TRANSIENT GLOBAL AMNESIA (TGA) is a sudden-onset, temporary loss of memory function characterized by anterograde amnesia. Because the episode usually lasts only 6–8 h, it is difficult to perform detailed analysis of cerebral blood flow (CBF) during TGA. We report a TGA patient who had unique hemodynamics after the attack. Written informed consent was obtained.
A 58-year-old right handed man with unremarkable medical history came to the emergency room complaining of difficulty in retaining new information at 2:30 a.m. The amnestic episode abruptly started at around 10:00 p.m. the previous day soon after he took a bath. Physical work-up indicated no abnormalities. On Mini-Mental State Examination (MMSE) he lost 3 points for not recalling three words. At 10:30 a.m. he was examined on technetium-99m ethyl cysteinate dimer brain perfusion single-photon emission computed tomography (99mTc-ECD-SPECT) followed by magnetic resonance imaging/angiography (MRI/MRA) and diffusion-weighted imaging (DWI), which ruled out any cerebrovascular events. The episode of TGA resolved suddenly at 1:00 p.m., that is, his amnestic episode continued for 15 h. He achieved full marks on the MMSE soon after the termination of TGA. Follow-up evaluation was performed 4 weeks later. The results of ECD-SPECT during and after TGA were investigated using an automated blood-flow-analysis program 3DSRT.1 3DSRT indicated a relatively diffuse and unilateral decrease of CBF in the right hemisphere during the acute phase. However, regional CBF of the bilateral hippocampi, in which most previous reports demonstrated significant changes during TGA attacks, was preserved as was that of the thalamus as well. The most notable finding was significantly decreased CBF of the right thalamus at follow up despite no visible regions on MRI/MRA and DWI. The quantified CBF values in thalamus were as follows (right/left): during attack, 44.2/44.1 mL/100 g per min; follow up, 37.4/45.5 mL/100 g per min; lower limits of normal database, 43.9/42.7 mL/100 g per min.
The true pathophysiology of TGA remains unclear. Among several possible causes proposed so far, an ischemic event is the most feasible mechanism at the moment. In the present case DWI during and after the episode indicated no abnormalities. Quinette et al. reviewed previous publications and demonstrated that 42 of 43 patients who were examined on SPECT during the acute phase had transient dysfunction in the medial temporal lobe.2 The analysis of ECD-SPECT in the present case using 3DSRT indicated atypical findings of significantly decreased CBF of the right thalamus after the TGA attack in spite of preserved blood flow during the acute phase. At the present time we cannot clearly explain the mechanisms of these findings. It is possible that we captured a moment of some sort of reversible vascular event, therefore it is necessary to accumulate more cases.