Perfusion- and diffusion-weighted magnetic resonance imaging in transient global amnesia

To the Editor:

We read with interest the report by Ay et al. [1] regarding diffusion-weighted imaging (DWI) in transient global amnesia (TGA). We have also studied a case of TGA with DWI, but found no abnormalities on the DWI or its quantitative expression, the apparent diffusion coefficient (ADC) images.

A 53-year-old woman without a history of migraine or vascular risk factors presented with an acute amnestic syndrome. Last seen at 3:30 PM, she was found at 8:30 PM repetitively asking "Where am I?" and "How did I get here?", having apparently just finished bathing. She had a dense anterograde amnesia, as well as a retrograde amnesia for about 48 hours. The remainder of her cognitive, physical, and neurologic examinations; laboratory data, urine, and serum toxic screens; and electrocardiogram were normal. A preliminary diagnosis of TGA was made. MR images were acquired at 1 AM the following day. Standard T1- and T2-weighted images were within normal limits. DWI [2] did not show any signal abnormalities (figure, top row); similarly, there was no alteration of the ADC. We also performed MR perfusion-weighted imaging (PWI) using a T2-weighted gadolinium bolus technique that provided a measurement of regional cerebral perfusion owing to the effect of contrast media on susceptibility, [3] which did not reveal any notable abnormalities of cerebral perfusion (figure, bottom row). The patient's amnesia was present throughout the imaging procedures. Upon being retested at 8 AM, her anterograde and retrograde memory deficits had resolved, with the exception of a several-hour period. Her EEG was normal. She was discharged without further sequelae.

Figure. Four consecutive MRI slices at 7-mm thickness covering thalamus and both medial and lateral temporal lobe structures. Images were obtained on a Siemens Vision 1.5-Tesla echo planar imaging capable system. The top row shows the diffusion-weighted images that were acquired using a time to echo of 118 msec and b-values of 0 and 1,000. The MR diffusion sequence at b = 1,000 was run 3 times with gradients applied in each of the x, y, and z directions. An average of the 3 diffusion directions was calculated to minimize the effects of diffusion anisotropy. Artifacts commonly seen are present in the posterior temporal/petrous bone region. Cerebral blood volume (CBV) maps are shown in the bottom row. Dynamic time series perfusion-weighted images were obtained during the gadolinium bolus to calculate CBV maps. Higher signal intensity is associated with higher blood volume, and vice versa. There was no apparent signal abnormality or regional hemispheric asymmetry in these blood vol- ume maps. Note that owing to the relatively high time to echo susceptibility artifacts are seen in the orbitofrontal and posterior-lateral temporal regions near the petrous bones.

Strupp et al. [4] reported that of 10 patients with TGA and normal T1 and T2 images, 7 had abnormal DWI signals in mesial temporal regions, whereas 3 did not. Ay et al. [1] presented a patient, as did Woolfenden et al., [5] who had abnormalities on both DWI and T2 images, suggesting stroke as the etiology of their transient amnesic episodes. Ay et al. [1] concluded that TGA is a pathophysiologically heterogeneous syndrome, sometimes caused by strokes. Case-control studies of patients with TGA, however, do not find an increased frequency of cerebral ischemic events, such as transient ischemic attacks or strokes. [6] Some would argue that a stroke whose symptoms mimic TGA should not be considered part of this clinically benign syndrome, as the evaluation, treatment, and prognosis are different. [7] We advocate that every patient who presents with an acute amnestic syndrome be imaged with MRI, but in general reserve the term "TGA" for those in whom etiologies such as seizure and stroke have been ruled out. We suggest that in some patients with this kind of "typical" TGA, there is no disruption in cerebral perfusion or energy-dependent processes (assuming that PWI, which rapidly returns to normal when reperfusion occurs, and DWI, which usually becomes abnormal when blood flow is reduced to less than one-third of normal levels, are sensitive enough to detect potentially relevant alterations). Further studies will be necessary to determine the underlying etiology of "typical" TGA.

Andrew E. Budson MD, Gottfried Schlaug MD, Hannah R. Briemberg MD, Boston, MA


1. Ay H, Furie KL, Yamada K, Koroshetz WJ. Diffusion-weighted MRI characterizes the ischemic lesion in transient global amnesia. Neurology 1998; 51: 901-903.

2. Warach S, Gaa J, Siewert B, Wielopolski P, Edelman RR. Acute human stroke studied by whole brain echoplanar diffusion-weighted magnetic resonance imaging. Ann Neurol 1995; 37: 231-241.

3. Edelman RR, Mattle HP, Atkinson DJ, et al. Cerebral blood flow : assessment with dynamic contrast-enhanced T2*-weighted MR imaging at 1.5 T. Radiology 1990; 176: 211-220.

4. Strupp M, Bruning R, Wu RH, Deimling M, Reiser M, Brandt T. Diffusion-weighted MRI in transient global amnesia : elevated signal intensity in the left mesial temporal lobe in 7 of 10 patients. Ann Neurol 1998; 43: 164-170.

5. Woolfenden AR, O'Brien MW, Schwartzberg RE, Norbash AM, Tong DC. Diffusion-weighted MRI in transient global amnesia precipitated by cerebral angiography. Stroke 1997; 28: 2311-2314.

6. 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.

7. Hodges JR. Unraveling the enigma of transient global amnesia. Ann Neurol 1998; 43: 151-153.


Last updated November, 2002.