вход по аккаунту


Diffusion-weighted MRI in transient global amnesia Elevated signal intensity in the left mesial temporal lobe in 7 of 10 patients.

код для вставкиСкачать
Diffusion-Weighted MRI in Transient
Global Amnesia; Elevated Signal Intensity
in the Left Mesial Temporal Lobe in
7 of 10 Patients
Michael Strupp, MD,* Roland Bruning, MD,? Ren Hua Wu, M D , t Michael Deimling, PhD,$
Maximilian Reiser, MD,t and Thomas Brandt, MD, FRCP"
Prompted by the findings of previous studies with positron emission tomography and single photon emission computed tomography, which demonstrated hypoperfusion or hyperperfusion in the left temporal lobe in isolated patients
with transient global amnesia (TGA), we compared the sensitivity of diffusion-weighted (DW) magnetic resonance
imaging (MRI) with that of conventional T1- and T2-weighted MRI in patients with TGA. Ten patients with the
typical syndrome of a pure TGA were included in the study. For all patients, a coronal DW sequence, a steady-state
free precession (SSFP) sequence, and conventional T I - and T2-weighted turbo spin-echo sequences were obtained.
Seven of the 10 patients had elevated signal intensity in the left hippocampal region on DW MRI; moreover, 3 of
these 7 patients exhibited bilateral signal abnormality in this sequence. All conventional T1- and T2-weighted images
as well as all follow-up studies were normal. The signal elevation in DW MRI correlates with a decrease in the
interstitial space and with cellular edema in the temporal lobe during TGA. The underlying pathomechanism causing
this cellular edema cannot be clearly outlined using DW MRI. Our data are, however, compatible with spreading
depression. This is the first study to show that DW MRI is a sensitive MRI method for evaluating TGA, especially in
the early stage of the disease.
Strupp M, Bruning R, Wu RH, Deimling M, Reiser M, Brandt T. Diffusion-weighted MRI
in transient global amnesia: elevated signal intensity in rhe left mesial remporal lobe
in 7 of 10 patients. Ann Neurol 1998;43:164-170
Transient global amnesia (TGA) is a benign neurological syndrome. It is characterized by an abrupt,
clear-cut onset of anterograde amnesia, usually accompanied by repetitive questioning, and generally
resolves within 24 hours.'-3 During the attack the patient remains alert and communicative. Other accompanying focal neurological symptoms or epileptic features are absent. General cognition appears to be
intact, and cognitive impairment is limited to amnesia (eg, no aphasia or apraxia). The pathogenesis and
etiology of TGA are still unclear. Several hypotheses
about its cause (eg, thromboembolism, epilepsy, or
migraine) have been proposed. Most studies of patients with pure TGA (definition criteria given by
Hodges')), however, have found no evidence for cerebrovascular disease2,* or for
Pure TGA
does not have a morphological correlate on conven-
tional (TI-, T2-, or proton-weighted) magnetic resonance imaging (MRI); the electroencephalogram of
these patients is also normal in most cases. Previous
studies of single cases using single photon emission
computed tomography and positron emission tomography, however, demonstrated hypoperfusion as well
as hyperperfusion in the left temporal
So far,
bitemporal hypoperfusion has been described during a
TGA in only 2 patients.','
Diffusion-weighted (DW) MRI allows measurement
of the self-diffusion of water predominantly within the
interstitial space.'-'' DW sequences are sensitive to the
movements of water molecules and transport in biological tissues. So far, DW MRI has shown promise in the
early detection of brain i n f a r ~ t i o n . 'This
~ ~ ~study
the first to evaluate the sensitivity of D W MRI in patients with acute or subacute pure TGA.
From the Departments of *Neurology and ?Radiology, Klinikum
Grosshadern, University of Munich, and Siemens Medical Systerns, Erlangen, Germany.
Address correspondence ro Dr Strupp, Department of Neurology,
Klinikum Grosshadern, University of Munich, Marchioninimage
15, 81366 Munich, Germany.
Received Jul 25, 1997, and in revised form Sep 23. Accepted for
publication Sep 23, 1997.
164 Copyright 0 1998 by the American Neurological Association
Subjects and Methods
Magnetic Resonance Imaging
Magnetic resonance imaging was performed with a superconductive 1.5-Tesla scanner (Vision, Siemens Medical Systems). The coronal D W sequence was a steady-state free precession (SSFP) sequence; the parameters used were as
follows: TR, 22 msec; voxel size, 1.1 X 1.1 X 6 mm; 10
averages; and imaging time per slice, 57 seconds; and three
slices were obtained at each diffusion weighting.
The diffusion gradient was applied along the frequency
encoding direction. The technique for measuring diffusion
has been described earlierI5 as well as the modality for using
DW MRI in conjunction with a steady-state sequence.' ',16217
The DW MRI sequence used is based on a sequence design,
which utilizes the echo part of the SSFP signals.I8 With flip
angles less than 90°, the MR signal should be composed of a
sum of primary echoes, stimulated echoes, and higher-order
echoes. Thus, the application of a diffusion-sensitizing gradient within this sequence makes this technique attractive,
for the diffusion time could be in the range of T 1 time.20.21
Because of the complicated signal generation in the nonbalanced D W SSFP sequence, it is not possible to characterize
the diffusion sensitivity by a plain b value as in conventional
SE sequences based on the Stejskal-Tanner s ~ h e m e . ' ~
Diffusion measurements were made on a tissue equivalent
phantom of fluid dimethylsulfoxide (DMSO). Signal intensity was measured for a given pluse length 6, the gradient
strength Gdi, was varied in eight steps from 3 to 23 mTIm,
a TR of 22 msec was used, and the applied flip angle was
= 1.2 X
50". With the known diffusion constant of D,
l o p 3 mm'lsec at 295°K from the fitted signal decay, b values were derived as follows: 6 = 2 msec: b = 165, 6 = 3
msec: b = 280, 6 = 5 msec: b = 595 sec/mm2. T o describe
a corresponding b value in DW MRI with SSFP requires
knowledge of the specific tissue relaxation times T 1 and
T2, and the effective flip angle a. Consequently, no diffusion coefficient could be easily calculated from an SSFP
Between 1994 and 1996, we conducted a prospective study
on 10 patients who presented with pure TGA, as defined by
the following criteria3:
1. There must be a clear-cut anterograde amnesia during the
2. Clouding of consciousness and loss of personal identity
must be absent, and the cognitive impairment must be
limited to amnesia.
3. There should be no accompanying focal neurological
symptoms or functionally relevant focal signs.
4. Epileptic seizures must be absent.
5. Attacks must be resolved within 24 hours.
6. Patients with recent head injury or known active epilepsy
are excluded.
7. There must be a witness of the attacks or information available from a capable observer present during the attack.
Our patients presented with clinical symptoms of TGA,
which had begun 2 to 144 hours before the radiological examination. All patients underwent a neurological and neuropsychological examination, T1- or T2-weighted and D W
MRI, as well as electroencephalography. Clinical follow-up
was done 1 week after the TGA. This study was conducted
according to the Helsinki I1 declaration, and all patients and
control subjects gave their informed consent to participate.
Ten patients (7 men, 3 women) participated; their ages
ranged from 53 to 71 years (mean age, 59.2 years). The clinical details and time of MRI after TGA are summarized in
Table 1. All patients had had their first TGA at the time of
admission. All symptoms lasted less than 24 hours and disappeared without any disability remaining, other than partial
amnesia for the episode. During the follow-up period (2-26
months) none of the patients presented with transient ischemic attack, stroke, epileptic seizure, or recurrent TGA.
Table I . Clinical Details of Patients with Transient Global Amnesia
Elevated Signal
Intensity in
Duration of
Episode (hr)
Precipitating Factors
Vascular Risk
Time of MRI
After TGA (hr)
Physical exertion
Sexual intercourse;
immersion in
Immersion in water
Emotional stress
Immersion in water
BH = both hippocampal regions; LH
2, after onset of
symptoms (ie,
during TGA)
Immersion in water
Immersion in water
N, artifacts
N, artifacts
left hippocampal region; N = normal; - = none.
Strupp et al: Diffusion-Weighted MRI in Transient Global Amnesia
Conventional TI- and T2-weighted turbo spin-echo
(TSE) sequences were also obtained (TWTE = 550/16 msec
and TWTE = 2,500/90 msec). Special care was taken to
angulate the slices perpendicular to the hippocampus, and all
sequences were obtained in identical orientations. In the film
review process to ensure consensus reading, two readers
(R.B., R.H.W.) compared the corresponding slices of DW
MRIs and TSE images for abnormalities in signal intensity
or changes in morphology. The presence of an abnormal signal intensity was assumed, if signal intensity of an area exceeded that of other regions of gray or white matter. Signalto-noise ratios were measured on the right hemisphere as a
region of interest (ROI) and divided by the standard deviation of background noise, measured for a large ROI.
Four of 10 patients with clinically confirmed pure
TGA showed elevated signal intensity in the left mesial
temporal region unilaterally, 3 of 10 patients showed
elevated signal intensity in both mesial temporal lobes,
and the D W images of 3 of 10 patients were normal.
Two of the latter exhibited gross motion, which degraded the image quality of the DW image. An example of a unilaterally increased signal intensity in the left
hippocampal region is given in Figure 1. The DW
MRI and T2-weighted MRI in this patient were performed 24 hours after TGA onset. In 4 of the 10 patients with unilaterally increased signal intensity, the
DW MRIs and T2-weighted MRIs were obtained after
a mean of 27 hours (18, 24, 28, 38 hours) after the
TGA. Although the changes in signal intensity were
subtle, probably due to the low signal-to-noise ratio of
the DW sequence (signal-to-noise ratios of the DW
MRI were 30.7 for 6 = 2 msec, 28.5 for 6 = 3 msec,
and 23.2 for 6 = 5 msec), they were easily identified
by readers during the film reading process. No decrease
of the diffusion signal was observed in the lateral
temporal or other temporal or parietal regions of any
T o enhance clarity of the visual impression, the ratio
of signal intensities of the diseased left mesial temporal
region versus lateral gray matter of the left temporal
lobe was calculated. The mean value ( 2 SD) for the
left mesial temporal lobe in the 7 patients with TGA
was 1.33 -+ 0.14; the mean value (2SD) for the lateral gray matter of the left lateral temporal lobe was
1.09 ? 0.08 (Fig 2). The means differed significantly
( p < 0.0001, unpaired Student’s t test). Conventional
T2-weighted imaging, measured in the same orientation, was normal in signal intensity in all parients. Furthermore, the size of both the hippocampus and the
parahippocampal fissure was considered unremarkable,
and no morphological changes were observed in the
mesial hippocampal region or elsewhere. All follow-up
examinarions of the 10 patients were negative, thus
supporting the impression that the observed changes
Fig 1. Patient 5 with transient global amnesia 24 hours befDre the scan (see Table 1). (A) Conventional T2-weighted imaging.
(B) Dzfision-weighted sequence demonstrates elevated signal intenrig in the l e j mesial temporal region, mainly in the hippocampus.
166 Annals of Neurology Vol 43
No 2 February 1998
temporal lobe structures, (2) the diencephalon, or (3)
the basal forebrain n ~ c l e i . ~ , ” , ’Seven
of our 10 patients with TGA had elevated signal intensity in the
left hippocampal region in the DW MRI SSFP; only
the right temporal lobe showed no isolated abnormalities. The bilateral signal changes found in 3 of 10 patients were most pronounced in the patient in whom
the MRI was performed 2 hours after the onset of
TGA. Stillhard’ and Evans and colleagues’ also demonstrated bitemporal changes (hypoperfusion) on a single photon emission computed tomogram of 1 patient.
These changes returned to normal within 3 weeks.
Therefore, our data suggest (1) that initially there are
bitemporal disturbances during TGA, which explain
global amnesia; ( 2 ) that the disturbances in the left
temporal lobe last longer than in the right temporal
lobe, which agrees with a study that showed that patients after TGA may have long-lasting if not permanent but subtle deficits in verbal episodic memoryz4;
and (3) that an involvement of the right temporal lobe
alone is not sufficient to cause amnesia.
Fig 2. Signal intensity ratios (region of interesthegion of interest of the white matter) of the lateral gray matter (LGM) and
the l e j mesial hippocampus (MH) of the 7patients who had
unilaterally increased signal intensity of the lefi mesial temporal region (identijed by two independent readers). The mean
value (2 SD)for the MH was 1.33 ? 0.14 and 1.09 ?
0.08 f i r the gray matter of the lateral temporal lobe (p <
0.0001, unpaired Student? t test).
had been reversible and also indicating the absence of a
structural lesion.
Three of the 10 patients showed bilateral signal abnormality in the hippocampal region on the DW sequence. In this subgroup the mean time difference between the TGA and the MRI was 18 hours (0, 23, 33
hours). The DW MRI shown in Figure 3 was performed very early in this patient (ie, during the TGA;
2 hours after manifestation of the symptoms). Conventional T2-weighted imaging, measured in the same orientation, and a follow-up DW MRI performed 2
weeks later were normal (see Fig 3).
The mean follow-up interval for 3 of 10 patients
with negative DW MRIs was 98 hours (28, 120, 144
hours). Two of these patients (28 and 120 hours after
TGA onset) moved their head in the scanner during
image acquisition, thus causing motion artifacts that
hinder the film reading process. The other patient with
a negative DW MRI was scanned 144 hours after the
TGA (not shown).
Locution of the Functional Disturbance in TGA
It is generally agreed that severe amnesia whether acute
or transient is related to biluteral damage of (1) mesial
Mechanism of TGA
The pathomechanism and etiology of TGA are still unclear. No epidemiological evidence is available from
studies of risk factor prevalence or natural history to
support a cerebrovascular etiology or an epileptic
m e ~ h a n i s m .Pure
~ TGA, however, has been shown to
be associated with migraine, which also has precipitating factors and a benign prognosis. Studies have shown
that the prevalence of migraine in patients with TGA is
about 25%,2 which is significantly larger than in
matched control groups. Furthermore, changes in regional cerebral blood flow have been reported in both
migraine and TGA.5,6,25,z6
Spreading depression of Leao27-29 may provide an
explanation for both diseases, The similarities between
TGA, migraine, and spreading depression are summarized in Table 2. Spreading depression-like
and migraine-can be provoked by several stimuli, for
instance, intense volleys of sensory inputs into the hippocampus, which releases the neurotransmitter glutamate, known to be a cause of spreading depression in
animals. Spreading depression can also be elicited in
the hippocampus of humans.30 Consequently, it has
been postulated that glutamate, the most important
transmirter in the hippocampus, may evoke spreading
depression and cause amnesia in humans.31 Spreading
depression causes changes in extracellular and intracellular ion concentrations (eg, rise in extracellular Kf
and intracellular Ca2+) with corresponding electrochemical gradient changes, depolarization of neurons
and glial cells, cell swelling, and shrinkage of the interstitial space.32 It is not followed by permanent neuronal damage (under normoxic condition^)^^ and is compensated by increased glucose metabolism and blood
Strupp et al: Diffusion-Weighted MRI in Transient Global Amnesia
Fig 3. Patient with ongoing transient global amnesia during the scan (Patient 1, see Table 1). (A) Conventional T2-weighted imaging. (B) Difision-weighted sequence demonstrates elevated signal intensity in both mesial temporal regions, mainly in the hippocampus. Follow-up scans with conventional T2-weighted imaging (C) and diffusion-weighted sequence ( 0 ) showed normal signal
intensity 2 weeks later.
flow during the repolarization phase.34 Several animal
studies (eg, with rats) have shown that D W MRI can
detect spreading depre~sion.’~
The signal abnormalities in the mesial temporal lobe
168 Annals of Neurology
Vol 43
No 2 February 1998
seen in D W MEU of our TGA patients seem to be
caused by reduced diffusion of interstitial water, owing
to a relative decrease of the interstitial space, which
suggests cell swelling. Changes of relative signal inten-
Table 2. Similarities between TGA, Migraine, and Spreading Depression
Precipitating factors
Transient symptoms
No persisting deficits and
good prognosis
Recurrent episodes
Changes in regional cerebral
blood flow
Signal changes in TI- and T2weighted images
Increased signal intensity in
(eg, stress, immersion in
water, physical activity)
(eg, stress, certain nutrients,
hormonal changes)
in 15 to 20%
(in most cases)
(in most cases, except
migraine-associated stroke)
Spreading Depression
(increased sensory input)
See Discussion for references.
sity ratios seen in the present study were highly significant, compared with nonaffected regions of the lateral
temporal lobe. Changes in signal intensity on DW images, however, are not specific for a single pathological
entity and do not discriminate between fluid shifts due
to cerebral ischemia and those arising secondary to effects of neuronal dysfunction, being positive under
both conditions. Baird and associate^^^ postulate that
DW MRI signal alterations may be induced by spreading waves of restricted diffusion or by intermittent depolarizations. Although this has never been observed in
human stroke, one can speculate that the lesion volume
may have been underestimated, because imaging was
performed between waves of the spreading depression,
especially in patients examined days after the onset. Although our study design did not allow serial scanning
of the patients within the expected propagation time of
diffusion coefficient changes in spreading depression ( 3
mm/min), 70% of our patients had left signal changes
in DW MRI and all T2-weighted images were normal
as well as all follow-up examinations of T 2 and diffusion weighting. Thus, a permanent damage of the neuronal structures, as occurs in most ischemic lesions,
seems unlikely. Lutsep and co-workersl* evaluated the
clinical utility of DW MRI for assessing ischemic
stroke in 103 patients. Compared with T2-weighted
images alone, D W MRI provided additional clinically
relevant findings in only 8% of all patients. This is in
contrast to our patients with TGA and thus supports
the hypothesis that TGA is not caused by ischemia.
In conclusion, the results of MRI/DW MRI in combination with the clinical findings and the similarities
between TGA and spreading depression (see Table 2)
support our interpretation that spreading depression
may be the underlying mechanism of TGA, but they
do not prove it.
We thank J. Benson for critically reading the manuscript.
1. Bender MB. Syndrome of isolated episodes of confusion with
amnesia. J Hillside Hospital 1956;5:212-215
2. Hodges JR, Warlow CP. The aetiology of transient global
amnesia: a case-control study of 114 cases with prospective
follow-up. Brain 1990;113:639-657
3. Hodges JR. Transient amnesia: clinical and neuropsychological
aspects. London: WB Saunders, 1991
4. Shuping JR, Rollinson RD, Toole JF. Transient global amnesia.
Ann Neurol 1980;7:281-285
5. Laloux P, Brichant C, Cauwe F, Decoster P. Technetium-99m
HM-PA0 single photon emission computed tomography imaging in transient global amnesia. Arch Neurol 1992;49:543-546
6. Matsuda H , Higashi S, Tsuji S, et al. High resolution Tc-99m
HMPAO SPECT in a patient with transient global amnesia.
Clin Nucl Med 1993;18:46-49
7. Stillhard G, Landis T , Schiess R, et al. Bitemporal hypoperfusion in transient global amnesia: 99m-Tc-HM-PA0 SPECT
and neuropsychological findings during and after an attack.
J Neurol Neurosurg Psychiatry 1990;53:339-342
8. Evans J, Wilson B, Wraight EP, Hodges JR. Neuropsychological and SPECT scan findings during and after transient global
amnesia: evidence for the differential impairment of remote episodic memory. J Neurol Neurosurg Psychiatry 1993;56:12271230
9. Le Bihan D, Turner R, Douek P, Patronas N. Diffusion MR
imaging: clinical application. AJNR Am J Neuroradiol 1996;
10. Le Bihan D, Breton E, Lallemand D, et al. Separation of diffusion and perfusion in intravoxel incoherent motion MR imaging. Radiology 1988;168:497-505
11. Le Bihan D, Breton E, Lallemand D, et al. MR imaging of
intravoxel incoherent motions: application to diffusion and perfusion in neurologic disorders. Radiology 1986;161:401-407
12. Mintorovitch J, Moseley ME, Chileuitt L, et al. Comparison of
diffusion and T2-weighted MRI for early detection of cerebral
ischemia and reperfusion in rats. Magn Reson Med 1991;18:
13. Chien D, Kwong KK, Cress DR, et al. MR diffusion imaging
Strupp et al: Diffusion-Weighted MRI in Transient Global Amnesia
of cerebral infarctions in humans. AJNR Am J Neuroradiol
1992;13:1097-1 102
Lutsep HL, Albers GW, Decrespigny A, et al. Clinical utility of
diffusion weighted magnetic resonance imaging in the assessment of ischemic stroke. Ann Neurol 1997;41:574-580
Stejskal EO, Tanner JE. Spin diffusion measurements: Spin
echos in the presence of a time-dependent field gradient. J
Chem Phys 1965;42:288-292
Deimling M, Heid 0. High resolution SSFP diffusion imaging.
In: Proceedings of the Second Annual Meeting of the Society of
Magnetic Resonance, San Francisco, 1994;1033
Bruning R, Wu RH, Deimling M, Reiser M. Diffusion measurements in the ischemic human brain with steady state sequence. Invest Radio1 1996;31:709-715
Le Bihan D. Intravoxel incoherent motion imaging using
steady-state free precession. Magn Reson Med 1988;7:346-35 1
Woessner DE. Effects of diffusion in nuclear magnetic resonance spin-echo experiments. J Chem Phys 1961;34:20572061
Merboldt KD, Bruhn H, Frahm J, et al. MRI of “diffusion” in
the human brain: new results using a modified CE-FAST sequence. Magn Reson Med 1989;9:423-429
Le Bihan D, Turner R, MacFall JR. Effects of intravoxel incoherent motions (IVIM) in steady-state free precession (SSFP)
imaging: application to molecular diffusion imaging. Magn Reson Med 1989;10:324-337
Markowitsch HJ. Can amnesia be caused by damage of a single
brain structure? Cortex 1984;20:27-45
Milner B. Memory and the medial temporal regions of the
brain. In: Pribram KH, Broadbent DE, eds. Biology of memory. New York: Academic Press, 1970
Hodges JR, Oxbury SM. Persistent memory impairment following transient global amnesia. J Clin Exp Neuropsychol
Weiller C , May A, Limmroth V, et al. Brain stem activation in
770 AiinaIs of Neurology
Vol 43
No 2
February 1998
spontaneous human migraine attacks. Nat Med 1995;1:658660
Ferrari MD, Haan J, Blokland JA, et al. Cerebral blood flow
during migraine attacks without aura and effect of sumatriptan.
Arch Neurol 1995;52:135-139
Lea0 AAP. Spreading depression of activity in [he cerebral cortex. J Neurophysiol 1944;7:359-390
Lauritzen M. Pathophysiology of the migraine aura: the spreading depression theory. Brain 1994;117:199-210
Olsen TSJ. Pathophysiology of the migraine aura: the spreading
depression theory. Brain 1995;118:307-308
Sramka M, Brobek G, Bures J, Nadvornik P. Functional ablation by spreading depression: possible use in human stereotactic
neurosurgeiy. Appl Neurophysiol 1977;40:48-61
Olesen J, Jorgensen MB. Leao‘s spreading depression in the
hippocampus explains transient global amnesia: a hypothesis.
Acta Neurol Scand 1986;73:219-220
Jing J, Aitken PG, Somjen GG. Interstitial volume changes
during spreading depression (SD) and SD-like hypoxic depolarization in hippocampal tissue slices. J Neurophysiol 1994;71:
Nedergaard M, Hansen A. Spreading depression is not associated with neuronal injury in the normal brain. Brain Res 1988;
Kocher M. Metabolic and hemodynamic activation of postischemic rat brain by cortical spreading depression. J Cereb
Blood Flow Metab 1990;10:564-571
Rother J, De Crespigny AJ, D‘Arceuil H, Moseley ME. MR
detection of cortical spreading depression immediately after focal ischemia in the rat. J Cereb Blood Flow Metab 1996;16:
214 -220
Baird AE, Benfield A, Schlaug G, et al. Enlargement of human
cerebral ischemic lesion volumes measured by diffusion
weighted magnetic resonance imaging. Ann Neurol 1997;41:
Без категории
Размер файла
794 Кб
patients, global, diffusion, amnesia, lobel, weighted, intensity, transiente, mri, signali, mesial, elevated, temporal, left
Пожаловаться на содержимое документа