This Article Abstract Full Text (PDF) All Versions of this Article: 38/10/2738    most recent STROKEAHA.106.480848v1 Alert me when this article is cited Alert me if a correction is posted Citation Map Services Email this article to a friend Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Request Permissions Citing Articles Citing Articles via HighWire Citing Articles via Google Scholar Google Scholar Articles by Fiehler, J. Search for Related Content PubMed PubMed Citation Articles by Fiehler, J. Pubmed/NCBI databases Medline Plus Health Information MRI Scans Stroke Related Collections CT and MRI Acute Cerebral Infarction Emergency treatment of Stroke Computerized tomography and Magnetic Resonance Imaging Intracerebral Hemorrhage Thrombolysis

(Stroke. 2007;38:2738.)
© 2007 American Heart Association, Inc.

Original Contributions

Bleeding Risk Analysis in Stroke Imaging Before ThromboLysis (BRASIL)

Pooled Analysis of T2*-Weighted Magnetic Resonance Imaging Data From 570 Patients

Jens Fiehler, MD; Gregory W. Albers, MD; Jean-Martin Boulanger, MD; Laurent Derex, MD; Achim Gass, MD; Niels Hjort, MD; Jong S. Kim, MD; David S. Liebeskind, MD; Tobias Neumann-Haefelin, MD; Salvador Pedraza, MD; Joachim Rother, MD; Peter Rothwell, MD, PhD; Alex Rovira, MD; Peter D. Schellinger, MD; Johannes Trenkler, MD for the MR STROKE Group

From the Department of Neuroradiology (J.F.), University Medical Centre, Hamburg, Germany; Stanford Stroke Center (G.W.A.), Palo Alto, Calif; Seaman Family MR Research Centre (J.-M.B.), Foothills Medical Centre, Calgary, Canada; Hôpital Neurologique (L.D.), Laboratoire CREATIS UMR 5515 CNRS and INSERM U 630, Hôpital Neurologique, Lyon, France; Neurologische Klinik (A.G.), Universitaetsklinikum Mannheim, Mannheim, Germany; Center for Functionally Integrative Neuroscience (N.H.), Department of Neuroradiology, Aarhus University Hospital, Aarhus, Denmark; Asan Medical Centre (J.S.K.), Seoul, Korea; UCLA Stroke Center (D.S.L.), University of California, Los Angeles, Calif; Department of Neurology (T.N.-H.), Institute of Neuroradiology, University Hospital, Frankfurt, Germany; Departments of Radiology and Neurology (S.P.), University Hospital, Girona, Spain; Neurologische Klinik (J.R.), Klinikum Minden, Minden, Germany; Stroke Prevention Research Unit (P.R.), Department of Clinical Neurology, University of Oxford, Oxford, England; MR Unit (A.R.), Department of Radiology, Institution Hospital Universitari Vall d’Hebron, Barcelona, Spain; Neurologische Klinik, Universitaetsklinikum Heidelberg, Kopfklinik, Heidelberg, and Neurologische Klinik, Universitaetsklinikum Erlangen (P.D.S.), Erlangen, Germany; and Institut für Radiologie (J.T.), Klinik für Neurologie, Landes-Nervenklinik Wagner-Jauregg, Linz, Austria.

Correspondence to Jens Fiehler, MD, Department of Neuroradiology, University Medical Center Hamburg-Eppendorf, Martinistrasse 52, 20246 Hamburg, Germany. E-mail fiehler{at}uke.uni-hamburg.de

	Abstract

Top Abstract Introduction Methods Results Discussion Appendix: References

 
Background and Purpose— There has been speculation that the risk of secondary symptomatic intracranial hemorrhage (SICH) may be increased after thrombolytic therapy in ischemic stroke patients who have cerebral microbleeds (CMBs) on T2*-weighted magnetic resonance imaging. Because of this concern, some centers withhold potentially beneficial thrombolytic therapy from these patients.

Methods— We analyzed magnetic resonance imaging data acquired within 6 hours after symptom onset from 570 ischemic stroke patients treated with intravenous tissue plasminogen activator in 13 centers in Europe, North America, and Asia. Baseline T2*-weighted magnetic resonance images were evaluated for the presence of CMBs. The primary end point was SICH, defined as clinical deterioration with an increase in the National Institutes of Health Stroke Scale score by 4 points, temporally related to a parenchymal hematoma on follow-up-imaging.

Results— A total of 242 CMBs were detected in 86 of 570 patients (15.1%). The number of CMBs ranged from 1 to 77 in the individual patient, with 5 CMBs in 6 of 570 patients (1.1%). Proportions of patients with SICH were 5.8% (95% CI, 1.9 to 13.0) in the presence of CMBs and 2.7% (95% CI, 1.4 to 4.5) in patients without CMBs (P=0.170, Fisher’s exact test), resulting in no significant absolute increase in the risk of SICH of 3.1% (95% CI, –2.0 to 8.3).

Conclusions— The data suggest that if there is any increased risk of SICH attributable to CMBs, it is likely to be small and unlikely to exceed the benefits of thrombolytic therapy. No reliable conclusion regarding risk in the rare patient with multiple CMBs can be reached.

Key Words: hemorrhage, intracranial • stroke • stroke management • therapy • thrombolysis

	Introduction

Top Abstract Introduction Methods Results Discussion Appendix: References

 
Thrombolytic therapy in acute stroke patients has been shown to be beneficial in large randomized trials.¹ Secondary postischemic symptomatic intracranial hemorrhage (SICH) remains the primary concern associated with the administration of thrombolytic drugs. In contrast, the risk of SICH after systemic thrombolytic therapy in myocardial infarction is low.² Several major stroke centers now use magnetic resonance imaging (MRI) as the primary imaging modality in acute stroke,^3–7 which may also represent the new "gold standard" for detecting both primary brain hemorrhage and secondary hemorrhagic transformation.⁸

The identification and exclusion of patients with increased risk for SICH are major challenges for diagnostic imaging before treatment. There is evidence that hypointense lesions on T2*-weighted MRI ("cerebral microbleeds" [CMBs]) generally indicate an increased risk of both primary brain hemorrhage^9,10 and hemorrhagic transformation after ischemic stroke.¹¹ Based on these observations, it has been hypothesized that CMBs may become a marker of increased risk of secondary hemorrhage after thrombolytic therapy in stroke patients.^9,11–13 Secondary hemorrhagic transformation of ischemic infarction, however, is not necessarily associated with a poor prognosis,¹⁴ and the key questions are "Is there an increased risk of hemorrhagic transformation after tissue plasminogen activator (tPA) when CMBs are present?" and "Does the risk of clinically relevant intracranial hematoma exceed the benefit of thrombolytic agents in the subgroup of patients with CMBs?"

Anecdotal case reports have described patients with CMBs who developed catastrophic postischemic hemorrhage, related¹⁵ or unrelated¹⁶ to therapy. On the basis of these observations, patients with CMBs are already excluded from thrombolytic treatment.^4,17 Evidence for these decisions is weak: Previous reports on CMBs in patients treated with thrombolytic therapy include a total of 27 to 70 patients with or without CMBs and differ considerably in the time points and methods of treatment, as well as on the definitions of hemorrhage.^11,12,18 These series showed contradictory results for the risk of hemorrhage and allow no general conclusion for the time window of 0 to 6 hours after symptom onset. This uncertainty about the relevance of CMBs in candidates for thrombolytic therapy requires urgent clarification from a large database^19,20 because exclusion from potentially beneficial therapy is a serious consequence for patients with CMBs. Because the true clinical significance of CMBs is still unknown, a study of this subject will change the management of acute stroke patients and will establish a basis for the planning of future clinical trials.

This article describes the protocol, methods, and outcome results of the Bleeding Risk Analysis in Stroke by T2*-Weighted Imaging before thromboLysis (BRASIL) study. BRASIL is a large, international, multicenter, nonrandomized analysis of patients from 13 centers and compares the risk of SICH after IV thrombolytic therapy in patients with and without CMBs by applying a predefined common end point. The primary objective of BRASIL was to obtain a more reliable risk estimate for clinically relevant SICH in the presence of CMB patients treated with IV thrombolytic therapy within the first 6 hours after symptom onset, which is relevant for many centers worldwide.⁴

	Methods

Top Abstract Introduction Methods Results Discussion Appendix: References

 
Patients
We analyzed datasets of consecutive patients with ischemic stroke from 13 centers in Europe, North America, and Asia. The first MRI was obtained within 6 hours after symptom onset. Follow-up imaging (MRI or computed tomography [CT]) was done either routinely or because of any clinical deterioration up to 10 days after thrombolysis. The National Institutes of Health Stroke Scale (NIHSS) score was assessed by a stroke neurologist at each imaging time point. Criteria for thrombolytic therapy varied slightly among centers; the absence of hemorrhage on brain CT and identification of a diffusion-weighted imaging (DWI) and/or a perfusion lesion on the stroke MRI was used as the only imaging criterion at most sites within the first 3 hours. Beyond 3 hours, thrombolytic therapy was mainly based on the presence of a mismatch (DWI lesion larger than perfusion lesion by >20%) and preferably no large DWI lesion. The presence of CMBs was not included for therapeutic decisions during the sampling period at any center. The standard dosage for IV tPA therapy was 0.9 mg/kg bodyweight and a maximum dose of 90 mg at all centers. The study was approved by the local ethics committees at all centers that performed prospective data sampling. Informed consent was obtained for all patients where required by local legislation.

Imaging Methods
MRI studies were performed on clinical MRI scanners equipped with a standard head coil. The imaging protocol included an axial with a DWI and perfusion-weighted imaging sequence, MR angiography, and a T2 sequence for most patients. Image acquisition time was <20 minutes for the entire protocol in the majority of cases. Dedicated T2*-weighted (T2*w) images (repetition time=0.8 to 2140 ms, echo time=14 to 49 ms) were acquired with a matrix of 128x128, 256x256, or 512x192. The field of view ranged from 230x230 mm to 320x192 mm. Slice thicknesses was 5 to 7 mm; the interslice gap was 0 to 2 mm. Follow-up imaging was performed 1 to 10 days after stroke onset either because of clinical deterioration or after a predefined interval of 5 to 10 days.

Definition of CMBs and SICHs
Baseline T2*w images were rated locally for the presence of CMBs and the existence of subsequent hemorrhage on follow-up imaging in the case of clinical deterioration. The baseline and follow-up images were read locally, and only the follow-up images that could be considered to represent SICH (any hemorrhage temporally related to clinical deterioration of >2 points on the NIHSS) were subject to central rereading by a neuroradiologist (J.F.) for the presence of space-occupying parenchymal hematomas to standardize the definition of parenchymal hematoma type 2 (PH2) and thus, of SICH (the Figure). Local ratings were performed by strictly adhering to predefined specifications either by an experienced neuroradiologist or a stroke neurologist. Readers were blinded to clinical and follow-up imaging data. A CMB was defined as a focal area of signal loss within the brain parenchyma within or remote from the ischemic lesion (as defined by DWI) measuring <5 mm on T2*w imaging. Areas of symmetric hypointensity of the globus pallidus and loss of signal in the middle cerebral artery and its branches were disregarded. The primary end point was the rate of SICH in patients with and without CMBs. SICH was defined as an increase of 4 NIHSS points^21,22 in the presence of a space-occupying effect of an intracerebral hematoma (space-occupying PH2^1,22).

View larger version (83K): [in this window] [in a new window]   Figure. Representative MR images demonstrate the complexity of designating a hemorrhage "symptomatic." The time interval from symptom onset is noted in parentheses. Upper row (a) of images is from a 64-year-old patient with occlusion of the intracranial bifurcation of the left internal carotid artery who experienced secondary clinical deterioration after 21 hours because of the presence of a considerable lesion growth leading to a small frontal hemorrhage. However, the hemorrhage was certainly not the cause of the deterioration. Lower row (b) of images is from a 71-year-old patient with occlusion of the right middle cerebral artery and secondary deterioration due to a PH with later intraventricular hemorrhage, as documented in the 2 images obtained after 22 and 28 hours, and that was classified as SICH. FLAIR indicates fluid-attenuated inversion recovery.

Statistical Analysis
The primary end point was SICH in patients with and without CMBs. Based on the percentage of patients with CMBs in a stroke population eligible for thrombolytic therapy from the literature (range, 12.2% to 18.2%^12,18) and a rate of SICH of 5.9% in CT-selected, tPA-treated patients,¹ we assumed CMBs in 14% of the cohort and the risk of SICH in patients without microbleeds to be 5% for the power analysis. Based on these assumptions, a study to rule out an odds ratio (OR) of 3.5 for increased risk for SICH in patients with CMBs versus patients without would require a total sample of 568 patients (80% power at the 95% level of confidence). For standard statistical analysis, we used the SPSS 13.0 for Windows software package (SPSS Inc, Chicago, Ill). Demographic data, time intervals of examinations, and clinical scale scores are given as medians with interquartile range. Fisher’s exact test, or a corresponding ² test for large samples, was used for comparing categorical variables. A Mann–Whitney U test was used to compare a continuous variable with a dichotomous outcome. Mantel-Haenszel analysis was used to obtain pooled estimates of ORs for the effect of CMBs. Based on a threshold value from the literature, the ORs for SICH were calculated for >0 and >1¹⁷ CMBs.

	Results

Top Abstract Introduction Methods Results Discussion Appendix: References

 
Patient Characteristics
Datasets from 873 patients treated consecutively with thrombolytic therapy in 13 centers in Europe, North America, and Asia were acquired for central analysis (Table 1). In a second step, all patients were excluded who had received sole or combined intra-arterial thrombolytic therapy, when local rating of CMBs was done on source images on perfusion-weighted imaging, b=0 images of DWI, or any combination of these exclusion criteria. For the final analysis presented here, only those patients who received IV tPA and were imaged with dedicated T2*w images for detection of hemorrhage were included (n=570; 341 males and 229 females). Imaging modality for follow-up in patients with SICH was CT (n=8) or MRI (n=10).

View this table: [in this window] [in a new window]   Table 1. Numbers of CMBs and SICHs by Center

The median age of the patients was 69 years (first and third quartile, 59 and 77 years), the median time from symptom onset to MRI was 136 minutes (first and third quartile, 100 and 193 minutes), and the median initial NIHSS score was 13 (first and third quartile, 8 and 17; data available for 563 of 570 patients). Thrombolytic therapy was initiated before MRI in 29 of 570 patients (1 of these patients had a single CMB but no SICH). Therapy was initiated within 3 hours after symptom onset in 381 of 570 patients (66.8%), within 3 to 6 hours in 180 of 570 (31.6%), and at an unknown time but within 6 hours in 9 of 570 (1.6%). The NIHSS score was available at the follow-up time point only in 451 of 570 patients but was available for all patients with clinical deterioration. Therefore, the primary end point of SICH was not affected by the missing follow-up NIHSS scores.

CMBs and ICH
A total of 242 CMBs were detected in 86 of 570 patients (15.1%). The number of CMBs ranged from 1 to 77 in individual patients (Table 2). An increase in the NIHSS score of 2 points, together with any type of hemorrhage on follow-up imaging, was observed in 37 patients (6.5%). Among these, a parenchymal hematoma with clinical deterioration of at least 4 points on the NIHSS (SICH) was found in 18 (3.2%) patients.

View this table: [in this window] [in a new window]   Table 2. Numbers of CMBs in Patients Without Clinical Deterioration and Hemorrhage (No HE), With Clinical Deterioration but Without PH (Any HE), and With SICH on Follow-Up Imaging

Proportions of patients with SICH were 5.8% (95% CI, 1.9 to 13.0) in the presence of CMBs and 2.7% (95% CI, 1.4 to 4.5) in patients without CMBs (P=0.170, Fisher’s exact test), resulting in a nonsignificant absolute increase in the risk of SICH of 3.1% (95% CI, –2.0 to 8.3). The OR for SICH in patients with CMBs versus patients without CMBs was 2.23 (95% CI, 0.67 to 6.97). The proportions of patients with any hemorrhage were 9.3% (95% CI, 4.1 to 17.5) in the presence of CMBs and 6.0% (95% CI, 4.1 to 8.5) without CMBs (P=0.240, Fisher’s exact test). The OR for any hemorrhage in patients with CMBs versus patients without CMBs was 1.61 (95% CI, 0.66 to 3.85). The OR for SICH in the presence of CMBs when therapy was initiated 3 hours after symptom onset, which is recommended by many local authorities, was 2.2 (95% CI, 0.6 to 8.0) and 0.1 (95% CI, 0.0 to 23.3) in the extended time window of >3 to 6 hours.

By Mantel-Haenszel analysis, a method that better accounts for heterogeneity among centers, we also found no significant increase in the risk of SICH (P=0.10). The overall OR for SICH was 2.76 (95% CI, 0.82 to 9.30; P=0.1). A separate analysis for a predefined threshold of >1 CMB showed similar results: ORs of 2.60 (0.57 to 9.11, P=0.24) and 2.38 (0.23 to 24.98, P=0.47), respectively.

Other Variables
Because of significant variation in the prevalence of CMBs among centers (P=0.0005), a logistic-regression analysis would had have to been stratified by center and thus, was omitted owing to the lack of statistical power. Furthermore, because the presence of CMBs was not a risk factor for SICH, performing a multivariate analysis was unnecessary. No association was observed between the presence of CMBs and sex (P=0.881, Fisher’s exact test), time from symptom onset to MRI (P=0.770, Mann–Whitney U test), and NIHSS score (P=0.770, Mann–Whitney U test). Patients with CMBs were significantly older (median, 72 years; first and third quartile, 65 and 79) than patients without CMBs (median, 69 years; first and third quartile, 58 and 77; P=0.001, Mann-Wilcoxon test).

	Discussion

Top Abstract Introduction Methods Results Discussion Appendix: References

 
Histopathologic studies suggest that most focal areas of MRI signal loss on T2*w imaging indicate focal hemosiderin deposits from previous small bleeds.²³ The differential diagnosis of CMBs, however, also includes microcalcifications¹⁰ and even the release of prosthetic heart valve material,²⁴ both of which certainly do not increase the risk for SICH. The spatial distribution of CMBs is very similar to the usual sites of ICH, and the existence of higher numbers of CMBs represents an independent factor for the overall incidence of ICH.²⁵ Based on these observations, it has been hypothesized that CMBs may represent an independent risk factor for devastating hemorrhage after thrombolytic therapy.⁹ Besides anecdotal observations of devastating postischemic hemorrhages in patients with CMBs,^15,16 previously published case series included 27¹¹ and 44¹⁸ stroke patients who were treated with IV tPA within 6 hours, 70 patients who were treated with IV tPA within 3 to 6 hours,²⁶ and 41 patients who were treated with combined IV/intra-arterial tPA within 3 hours of symptom onset or with intra-arterial thrombolytics only within 12 hours from symptom onset.¹² The inhomogeneity of the investigated time points, the different treatment methods, and the different definitions of hemorrhagic transformation hamper both the generalization of results and the estimation of risk for SICH after tPA application within the first 6 hours, which is most relevant for several international stroke centers. This drawback has been overcome in BRASIL, wherein all 570 patients were treated with IV tPA within the first 6 hours after symptom onset, imaging evaluation was performed according to a predefined scheme, the end point of SICH was conservatively defined, and follow-up imaging was evaluated at a central reading center.

Clinical Relevance of CMBs
In any case, the risk of SICH must be balanced against the potential benefit of the withheld therapy and needs to be compared with the beneficial effects of tPA. BRASIL was not designed to evaluate the degree of benefit of tPA in MRI-selected patients. However, the rate of SICH of 5.8% (95% CI, 1.9 to 13.0) in patients with CMBs and of 2.7% (95% CI, 1.4 to 4.5) in patients without CMBs can be compared with data from the literature. In studies based on CT-selected patients, the rate of favorable outcome improved from 29.2% (95% CI, 25.1 to 33.5; placebo group) to 42.3% (95% CI, 37.8 to 47.0; treated with tPA) within the first 3 hours, resulting in an absolute increase of 10%.²⁷ The decreasing benefit of thrombolytic therapy for later time points, which is known for CT-selected patients,¹ is not necessarily observed in patients selected according to MRI criteria.^3,28,29 Furthermore, in line with our finding of the general rate of 3.2% for SICH, the rate of substantial hemorrhage seems to be lower in MRI-selected patients (3%)^29,30 than expected from CT-based trials (5.9%)¹ or registries (3.3% to 15.7%).³¹ Whether advanced MRI is more effective in selecting those most likely to benefit from thrombolytic therapy should be evaluated in a randomized study.³² Nevertheless, the aforementioned benefit of thrombolytic therapy as known from CT-based studies¹ will exceed the risk, even for the maximum rate of SICH, which is still consistent with the BRASIL data (3.1%; 95% CI, –2.0 to 8.3). Even if not based on a randomized trial, this information reflects the highest available level of evidence and might support clinicians in their individual therapy decisions.

Based on our data, a minor increase in hemorrhage risk cannot be ruled out. A study of 7664 patients with dedicated T2*w MRI and IV thrombolytic therapy would be necessary to exclude an increased risk of hemorrhage with an OR of 1.5 (80% power, 95% level of confidence). We suspect that such a population has probably not yet been investigated with sufficient documentation worldwide, and funding and recruitment for such a prospective study seem virtually impossible. However, even a slight increase in the SICH rate, if indeed it exists, would not justify withholding potentially beneficial tPA therapy from patients. Thrombolysis has been proven to be effective even when the rate of hemorrhage is significantly increased.^1,14

Relevance of the Number of CMBs for the Risk of SICH
The number of CMBs might be valuable for predicting subsequent hemorrhage after thrombolysis therapy, because the number of CMBs has been significantly associated with an increased risk of primary ICH.³³ Kang et al¹⁷ reported an acute stroke patient with multiple CMBs on the pretreatment scan who developed a hemorrhage after thrombolytic therapy. These authors concluded that 2 or more CMBs should be considered a relative contraindication for thrombolytic treatment. The clinical relevance of this threshold was not confirmed in our study.

In the present study, we were able to analyze a large number of stroke patients with CMBs treated with thrombolytic therapy (n=86). Nevertheless, assessment of the association of the individual number of CMBs with bleeding risk was complicated by the low rate of patients with a large number of CMBs. In our study, only 6 of 570 patients (1.1%) had 5 or more CMBs. In contrast, the proportion of patients with >10 CMBs was reported to be 14.6% to 31.5% in patient groups predominantly presenting with primary hemorrhage, who were not the target population for our study.^9,34 However, only 1 of 41 (2.4%) had >5 CMBs in a comparable sample of patients eligible for thrombolytic therapy.¹²

Our data allow no conclusions to be drawn concerning the rare patients with multiple CMBs in whom postischemic brain hemorrhages have been reported anecdotally after tPA therapy.¹⁵ The low number of detected CMBs in the present study may be explained by both differences among patient populations³⁵ and the use of fast sequences in the hyperacute setting with more artifacts and lower spatial resolution.³⁶ It must be considered that the rate of CMBs increases with age,³⁷ which is a known risk factor for SICH.³⁸ The first observation was confirmed in our study, indicating a possible environmental fallacy in the interpretation of a possible association of CMBs with bleeding risk.

Role of the Definition of SICH
The definition of SICH used profoundly affects its detection rate. Slight hemorrhagic transformation of ischemic brain tissue is associated with relatively small infarcts and often has a good prognosis.^39,40 The space-occupying effect of the hemorrhage plays a major role in clinical worsening: Compared with absent and milder forms of hemorrhagic transformation, only space-occupying parenchymal hematomas (ie, PH2) had a devastating impact on the clinical course.²² Because even a tiny hemorrhage within a growing infarct lesion might fulfill the criterion of "SICH" (the Figure), finding of a PH2 was incorporated into the definition of SICH. Of the 37 of 570 patients with any type of hemorrhage in the presence of clinical deterioration, only 18 had a PH2-type hemorrhage (SICH).

Limitations
Limitations of the present study include possible patient selection, missing randomization, no control group without thrombolytic therapy, retrospective analysis in some centers, and no control for several clinical and laboratory variables. Therapy randomization for the presence of CMBs or some other control group without therapy could not be performed because it would have been unethical to withhold beneficial therapy (as proven in CT-based trials¹) from patients with CMBs. A control group would not have profoundly enhanced the results of our analysis, because BRASIL aimed to compare the risk of SICH after thrombolytic therapy in patients with and without CMBs. Bias caused by unwanted patient selection is unlikely for 2 reasons: First, all centers provided data from a consecutive patient cohort treated with thrombolytic therapy uninfluenced by the findings on T2*w images. Second, the rate of CMBs of 15.1% in our study is very comparable to that in other thrombolytic therapy studies (12.2% to 20.0%).^11,12,26 Thus, exclusion of CMB-positive patients who would have hemorrhaged while maintaining the typical rate of stroke patients showing any number of CMBs is unlikely. Unexpected heterogeneity in the prevalence of CMBs among centers might have had an impact on the precision of the estimates of differences between CMB and non-CMB patients.

We suspect that the considerable discrepancy in the prevalence of any CMB per site can be explained in part by chance effects and other variables. We were able to control for demographic and clinical data at presentation but not for laboratory data, other imaging information like degree of leukoaraiosis,⁴¹ and clinical history, which have already been extensively analyzed by others.³⁸ Lack of a central reading of pretreatment T2*w images should not have influenced the analysis because interobserver agreement for reading CMBs is known to be very high.⁴² Heterogeneity in the prevalence of CMBs among centers also underlines the concept that translating the presence of CMBs into risk of SICH in these patients might be dependent on both the local patient population and the MRI sequence parameters and that findings from single-center studies cannot be easily transferred to other stroke centers.

Conclusions
Based on data from 13 high-volume stroke centers, BRASIL data represent the highest available level of evidence on the prognostic value of CMBs for clinically relevant secondary bleeding complications in acute ischemic stroke patients after thrombolytic therapy. The data suggest that the risk estimate of hemorrhage in the presence of CMBs is unlikely to exceed the benefit of thrombolytic treatment in MRI-selected patients, as expected from the literature. However, a minor increase in hemorrhagic risk cannot be excluded, and conclusions concerning the rare patient with multiple CMBs cannot be drawn. Therefore, a prospective study is warranted to acquire more information on the prognostic value of single and multiple CMBs for outcome in stroke patients.

	Appendix:

Top Abstract Introduction Methods Results Discussion Appendix: References

 
BRASIL Coinvestigators
Hamburg, Germany: Michael Bubenheim, Thomas Kucinski, Einar Goebell, Goetz Thomalla, Hermann Zeumer; Stanford, Calif: Wataru Kakuda; Calgary, Canada: Andrew M. Demchuk, Shelagh B. Coutts; Lyon, France: Norbert Nighoghossian, Marc Hermier; Mannheim, Germany: Olivera Lecei; Aarhus, Denmark: Mahmoud Ashkanian, Christine Solling, Grethe Andersen, Leif Østergaard; Seoul, Korea: Dong-Wha Kang, Sang-Beom Jeon; Los Angeles, Calif: Jeffrey L. Saver, Jeffry R. Alger, Latisha K. Ali, Brian H. Buck, Gary R. Duckwiler, Reza Jahan, Doojin Kim, Bruce Ovbiagele, Noriko Salamon, Nerses Sanossian, Sidney Starkman, Paul M. Vespa, J. Pablo Villablanca, Fernando Viñuela; Frankfurt, Germany: Silke Hölig, Marek Humpich, Joachim Berkefeld, Bernard Yan; Girona, Spain: Joaquin Serena, Antoni Davalos, Victoria Medina, Mar Castellanos, Yolanda Silva, Ana Quiles, Iratxe Diez, Sebastian Remollo; Barcelona, Spain: Raquel Delgado, Carlos Molina, Jose Alvarez-Sabin; Heidelberg, Germany: Jochen Fiebach, Hagen Huttner, Eric Juettler, Martin Koehrmann, Peter A. Ringleb; Linz, Austria: Edda Biedermann, Hans-Peter Haring.

	Acknowledgments

 
Disclosures

J.F. has received speaker fees from Bracco ALTANA; D.S.L. is on an Advisory Board for Coaxia Inc; J.R. received honoraria from Boehringer-Ingelheim; P.D.S. is on a speakers bureau and advisory board for Boehringer Ingelheim.

Received January 9, 2007; revision received March 19, 2007; accepted April 12, 2007.

	References

Top Abstract Introduction Methods Results Discussion Appendix: References

 
1. Hacke W, Donnan G, Fieschi C, Kaste M, von Kummer R, Brott T, Frankel M, Grotta J, Haley EJ, Kwiatkowski T, Levine S, Lewandowski C, Lu M, Lyden P, Marler J, Patel S, Tilley B, Albers G, Bluhmki E, Wilhelm M, Hamilton RL, for the ATLANTIS E and NINDS rt-PA Study Group Investigators. Association of outcome with early stroke treatment: pooled analysis of ATLANTIS, ECASS, and NINDS rt-PA stroke trials. Lancet. 2004; 363: 768–774.[CrossRef][Medline] [Order article via Infotrieve]

2. Gurwitz JH, Gore JM, Goldberg RJ, Barron HV, Breen T, Rundle AC, Sloan MA, French W, Rogers WJ. Risk for intracranial hemorrhage after tissue plasminogen activator treatment for acute myocardial infarction: participants in the National Registry of Myocardial Infarction 2. Ann Intern Med. 1998; 129: 597–604.[Abstract/Free Full Text]

3. Rother J, Schellinger PD, Gass A, Siebler M, Villringer A, Fiebach JB, Fiehler J, Jansen O, Kucinski T, Schoder V, Szabo K, Junge-Hulsing GJ, Hennerici M, Zeumer H, Sartor K, Weiller C, Hacke W. Effect of intravenous thrombolysis on MRI parameters and functional outcome in acute stroke <6 hours. Stroke. 2002; 33: 2438–2445.[Abstract/Free Full Text]

4. Hjort N, Butcher K, Davis SM, Kidwell CS, Koroshetz WJ, Rother J, Schellinger PD, Warach S, Ostergaard L. Magnetic resonance imaging criteria for thrombolysis in acute cerebral infarct. Stroke. 2005; 36: 388–397.[Abstract/Free Full Text]

5. Schellinger PD, Hacke W. Stroke: advances in therapy. Lancet Neurol. 2005; 4: 2.[CrossRef][Medline] [Order article via Infotrieve]

6. Rowley HA. Extending the time window for thrombolysis: evidence from acute stroke trials. Neuroimaging Clin North Am. 2005; 15: 575–587.[CrossRef]

7. Muir KW, Buchan A, von Kummer R, Rother J, Baron JC. Imaging of acute stroke. Lancet Neurol. 2006; 5: 755–768.[CrossRef][Medline] [Order article via Infotrieve]

8. von Kummer R. MRI: the new gold standard for detecting brain hemorrhage? Stroke. 2002; 33: 1748–1749.[Free Full Text]

9. Kato H, Izumiyama M, Izumiyama K, Takahashi A, Itoyama Y. Silent cerebral microbleeds on T2*-weighted MRI: correlation with stroke subtype, stroke recurrence, and leukoaraiosis. Stroke. 2002; 33: 1536–1540.[Abstract/Free Full Text]

10. Fiehler J. Cerebral microbleeds: old leaks and new haemorrhages. Int J Stroke. 2006; 1: 122–130.[CrossRef][Medline] [Order article via Infotrieve]

11. Nighoghossian N, Hermier M, Adeleine P, Blanc-Lasserre K, Derex L, Honnorat J, Philippeau F, Dugor JF, Froment JC, Trouillas P. Old microbleeds are a potential risk factor for cerebral bleeding after ischemic stroke: a gradient–echo T2*-weighted brain MRI study. Stroke. 2002; 33: 735–742.[Abstract/Free Full Text]

12. Kidwell CS, Saver JL, Villablanca JP, Duckwiler G, Fredieu A, Gough K, Leary MC, Starkman S, Gobin YP, Jahan R, Vespa P, Liebeskind DS, Alger JR, Vinuela F. Magnetic resonance imaging detection of microbleeds before thrombolysis: an emerging application. Stroke. 2002; 33: 95–98.[Abstract/Free Full Text]

13. Senior K. Microbleeds may predict cerebral bleeding after stroke. Lancet. 2002; 359: 769.[Medline] [Order article via Infotrieve]

14. von Kummer R. Brain hemorrhage after thrombolysis: good or bad? Stroke. 2002; 33: 1446–1447.[Free Full Text]

15. Chalela JA, Kang DW, Warach S. Multiple cerebral microbleeds: MRI marker of a diffuse hemorrhage-prone state. J Neuroimaging. 2004; 14: 54–57.[Medline] [Order article via Infotrieve]

16. Smith EE, Fitzsimmons AL, Nogueira RG, Singhal AB. Spontaneous hyperacute postischemic hemorrhage leading to death. J Neuroimaging. 2004; 14: 361–364.[CrossRef][Medline] [Order article via Infotrieve]

17. Kang DW, Chalela JA, Dunn W, Warach S. MRI screening before standard tissue plasminogen activator therapy is feasible and safe. Stroke. 2005; 36: 1939–1943.[Abstract/Free Full Text]

18. Derex L, Nighoghossian N, Hermier M, Adeleine P, Philippeau F, Honnorat J, Dardel P, Froment JC, Trouillas P. Thrombolysis for ischemic stroke in patients with old microbleeds on pretreatment MRI. Cerebrovasc Dis. 2004; 17: 238–241.[CrossRef][Medline] [Order article via Infotrieve]

19. McCarron MO, Nicoll JA. Cerebral amyloid angiopathy and thrombolysis-related intracerebral haemorrhage. Lancet Neurol. 2004; 3: 484–492.[CrossRef][Medline] [Order article via Infotrieve]

20. Viswanathan A, Chabriat H. Cerebral microhemorrhage. Stroke. 2006; 37: 550–555.[Abstract/Free Full Text]

21. Brott TG, Haley ECJ, Levy DE, Barsan W, Broderick J, Sheppard GL, Spilker J, Kongable GL, Massey S, Reed R. Urgent therapy for stroke, part I: pilot study of tissue plasminogen activator administered within 90 minutes. Stroke. 1992; 23: 632–640.[Abstract/Free Full Text]

22. Fiorelli M, Bastianello S, von Kummer R, del Zoppo GJ, Larrue V, Lesaffre E, Ringleb AP, Lorenzano S, Manelfe C, Bozzao L. Hemorrhagic transformation within 36 hours of a cerebral infarct: relationships with early clinical deterioration and 3-month outcome in the European Cooperative Acute Stroke Study I (ECASS I) cohort. Stroke. 1999; 30: 2280–2284.[Abstract/Free Full Text]

23. Fazekas F, Kleinert R, Roob G, Kleinert G, Kapeller P, Schmidt R, Hartung HP. Histopathologic analysis of foci of signal loss on gradient-echo T2*-weighted MR images in patients with spontaneous intracerebral hemorrhage: evidence of microangiopathy-related microbleeds. AJNR Am J Neuroradiol. 1999; 20: 637–642.[Abstract/Free Full Text]

24. van Gorp MJ, van der Graaf Y, de Mol BA, Bakker CJ, Witkamp TD, Ramos LM, Mali WP. Bjork-Shiley convexoconcave valves: susceptibility artifacts at brain MR imaging and mechanical valve fractures. Radiology. 2004; 230: 709–714.[Abstract/Free Full Text]

25. Lee SH, Kwon SJ, Kim KS, Yoon BW, Roh JK. Cerebral microbleeds in patients with hypertensive stroke: topographical distribution in the supratentorial area. J Neurol. 2004; 251: 1183–1189.[CrossRef][Medline] [Order article via Infotrieve]

26. Kakuda W, Thijs VN, Lansberg MG, Bammer R, Wechsler L, Kemp S, Moseley ME, Marks MP, Albers GW. Clinical importance of microbleeds in patients receiving IV thrombolysis. Neurology. 2005; 65: 1175–1178.[Abstract/Free Full Text]

27. Warlow C, Sudlow C, Dennis M, Wardlaw J, Sandercock P. Stroke. Lancet. 2003; 362: 1211–1224.[CrossRef][Medline] [Order article via Infotrieve]

28. Hacke W, Albers G, Al-Rawi Y, Bogousslavsky J, Davalos A, Eliasziw M, Fischer M, Furlan A, Kaste M, Lees KR, Soehngen M, Warach S. The Desmoteplase in Acute Ischemic Stroke Trial (DIAS): a phase II MRI-based 9-hour window acute stroke thrombolysis trial with intravenous desmoteplase. Stroke. 2005; 36: 66–73.[Abstract/Free Full Text]

29. Thomalla G, Schwark C, Sobesky J, Bluhmki E, Fiebach JB, Fiehler J, Zaro Weber O, Kucinski T, Juettler E, Ringleb PA, Zeumer H, Weiller C, Hacke W, Schellinger PD, Rother J. Outcome and symptomatic bleeding complications of intravenous thrombolysis within 6 hours in MRI-selected stroke patients: comparison of a German multicenter study with the pooled data of ATLANTIS, ECASS, and NINDS tPA trials. Stroke. 2006; 37: 852–858.[Abstract/Free Full Text]

30. Kohrmann M, Juttler E, Fiebach JB, Huttner HB, Siebert S, Schwark C, Ringleb PA, Schellinger PD, Hacke W. MRI versus CT-based thrombolysis treatment within and beyond the 3 h time window after stroke onset: a cohort study. Lancet Neurol. 2006; 5: 661–667.[CrossRef][Medline] [Order article via Infotrieve]

31. Graham GD. Tissue plasminogen activator for acute ischemic stroke in clinical practice: a meta-analysis of safety data. Stroke. 2003; 34: 2847–2850.[Abstract/Free Full Text]

32. Lindley RI, Wardlaw JM, Sandercock PA. Alteplase and ischaemic stroke: have new reviews of old data helped? Lancet Neurol. 2005; 4: 249–253.[CrossRef][Medline] [Order article via Infotrieve]

33. Lee SH, Bae HJ, Kwon SJ, Kim H, Kim YH, Yoon BW, Roh JK. Cerebral microbleeds are regionally associated with intracerebral hemorrhage. Neurology. 2004; 62: 72–76.[Abstract/Free Full Text]

34. Naka H, Nomura E, Wakabayashi S, Kajikawa H, Kohriyama T, Mimori Y, Nakamura S, Matsumoto M. Frequency of asymptomatic microbleeds on T2*-weighted MR images of patients with recurrent stroke: association with combination of stroke subtypes and leukoaraiosis. AJNR Am J Neuroradiol. 2004; 25: 714–719.[Abstract/Free Full Text]

35. Koennecke HC. Cerebral microbleeds on MRI: prevalence, associations, and potential clinical implications. Neurology. 2006; 66: 165–171.[Abstract/Free Full Text]

36. Gagnon A, Simon J, Sohn C, Coutts S, Sevick R, Eliasziw M, Hill M, Demchuk A. Microbleeds detected by perfusion-weighted MR imaging at 3 T independently predict subsequent hemorrhagic transformation in acute stroke. Stroke. 2004; 35: 330.

37. Jeerakathil T, Wolf PA, Beiser A, Hald JK, Au R, Kase CS, Massaro JM, DeCarli C. Cerebral microbleeds: prevalence and associations with cardiovascular risk factors in the Framingham Study. Stroke. 2004; 35: 1831–1835.[Abstract/Free Full Text]

38. Tanne D, Kasner SE, Demchuk AM, Koren-Morag N, Hanson S, Grond M, Levine SR. Markers of increased risk of intracerebral hemorrhage after intravenous recombinant tissue plasminogen activator therapy for acute ischemic stroke in clinical practice: the Multicenter rt-PA Stroke Survey. Circulation. 2002; 105: 1679–1685.[Abstract/Free Full Text]

39. Molina CA, Montaner J, Abilleira S, Ibarra B, Romero F, Arenillas JF, Alvarez-Sabin J. Timing of spontaneous recanalization and risk of hemorrhagic transformation in acute cardioembolic stroke. Stroke. 2001; 32: 1079–1084.[Abstract/Free Full Text]

40. Fiehler J, Remmele C, Kucinski T, Rosenkranz M, Thomalla G, Weiller C, Zeumer H, Rother J. Reperfusion after severe local perfusion deficit precedes hemorrhagic transformation: an MRI study in acute stroke patients. Cerebrovasc Dis. 2005; 19: 117–124.[CrossRef][Medline] [Order article via Infotrieve]

41. Neumann-Haefelin T, Hoelig S, Berkefeld J, Fiehler J, Gass A, Humpich M, Kastrup A, Kucinski T, Lecei O, Liebeskind D, Rother J, Rosso C, Samson Y, Saver J, Yan B; MR Stroke Group. Leukoaraiosis is a risk factor for symptomatic intracerebral hemorrhage after thrombolysis for acute stroke. Stroke. 2006; 37: 2463–2466.[Abstract/Free Full Text]

42. Lee SH, Kim BJ, Roh JK. Silent microbleeds are associated with volume of primary intracerebral hemorrhage. Neurology. 2006; 66: 430–432.[Abstract/Free Full Text]

This article has been cited by other articles:

				O. C. Singer, W. Kurre, M. C. Humpich, M. W. Lorenz, A. Kastrup, D. S. Liebeskind, G. Thomalla, J. Fiehler, J. Berkefeld, T. Neumann-Haefelin, et al. Risk Assessment of Symptomatic Intracerebral Hemorrhage After Thrombolysis Using DWI-ASPECTS Stroke, August 1, 2009; 40(8): 2743 - 2748. [Abstract] [Full Text] [PDF]

				S. M. Greenberg, R. N. K. Nandigam, P. Delgado, R. A. Betensky, J. Rosand, A. Viswanathan, M. P. Frosch, and E. E. Smith Microbleeds Versus Macrobleeds: Evidence for Distinct Entities * Supplemental Materials and Methods Stroke, July 1, 2009; 40(7): 2382 - 2386. [Abstract] [Full Text] [PDF]

				M. Wintermark, H. A. Rowley, and M. H. Lev Acute Stroke Triage to Intravenous Thrombolysis and Other Therapies with Advanced CT or MR Imaging: Pro CT Radiology, June 1, 2009; 251(3): 619 - 626. [Full Text] [PDF]

				M. W. Vernooij, M. D. M. Haag, A. van der Lugt, A. Hofman, G. P. Krestin, B. H. Stricker, and M. M. B. Breteler Use of Antithrombotic Drugs and the Presence of Cerebral Microbleeds: The Rotterdam Scan Study Arch Neurol, June 1, 2009; 66(6): 714 - 720. [Abstract] [Full Text] [PDF]

				W.-D. Heiss and A. G. Sorensen Advances in Imaging Stroke, May 1, 2009; 40(5): e313 - e314. [Full Text] [PDF]

				A.A. Konstas, G.V. Goldmakher, T.-Y. Lee, and M.H. Lev Theoretic Basis and Technical Implementations of CT Perfusion in Acute Ischemic Stroke, Part 1: Theoretic Basis AJNR Am. J. Neuroradiol., April 1, 2009; 30(4): 662 - 668. [Abstract] [Full Text] [PDF]

				M. W. Vernooij, J. Heeringa, G. J. de Jong, A. van der Lugt, and M.M.B. Breteler CEREBRAL MICROBLEED PRECEDING SYMPTOMATIC INTRACEREBRAL HEMORRHAGE IN A STROKE-FREE PERSON Neurology, February 24, 2009; 72(8): 763 - 765. [Full Text] [PDF]

				S.-H. Lee, W.-S. Ryu, and J.-K. Roh Cerebral microbleeds are a risk factor for warfarin-related intracerebral hemorrhage Neurology, January 13, 2009; 72(2): 171 - 176. [Abstract] [Full Text] [PDF]

				C. Cordonnier, G. M. Potter, C. A. Jackson, F. Doubal, S. Keir, C. L.M. Sudlow, J. M. Wardlaw, and R. A.-S. Salman Improving Interrater Agreement About Brain Microbleeds: Development of the Brain Observer MicroBleed Scale (BOMBS) Stroke, January 1, 2009; 40(1): 94 - 99. [Abstract] [Full Text] [PDF]

				L Derex and N Nighoghossian Intracerebral haemorrhage after thrombolysis for acute ischaemic stroke: an update J. Neurol. Neurosurg. Psychiatry, October 1, 2008; 79(10): 1093 - 1099. [Abstract] [Full Text] [PDF]

				P. D. Schellinger and M. Kohrmann MRA/DWI Mismatch: A Novel Concept or Something One Could Get Easier and Cheaper? Stroke, September 1, 2008; 39(9): 2423 - 2424. [Full Text] [PDF]

				J. Fiehler and on behalf of the BRASIL investigators Response to Letter by Vernooij et al Stroke, July 1, 2008; 39(7): e116 - e116. [Full Text] [PDF]

				M. W. Vernooij, A. van der Lugt, and M. M.B. Breteler Risk of Thrombolysis-Related Hemorrhage Associated With Microbleed Presence Stroke, July 1, 2008; 39(7): e115 - e115. [Full Text] [PDF]

				B R Thanvi, S Treadwell, and T Robinson Haemorrhagic transformation in acute ischaemic stroke following thrombolysis therapy: classification, pathogenesis and risk factors Postgrad. Med. J., July 1, 2008; 84(993): 361 - 367. [Abstract] [Full Text] [PDF]

				A. G. Sorensen and W.-D. Heiss Advances in Imaging 2007 Stroke, February 1, 2008; 39(2): 276 - 278. [Full Text] [PDF]

This Article Abstract Full Text (PDF) All Versions of this Article: 38/10/2738    most recent STROKEAHA.106.480848v1 Alert me when this article is cited Alert me if a correction is posted Citation Map Services Email this article to a friend Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Request Permissions Citing Articles Citing Articles via HighWire Citing Articles via Google Scholar Google Scholar Articles by Fiehler, J. Search for Related Content PubMed PubMed Citation Articles by Fiehler, J. Pubmed/NCBI databases Medline Plus Health Information MRI Scans Stroke Related Collections CT and MRI Acute Cerebral Infarction Emergency treatment of Stroke Computerized tomography and Magnetic Resonance Imaging Intracerebral Hemorrhage Thrombolysis