J.L.S.
 |
Abstract |
|---|
 TOP Abstract IntroductionMATERIALS AND METHODSRESULTSDISCUSSIONReferences |
|---|
MATERIALS AND METHODS: T2-weighted turbo gradient- and spin-echo images and echo-planar diffusion- and perfusion-weighted images were obtained. Trace diffusion-weighted images and time-to-peak perfusion maps were automatically postprocessed and immediately available for interpretation.
RESULTS: Forty-one patients with acute stroke symptoms underwent imaging within 6 hours of symptom onset; 35 were eligible for the therapy protocol. The mean time from entering the emergency department to beginning MR imaging was 45 minutes; the mean total MR imaging time was less than 15 minutes. Immediate image analysis directly affected individual clinical management. Four patients showed evidence of no infarct; seven, of lacunar infarct; and 24, of acute cortical infarct. Sixteen patients underwent angiography, thirteen had large-vessel occlusion, eleven were treated intraarterially, and in seven, recanalization was achieved.
CONCLUSION: Echo-planar diffusion- and perfusion-weighted MR imaging for acute stroke is feasible and applicable before therapy decisions. Ultrafast MR imaging permitted immediate triage of 35 patients with symptoms of hyperacute stroke and thus helped avoid the risks from angiography and thrombolytic agents in some or spurred the judicious use of more aggressive intervention in others.
Index terms: Brain, infarction, 13.781, 13.782 • Brain, MR, 13.121411, 13.121412, 13.121413, 13.121416, 13.12143, 13.12144 • Magnetic resonance (MR), comparative studies, 13.121411, 13.121412, 13.121413, 13.121416, 13.12143, 13.12144 • Magnetic resonance (MR), diffusion study, 13.12144 • Magnetic resonance (MR), perfusion study, 13.12144
|
Introduction |
|---|
|
TOPAbstractIntroductionMATERIALS AND METHODSRESULTSDISCUSSIONReferences |
|---|
The addition of perfusion data in the evaluation of stroke was suggested even as the efficacy of diffusion-weighted imaging of stroke in humans was becoming apparent (9–12). Several groups (7,13,14) then showed the ability of ultrafast echo-planar diffusion- and perfusion-weighted MR imaging to demonstrate cerebral infarct and ischemia during the early hours after onset of stroke symptoms. The potential to illuminate an ischemic penumbra, or region of ischemic tissue at risk of infarct, by using relative subtraction of diffusion disruptions from perfusion deficits has been hypothesized previously (13,15).
First, we sought to test the feasibility of using diffusion- and perfusion-weighted MR imaging techniques within the extreme time constraints of stroke evaluation prior to the initiation of therapeutic intervention. We strongly believed that because every moment of delay carried the potential for greater brain tissue death, the total time devoted to MR imaging—including the time for transporting the patient to and from imaging, transferring the patient on and off the imager, imager setup, and data acquisition—must never exceed 20 minutes and preferably should conclude in 15 minutes. We believed equally strongly that if the information was to have any use in acute management decisions, then all postprocessing algorithms had to be online and capable of delivering output to the console within that same allotted time.
Then, we sought to assess the potential added diagnostic value these tests offered to patients in the setting of emergent stroke once we had achieved the technical and logistic goals. We hoped to prospectively stratify patients into the following three groups: those without MR imaging–confirmed cerebral ischemia in whom no additional risks were justified, those with evidence of ischemia exclusive to the cortical gray matter in whom we believed that risks of intravenous recombinant tissue-type plasminogen activator only were reasonable, and those with cortical ischemia for whom the greatest potential benefit existed and thus catheter angiography and potential intraarterial thrombolytic therapy were justified.
|
MATERIALS AND METHODS |
|---|
|
TOPAbstractIntroductionMATERIALS AND METHODSRESULTSDISCUSSIONReferences |
|---|
The diffusion-weighted images were obtained with a single-shot, gradient-echo, echo-planar pulse sequence with diffusion gradient b values of 0 and 1,000 sec/mm2 along all three orthogonal axes over 18 axial sections (TR, 6,000 msec; 18 7-mm-thick sections with no intersection gap; matrix, 96 x 128; field of view, 240 mm with one acquisition). Perfusion-weighted images were obtained during the first pass of a 0.1 mmol/kg bolus of gadolinium-based contrast material (Omniscan; Nycomed, Oslo, Norway or Magnevist; Berlex Laboratories, Wayne, NJ) by using a gradient-echo, echo-planar sequence set for a selected 12 of the 18 section positions measured 30 times sequentially, for a total imaging time of 49 seconds to ensure imaging during the entire duration of the first passage of contrast material. The following imaging parameters were used: per-section image acquisition rate, approximately seven axial images per second; TR, 1,680 msec; 4-mm-thick sections with a 0.5 intersection gap; matrix, 64 x 128; field of view, 300 mm with one acquisition.
The total image acquisition and reconstruction processing time with the described sequences was approximately 6 minutes. This permitted a total imaging time, including patient transfer to and from the table, patient positioning, and coil tuning, of 15–20 minutes. Diffusion-weighted image data were automatically postprocessed to yield trace images at each section position from the primary data. The perfusion-weighted sequence generated a time-to-peak map for each section position. These maps were immediately available for interpretation at the console with all the other images. All image interpretations were performed by a single attending-level board-certified radiologist (J.L.S., C.F.L., R.W.T., or J.S.L.) who has added qualifications in neuroradiology and training in reading diffusion- and perfusion-weighted MR images. The areas of hyperintensity were noted on the diffusion-weighted trace images, and areas of decreased (hyperintense) or increased (hypointense) perfusion were noted on the time-to-peak maps. The times of symptom onset, arrival in the emergency department, CT, MR, and angiography were tabulated for each patient.
The data collected included the site and distribution of ischemic changes on diffusion-weighted images, presence of concordant or discordant perfusion abnormality, and changes on T2-weighted images. These data were then used to decide for or against anticoagulation therapy, guide catheter placement toward the involved distribution, or decide whether to proceed with intraarterial thrombolysis. The images obtained in the patients in the stroke protocol group were immediately evaluated at the console and used to place these patients in one of the following three categories: no MR imaging evidence of infarct, evidence of infarct limited to the perforating arterial distributions only, or evidence of acute infarct involving cortical or brain stem tissue.
Thrombolytic therapy was administered to patients after MR imaging within the following guidelines. Intravenous recombinant tissue-type plasminogen activator was administered only within the first 3 hours following symptom onset. This therapy was used with a full dose of 0.9 mg/kg gadolinium-based contrast material in those patients with small-vessel (perforating arteries) disease. Patients with evidence of large-vessel (first- through fourth-order arteries) occlusion were given an initial dose of 0.6 mg/kg. These patients then underwent angiography for potential additional intraarterial lysis. Patients who presented with symptoms after 3 hours but before 6 hours were limited to intraarterial therapy. Urokinase, in aliquots of 125,000–250,000 U, was used in all intraarterial treatments, with an expected maximum of 1,000,000 U.
|
RESULTS |
|---|
|
TOPAbstractIntroductionMATERIALS AND METHODSRESULTSDISCUSSIONReferences |
|---|
Diffusion-weighted imaging was far more temporally and spatially sensitive for the earliest changes in cerebral ischemia in the acute setting relative to conventional T2-weighted or FLAIR imaging. Diffusion-weighted imaging alone demonstrated 26 cortical infarcts in 24 patients (Fig 1a). In 10 patients, the images showed ischemia in 13 deep gray matter structures. Abnormally increased signal intensity appeared in the deep white matter regions consistent with acute small-vessel ischemia in six patients (Fig 1b). Six patients had no visible abnormalities. Signal intensity changes in the same regions were apparent prospectively on T2-weighted images (or turbo FLAIR images when acquired) in two patients (Fig 1).
|
|
|
|
|
For the 35 patients entered into our stroke protocol group for potential therapy, imaging represented a potential delay in therapy, so we tracked our time efficiency. For patients transferred from the emergency department as candidates for potential intraarterial thrombolysis, we summed the time spent for taking the clinical history, performing the physical examination, taking vital signs, performing blood work, performing electrocardiography, and performing nonenhanced head CT prior to MR imaging. The average time expended, from entering the emergency department to beginning MR imaging, was 45 minutes (range, 15–75 minutes).
Furthermore, we logged the total time in the MR suite, from leaving the emergency department to returning from MR imaging. In all but three patients, the total MR imaging time was less than 15 minutes. These three patients had extreme morbidity and were delayed because of ventilation constraints prior to adequate training of all respiration therapists in the use of MR-compatible units. Thus, those patients who required angiography arrived in the interventional angiography suite 1 hour or sooner after arriving in the emergency department, with all evaluations, including CT and MR imaging, completed. In sum, we were able to perform imaging in 35 patients during their acute assessment period and prior to the final therapeutic decision, with only a limited time delay.
In the 35 stroke patients, we were able to compare perfusion deficits to regions with abnormal diffusion to produce an estimate of brain tissue at risk, or penumbra. Immediate imaging analysis directly affected individual clinical management; we were able to triage critical patients into groups directed toward and away from aggressive therapy with its inherent risks (Table). Four patients were triaged into group 1, in which no infarct was evident following image interpretation, and our therapy protocol was effectively terminated despite suggestive symptoms. As an example, one of these four patients subsequently demonstrated seizure activity, which confirmed the presence of a Todd paralysis that had mimicked an acute stroke, and further risk associated with therapy was avoided. Seven patients were triaged into group 2, in which imaging demonstrated the presence of acute lacunar infarcts in deep gray or white matter structures (Fig 3). For these individuals, treatment was limited to intravenous therapy for two patients, who presented with symptoms within the initial 3-hour window, or to conservative options for five patients, who presented at later time periods.
|
|
|
|
|
|
|
DISCUSSION |
|---|
|
TOPAbstractIntroductionMATERIALS AND METHODSRESULTSDISCUSSIONReferences |
|---|
To our knowledge, for the first time, the early availability of this information permitted the immediate triage of 35 patients with hyperacute symptoms and thus influenced the decisions of whether to treat, whether to administer only intravenous thrombolytic agents, and whether to advance to angiography and attempt intraarterial thrombolysis. Therefore, diffusion- and perfusion-weighted MR imaging can be performed within the time constraints of an acute stroke protocol and provide valuable data to guide critical patient triage and therapeutic decisions.
This study advances previous efforts along several lines. Foremost, we performed imaging in the majority of our patients prospectively, during the interval of the therapeutic decision rather than following it. This allowed the MR images to be of additional diagnostic value to the acutely ill patient, and in some cases, it enabled us to avoid risky interventions when they were not necessary. Furthermore, compared with previous work (14), multisection analysis of each patient's brain, in both the diffusion- and perfusion-weighted modes, was used in this study, which minimized sampling errors. It is worth noting that we administered intraarterial thrombolytic agents to 11 (69%) of the patients who underwent angiography in the population reported herein. In comparison, the findings in a previous series of acute stroke angiograms, which were obtained without MR imaging being performed beforehand, resulted in the administration of intraarterial thrombolytic agents to 51 (38%) of the 136 patients who underwent angiography. Our use of MR imaging to triage patients resulted in a much higher rate of true-positive angiograms and thus helped us to clearly exclude a number of patients in whom risky invasive treatment was not necessary.
Diffusion-weighted images can be obtained and viewed variably. We used a set of parameters designed to maximize lesion conspicuity and minimize both acquisition and processing times. For these reasons, we obtained images in only three orthogonal planes, and from these, trace images were generated. Although the use of apparent diffusion coefficient maps in the evaluation of acute stroke has been reported (13,16), these cannot yet be generated on our system in a manner that is applicable to our time constraints.
The reader should note that perfusion data obtained during the first passage of the contrast material bolus can be evaluated by examining multiple variables other than the time-to-peak effect, as was used here. Currently at our institution, a time-efficient processing algorithm for perfusion-weighted MR imaging is available for the time-to-peak option only. Measurements of mean transit time, as well as relative cerebral blood volumes or flow, are possible (12,13,17). However, in our experience, these have not proved to be as sensitive or conspicuous in the acute setting. For some, the transfer of data offline for processing, which is not practical for hyperacute triage, has been required to obtain these measurements. For others, these measurements have involved longer reconstruction times—for example, 10 minutes for one group (13), which alone exceeds our total acquisition and processing time. However, in at least one center (18), relative cerebral blood flow and mean transit time maps are generated from multiple sets of 11 sections in less than 4 minutes. Although this is not truly time prohibitive, it is approximately twice the time required to process our image data. We continue to obtain high-speed turbo GRASE images because of their sensitivity to blood susceptibility effects and because of the requirement to exclude the "shine through" of T2-weighted signal intensity that occurs on diffusion-weighted images (19).
We have encountered patients with no or minimal abnormal signal intensity on diffusion-weighted images who ultimately had very large infarcts that involved whole cerebral vessel distributions. These patients had marked perfusion abnormalities at these distributions. Thus, in our series, diffusion-weighted imaging alone was, on occasion, insufficiently sensitive to allow exclusion of substantial risk of infarct (Fig 2). This finding confirms those in other reports (13,20) yet demonstrates further the potential to miss large volumes of tissue that may ultimately become infarcted if only diffusion-weighted images are relied on.
Our study results confirm the finding in previous reports (16,20) that most unremarkable diffusion-weighted images help delineate patients who present with stroke symptoms but do not have cerebral ischemia and will recover without intervention, but only when these images are coupled with normal perfusion-weighted images. Furthermore, in several instances in our study, MR images alone demonstrated the presence of cortical tissue at risk of infarct when the clinical assessment suggested the presence of only small deep white matter at risk. Although these patients may be treated with intravenous thrombolytic therapy efficaciously within the first 3 hours after symptom onset (21), we currently will not use the invasive, aggressive, and ultimately successful measures described without MR imaging evidence of their necessity (Fig 4).
Although MR imaging offers clear advantages over the more widely available CT scanning (22,23), CT still can be used to characterize early stroke. The "dense middle cerebral artery sign" at CT continues to indicate a poor prognosis that possibly can be improved with aggressive therapy (24,25). CT, in combination with xenon gas inhalation, can be used for imaging and quantitative measurement of cerebral perfusion (26). Furthermore, in settings where MR images may not be obtainable, CT data may help predict outcome (27). Alternatively, because MR imaging can help identify hemorrhagic stroke on T2-weighted images (28), there may come a time when CT can be safely bypassed in the hyperacute evaluation of stroke.
Currently, MR perfusion and diffusion techniques demonstrate cerebral infarct and ischemia earliest and best. For example, FLAIR imaging techniques have shown improvement over conventional T1- and T2-weighted imaging in cerebral infarct evaluation (29), but the FLAIR sequence has not shown the early temporal sensitivity of diffusion- and perfusion-weighted sequences when performed in our patient populations. Furthermore, MR spectroscopy can help identify the abnormal accumulation of intracellular lactate and the associated drop in the pH level that occurs with ischemic damage to cellular sodium-potassium pumps (18). However, this technique's current lack of spatial resolution renders it much less useful in delineating the boundaries of parenchymal ischemia.
Our work could be strengthened in several ways in the future. First, a greater degree of follow-up imaging would better test the estimation of penumbra. In those patients who do not undergo successful therapeutic intervention, images obtained later would be expected to demonstrate an infarct size that approaches that of the perfusion deficit. Similarly, in those patients in whom arterial therapy produces successful recanalization, later images should show infarcts that are no larger than the region of diffusion abnormality. The addition of online, rapidly available apparent diffusion coefficient maps (13,30) potentially could allow better identification of ischemic cores because of the additional demarcation of threshold boundaries within the regions of abnormal diffusion. Absolute quantification may be possible (31).
Similarly, the quantification of perfusion and regions of deficit (32), and, thus, the acquisition of better images of the degrees of ischemic risk, may be possible in the future. In addition, if our treatment populations become larger, then a certain number of parenchymal hemorrhages will be encountered. Review of initial imaging findings from the subset of patients with parenchymal hemorrhage may allow the identification of imaging conditions that are suggestive of a higher risk of bleeding, and this could lead to improved pretreatment stratification of patients. The results of current animal studies (33) suggest that early gadolinium-based contrast material enhancement following reperfusion may represent an increased risk of hemorrhagic transformation. Similarly, because an underlying threat of increased tissue damage from reperfusion injury exists (34), we should encounter a group of patients whose condition worsens despite adequate therapy. Again, a review of their initial images may reveal predictive characteristics.
In conclusion, rapidly acquired and processed echo-planar diffusion- and perfusion-weighted MR imaging can be used successfully in patients with the most acute symptoms of stroke. The available image and image map data can then be incorporated into emergency triage protocols. This may exclude some patients from the risks of invasive testing and hemorrhage from thrombolytic agents or spur the judicious use of more aggressive interventions to improve the outcome of patients with the greatest burden of risk based on the natural history of their disease.
|
Acknowledgments |
|---|
|
Footnotes |
|---|
Author contributions: Guarantor of integrity of entire study, J.L.S.; study concepts, J.L.S.; study design, J.L.S., R.W.T., J.S.L.; definition of intellectual content, J.L.S., R.W.T.; literature research, J.L.S.; clinical studies, all authors; data acquisition, J.L.S., D.M.D.L.; data and statistical analyses, J.L.S.; manuscript preparation, J.L.S.; manuscript editing, J.L.S., R.W.T., J.S.L.; manuscript review, J.L.S., R.W.T., C.F.L., J.S.L.
|
References |
|---|
|
TOPAbstractIntroductionMATERIALS AND METHODSRESULTSDISCUSSIONReferences |
|---|
This article has been cited by other articles:
|
|
 |
|
  S. W. Youn, J. H. Kim, Y.-C. Weon, S. H. Kim, M.-K. Han, and H.-J. Bae Perfusion CT of the Brain Using 40-mm-Wide Detector and Toggling Table Technique for Initial Imaging of Acute Stroke Am. J. Roentgenol., September 1, 2008; 191(3): W120 - W126. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 K. Shanmuganathan, R.P. Gullapalli, J. Zhuo, and S.E. Mirvis Diffusion Tensor MR Imaging in Cervical Spine Trauma AJNR Am. J. Neuroradiol., April 1, 2008; 29(4): 655 - 659. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
H.S. Kim, S.Y. Kim, and J.M. Kim Underestimation of Cerebral Perfusion on Flow-Sensitive Alternating Inversion Recovery Image: Semiquantitative Evaluation with Time-to-Peak Values AJNR Am. J. Neuroradiol., November 1, 2007; 28(10): 2008 - 2013. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
R.G. Gonzalez Imaging-guided acute ischemic stroke therapy: From "time is brain" to "physiology is brain". AJNR Am. J. Neuroradiol., April 1, 2006; 27(4): 728 - 735. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 P L Tan, D King, C J Durkin, T M Meagher, and D Briley Diffusion weighted magnetic resonance imaging for acute stroke: practical and popular. Postgrad. Med. J., April 1, 2006; 82(966): 289 - 292. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
P.W. Schaefer, L. Roccatagliata, C. Ledezma, B. Hoh, L.H. Schwamm, W. Koroshetz, R.G. Gonzalez, and M.H. Lev First-Pass Quantitative CT Perfusion Identifies Thresholds for Salvageable Penumbra in Acute Stroke Patients Treated with Intra-arterial Therapy AJNR Am. J. Neuroradiol., January 1, 2006; 27(1): 20 - 25. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 R. J. Seitz, S. Meisel, P. Weller, U. Junghans, H.-J. Wittsack, and M. Siebler Initial Ischemic Event: Perfusion-weighted MR Imaging and Apparent Diffusion Coefficient for Stroke Evolution Radiology, December 1, 2005; 237(3): 1020 - 1028. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 J M U-King-Im, R A Trivedi, M J Graves, K Harkness, H Eales, I Joubert, B Koo, N Antoun, E A Warburton, J H Gillard, et al. Utility of an ultrafast magnetic resonance imaging protocol in recent and semi-recent strokes J. Neurol. Neurosurg. Psychiatry, July 1, 2005; 76(7): 1002 - 1005. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 N. Hjort, K. Butcher, S.M. Davis, C.S. Kidwell, on behalf of the UCLA Thrombolysis Investigators, W.J. Koroshetz, J. Rother, P.D. Schellinger, S. Warach, and L. Ostergaard Magnetic Resonance Imaging Criteria for Thrombolysis in Acute Cerebral Infarct Stroke, February 1, 2005; 36(2): 388 - 397. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
K. Shanmuganathan, R. P. Gullapalli, S. E. Mirvis, S. Roys, and P. Murthy Whole-Brain Apparent Diffusion Coefficient in Traumatic Brain Injury: Correlation with Glasgow Coma Scale Score AJNR Am. J. Neuroradiol., April 1, 2004; 25(4): 539 - 544. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
C. S. Kidwell, J. R. Alger, and J. L. Saver Beyond Mismatch: Evolving Paradigms in Imaging the Ischemic Penumbra With Multimodal Magnetic Resonance Imaging Stroke, November 1, 2003; 34(11): 2729 - 2735. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
L. Restrepo, G. Pradilla, R. Llinas, and N. J. Beauchamp Perfusion- and Diffusion-Weighted MR Imaging-Guided Therapy of Vertebral Artery Dissection: Intraarterial Thrombolysis through an Occipital Vertebral Anastomosis AJNR Am. J. Neuroradiol., October 1, 2003; 24(9): 1823 - 1826. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
R. T. Higashida and A. J. Furlan Trial Design and Reporting Standards for Intra-Arterial Cerebral Thrombolysis for Acute Ischemic Stroke Stroke, August 1, 2003; 34 (8): e109 - e137. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
R. E. Latchaw, H. Yonas, G. J. Hunter, W. T.C. Yuh, T. Ueda, A. G. Sorensen, J. L. Sunshine, J. Biller, L. Wechsler, R. Higashida, et al. Guidelines and Recommendations for Perfusion Imaging in Cerebral Ischemia: A Scientific Statement for Healthcare Professionals by the Writing Group on Perfusion Imaging, From the Council on Cardiovascular Radiology of the American Heart Association Stroke, April 1, 2003; 34(4): 1084 - 1104. [Full Text] [PDF]
|
|
|
|
 |
|
 T. W. Stadnik, P. Demaerel, R. R Luypaert, C. Chaskis, K. L. Van Rompaey, A. Michotte, and M. J. Osteaux Imaging Tutorial: Differential Diagnosis of Bright Lesions on Diffusion-weighted MR Images RadioGraphics, January 1, 2003; 23(1): e7 - e7. [Abstract] [Full Text]
|
|
|
|
|
|
L. Restrepo, R. J. Wityk, M. A. Grega, L. Borowicz Jr, P. B. Barker, M. A. Jacobs, N. J. Beauchamp, A. E. Hillis, and G. M. McKhann Diffusion- and Perfusion-Weighted Magnetic Resonance Imaging of the Brain Before and After Coronary Artery Bypass Grafting Surgery Stroke, December 1, 2002; 33(12): 2909 - 2915. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
H. J. Cloft, T. A. Tomsick, D. F. Kallmes, J. H. Goldstein, and J. J. Connors Assessment of the Interventional Neuroradiology Workforce in the United States: A Review of the Existing Data AJNR Am. J. Neuroradiol., November 1, 2002; 23(10): 1700 - 1705. [Full Text] [PDF]
|
|
|
|
