Case Reports
Diffusion-Weighted Magnetic Resonance Imaging in a Case
of Cerebral Venous Thrombosis
J.C. Corvol, MD; C. Oppenheim, MD; R. Manaı̈, MD; M. Logak, MD; D. Dormont, MD;
Y. Samson, MD; C. Marsault, MD; G. Rancurel, MD
Background—Diffusion-weighted imaging (DWI) is the most sensitive MR sequence in acute arterial ischemic stroke but
has not yet been evaluated in venous cerebral ischemia. We describe a patient with DWI performed at the acute phase
of a venous ischemic stroke.
Case Description—A rapid cerebral MRI including DWI and fast fluid-attenuated inversion recovery (FLAIR) sequences
was performed at the acute phase of a venous stroke confirmed by conventional angiography. DWI showed a slight
decrease in apparent diffusion coefficient values 3 hours after onset (0.5360.0731023 mm2/s) and was normal 48 hours
later (0.06460.1531023 mm2/s). Fast FLAIR sequences showed large left frontoparietal hyperintensities. The lack of
a clear decrease in apparent diffusion coefficient values associated with marked FLAIR abnormalities may suggest
prominent or early associated vasogenic edema. Physiopathological differences between arterial and venous ischemia
may explain the different type of DWI FLAIR abnormalities during the acute phase as well as the better recovery of
neurological deficit in venous stroke than in arterial ischemic stroke.
Conclusions—In the context of an acute stroke, the contrast between marked FLAIR and subtle DWI abnormalities on MRI
may reflect the venous mechanism of cerebral ischemia. (Stroke. 1998;29:2649-2652.)
Key Words: cerebral veins n magnetic resonance imaging n magnetic resonance imaging, diffusion-weighted

T

Downloaded from http://ahajournals.org by on April 11, 2024

had right hemiplegia and hypoesthesia and nonfluent aphasia.
She was apyretic, and blood pressure was 140/80 mm Hg
with regular pulse of 70/min. CT scan performed 2 hours and
50 minutes after onset was normal. Three hours after deficit
onset, a rapid MRI was obtained with a total acquisition time
of only 5 minutes. Two sequences were performed on 1.5-T
MR equipment (GE Signa Horizon Echospeed). The first
sequence featured multislice single-shot, spin-echo diffusion,
echo-planar imaging with a pair of diffusion gradients centered around the 180° pulse with 6-mm axial slice thickness,
1.5-mm gap, 96364 matrix, 28321-cm field of view, repetition time of 4000 ms, and effective echo time of 120 ms.
Five sets of 17 slices were acquired with 5 values of b (d525
ms, D528 ms), from b50 to 800 s/mm2, with diffusion
gradients applied simultaneously in 3 orthogonal spatial
directions. The DW raw images were transferred to an
independent workstation. Dedicated software (Functool, GE)
allowed the quantification of the diffusion changes. Calculation of the apparent diffusion coefficient (ADC) was based on
the Stejskal and Tanner equation11 as the negative slope of the
linear regression best fitting the point for b versus ln (SI),
where SI is the signal intensity from a region of interest of the
images acquired at each b value. Performing this calculation
on a pixel-by-pixel basis created maps of ADC. The second

he arterial origin of an ischemic stroke must be confirmed at the acute phase because thrombolysis indication depends on an early diagnosis and is proposed within 3
hours after onset.1 Although progress in neuroimaging makes
the diagnosis of cerebral venous thrombosis (CVT) easier,2– 4
the acute phase of a venous stroke may be difficult to
differentiate from arterial ischemic stroke (AIS) because of
the polymorphism of clinical presentation,2,5 which sometimes mimics an AIS. Diffusion-weighted imaging (DWI) is
a new MR technique that shows early abnormalities within
minutes after stroke onset, related to early cytotoxic edema in
AIS.6 –10 DWI performed at the acute phase of CVT has not
yet been evaluated.
We describe a patient with CVT whose initial clinical
presentation suggested AIS. The early DWI pattern was
unusual for AIS, and the venous mechanism of ischemia was
confirmed by angiography. The MR data are suggestive of
prominent vasogenic edema, which may be the basis for the
pathophysiology of venous ischemia.

Case Report
A 64-year-old woman was admitted to the hospital 2 hours
and 30 minutes after the sudden onset of right hemiplegia and
nonfluent aphasia. On admission, this right-handed woman

Received June 23, 1998; final revision received July 31, 1998; accepted August 18, 1998.
From the Service d’Urgences Cérébro-Vasculaires, Groupe hospitalier Pitié-Salpétrière (J.C.C., R.M., M.L., Y.S., G.R.), and Service de
Neuroradiologie Marsault, Groupe hospitalier Pitié-Salpétrière (C.O., D.D., C.M.), Paris, France.
Reprint requests to G. Rancurel, MD, Service d’Urgences Cérébro-Vasculaires, Groupe hospitalier Pitié-Salpétrière, 47 Boulevard de l’hôpital, 75631
Paris Cedex 13, France.
© 1998 American Heart Association, Inc.
Stroke is available at http://www.strokeaha.org

2649

2650

Diffusion-Weighted Imaging in Cerebral Venous Thrombosis

Downloaded from http://ahajournals.org by on April 11, 2024

Figure 1. Left, Axial cerebral DWI (b5800 s/mm2) obtained 3 hours after onset shows a barely visible hypersignal limited to the left
rolandic area (arrow). There is a corresponding slight decrease of ADC values (0.5360.0731023 mm2/s) on the ADC maps compared
with a matching location in the uninvolved contralateral hemisphere (0.6160.0931023 mm2/s). Right, Axial cerebral fast FLAIR
sequences obtained on the same date at the same level show large left frontoparietal hyperintensities. Note the contrast between subtle abnormalities in DWI and marked fast FLAIR abnormalities.

sequence featured fast fluid-attenuated inversion recovery
(FLAIR) acquisition with 5-mm axial interleaved slices,
2563256 matrix, 24324-cm field of view, repetition time of
10 002 ms, effective echo time of 148 ms, inversion time of
2200 ms, and 32-kHz bandwidth; direction of the frequency
encoding gradient was superior to inferior.
DWI showed a discrete hypersignal limited to the left
rolandic area in only 1 slice (5 mm). There was a corresponding very slight decrease of ADC values on the ADC maps
(0.5360.0731023 mm2/s) compared with a matching location
in the uninvolved contralateral hemisphere (0.6160.093
1023 mm2/s) (Figure 1, left panel). This slight decrease of
ADC values contrasted with the clear ADC drop usually
observed at the acute phase of AIS. Conversely, the fast
FLAIR sequences showed large left frontoparietal hyperintensities visible on 7 slices (35 mm) (Figure 1, right panel).
The patient had a generalized seizure just after the MRI
was performed. She was therefore rejected for thrombolysis,
and a second MRI was performed 48 hours later. The
hypersignal on DWI completely disappeared, and ADC
values returned to normal (0.6460.1531023 mm2/s) compared with the uninvolved contralateral hemisphere
(0.6360.1931023 mm2/s) (Figure 2). Signal abnormalities on
FLAIR sequences were unchanged.
These results, further supported by the generalized seizure,
were unusual for AIS and led us to suspect a venous stroke.
This was confirmed by MRI angiography and conventional

angiography, which showed extensive thrombosis of the
superior sagittal and left lateral sinuses (Figure 3). D-dimer
values were .1000 ng/mL (enzyme-linked immunosorbent
assay method). The patient had a hemoglobin concentration
of 61 g/L, with anemia related to a previously unknown
gastric ulcer. Platelets and leukocytes were normal; coagulation tests, including normal levels of protein C, protein S, and
antithrombin III, were normal; and antinuclear and antiphospholipid antibodies were negative. The patient’s family later
reported a history of 8 days of continuous headaches preceding the stroke. She was treated with intravenous heparin
therapy (adjusted according to partial thromboplastin time 2
to 3 times the partial thromboplastin time before administration of heparin), phenytoin 300 mg/d per os, omeprazole 20
mg/d, and eradication of Helicobacter pylori. The patient’s
outcome was favorable, with marked regression of neurological deficit within 3 days. She was discharged 3 weeks later
with only a moderate right ataxia.

Discussion
The context of a sudden neurological deficit with a normal
initial CT scan was compatible, in this patient, with AIS in
the left superficial middle cerebral artery territory. Initially,
this patient was considered to have AIS and was admitted to
our stroke center for possible inclusion in a thrombolysis
protocol (European Cooperative Acute Stroke Study II).

Corvol et al

Figure 2. Axial cerebral DWI (b5800 s/mm2) obtained 48 hours
after onset shows the disappearance of the left rolandic hypersignal. ADC value returned to normal (0.6460.1531023 mm2/s)
on the ADC maps compared with a matching location in the
uninvolved hemisphere (0.6360.1931023 mm2/s).
Downloaded from http://ahajournals.org by on April 11, 2024

A 5-minute MRI including DWI and a fast FLAIR sequence was obtained. DWI, first developed by Le Bihan et
al,6 is a new MR technique based on the molecular motion of
water. It is extremely sensitive at the acute phase of AIS since
abnormalities appear within minutes after onset in animals,12
related to slower diffusion motion in intracellular cytotoxic
edema induced by ischemia, resulting in a drop of ADC
values.8 –10 In our patient, the slight decrease of ADC values
observed on DWI 3 hours after onset (Figure 1, left panel)

December 1998

2651

was inconsistent with the clear ADC drop usually observed in
AIS.13 Moreover, ADC values returned to normal at 48 hours
after onset (Figure 2), while they should reach minimum
values in AIS.9
FLAIR sequences, developed by Rydberg et al14 allow the
acquisition of heavily T2-weighted, cerebrospinal fluid–
nulled images. These sequences are highly sensitive to
ischemic strokes,15,16 but, as shown for conventional T2
sequence,17 FLAIR abnormalities may be delayed compared
with DWI abnormalities in AIS. However, the hyperintensities were marked and larger than on DWI only 3 hours after
the sudden clinical deficit, suggesting preexisting but clinically asymptomatic focal abnormalities. Overall, these data
were compatible with a venous ischemic stroke developed on
a preexisting cortical venous and superior sagittal sinuses
thrombosis, as confirmed later by angiography.
The venous strokes share some common features with
other types of focal ischemia, such as a reduced capillary
perfusion pressure,18,19 increased cerebral blood volume,20
and decreased cerebral blood flow.21 However, venous ischemia also has specific characteristics, such as an early bloodbrain barrier disruption and an increase of the net capillary
filtration, that may result in early vasogenic edema.19,24 In our
patient, the contrast between slight DWI abnormalities and
marked FLAIR abnormalities may reflect prominent and
early vasogenic edema associated with mild cytotoxic edema.
This contrast may explain the excellent and rapid recovery of
neurological deficit of most cases of venous stroke that is
treated early.22 A relative preservation of neuronal tissue has
also been documented recently by Hsu et al,23 using proton
MR spectroscopy in a case of deep CVT. The authors
suggested an impaired but viable tissue in that case.
Another argument for an important vasogenic rather than a
pure cytotoxic component of the edema in venous ischemia is
derived from the experimental rat model of CVT described by
Röther et al 24 Although a strict comparison is difficult to
make, the authors, who also used DWI, showed an early
marked decrease of ADC values, followed by an increase
during 48 hours. They concluded that early cytotoxic edema
was closely followed by vasogenic edema.
To our knowledge, this is the first reported case of early
DWI after onset of venous stroke in humans. Although the
diversity of CVT patterns of presentation precludes definitive
conclusions from a single case, the present findings clearly
show that a 5-minute MRI protocol, which can easily be
performed within the therapeutic window of cerebral ischemia, may greatly improve the accuracy of diagnosis in acute
stroke. Moreover, our results suggest that new MR techniques
may improve knowledge of the pathophysiology of cerebrovascular disorders.

References

Figure 3. Cerebral digitalized cerebral angiography. Lateral projection shows thrombosis of the superior sagittal sinus (arrows)
with flow inversion in cortical veins (arrowheads).

1. 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;333:1581–1587.
2. Anxionnat R, Blanchet B, Dormont D, Bracard S, Chiras J, Maillard S,
Louail C, Moret C, Braun M, Roland J, Marsault CL, Picard L. Present
status of computerized tomography and angiography in the diagnosis of
cerebral thrombophlebitis cavernous sinus thrombosis excluded. J Neuroradiol. 1994;21:59 –71.

2652

Diffusion-Weighted Imaging in Cerebral Venous Thrombosis

3. Dormont D, Anxionnat R, Evrard S, Louaille C, Chiras J, Marsault C.
MRI in cerebral venous thrombosis. J Neuroradiol. 1994;21:81–99.
4. Chelly D, Levy C, Ameri A, Brunereau L. Imaging of cerebral thrombophlebitis. Ann Radiol (Paris). 1994;37:108 –117.
5. Bousser MG, Ross Russell R. Cerebral Venous Thrombosis. Major Problems in
Neurology Series, No. 13. London, England: WB Saunders; 1997.
6. Le Bihan D, Breton E, Lallemand D, Grenier P, Cabanis E, Lava-Jeantet
M. MR imaging of intravoxel incoherent motions: application to diffusion
and perfusion in neurologic disorders. Radiology. 1986;161:401– 407.
7. Basser PJ, Mattiello J, Le Bihan D. Estimation of the effective selfdiffusion tensor from the NMR spin echo. J Magn Reson B. 1994;103:
247–254.
8. Fisher M, Prichard JW, Warach S. New magnetic resonance techniques
for acute ischemic stroke. JAMA. 1995;274:908 –911.
9. Warach S, Chien D, Li W, Ronthal M, Edelman RR. Fast magnetic
resonance diffusion-weighted imaging of acute human stroke. Neurology.
1992;42:1717–1723.
10. Bahn MM, Oser AB, Cross DT. CT and MRI of stroke. J Magn Reson
Imaging. 1996;6:833– 845.
11. Stejskal E, Tanner J. Spin diffusion measurements: spin echos in the
presence of a time dependent field gradient. J Chem Phys. 1965;42:
288 –292.
12. Schäbitz WR, Fisher M. Diffusion-weighted imaging for acute cerebral
infarction. Neurol Res. 1995;17:270 –274.
13. Ulug AM, Beauchamp N, Bryan RN, van Zijl PC. Absolute quantitation
of diffusion constants in human stroke. Stroke. 1997;28:483– 490.
14. Rydberg JN, Hammond CA, Grimm RC, Erickson BJ, Jack CR Jr, Huston
J, Riederer SJ. Initial clinical experience in MR imaging of the brain with
a fast fluid-attenuated inversion recovery pulse sequence. Radiology.
1994;193:173–180.

15. Alexander JA, Sheppard S, Davis PC, Salverda P. Adult cerebrovascular
disease: role of modified rapid fluid-attenuated inversion-recovery
sequences. AJNR Am J Neuroradiol. 1996;17:1507–1513.
16. Noguchi K, Ogawa T, Inugami A, Fujita H, Hatazawa J, Shimosegawa E,
Okudera T, Uemura K, Seto H. MRI of acute cerebral infarction: a
comparison of FLAIR and T2-weighted fast spin-echo imaging. Neuroradiology. 1997;36:406 – 410.
17. Lutsep HL, Albers GW, de Crespigny A, Kamat GN, Marks MP, Moseley
ME. Clinical utility of diffusion-weighted magnetic resonance imaging in
the assessment of ischemic stroke. Ann Neurol. 1997;41:574 –580.
18. Wagner E, Traystman R. Effects of cerebral venous and cerebrospinal
fluid pressure on cerebral blood flow. In: Auer L, Löw F, eds. The
Cerebral Veins. New York, NY: Springer; 1983:223–230.
19. Villringer A, Mehraein S, Einhäupl KM. Pathophysiological aspects of
cerebral sinus venous thrombosis. J Neuroradiol. 1994;21:72– 80.
20. Gotoh M, Ohmoto T, Kuyama H. Experimental study of venous circulatory disturbance by dural sinus occlusion. Acta Neurochir (Wien).
1993;124:120 –126.
21. Ungersböck K, Heimann A, Kempski O. Cerebral blood flow alterations
in a rat model of cerebral sinus thrombosis. Stroke. 1993;24:563–569.
22. Einhaüpl KM, Villringer A, Meister W, Mehraein S, Garner C, Pellkofer
M, Haberl RL, Pfister HM, Schmiedek P. Heparin treatment in sinus
venous thrombosis. Lancet. 1991;338:597– 600.
23. Hsu LC, Lirng JF, Fuh JL, Wang SJ, Shyu HY, Liu HC. Proton magnetic
resonance spectroscopy in deep cerebral venous thrombosis. Clin Neurol
Neurosurg. 1998;100:27–30.
24. Röther J, Waggie K, van Bruggen N, de Crespigny AJ, Moseley ME.
Experimental cerebral venous thrombosis: evaluation using magnetic
resonance imaging. J Cereb Blood Flow Metab. 1996;16:1353–1361.

Downloaded from http://ahajournals.org by on April 11, 2024