| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Neuroradiology |
1 From the Department of Neurology (R.J.S., S.M., P.W., U.J., M.S.) and Institute of Diagnostic Radiology (H.J.W.), Heinrich-Heine University of Düsseldorf, Moorenstrasse 5, D-40225 Düsseldorf, Germany. Received August 19, 2004; revision requested October 28; revision received November 8; accepted February 1, 2005. Supported by SFB 194 (TP A13), Kompetenznetz-Schlaganfall (BMBF, TP B5, TP C4), the Brain Imaging Center West (BMBF, TP4), and the Biomedizinisches Forschungszentrum of the Heinrich-Heine-University Düsseldorf. Address correspondence to R.J.S. (e-mail: seitz{at}neurologie.uni-duesseldorf.de).
 |
ABSTRACT |
|---|
 TOP ABSTRACT INTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONReferences |
|---|
MATERIALS AND METHODS: Local ethics committee approval and informed consent were obtained. On parametric maps obtained in 64 patients (mean age, 64 years ± 13 [standard deviation]; 37 men and 27 women) with acute middle cerebral artery infarction, lesion volumetry was performed to determine time to peak, mean transit time, cerebral blood volume, and apparent diffusion coefficient obtained within 3 hours of symptom onset. The infarct lesions were assessed on T2-weighted MR images obtained at follow-up on day 8. Cerebrovascular changes were determined on MR angiograms. Inferential and correlation statistics were used.
RESULTS: A perfusion delay of more than 6 seconds relative to the nonaffected hemisphere on time-to-peak maps helped to predict the lesion volume on T2-weighted images (r = 0.686, P < .001). In contrast, neither the volume nor the degree of the diffusion abnormality helped to predict the infarct volume (r < 0.46). This was because in one subgroup of patients there was an increase and in one subgroup there was a decrease in infarct volume on the T2-weighted images (P < .001). There was a greater prevalence (P < .02) of cerebral artery abnormalities in the patients with larger infarcts. Clinically, the neurologic impairment was more severe (P < .01) and the mean arterial pressure higher (P < .04) in these patients.
CONCLUSION: The results suggest that in acute stroke the severity of the initial ischemic event as determined on time-to-peak maps indicates hemodynamic compromise in addition to internal carotid artery or middle cerebral artery occlusion, because of abnormalities in other cerebral arteries.
© RSNA, 2005
|
INTRODUCTION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONReferences |
|---|
Nevertheless, it is difficult to make a prognosis based on these imaging findings since the "perfusion-diffusion mismatch" is not a stable phenomenon but rather is subject to a complicated time course of ischemic lesion evolution (9,11). This is owing to changes at both ends (ie, water diffusion and tissue perfusion). First, it has been shown that the ADC in the initial stage of ischemia continuously drops to reach its minimum value of about 50% between 24 and 48 hours after symptom onset. Thereafter, the ADC rises again and reaches values even higher than its normal value, and the lesions depicted on diffusion-weighted images also expand spatially after about 8 days (12,13). Perfusion-weighted images, by contrast, may depict a variety of temporal patterns depending on the time and degree of spontaneous or treatment-induced reperfusion of the tissue (9). Nevertheless, there is the notion that the perfusion-diffusion mismatch region may be a potential target for early reperfusion procedures, such as systemic thrombolysis with recombinant tissue plasminogen activator (rtPA) (14–16), a combination of rtPA with platelet glycoprotein IIb/IIIa receptor antagonists (17,18), or emergency carotid endarterectomy (19,20).
Results of recent observations have shown that brain infarcts may become as large as predicted from the ischemic abnormalities on the initial MR images (7,21), or they may regress (5,21) or deteriorate despite recanalization (18,22). We hypothesized that there are patients in whom ischemic lesion size decreases subsequent to treatment onset, while there are others in whom lesion size increases. Thus, the purpose of our study was to prospectively determine if the degree of acute perfusion or diffusion abnormalities measured prior to treatment onset help to predict the evolution of brain infarction on MR images.
|
MATERIALS AND METHODS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONReferences |
|---|
Subjects
Sixty-four consecutive patients (mean age, 64 years ± 13 [standard deviation]; age range, 27–91 years; 37 men and 27 women) experiencing a first stroke participated in this study. Inclusion criteria were as follows: (a) a first hemiparetic brain infarct in the middle cerebral artery territory, (b) acute stroke MR imaging protocol (described later) performed before treatment onset, with perfusion-weighted images that showed a lesion within the middle cerebral artery territory, and (c) follow-up T2-weighted MR imaging performed between day 6 and day 11 (mean, day 8), with images that showed only one brain lesion in the middle cerebral artery territory.
All patients underwent continuous long-term blood pressure monitoring for at least 24 hours and were administered medication according to their individual requirements. In all patients, C-reactive protein, blood glucose, fibrinogen, hematocrit, and platelet levels were assessed as part of our routine investigation. Five different subgroups of patients were identified according to the therapeutic regimen they underwent, as follows:
Acute carotid endarterectomy.—Ten patients (58 years ± 10; five men and five women) underwent emergency carotid endarterectomy, which is detailed elsewhere (20). The patients had acute stroke syndrome characterized by sudden onset of a profound hemiparesis, as assessed with the European Stroke Scale (23) (score: range, 34–96; median, 88), that was caused by occlusion of the contralateral internal carotid artery. These patients underwent surgery within 48 hours after symptom onset.
Thrombolysis.—Sixteen patients aged 71 years ± 14 (nine men and seven women) with severe neurologic impairment (European Stroke Scale score: range, 11–95; median, 31) underwent systemic thrombolysis, which was initiated in the stroke unit at our institution an average of 153 minutes (range, 100–210 minutes) after symptom onset. They were administered a body weight–adjusted dose of intravenous rtPA (Boehringer, Ingelheim, Germany) over 1 hour, with an initial bolus of 10% (24–26). Within the following 24 hours, these patients received 10 000 IU unfractionated heparin (Roche, Basel, Switzerland) intravenously and 3 g magnesium (Verla-Pharm, Tutzing, Germany) in saline per 24 hours.
Thrombolysis with rtPA and tirofiban.—Thirteen patients aged 63 years ± 15 (six men and seven women) were treated with intravenous rtPA followed immediately with tirofiban (MSD, Haar, Germany) as detailed elsewhere (the rtPA and tirofiban group) (18). These patients also had severe neurologic impairment (European Stroke Scale score: range, 12–84; median, 35). These patients were first administered an intravenous bolus of 20 mg rtPA, which was delivered 121 minutes (range, 60–270 minutes) after symptom onset; one patient was administered a bolus of 10 mg followed by a 1-hour infusion of an additional 40 mg rtPA. The rtPA treatment was followed with intravenous tirofiban. Tirofiban was administered according to a body weight–adjusted dose starting with a bolus of 0.4 (µg · kg–1)/min for 30 minutes, which was followed by the continuous infusion of 0.1 (µg · kg–1)/min, along with 10 000 IU unfractionated heparin and 3 g magnesium in saline per 24 hours. The tirofiban treatment was continued for at least 24 hours. It should be noted that the action of tirofiban can be well controlled because of its short half-life in plasma, approximately 1.6 hours, and the small risk of fatal or symptomatic intracranial hemorrhage (17,18).
Prevention of microemboli with tirofiban.—Eleven patients aged 63 years ± 7 (eight men and three women) had recurring or progressive stroke syndrome related to microemboli detected at transcranial Doppler sonography and therefore underwent treatment with tirofiban for at least 24 hours (up to 104 hours). As described previously, tirofiban was administered by means of a body weight–adjusted dose starting with a bolus of 0.4 (µg · kg–1)/min for 30 minutes, which was followed by continuous infusion of 0.1 (µg · kg–1)/min (18). The neurologic deficit was profound in all patients (European Stroke Scale score: range, 58–72; median, 61).
Patients not eligible for thrombolytic or surgical stroke therapy.—Fourteen patients aged 62 years ± 13 (nine men and five women) were moderately neurologically impaired (European Stroke Scale score: range, 39–99; median, 86). These patients had arrived at the hospital within 8 hours after stroke onset, but, because arrival was more than 3 hours after stroke onset or because of other medical contraindications, they were not eligible for systemic thrombolysis or acute carotid endarterectomy. These patients received a continuous infusion of 10 000 IU unfractionated heparin or high-dose heparin treatment because cardiac thromboembolism was suspected owing to atrial fibrillation and definite arrhythmia (n = 4); these patients also received 3 g magnesium in saline per 24 hours.
MR Imaging
The MR imaging protocol was part of our routine clinical protocol and was performed in patients as an initial investigation in the initial stage before any medical or surgical treatment. The protocol comprised the performance of T2- and diffusion-weighted sequences and an MR angiographic investigation and the acquisition of serial T2*-weighted image measurements, with 40 images obtained for the measurement of tissue perfusion (7,27). Imaging was performed by using a 1.5-T whole-body MR imager (Magnetom Vision; Siemens, Erlangen, Germany), and the total imaging time was approximately 20 minutes. All perfusion-, diffusion-, and T2-weighted MR imaging data were transferred to a workstation (Sparc II; Sun Microsystems, Santa Clara, Calif) with Windows NT (Microsoft, Redmond, Wash) and were analyzed offline by using computer programs that we have developed in-house (Stroketool; DIS-Digital Image Solutions, Cologne, Germany; available at www.digitalimagesolutions.de).
Perfusion-weighted MR imaging.—Perfusion-weighted imaging measurements were obtained by performing serial T2*-weighted single-shot echo-planar MR imaging sequences, with imaging of 12 sections 40 times every 2 seconds. The parameters were as follows: echo time, 54 msec; 12 sections with 5-mm section thickness; intersection gap, 1.5 mm; matrix, 128 x 128; field of view, 240 mm. At the time of the fourth imaging, a 15-mL bolus of gadopentetate dimeglumine (Magnevist; Schering, Berlin, Germany) was injected a rate of 5 mL/sec and was immediately followed with 15 mL of NaCl administered at the same rate. From the perfusion-weighted imaging data, time-to-peak (TTP) parameter maps were calculated as detailed previously (7). TTP here refers to the time between the start of imaging and the time at which the signal intensity reaches its minimum produced by the bolus of contrast agent administered at the T2*-weighted series of imaging.
By using the singular value decomposition algorithm described by Ostergaard et al (28–30), we also calculated maps of the mean transit time (MTT), the regional cerebral blood flow (rCBF), and the regional cerebral blood volume (rCBV) by using the logarithmic relation c(t) 
–ln[S(t)/S0]/TE, where c(t) is the concentration of contrast agent at time t, S(t) is the signal intensity at time t, S0 is the precontrast signal intensity, and TE is the echo time. The arterial input functions needed for the calculation based on the singular value decomposition algorithm to determine rCBF, MTT, and rCBV were defined interactively by selecting 5 pixels in the supraclinoidal portion of the internal carotid artery in the nonaffected hemisphere. The time course of this mean signal intensity was used as an arterial input function for the deconvolution of the tissue signal intensity–time curves, as suggested previously (29). To minimize interindividual variations, image intensity in supraventricular white matter in the nonaffected hemisphere was set to 22 mL/100 g/min (30,31). For this purpose, regions of interest on image sections through the basal ganglia and thalamus (38 mL) and dorsal to the lateral ventricle (33 mL) were used to determine the mean rCBF in the contralesional hemisphere (ie, the nonaffected hemisphere contralateral to the infarct) (S.M. and P.W., 10 and 3 years of experience in MR imaging, respectively).
Diffusion-weighted MR imaging.—Diffusion-weighted MR imaging was performed by using a single-shot echo-planar sequence with two b values, 0 and 1000 sec/mm2, in three dimensions in space—readout, phase-encoding, and section-selection directions—to result in four images per section. The imaging parameters were echo time, 100 msec; 20 transverse sections with 5-mm section thickness; intersection gap, 1.5 mm; matrix, 96 x 128; field of view, 240 mm; and imaging time, 26 seconds. Also, diffusion-weighted imaging maps were converted to quantitative maps of ADC according to the exponential relation S(b) = S(0) exp(–b · ADC), where S(b) is the signal intensity by using a diffusion weighting with the value b, and S(0) is the signal intensity with b = 0. To get the quantitative three-dimensional trace ADC image map, we calculated the average from the three ADC maps resulting from the images with diffusion weighting in the different directions in space: readout, section-selection, and phase-encoding directions.
T2-weighted MR imaging.—For T2-weighted imaging, we used a turbo gradient-echo spin-echo sequence, with 7040/115 (repetition time msec/echo time msec), echo train length of 69, 20 transverse sections obtained parallel to the base of the skull, section thickness of 5 mm, intersection gap of 1.5 mm, flip angle of 160°, field of view of 230 mm, matrix of 345 x 512, and imaging time of 1 minute 17 seconds.
MR angiography.—For MR angiography, we used a three-dimensional time-of-flight investigation with the following sequence parameters: 35/7.2; flip angle, 20°; 1.5-mm section thickness; matrix, 200 x 512; 108 sections acquired; field of view, 200 mm; total imaging time, 6 minutes 44 seconds. Data of the entire three-dimensional data set were computed to show six projections with a rotation angle of 180°, each in a lateral and a sagittal plane. The field of view included the distal and siphon parts of the internal carotid arteries, the distal parts of the vertebral arteries, the basilar artery, the circle of Willis, and the intracranial arteries up to the branches of their third portions.
Lesion Volumetry on Perfusion-weighted MR Images
Volumetric analysis of the ischemic brain tissue was performed on TTP, MTT, and rCBV maps from each patient independently by two of the investigators (S.M., P.W.) who were blinded to any treatment, clinical information, and timing of the MR imaging examinations. Mean voxel signal intensity determined by means of manually drawn regions of interest in the unaffected contralateral hemisphere excluding artifacts and ventricles (30–40 mL) served as a reference standard (7). The method of referring to the contralateral tissue minimized the variation among the individual patients. Three different parameters were chosen to assess the degree of ischemic perfusion impairment, as follows:
First, to delineate severe ischemia, we used the model-independent TTP maps and applied a time delay of 6 seconds after bolus arrival relative to the unaffected contralesional hemisphere. This threshold enabled prediction of the subsequent infarct lesion, as reported previously (7,27). Specifically, it was found that a delay of 4 seconds led to overestimation of the ischemic lesion volume, while a delay of 8 seconds or more depicted the ischemic lesion only partially. By referring to the nonaffected hemisphere, differences in bolus delay within the brain owing to individual cardiac output were treated as statistical fluctuations.
Second, we wanted to determine the entire portion of brain tissue with impaired perfusion. Therefore, we analyzed the MTT images. The MTT values, determined by means of regions of interest (30–40 mL) manually drawn in gray and white matter dorsal to the lateral ventricle in the nonaffected hemisphere, were in the range of 3–5 seconds in individual patients. To delineate the ischemic area on the affected image sections, we used a threshold of 5–7 seconds, which was 2 seconds longer than that of the reference. This step of 2 seconds equaled the temporal resolution of the dynamic MR data acquisition and, therefore, was the most sensitive measure of abnormal perfusion.
Third, the areas of maximal depression of perfusion in white and gray matter structures were identified best on the rCBV images. For volumetry, regions of interest drawn to cover gray and white matter dorsal to the lateral ventricle in the nonaffected hemisphere (30–40 mL) typically had a mean value of 3–4 mL/100 g. A threshold of 2 mL/100 g, which was below that of normal white matter, was used to indicate the maximal ischemia in both the cerebral cortex and the white matter on the affected image sections.
These automatically defined areas of impaired perfusion were superimposed onto the rCBF maps to measure the mean rCBF separately for the abnormal TTP, MTT, and rCBV areas across the image sections affected.
ADC Evaluation
On diffusion-weighted images (S.M., P.W.), a threshold of 20% above the signal intensity in the region of interest in the nonaffected hemisphere (30–40 mL) was used for delineating the lesion volume, as detailed previously (12). In addition, the mean ADC in lesions on diffusion-weighted images was determined. Also, the maximal ADC decrease was determined as follows: The individual voxels on each image section with the lowest ADC within the ischemic lesion were identified and were averaged over all affected images in each patient. To control for global variation of ADC among the patients, the percentage change of the lowest ADC (x10–3 mm2/sec) within the ischemic lesion was also calculated in comparison with the mean ADC (x10–3 mm2/sec) obtained from a region of interest in the nonaffected hemisphere (30–40 mL) in each patient.
For follow-up evaluation on day 8, lesions on the T2-weighted MR images were defined interactively in each patient by two independent observers (S.M. and P.W.) and were manually outlined. The lesions had an interobserver reliability of less than 7% (four of 64 patients).
MR Angiogram Analysis
The MR angiograms from the initial imaging protocol were analyzed independently by two investigators (R.J.S. and U.J., 7 and 5 years of experience in MR angiography, respectively) who were blinded to the patients and the treatment regimens. Evaluation comprised the distal parts of the internal carotid artery, the distal vertebral arteries, the basilar artery, and the anterior, middle, and posterior cerebral arteries in both cerebral hemispheres. Abnormalities were classified with respect to presence of occlusion or subtotal signal intensity loss in an artery, presence of circumscribed arterial stenosis, and reduced arterial flow signal intensity. Abnormalities were graded as follows: score of 0, no abnormality; score of 1, occlusion of the internal carotid or middle cerebral artery; score of 2, occlusion of the internal carotid and middle cerebral arteries; and score of 3, multiple abnormalities in at least three arteries in one or both cerebral hemispheres.
Statistical Analysis
Statistical analysis was performed (S.M., R.J.S.) by using statistical software (SPSS 10.1 for Windows; SPSS, Chicago, Ill). Two groups of patients were defined on the basis of enlargement (group 1) or shrinkage (group 2) of the infarct lesions on T2-weighted images on day 8 in comparison with the size of the acute ischemic lesions on diffusion-weighted images obtained before treatment onset. Explorative evaluation of the perfusion-weighted imaging and ADC data and the laboratory findings in patients was subsequently performed to determine possible predictors by using a two-tailed t test for independent samples and within group comparisons. Only group differences with P < .05 were considered significant. Also, prediction of the lesion size at T2-weighted imaging by using the diffusion- and perfusion-weighted imaging data was determined with use of correlation coefficients. Evaluation of the neurologic impairment, as assessed with the European Stroke Scale, among the patient groups was performed with the nonparametric Mann-Whitney test. Group comparison of the abnormalities in the cerebral arterial circulation was performed with the distribution-free Wilcoxon signed rank test. Interobserver errors of the T2-weighted imaging lesion data were calculated from mean square deviations and are given as percentages. Those percentages were averages over the individual square deviations relative to the corresponding mean values.
|
RESULTS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONReferences |
|---|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
DISCUSSION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONReferences |
|---|
In contrast, neither the volume of the pretreatment lesions on diffusion-weighted images nor the magnitude of the ADC changes in lesions on diffusion-weighted images were discriminating in the first 3 hours after stroke onset. This was an unexpected finding given the reported association between ADC changes and perfusion abnormalities (32). However, in the initial stage of ischemia, the lesions on diffusion-weighted images are inhomogeneous and differ from normal tissue in various ways, showing different dynamics of lesion evolution toward infarction (11). In the 6-hour period after stroke onset, even a small decrease in the ADC, in the range of a few percent, relative to the contralateral hemisphere was shown to accommodate infarct growth (21). In contrast, the comparable extent of the lesions on diffusion-weighted imagines and of the ADC depressions across the two patient groups in our study suggests that the 3-hour interval between stroke onset and MR imaging determined the manifestation of diffusion abnormalities in the ischemic brain lesions. Accordingly, diffusion-weighted imaging changes develop further in the presence of extensive ischemia, which probably requires a longer period of time than 3 hours.
As known from clinical and pathophysiologic studies, blood flow obstruction by means of vessel stenosis or thromboembolism impairs cerebral perfusion. Compared with other measures of cerebral perfusion, such as rCBF, MTT, or rCBV, TTP maps show the perfusion abnormality readily, since they are model-independent and can be calculated fast. Notably, the effect of the TTP lesion measurements for determining the acute neurologic deficit as well as of the infarct volume for determining the resulting neurologic deficit found in this study is also well in line with results of previous studies from our and from other laboratories (33–35).
In this study, we were able to show that the initial ischemic event resulted in addition to acute middle or internal cerebral artery occlusion from widespread arterial changes, such as vessel stenosis or occlusion in multiple cerebral arteries at MR angiography. MR angiography has recently been established as a valid and sensitive tool to demonstrate pathologic findings in extra- and intracranial arteries in acute stroke (36,37). Our findings suggest that the acute macrovascular occlusion of the middle cerebral artery causing the acute neurologic stroke syndrome apparently induced a particularly devastating perfusion deficit when a compensatory redistribution of arterial blood along collateral vessels was impaired. This supports earlier findings determined with transcranial Doppler sonography (38), conventional angiography (39), and positron emission tomography (PET) measurements of oxygen extraction (40) that showed the importance of viable collateral vessels in the circle of Willis for a beneficial outcome after ischemia. Recently, the importance of viable collateral vessels was also shown for intraarterial thrombolysis (41).
Surprisingly, the normal values of hematocrit, platelet counts, fibrinogen, C-reactive protein, and blood glucose, which are well-recognized cerebrovascular risk indicators, did not separate the patient groups with larger infarctions from those with smaller infarctions. However, an elevated mean arterial pressure was found in the patients with lesion volume increase at T2-weighted imaging (group 1). Although our blood pressure recordings were performed after stroke and therefore were influenced by it, our observation supports high blood pressure as a risk factor for stroke and stroke outcome, although there is no linear relationship (42). Our findings of an excess of the area of severe hypoperfusion relative to the early lesion on diffusion-weighted images substantiates the concept of the perfusion-diffusion mismatch as a quantitative and reliable measure (10,27) with important pathophysiologic and therapeutic implications, as we will discuss.
Perfusion imaging is the method of choice for demonstrating ischemic brain tissue in the hyperacute stage of stroke (27,43). We used parametric maps of cerebral perfusion that were calculated from the dynamically recorded MR imaging data, as established by Ostergaard et al (28,29). The maps closely corresponded to rCBF measurements obtained with PET in healthy subjects (30) and allowed us to estimate a normal rCBF of 58 mL/100 g/min in the nonaffected hemisphere (44) and severely depressed rCBF values in the different portions of the acute ischemic lesions. We found that the central infarct portions with the maximal perfusion abnormality, which progress irreversibly to infarction (44,45), and the outer areas with only a slight perfusion abnormality were similar among the two patient groups. To determine the area of maximal perfusion impairment, we used parametric rCBV maps since they show a lack of perfusion affecting both gray and white matter to a similar degree. Conversely, the entire area of impaired perfusion was sensitively assessed on parametric MTT maps without difference between gray and white matter. A similar grading of the perfusion impairment can be obtained from the model-independent TTP maps by using different delays in comparison with the nonaffected hemisphere (7,27), but the estimates of this semiquantitative measure have no biologic scale.
However, the volume of severe ischemia determined at a TTP delay of some 6 seconds relative to the nonaffected hemisphere helped to predict the definite infarct lesion (7,27,44,46,47). The ischemic lesions at this threshold were different among the two groups. The rCBF in the TTP lesions corresponded to a severely depressed rCBF of 25 mL/100 g/min, which is in the range of normal white matter. At this degree of hypoperfusion, cortical neurons are known to release the excitatory glutamate as a critical step in the initiation of excitotoxicity (44,48). Thus, irrespective of the different treatment regimens in our patients, the TTP threshold of 6 seconds seemed to indicate the viability threshold of ischemic brain tissue (14,49,50). Our data are consistent with recent observations (43,51) of lesions of considerable size on T2-weighted images on day 8 in patients who were successfully treated with rtPA because of thromboembolic stroke. Thus, despite the clinical improvement, thrombolysis apparently cannot completely prevent the transition of ischemic tissue to necrosis. One may speculate that an additional application of antiglutamatergic or other neuroprotective drugs may enhance the amount of salvageable brain tissue threatened by ischemia.
Finally, the ischemic brain lesions in which smaller regions had TTP greater than 6 seconds induced a lesser acute neurologic deficit and a faster and better neurologic recovery than did ischemic brain lesions in which larger regions had TTP greater than 6 seconds. This corresponded to previous observations showing that larger middle cerebral artery infarctions induce a more severe neurologic deficit (33,35,52). However, even the patients with more severe ischemic lesions improved to a degree comparable to that of the patients with smaller ischemic lesions. It should be noted that recovery from hemiplegia and hemiparesis continues for at least 4–6 weeks after stroke (33).
Our findings are limited by the fact that this was an analysis of two relatively small patient groups subjected to different treatment regimens. Moreover, follow-up perfusion studies are expected to show whether the onset or rate of reperfusion may be different between the patients with lesion regression (group 2) and those with volume increase (group 1) at T2-weighted imaging. Finally, voxel-by-voxel coregistrations of the different MR images should be suited to show which portions of the ischemic lesions are actually salvaged from tissue necrosis.
In conclusion, our data showed that ischemic lesions with a large proportion of severe ischemia (TTP, >6 seconds) exhibited widespread MR angiographic abnormalities in several cerebral arteries, which resulted in hemodynamic compromise. Consequently, systemic thrombolysis may act on the ischemic area not only by resolving the thromboembolus that has obstructed the middle cerebral artery but also by improving blood circulation in the net of collateral vessels. This is probably why the combination of systemic thrombolysis with platelet receptor antagonists has been shown to be particularly effective in acute thromboembolic infarction of the middle cerebral artery (18,53). Thus, for future trials, a TTP of more than 6 seconds appears to be an easy-to-determine parametric surrogate marker for efficacy of acute stroke treatment.
|
ACKNOWLEDGMENTS |
|---|
|
FOOTNOTES |
|---|
Abbreviations: ADC = apparent diffusion coefficient • MTT = mean transit time • rCBF = regional cerebral blood flow • rCBV = regional cerebral blood volume • rtPA = recombinant tissue plasminogen activator • TTP = time to peak
Authors stated no financial relationship to disclose.
Author contributions: Guarantors of integrity of entire study, R.J.S., H.J.W., M.S.; study concepts/study design or data acquisition or data analysis/interpretation, all authors; manuscript drafting or manuscript revision for important intellectual content, all authors; approval of final version of submitted manuscript, all authors; literature research, R.J.S., H.J.W., M.S.; clinical studies, R.J.S., U.J., M.S.; statistical analysis, S.M., R.J.S.; and manuscript editing, R.J.S., U.J., M.S.
|
References |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONReferences |
|---|
This article has been cited by other articles:
|
|
 |
|
  O. Zaro-Weber, W. Moeller-Hartmann, W.-D. Heiss, and J. Sobesky The Performance of MRI-Based Cerebral Blood Flow Measurements in Acute and Subacute Stroke Compared With 15O-Water Positron Emission Tomography: Identification of Penumbral Flow Stroke, July 1, 2009; 40(7): 2413 - 2421. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 P.G. Kranz and J.D. Eastwood Does Diffusion-Weighted Imaging Represent the Ischemic Core? An Evidence-Based Systematic Review AJNR Am. J. Neuroradiol., June 1, 2009; 30(6): 1206 - 1212. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
E. S. Roach, M. R. Golomb, R. Adams, J. Biller, S. Daniels, G. deVeber, D. Ferriero, B. V. Jones, F. J. Kirkham, R. M. Scott, et al. Management of Stroke in Infants and Children: A Scientific Statement From a Special Writing Group of the American Heart Association Stroke Council and the Council on Cardiovascular Disease in the Young Stroke, September 1, 2008; 39(9): 2644 - 2691. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 P. Weller, H.-J. Wittsack, M. Siebler, V. Homberg, and R. J. Seitz Motor Recovery as Assessed with Isometric Finger Movements and Perfusion Magnetic Resonance Imaging after Acute Ischemic Stroke Neurorehabil Neural Repair, September 1, 2006; 20(3): 390 - 397. [Abstract] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| RADIOLOGY | RADIOGRAPHICS | RSNA JOURNALS ONLINE |
) and group 2 (
). (a) Correlation of ischemic lesion volumes on the TTP maps at a delay of 6 seconds (TTP6s-Lesion) relative to the nonaffected contralateral hemisphere and the lesion on T2-weighted images (T2-lesions) (r = 0.686, P < .001; regression formula: y = 0.6x + 23.77). (b) No correlation of the acute lesion volumes on diffusion-weighted images (DWI-Lesion) and the lesion volumes on T2-weighted images (T2-Lesion) (r = 0.458). (c) No correlation of the maximal ADC depression in the ischemic lesions and the lesion volumes on T2-weighted images (T2-Lesion) (r = 0.286).
