A Standardized MRI Stroke Protocol

Comparison with CT in Hyperacute Intracerebral Hemorrhage

Peter D. Schellinger, MD; Olav Jansen, MD; Jochen B. Fiebach, MD;
Werner Hacke, MD; Klaus Sartor, MD
Background and PurposeDiagnostic imaging in hyperacute ischemic stroke has been revolutionized by the introduction
of diffusion- and perfusion-weighted MRI (DWI and PWI). CT, however, is still needed to exclude intracerebral
hemorrhage (ICH). The purpose of our study was to determine the diagnostic accuracy of a standardized, multimodal
MRI (mMRI) stroke protocol in the qualitative and quantitative assessment of hyperacute ICH (,6 hours).
MethodsWe investigated 9 patients with hyperacute ICH with CT followed immediately by a standardized mMRI stroke
protocol (DWI, PWI [T2*-WI], FLAIR, T2-WI, and MRA). The time interval between MRI and symptom onset ranged
from 3 hours to 5 hours 45 minutes. We analyzed and compared the size of the hematoma on CT and all mMRI images
by semiautomatic volumetry.
ResultsICH was unambiguously identified on the basis of all mMRI sequences. With increasing susceptibility effect
(T2*-WI), the ICH, appearing as an area of hyperintensity with central signal loss, became qualitatively most evident.
Regarding quantitation, T2*-WI overestimated (median and mean difference, 18.9%/17.8%; SD s524.4%) and DWI
correlated best (median and mean difference, 3.97%/24.36%; SD s537.42%) with hematoma size on CT.
ConclusionsMultimodal stroke MRI is as reliable as CT in the assessment of hyperacute ICH. Therefore, additional CT
is no longer necessary to rule out ICH in hyperacute stroke. The use of mMRI alone in the diagnostic workup of a
hyperacute stroke patient saves time and costs while rendering all the critical information needed to initiate an optimal
treatment. (Stroke. 1999;30:765-768.)
Key Words: intracerebral hemorrhage n magnetic resonance imaging n stroke n tomography, x-ray computed

he diagnosis of intracerebral hemorrhage (ICH) is still a

domain of CT rather than MRI, especially in acute
stroke.1,2 Most authors generally claim that the sensitivity of
MRI for detecting hyperacute ICH is poor.3 6 The key
substrate for MRI visualization of hemorrhage is deoxyhemoglobin, which causes a signal loss in T2-weighted imaging
(T2-WI) because of paramagnetic susceptibility effects, although usually not within the first 12 to 24 hours.79 There are
few publications that report a high diagnostic accuracy of
MRI for ICH within a time window of #6 hours after
symptom onset.10,11 For the very first time, a case cohort
study11 demonstrated early MRI changes in susceptibilityweighted T2*WI within 5 hours in 6 patients with acute ICH.
Unfortunately, these patients were not examined according to
a standardized MRI protocol. Furthermore, the time interval
between CT and MRI examinations was up to 4 days, and the
hemorrhage volumes were not quantified to assess the diagnostic accuracy of MRI. The recent introduction of new
multimodal MRI (mMRI) techniques such as diffusion- and
perfusion-weighted imaging (DWI and PWI) has improved
diagnostic imaging in hyperacute ischemic stroke, providing
important additional information2,1218; however, the need for

a CT scan to rule out acute ICH before administration of a

specific therapy, such as thrombolysis, is time-consuming
and reduces the feasibility and cost-effectiveness of stroke
MRI.19 Because there are not sufficient data about the
diagnostic accuracy of new MRI sequences in hyperacute
ICH, we report the imaging findings from 9 patients with
primary ICH who were examined with mMRI within 6 hours
after symptom onset as part of an open, prospective stroke
trial using a standardized MRI stroke protocol.

Subjects and Methods

Since December 1997 we have prospectively and longitudinally
evaluated acute ischemic stroke patients according to a standardized
clinical and mMRI protocol. All patients received a CT scan before
enrollment in the study. During this time we also examined 9 patients
(3 women, 6 men) aged 40 to 73 years (mean, 60 years) with acute
ICH within 6 hours after symptom onset with CT and MRI. To
guarantee patient safety, a neurological critical care fellow was
present throughout the MRI examination: the patients oxygen
saturation and a 1-lead ECG were monitored continuously, and BP
was measured intermittently. Patients with unstable vital signs or
general MRI contraindications were excluded from the study. Informed consent was obtained from all patients or their next of kin.

Received December 4, 1998; final revision received January 21, 1999; accepted January 22, 1999.
From the Departments of Neurology (P.D.S., W.H.) and Neuroradiology (O.J., J.B.F., K.S.), Medical Faculty, University of Heidelberg, Germany.
Correspondence to Peter D. Schellinger, MD, Abteilung Neuroradiologie, Universitatsklinik Heidelberg, Im Neuenheimer Feld 400, D-69120
Heidelberg, Germany. E-mail Peter_Schellinger@ukl.uni-heidelberg.de
1999 American Heart Association, Inc.
Stroke is available at http://www.strokeaha.org

765
Downloaded from http://stroke.ahajournals.org/
by guest on December 4, 2014

766

Standardized Stroke MRI in Hyperacute Intracerebral Hemorrhage

Figure 1. A 40-year-old man with no risk

factor other than untreated arterial hypertension presented with severe headache,
plegia of the right arm, grade 3 paresis
of the right leg, severe nonfluent aphasia, and forced gaze deviation to the left.
CT was performed 2 h and MRI 3 h after
symptom onset. Initial CT demonstrates
a left putaminal ICH (A) and T2-WI an
irregular hyperintense putaminal lesion
on the left (B). C through E, DWI, FLAIR,
and T2*-W PWI source images show an
increasing area of signal loss within the
core of the left-sided hematoma with a
heterogeneous appearance caused by
small amounts of deoxyhemoglobin.

All patients were examined with a fourth-generation CT scanner

(PQ 2000, Picker) and immediately thereafter with a 1.5-T wholebody MR imager (EDGE, Picker) equipped with enhanced gradient
hardware for echo planar imaging (EPI). For the MRI examination
we used a circular polarized head coil. The mMRI protocol includes
an axial T2-W fast-spin-echo sequence, an axial fluid-attenuated
inversion recovery (FLAIR) EPI sequence, an axial isotropic DWI
SE EPI sequence, time-of-flight MR angiography, and PWI with an
axial T2*-W gradient echo EPI sequence (40 data sets during and
after injection of 25 mL Gd-DTPA (Magnevist, Schering AG) with
a power injector (5 mL/s).
The diagnosis of ICH was established by CT. The acute hematoma
was identified on MRI on the basis of a heterogeneous region of
signal loss and focal hyperintensity characteristic of the MR appearance of ICH.9 To evaluate the sensitivity of different MRI sequences,
we compared the hematoma size in CT and MRI images by
performing an offline volumetric analysis of the hematoma on CT
images, FLAIR images, fast-spin-echo T2-WI, DWI source images
(b51000), and PWI source images. The area suspected as representing the hematoma was traced by hand with the aid of an image
analysis system (VISTAR and VOXEL, Picker) for each slice
separately, and these areas were subsequently used to calculate
lesion volumes. Mean, median, and SD are expressed in relation to
hematoma size on CT, ie, a negative sign stands for a larger
hematoma volume on MRI than on CT.

Results
Intracranial hemorrhage was qualitatively detected on all CT
and all MRI images (fast T2, FLAIR, DWI, and T2*-WI) in
all 9 patients. The mean time between CT examination and

symptom onset was 2 hours (range, 0:45 to 3:45), and the

mean time between MRI and symptom onset was 4 hours
(range, 3:00 to 5:45), which leaves a mean time interval
between CT and MRI of 2 hours (range, 1:00 to 3:45). The
scanning time for the entire MRI protocol was approximately
20 minutes with an additional 10 minutes for patient positioning. The typical appearance of ICH on the MRI images
was a heterogeneous focus of high and low signal intensities.
With increasing susceptibility weight, the central area of
hypointensity became more pronounced. The T2*-WI
showed no or only few areas of hyperintensity, or merely a
faint ring around a central core of signal loss. Figures 1 and
2 illustrate the appearance of a larger and a smaller hyperacute ICH on CT and MRI in 2 representative patients; these
findings were common to all 9 patients.
Volumetric analysis showed a good raw correlation of
hematoma volumes in all MRI images compared with CT. As
already reported by Rosen et al,20 images obtained with
sequences with a high sensitivity for susceptibility effects
(T2*-WI, FLAIR) generally overestimate the actual hematoma size in comparison to the lesion volume assessed on CT.
Hematoma volumes on DWI (median and mean difference,
3.97%/24.36%; SD s537.42%) followed by FLAIR (median and mean difference, 2.91%/26.25%; SD s528.39%)
corresponded best with lesion size on CT. Conventional
T2-WI substantially underestimated (median and mean dif-

Downloaded from http://stroke.ahajournals.org/ by guest on December 4, 2014

Schellinger et al

April 1999

767

Figure 2. A 65-year-old woman with

arterial hypertension and obesity presented 4 hours after symptom onset
with somnolence, grade 2 to 3 sensorimotor hemiparesis on the left side,
incomplete head and eye deviation to
the right, moderate dysarthrophonia,
and incomplete hemineglect to the left.
CT was performed 1.5 h and MRI
3.75 h after symptom onset. A, Initial
CT shows an ICH located in the right
external capsule; B, T2-WI shows the
same lesion as a hyperintense signal.
FLAIR (C) and DWI (D) revealed an
irregular hyperintense putaminal lesion
on the right, with a hypointense core
more prominent on the sequences with
a stronger weight on susceptibility
effects and strongest on T2*-WI (E).

ference, 17.24%/12.98%; SD s534.46%) and T2*-WI substantially overestimated (median and mean difference,
217.94%/18.86%; SD s524.45%) the hematoma size (see
the Table).

Discussion
Our MRI findings in 9 patients with primary hypertensive
ICH examined within 6 hours of symptom onset clearly
demonstrate the sensitivity of susceptibility-weighted MRI
sequences for fresh intracerebral blood. MRI performed with
a standardized multimodal protocol for stroke as also used by
others,15,16,18 therefore, is as good as CT in ruling out or
defining the extent of ICH. Older studies postulated a 24-hour
gap before detectable amounts of paramagnetic deoxyhemoglobin have accumulated.7,9 Our results, however, support the
hypothesis that small amounts of deoxyhemoglobin are present within the very first hours of ICH and are detectable by
susceptibility-weighted MRI sequences. Figures 1 and 2
suggest that T2*-WI are suited best for the diagnosis of ICH,
which holds true for the qualitative detection also of relatively small thalamic hematomas without mass effect (Figure
2). For quantitative analysis, DWI and FLAIR are the best
indicators of lesion volume compared with CT. Further
information, such as the presence of space-occupying edema,
midline shift, and ventricular hemorrhage, is best derived
from conventional T2-WI. In addition to the standardized
stroke protocol, postcontrast T1-WI scans may be obtained

after PWI if another primary disease is suspected (eg,

apoplectic glioma or metastases). There currently is no
information about the sensitivity of stroke MRI for subarachnoid hemorrhage (SAH) in patients. Animal experiments21
demonstrated early DWI changes in rats suggesting early
vasospasm, and spreading depression as indirect signs of
SAH. A signal loss due to susceptibility effects analogous to
ICH is not described. Although strokelike symptoms are
unusual in SAH, further studies aimed at the diagnostic reach
of stroke MRI in acute SAH are worthwhile.
Patel et al in 199611 reported a case cohort of 6 patients
evaluated with MRI within 6 hours after acute ICH and also
found susceptibility-weighted sequences to be sensitive for
acute ICH. In their patient series, however, CT was performed from 9.5 hours up to 4 days after symptom onset, so
that secondary hemorrhage after primary ischemia cannot be
excluded. In 3 patients no time interval for CT was given.
Furthermore, MRI examinations were conducted with 3
different types of MR scanners (with different field
strengths), and volumetric analysis was not performed.
In the emergency evaluation of acute stroke, time is
brain.22 It is thus of utmost importance that diagnostic efforts
are as specific and time efficacious as possible, especially
when considering aggressive treatment strategies such as
thrombolysis.23,24 MRI stroke protocols, including DWI and
PWI, are very promising with regard to the characterization
of acute stroke patients and the identification of patients

Downloaded from http://stroke.ahajournals.org/ by guest on December 4, 2014

768

Standardized Stroke MRI in Hyperacute Intracerebral Hemorrhage

Volumetric and Statistical Data

Lesion Volume, mL

CT

T2 FSE

DWI

FLAIR

T2* EPI

Patient 1

66.2

40.6

65.3

45.0

67.1

Patient 2

6.2

3.0

5.8

6.3

8.6

Patient 3

8.9

4.8

1.8

9.2

7.0

Patient 4

50.6

56.9

46.6

48.8

57.1

Patient 5

61.7

52.8

74.1

71.9

68.6

Patient 6

13.1

22.0

20.0

22.4

28.1

Patient 7

20.2

19.3

26.2

22.0

26.2

Patient 8

41.4

34.3

36.5

35.2

42.7

Patient 9

3.0

2.3

4.3

3.2

5.3

Mean difference, %

12.984

24.355

26.253

217.802

Median difference, %

17.235

3.971

22.912

218.864

Standard deviation s, %

34.461

37.419

26.762

24.445

Lesion volumes are shown for all patients and sequences. From left to right,
the sensitivity for susceptibility effects increases from zero (CT) to maximum
(T2*WI). Mean difference, median difference, and standard deviation of MRI
sequences with regard to CT are given.

suitable for specific therapy. Surprisingly, though, MRI is

still not generally considered to be the primary and only
diagnostic tool in acute stroke patients, because there is doubt
regarding the ability to detect hyperacute hemorrhage. Although a larger number of patients would be useful to confirm
these findings, our results show that mMRI is as sensitive as
CT in the diagnosis of ICH. The initial and exclusive use of
mMRI is therefore feasible, cost-effective, and time saving.
In conclusion, mMRI may be the single diagnostic tool of
choice in the initial assessment of patients with hyperacute
ischemic or hemorrhagic stroke.

Acknowledgments
We thank Olivia Pohlers and Henning Ryssel, Department of
Neuroradiology, University at Heidelberg, Medical Faculty, for their
help with the volumetric analysis of the image data, and Sabine
Heiland, PhD, Department of Neuroradiology, University at Heidelberg, Medical Faculty, for her assistance and expertise in optimizing
the multimodal MRI stroke protocol. Derk Krieger, MD, Associate
Professor of Neurology, Cleveland Clinic Foundation, Cleveland,
Ohio, strongly encouraged us to publish our observations.

References
1. Higer HP, Pedrosa P, Schaeben W, Bielke G, Meindl S. Intracranial
hemorrhage in MRT. Radiologe. 1989;29:297302.
2. Jansen O, Heiland S, Schellinger P. Neuroradiological diagnosis in acute
ischemic stroke: value of modern techniques. Nervenarzt. 1998;69:
465 471.
3. Hayman LA, Taber KH, Ford JJ, Bryan RN. Mechanisms of MR signal
alteration by acute intracerebral blood: old concepts and new theories.
AJNR Am J Neuroradiol. 1991;12:899 907.
4. Hayman LA, Pagani JJ, Kirkpatrick JB, Hinck VC. Pathophysiology of
acute intracerebral and subarachnoid hemorrhage: applications to MR
imaging. AJR Am J Roentgenol. 1989;153:135139.

5. Weingarten K, Zimmerman RD, Cahill PT, Deck MD. Detection of acute

intracerebral hemorrhage on MR imaging: ineffectiveness of prolonged
interecho interval pulse sequences. AJNR Am J Neuroradiol. 1991;12:
475 479.
6. Bradley WG Jr. MR appearance of hemorrhage in the brain. Radiology.
1993;189:1526.
7. Mittl RL Jr, Gomori JM, Schnall MD, Holland GA, Grossman RI, Atlas
SW. Magnetization transfer effects in MR imaging of in vivo intracranial
hemorrhage. AJNR Am J Neuroradiol. 1993;14:881 891.
8. Thulborn KR, Atlas SW. Intracranial Hemorrhage. In: Atlas SW, ed.
Magnetic Resonance Imaging of the Brain and Spine. New York, NY:
Raven Press; 1991:175222.
9. Mattle HP, Edelman RR, Schroth G, O9Reilly GV. Spontaneous and
Traumatic Hemorrhage in Clinical and Magnetic Resonance Imaging.
Vol 1. 2nd ed. Philadelphia, Pa: WB Saunders Co; 1996:652702.
10. Ebisu T, Tanaka C, Umeda M, Kitamura M, Fukunaga M, Aoki I, Sato H,
Higuchi T, Naruse S, Horikawa Y, Ueda S. Hemorrhagic and nonhemorrhagic stroke: diagnosis with diffusion-weighted and T2-weighted
echo-planar MR imaging. Radiology. 1997;203:823 828.
11. Patel MR, Edelman RR, Warach S. Detection of hyperacute primary
intraparenchymal hemorrhage by magnetic resonance imaging. Stroke.
1996;27:23212324.
12. Koroshetz WJ, Gonzalez G. Diffusion-weighted MRI: an ECG for brain
attack? Ann Neurol. 1997;41:565566.
13. Fisher M, Prichard JW, Warach S. New magnetic resonance techniques
for acute ischemic stroke. JAMA. 1995;274:908 911.
14. Mintorovitch J, Moseley ME, Chileuitt L, Shimizu H, Cohen Y,
Weinstein PR. Comparison of diffusion- and T2-weighted MRI for the
early detection of cerebral ischemia and reperfusion in rats. Magn Reson
Med. 1991;18:39 50.
15. Tong DC, Yenari MA, Albers GW, M OB, Marks MP, Moseley ME.
Correlation of perfusion- and diffusion-weighted MRI with NIHSS score
in acute (,6.5 hour) ischemic stroke. Neurology. 1998;50:864 870.
16. Sorensen AG, Buonanno FS, Gonzalez RG, Schwamm LH, Lev MH,
Huang Hellinger FR, Reese TG, Weisskoff RM, Davis TL, Suwanwela N,
Can U, Moreira JA, Copen WA, Look RB, Finklestein SP, Rosen BR,
Koroshetz WJ. Hyperacute stroke: evaluation with combined multisection
diffusion-weighted and hemodynamically weighted echo-planar MR
imaging. Radiology. 1996;199:391 401.
17. Warach S, Dashe JF, Edelman RR. Clinical outcome in ischemic stroke
predicted by early diffusion-weighted and perfusion magnetic resonance
imaging: a preliminary analysis. J Cereb Blood Flow Metab. 1996;16:5359.
18. Barber PA, Darby DG, Desmond PM, Yang Q, Gerraty RP, Jolley D, Donnan
GA, Tress BM, Davis SM. Prediction of stroke outcome with echoplanar
perfusion- and diffusion-weighted MRI. Neurology. 1998;51:418426.
19. Powers WJ, Zivin J. Magnetic resonance imaging in acute stroke: not
ready for prime time. Neurology. 1998;50:842 843.
20. Rosen BR, Belliveau JW, Chien D. Perfusion imaging by nuclear
magnetic resonance. Magn Reson Q. 1989;5:263281.
21. Busch E, Beaulieu C, de Crespigny A, Moseley ME. Diffusion MR
imaging during acute subarachnoid hemorrhage in rats. Stroke. 1998;29:
21552161.
22. The European Ad Hoc Consensus Group. Optimizing intensive care in
stroke: a European perspective: a report of an ad hoc consensus group
meeting. Cerebrovasc Dis. 1997;7:113128.
23. The National Institute of Neurological Disorders and Stroke rt-PA Stroke
Study Group. Tissue plasminogen activator for acute ischemic stroke.
N Engl J Med. 1995;333:15811587.
24. Hacke W, Kaste M, Fieschi C, Toni D, Lesaffre E, von Kummer R,
Boysen G, Bluhmki E, Hoexter G, Mahagne MH, Hennerici M. Intravenous thrombolysis with recombinant tissue plasminogen activator for
acute hemispheric stroke: the European Cooperative Acute Stroke Study.
JAMA. 1995;274:10171025.

Downloaded from http://stroke.ahajournals.org/ by guest on December 4, 2014

A Standardized MRI Stroke Protocol: Comparison with CT in Hyperacute Intracerebral

Hemorrhage
Peter D. Schellinger, Olav Jansen, Jochen B. Fiebach, Werner Hacke and Klaus Sartor
Stroke. 1999;30:765-768
doi: 10.1161/01.STR.30.4.765
Stroke is published by the American Heart Association, 7272 Greenville Avenue, Dallas, TX 75231
Copyright 1999 American Heart Association, Inc. All rights reserved.
Print ISSN: 0039-2499. Online ISSN: 1524-4628

The online version of this article, along with updated information and services, is located on the
World Wide Web at:
http://stroke.ahajournals.org/content/30/4/765

Permissions: Requests for permissions to reproduce figures, tables, or portions of articles originally published
in Stroke can be obtained via RightsLink, a service of the Copyright Clearance Center, not the Editorial Office.
Once the online version of the published article for which permission is being requested is located, click
Request Permissions in the middle column of the Web page under Services. Further information about this
process is available in the Permissions and Rights Question and Answer document.
Reprints: Information about reprints can be found online at:
http://www.lww.com/reprints
Subscriptions: Information about subscribing to Stroke is online at:
http://stroke.ahajournals.org//subscriptions/

Downloaded from http://stroke.ahajournals.org/ by guest on December 4, 2014