Transient Global Amnesia: Unlocking an Ischemic Etiology

Abstract & Commentary

Synopsis: Treatment of TGA patients with antiplatelet treatment is likely warranted, particularly if there are underlying vascular risk factors.

Source: Sedlaczek O, et al. Detection of Delayed Focal MR Changes in the Lateral Hippocampus in Transient Global Amnesia. Neurology. 2004:62:2165-3170.

Transient global amnesia (TGA) presents with acute and profound memory loss, but is nevertheless a benign condition that generally does not recur. It has been postulated that TGA may be explained by hippocampal venous congestion, occurring during valsalva maneuver, sexual activity, or excessive exercise. Ischemic lesions in the hippocampus have been demonstrated inconsistently on MRI with diffusion-weighted imaging (DWI), with the majority of reports showing negative findings. In the study of Sedlaczek and colleagues, using sequential DWI imaging, a much higher frequency of hippocampal ischemic lesions is demonstrated in individuals with images delayed 24-48 hours after symptom onset.

In 26 of 31 cases, small DWI-positive hippocampal lesions were observed on 48 hour scans. In 23 cases, there was a corresponding reduction in the absolute diffusion coefficient (ADC), confirming an ischemic etiology. Lesions were left sided in 15, with 6 right and 5 bilateral. Scans at 24 hours showed the lesion in 23 patients, but 11 of these could only be seen retrospectively. Hyperacute scans showed a lesion in only 2 patients. None of the scans showed signs consistent with venous stasis or thrombosis.

An embolic etiology could only be identified in one-fourth of patients, while 50% had more than 1 vascular risk factor. Sedlaczek et al identify a "watershed zone" within the hippocampal blood supply where the superior and inferior hippocampal arteries meet. This area corresponds to the CA1 sector of Sommer, which is the most susceptible portion of the hippocampus to O2 deprivation. Sedlaczek et al postulate that high metabolic demand, such as during exercise, might produce hypoperfusion in patients with existing microvascular changes in this region.


There have been many postulated, and unconfirmed, etiologies for TGA including migraine, seizure, and transient ischemic attack (TIA). These data suggest that TIA may be the most compelling of these theories. It has been well documented that DWI imaging may be positive in TIA, even when neurologic symptoms are fleeting. Many questions still remain unanswered however. Why do TGA associated DWI abnormalities appear on delayed but not hyperacute MRI scans? If there is truly an underlying vascular abnormality, why doesn’t TGA recur more frequently in susceptible individuals? As suggested in the editorial accompanying Sedlaczek’s report, treatment of TGA patients with antiplatelet treatment is likely warranted, particularly if there are underlying vascular risk factors. — Alan Z. Segal

Dr. Segal, Assistant Professor, Department of Neurology, Weill-Cornell Medical College, Attending Neurologist, New York Presbyterian Hospital, is Assistant Editor of Neurology Alert.