Chronic hippocampal subfield damage in transient global amnesia revealed by 7T MRI: All is not reversible?

Abstract

Transient global amnesia (TGA) is a neurological condition characterized by temporary memory loss and classically associated with a reversible unilateral punctate focus of restricted diffusion in the cornu ammonis 1 (CA1) region of the hippocampus. Historically, the lesions were considered to be transient in nature with no long-term imaging abnormality. However, more recent studies have challenged the concept that there are no long-term neurological sequelae. In line with this evidence, we explore the role of ultra-high-resolution imaging using 7 Tesla MRI to evaluate for long-term imaging abnormalities in a 63-year-old woman with a typical clinical course and acute TGA imaging findings. The 7 Tesla MRI revealed a residual lesion on susceptibility-weighted imaging (SWI) with evidence of gliosis and volume loss at the site of the acute lesion in CA1 eight months after the acute episode. This case challenges the traditional mantra of TGA as a fully reversible condition with no long-term imaging findings, suggesting the need for further research using ultra-high-field MRI to determine TGA's potential long-term imaging sequelae and any association with neurocognitive sequelae.

Introduction

Transient global amnesia (TGA) is a neurological condition that is characterized by a sudden and temporary loss of the ability to form new memories, usually lasting for several hours. ¹ TGA is often associated with a unilateral punctate hyperintense lesion in the cornu ammonis 1 (CA1) region of the hippocampus on diffusion-weighted imaging (DWI) obtained between 24 and 96 h after the onset of symptoms. ² This imaging finding has been widely considered to be fully reversible, with no visible residual effects observed in imaging obtained more than 30 days after the TGA episode. ² In light of its recent approval for clinical use, there is a growing interest in the potential clinical applications of 7 Tesla (T) MRI.^3–10

In this case, we utilized ultra-high-resolution 7T MRI to assess for chronic imaging findings in a patient presenting with typical clinical course and acute TGA imaging findings. To our knowledge, this is the first case of TGA to be assessed by this approach and reveals a residual lesion on susceptibility-weighted imaging (SWI) with evidence of gliosis and volume loss at the site of the acute lesion. Our case challenges the traditional mantra that there are no long-term imaging findings in TGA and shows the need for further studies of patients using ultra-high-field MRI to better determine the incidence of long-term injury to the hippocampus and the clinical implications.

Case presentation

A 63-year-old woman with a history of generalized anxiety disorder presented to a primary care physician for the evaluation of a 1-h episode of memory loss. Three days prior, she experienced a sudden episode of memory loss during sexual intercourse, which resulted in her inability to recall the continuation of the activity or subsequent events. Her partner reported that she repeatedly questioned him but that she did not lose consciousness. At the time of the event, he conducted a home evaluation for the typical symptoms of a stroke and reported no other abnormality. There were no associated facial weaknesses, changes in speech, visual disturbances, lateralized sensory-motor symptoms, or gait abnormalities, and her symptoms resolved within an hour. A neurologic examination conducted at the time of presentation showed no abnormalities. Two days later, a 3T MRI brain was performed, which revealed a hyperintense punctate focus on T2-weighted (T2W) imaging in the CA1 region of the left hippocampus with corresponding diffusion restriction on DWI (Figure 1). This finding and the patient’s symptoms were consistent with the diagnosis of TGA.

Figure 1.

3 Tesla brain MRI performed 2 days after onset of TGA symptoms. (a) Diffusion-weighted image and (b) apparent diffusion coefficient map show a focal area of diffusion restriction (arrow) in the CA1 region of the left hippocampus. (c) Susceptibility-weighted image shows no apparent abnormality in the same region.

At a follow-up visit with a neurologist 3 months later, the patient reported no symptoms suggestive of hemispheric, retinal, or brainstem ischemia, and there were no apparent long-term clinical effects of the TGA episode. To assess for potential chronic imaging abnormalities and to assess for other signs of cerebrovascular disease, 7T MRI brain was performed 8 months after the episode. 7T MRI revealed a hypointense punctate lesion on SWI, suggestive of iron deposition, in place of the prior lesion seen on DWI. This was associated with focal CA1 gliosis and volume loss, seen on coronal T2W imaging, at the level of the prior lesion (Figure 2). The patient displayed no clinical sequelae or signs 8 months post-TGA and was managed conservatively.

Figure 2.

7 Tesla brain MRI performed 8 months after resolution of symptoms. (a) Susceptibility-weighted image (repetition time (TR) = 21 ms, echo time (TE) = 14 ms, flip angle = 15°, receiver bandwidth = 140 Hz/Px, GeneRalized Autocalibrating Partial Parallel Acquisition (GRAPPA) acceleration factor = 3, phase and slice partial Fourier = 6/8, slice thickness = 1.2 mm, in-plane resolution = 0.3 × 0.3 mm, imaging time = 5 min 11 s) and (b) phase map show development of a punctate focus of iron deposition (arrow) in the left CA1 region corresponding to the location of the original acute lesion. (c) High-resolution coronal T2-weighted image at the level of the acute lesion (c) shows very focal volume loss (gliosis) in the CA1 region (arrow) which is essentially normal on the adjacent slice (d; arrow).

Discussion

TGA is a well-defined syndrome characterized by the sudden onset of significant anterograde amnesia, accompanied by variable degrees of retrograde amnesia, lasting no longer than 24 h in the absence of any other neurological abnormalities. ¹ Similar to our patient’s acute imaging results, TGA is often associated with a punctate hyperintense lesion in the CA1 region of the hippocampus on DWI obtained within 24–96 h of the TGA episode. ² It is crucial to note that these acute hippocampal DWI lesions in TGA, like the condition itself, are considered to be transient in nature and are believed to “vanish” on imaging performed more than 30 days after the onset of TGA. ¹¹ This apparent resolution of imaging findings aligns well with the conventionally held view that TGA has no long-term cognitive consequences. ¹² However, more recent studies have challenged this belief.^13,14

The imaging obtained 8 months after our patient’s episode of TGA revealed alterations in the SWI signal and evidence of gliosis and volume loss on coronal T2-weighted images. This conflicts with previous research, including a study by Paech et al., which concluded that acute DWI lesions in TGA have no detectable long-term sequelae, even when using ultra-high-field MRI.^2,15 However, it is essential to consider a few potential reasons for the discrepancy between the long-term findings of our patient and those reported by Paech et al. First, Paech et al. did not perform SWI, which may have led to the detection of focal iron deposition, as seen in our case. ¹⁵ Second, despite both studies utilizing a 7T MRI, Paech et al. acquired images along the longitudinal axis of the hippocampus with a slice thickness of 2 mm, which may not capture more focal subfield injury. In our case, high-resolution coronal images were acquired with a slice thickness of 1.2 mm allowing exquisite depiction of the hippocampal subfields and reducing the incidence of volume averaging that may obscure very focal injuries. ¹⁵ Lastly, Paech et al. also focused on detecting hippocampal sulcal cavities (HSCs) using T2W sequences, following Nakada et al., who found a higher incidence of HSCs in long-term follow-up imaging of TGA patients using a 3T MRI.^15,16 Their study did not assess more focal subfield (CA1) damage. This is important because HSCs are likely unrelated to TGA, while chronic focal subfield damage at the site of the acute imaging findings is almost certainly related.

The findings by Nakada et al. linking TGA to HSCs may have arisen from the limited understanding of HSCs at the time. ¹⁶ Nevertheless, recent studies, including the work of Paech et al., suggest that previous assessments of HSCs, including that of Nakada et al., may have underestimated their prevalence and that HSCs may, in fact, be benign variations in the embryonal folding of the hippocampus, with no correlation to TGA or cognitive decline.^13,15 Furthermore, Nakada et al. lacked documented evidence of the development of these HSCs through prior imaging and relied solely on matched control groups.

Wang et al. previously reported quantitative susceptibility increase throughout the affected CA1 region in TGA. ¹⁷ However, they did not report focal SWI lesions at the site of the injury nor evidence of volume loss and gliosis. ¹⁷ This may be related to their use of 3T MRI. The detection of alterations in SWI signal and evidence of gliosis on imaging performed 8 months after a TGA episode is significant as it raises the possibility of long-term sequelae associated with TGA. Our long-term imaging findings may align with recent studies that highlight the long-term neurological effects of TGA, including mood disorders, memory deficits, and an increased risk of dementia and epilepsy.^13,14

Further investigation is required to determine the potential long-term imaging sequelae of TGA using ultra-high-field MRI and any long-term neurocognitive sequelae, as well as any relationship between these outcomes. To this end, we recommend incorporating dedicated coronal images to assess for changes such as volume loss and gliosis.

Conclusion

We present a case of a patient who underwent ultra-high-field 7T MRI 8 months following her episode of TGA. The MRI showed novel long-term imaging findings of focal SWI lesions and evidence of gliosis and volume loss on T2W at the site of the acute DWI lesion. These findings, together with recent studies on the long-term neurocognitive sequelae of TGA, suggest that TGA may have more lasting effects than previously believed. Further research utilizing ultra-high-field MRI and high-resolution hippocampal imaging to assess hippocampal subfields is necessary to examine TGA’s potential chronic imaging findings and any association with neurocognitive sequelae.

ORCID iDs

Rahul B Singh https://orcid.org/0000-0002-4224-8026

Ahmed K Ahmed https://orcid.org/0000-0003-2886-3046

Erik H Middlebrooks https://orcid.org/0000-0002-4418-9605