revue neurologique 171 (2015) 282–288

Available online at

ScienceDirect www.sciencedirect.com

Hippocampus and epilepsy

Hippocampal modifications in transient global amnesia Modifications hippocampiques dans l’ictus amne´sique idiopathique P. Quinette a,b,c,d, J.M. Constans e, M. Hainselin a,b,c,d,f, B. Desgranges a,b,c,d, F. Eustache a,b,c,d,*, F. Viader a,b,c,d,g a

U1077, Inserm, 5, avenue de la Coˆte-de-Nacre, CS 30001, 14033 Caen Cedex 9, France UMR-S1077, University of Caen - Basse-Normandie, esplanade de la Paix, 14032 Caen Cedex 5, France c UMR-S1077, E´cole Pratique des Hautes E´tudes, 5, avenue de la Coˆte-de-Nacre, 14032 Caen Cedex 5, France d U1077, Caen University Hospital, 5, avenue de la Coˆte-de-Nacre, 14033 Caen Cedex 9, France e Radiology and Medical Imaging Department, Amiens University Hospital, place Victor-Pauchet, 80054 Amiens Cedex 1, France f CRPCPO, EA 7273, University of Picardie Jules Verne, chemin du Thil, 80000 Amiens, France g Neurology Department, Caen University Hospital, avenue de la Coˆte-de-Nacre, 14033 Caen Cedex 9, France b

info article

abstract

Article history:

Transient global amnesia (TGA) is an acute and transient syndrome with a remarkably

Received 22 December 2014

stereotypical set of signs and symptoms. It is characterized by the abrupt onset (no

Accepted 28 January 2015

forewarning) of massive episodic memory impairment, both anterograde and retrograde.

Available online 11 March 2015

Ever since it was first described, TGA has fascinated neurologists and other memory experts, and in recent years, there has been a surge of neuroimaging studies seeking to pin down the

Keywords:

brain dysfunction responsible for it. Several pathophysiological hypotheses have been put

Amnesia

forward, including the short-lived suggestion of an epileptic mechanism. All the available

Hippocampus

data indicate that the brain modifications are reversible, and that the mechanism behind

Diffusion magnetic resonance

TGA is of a functional nature. However, while diffusion-weighted imaging studies have

imaging

clearly identified the hippocampus and, more specifically, the CA1 area, as the locus of brain

Spectroscopy

modifications associated with TGA, researchers have yet to determine whether the origin of the mechanism is vascular or neurochemical. Spectroscopy may provide a means of settling

Mots cle´s : Amne´sie

this issue once and for all. # 2015 Elsevier Masson SAS. All rights reserved.

Hippocampe Imagerie par re´sonance magne´tique de diffusion Spectroscopie

* Corresponding author at: Inserm, EPHE, Universite´ de Caen/Basse-Normandie, Unite´ de recherche U1077, GIP Cyceron, boulevard Henri-Becquerel, BP 5229, 14074 Caen Cedex 05, France. E-mail address: [email protected] (F. Eustache). http://dx.doi.org/10.1016/j.neurol.2015.01.003 0035-3787/# 2015 Elsevier Masson SAS. All rights reserved.

revue neurologique 171 (2015) 282–288

283

r e´ s u m e´ L’ictus amne´sique idiopathique est un syndrome amne´sique aigu et transitoire dont la se´miologie est remarquablement ste´re´otype´e : il se caracte´rise par la survenue soudaine et sans signe avant-coureur d’une atteinte massive de la me´moire e´pisodique (amne´sie ante´rograde et re´trograde). Depuis ses premie`res descriptions, l’IA a suscite´ l’inte´reˆt de tous les neurologues et autres spe´cialistes de la me´moire, mais ces dernie`res anne´es ont vu l’e´mergence de nombreux travaux en neuro-imagerie dont l’objectif principal concerne la localisation et la nature du dysfonctionnement ce´re´bral responsable de l’IA. Plusieurs hypothe`ses physiopathologiques ont e´te´ propose´es, parmi lesquelles un me´canisme e´pileptique tre`s rapidement e´carte´. Si les e´tudes en IRM de diffusion ont clairement e´tabli l’hippocampe, et plus particulie`rement le champ CA1, comme le sie`ge des modifications ce´re´brales associe´es a` l’e´pisode, l’origine vasculaire ou neurochimique du me´canisme reste de´battue. La spectroscopie pourrait eˆtre une me´thode pertinente pour apporter des arguments en faveur de l’une ou l’autre des hypothe`ses. Enfin, l’ensemble des donne´es s’accorde sur la nature re´versible des alte´rations ce´re´brales et confirme le caracte`re fonctionnel du me´canisme en jeu dans l’ictus amne´sique idiopathique. # 2015 Elsevier Masson SAS. Tous droits re´serve´s.

1.

Introduction

It was in two studies published back in 1956 – one French [1], the other American [2] – that transient global amnesia (TGA) was described for the first time. Its low reported incidence in the general population, ranging from five to 11 cases per 100,000 per year, means that it is seldom encountered, although its true incidence is certainly somewhat higher, as the episodes are extremely short-lived and not always detected in A&E departments. The signs and symptoms of this amnesic syndrome are remarkably stereotypical. Of sudden onset, TGA generally occurs in individuals aged around 60 years, who display massive anterograde amnesia and a more variable degree of retrograde amnesia, but no disturbance of identity. The clinical picture includes temporal, and sometimes spatial, disorientation, and iterative questioning is almost always present. Apart from the transient memory impairment, there are no discernible neurological disturbances. Four to 6 hours later on average, the episode slowly starts to ebb, leaving behind lacunar amnesia for the episode and the period immediately preceding it. TGA is generally an isolated syndrome, and repeat episodes are rare. It has been shown to be benign, even if three cases of primary progressive aphasia have been described in individuals who had each had a recurrent attack of TGA [3]. A suggested link with Alzheimer’s disease so far remains unproven [4]. The turning point in the history of this syndrome came in 1990, when Hodges and Warlow [5] established strict diagnostic criteria. We therefore have a clear framework to work within, and where patients’ symptoms do not meet the TGA criteria, physicians need to look for a different diagnosis. Even so, despite the many advances that have been made over more than two decades, TGA still holds many unsolved mysteries. The neuropsychology of TGA has been the subject of extensive investigations, mainly focusing on patients’ memory capacity in the acute phase. As tests have confirmed the strictly episodic nature of the amnesia, TGA is now regarded as a pure model of episodic memory, offering

neuropsychologists a unique opportunity to explore human memory systems. Of particular interest are the handful of studies that have used brain imaging to identify the pathophysiological mechanisms and brain regions implicated in this syndrome.

2.

Pathophysiological hypotheses

Given the episodic nature of the amnesia, the fact that it is an isolated occurrence, and the topography of the anomalies that are regularly observed in magnetic resonance imaging (MRI) investigations, most contemporary authors are agreed that TGA’s underlying dysfunction lies in the hippocampus (see Section 3 ‘‘Neuroimaging’’ below). However, the precise mechanism behind this dysfunction is still largely an enigma, and several hypotheses have been advanced to resolve it.

2.1.

Epileptic mechanism

Within a decade of the first observations being made, researchers had raised the possibility of an epileptic mechanism [6]. However, despite reports of electroencephalography (EEG) anomalies, several studies have established a clear difference between TGA and episodes of epileptic amnesia, which are even more fleeting, and tend to be repeated at close intervals [7,8]. Epileptic amnesia is therefore a differential diagnosis, and not a mechanism or aetiology of TGA [7,9]. One of Hodges and Warlow’s criteria for TGA is precisely the absence of epileptic disorders.

2.2.

Vascular mechanisms

The abrupt onset of TGA has fuelled the hypothesis of a vascular problem, and anomalies picked up in diffusionweighted imaging (DWI) tend to support this. In a series of 28 patients, Winbeck et al. [10] found that these anomalies were correlated with more pronounced carotid atheroma, and suggested that some TGA attacks could have an underlying

284

revue neurologique 171 (2015) 282–288

vascular mechanism. However, most studies have shown that individuals who experience an episode of TGA have neither the risk factors nor the prognosis of patients with cerebrovascular disease [11–14]. For example, a recent study comparing 4299 TGA cases with more than 100,000 cases of transient ischaemic attack (TIA) found that the cumulative 1-year rate of stroke after TGA was just 0.54%, compared with 4.72% after TIA [15]. There is therefore nothing in the data that are currently available to suggest that TGA is a form of ischaemic stroke linked to atherosclerosis or cardiogenic embolism. This has not stopped authors coming up with pathophysiological hypotheses involving less common vascular mechanisms. These are mainly to do with elevated central venous pressure, as episodes frequently have trigger factors reminiscent of the Valsalva manoeuvre (e.g., involving the holding of breath and increased intrathoracic pressure). Some authors have wondered whether TGA might be the result of a paradoxical embolism, when a blood clot forms in a peripheral vein and reaches the brain via a patent foramen ovale (PFO; interatrial communication). An increase in intrathoracic and right-atrial pressure would cause the PFO to open, thus allowing the embolism to enter the systemic circulation. This hypothesis is predicated on a particularly high PFO frequency among individuals who have a TGA episode. However, although some authors have claimed that this is indeed the case [16], there has been no recent confirmation [17]. The other hypothesis involving elevated intrathoracic pressure is that TGA stems from a haemodynamic mechanism [18,19]. By increasing intrathoracic pressure, the Valsalva manoeuvre effectively blocks the flow of blood from the superior vena cava. This causes hyperpressure to build up, affecting the jugular veins. TGA episodes are frequently preceded by intensely emotional or stressful situations [20], and just like bouts of intense physical exertion, these could conceivably give rise to venous congestion, by causing hyperventilation and arterial vasodilation [21]. Several studies have also reported increased prevalence of jugular venous insufficiency among TGA patients, compared with controls, owing to valvular anomalies, especially when these patients have several other risk factors (see meta-analysis [22]). This jugular insufficiency could result in an abnormal backflow from the veins to the intracranial sinuses, possibly triggering venous ischaemia affecting the hippocampus in particular. However, although a recent study [23] confirmed a greater prevalence of jugular insufficiency among TGA patients than in controls, it failed to find any modification in intracranial circulation, even during the Valsalva manoeuvre, and thus questioned the relevance of this anomaly.

2.3.

Neurochemical mechanism

Whether or not TGA is triggered by an acute vascular event, and regardless of whether the mechanism is arterial or venous in all patients, we know that a temporary neural dysfunction within the hippocampus constitutes the final common pathway to the amnesic syndrome. The latter’s utterly stereotypical nature, varying only in duration, means we can safely say that even if several different mechanisms are in play, the outcome is the same in all individuals. We have

already seen the convincing arguments against an epilepsyrelated dysfunction. The most popular hypothesis at present is that of cortical spreading depression (CSD), which was described in animals back in 1944 [24,25]. It has been suggested that CSD is responsible for migraine auras, but the arguments – though solid – are all indirect, as nobody has ever found proof of its existence in humans. It is because a history of migraine or headache is frequently associated with TGA [26–28] that CSD, triggered by excessive glutamate release, is seen as the guilty party. According to Olesen and Jørgensen [29], it brings about a transient dysfunction of hippocampal structures. If CSD is indeed the neural dysfunction behind TGA, there must be a particular mode of initiation and propagation within the hippocampus, as the mean duration of a TGA episode is far longer than that of a migraine aura. Classic morphological imaging cannot inform this debate, but magnetic resonance spectroscopy (MRS) during the acute phase of TGA can (see below).

3.

Neuroimaging

As far as we know, TGA does not leave any permanent brain lesions in it wake, and standard imaging investigations are always normal. For this reason, authors are increasingly turning to more sophisticated and sensitive techniques to look for evidence of focal brain dysfunction. As yet, there have been few published studies, as it can be materially difficult to implement some of techniques (especially positron emission tomography, PET) within the very limited timeframe imposed by TGA.

3.1.

Positron emission tomography (PET)

As we have just indicated, the main explanation for the very small number of PET studies is the abrupt onset and transient nature of TGA, which calls for emergency imaging logistics and the instant availability of several different teams (researchers, A&E physicians, nuclear medicine specialists). Despite these hurdles, there have already been six such studies, in which a total of eight patients were investigated across the time course of the episode, with three others undergoing PET examinations shortly afterwards. The measures were systematically carried out in resting-state conditions. One of the studies [30] briefly reported the case of a female patient who underwent a PET scan in the midst of the acute phase and a DWI scan 48 hours later. These revealed hypometabolism in the left hippocampus, associated with elevated signal intensity on DWI and reduced apparent diffusion coefficient (ADC) of water molecules in this region. There is a dearth of cognitive data in studies reporting PET examinations carried out in the acute or subacute phase. Nevertheless, in their case study of a 60-year-old female patient, Baron et al. [31] drew a link between reduced cerebral blood flow in the right lateral cortex and difficulty retrieving information from episodic memory. In another case study of a female patient, this time aged 59 years, Eustache et al. [32] suggested that there might be a connection between hypometabolism of the left prefrontal regions and impaired episodic memory performances (verbal and visual) and semantic retrieval. When

revue neurologique 171 (2015) 282–288

Guillery et al. [33] examined a 68-year-old female patient (MLP) during the acute phase, and a 63-year-old female patient (NR) during the recovery phase, they observed reduced blood flow in the amygdala and hippocampus in both women. Hippocampal anomalies could explain the deficit in verbal information storage exhibited by MLP and the problems with episodic associations displayed by NR.

3.2.

MRI

In clinical practice, computer-assisted tomography (CT), and above all MRI, are sufficient to rule out other aetiologies of acute amnesia, not least stroke, epileptogenic lesions, limbic encephalitis and prolonged epileptic seizures [34]. Out of the 130 TGA patients studied by d’Agosti’s team [35], 13 (10%) had brain lesions that were visible on MRI. These patients did not differ in any way from the others in the series, and as the lesions did not concern internal temporal structures, they can be regarded as entirely unrelated, apart from the exceptional cases of TGA that were symptomatic of stroke [36]. As for the emergency brain CT scans that are often carried out when a patient presents with transient amnesia, experience has taught us that they are always normal in the case of a confirmed TGA episode, which is why researchers therefore now use MRI. This provides better anatomical and contrast resolution which, when coupled with numerous sequences, allows very small lesions to be detected. So far, four of the MRI techniques that can be used in TGA [anatomical MRI, diffusion MRI, functional MRI (fMRI), and proton MRS] have given rise to publications. For many years, anatomical imaging studies failed to reveal any specific damage, suggesting that TGA is a benign amnesic syndrome with no brain sequelae [37]. However, the persistence of minor memory problems several months after the episode [4,38,39], plus brain anomalies picked up in PET [40] led authors to wonder about possible structural modifications in the hippocampus after a TGA episode. In two group studies [41,42], visual inspection of anatomical MRI scans (3 tesla) of a total of 29 patients revealed particularly large hippocampal cavities (> 3 mm) in 23 of them. These anomalies were unrelated to their cognitive functioning. Although these results need to be confirmed, they suggest that despite its benign nature, TGA can result in hippocampal atrophy. It is important to note that while some of these results were observed in patients just 7 days after the episode, for others the interval was greater than 5 years. It was in 1998 that Strupp et al. [43] first reported elevated signal intensity in the hippocampus in diffusion MRI during a TGA episode. These elevated signals have since been described in numerous studies of patients with TGA. In optimum technical conditions (3T MRI and slice thickness  3 mm), the detection rate is around 80% [44,45], and has been as high as 88% in the most recent studies [46]. The hyperintense signals are dot-shaped, with a mean diameter of 3 mm, and in four out of five cases, they are unilateral – left lateralized more often than not [47–51]. They are located in the lateral part of the hippocampus, in the CA1 area [37,41,52,53]. Signal elevation is sometimes associated with changes in the ADC coefficient, although this information is not always provided. In contrast to ischaemic stroke, where they are present from the very outset,

285

hyperintense signals in diffusion MRI appear relatively late (24– 72 hours post-onset) in TGA, and the best time for seeing them is 48 hours post-onset [54]. They remain visible for up to around 10 days after the episode, but then disappear completely [9,55]. In these studies, they do not seem to be at all ominous, as the recovery path is the same, regardless of whether or not they are found, although they probably often go undetected, owing to insufficient technical means. The use of MRI in TGA studies has also allowed researchers to improve their knowledge of hippocampal involvement in a variety of cognitive functions. In particular, several authors have established a link between anomalies in the CA1 area of the hippocampus and deficits in spatial orientation, autobiographical memory and autonoetic consciousness [56,57]. So far, fMRI has been used in just two case studies. In the first one, featuring a visual scene encoding task, a male patient was examined both in the recovery phase [58] and 7 months later. Compared with the scan at 7 months and results for three control participants, the patient exhibited reduced brain activity in the temporolimbic regions, including the parahippocampal gyrus, during encoding in the recovery phase. This reduced activity was associated with the steep decline in memory performances. The authors also noted heightened activity in the frontoparietal regions, which they interpreted as reflecting the implementation of visuospatial or working memory strategies. The female patient in the second case study exhibited no temporal activation whatsoever during the encoding of new scenes or the recognition of old ones at the time of the TGA episode [59]. At 3-month follow-up, however, her profile was similar to that of five controls, with activation of the parahippocampal gyrus. An fMRI study recently reported modifications in functional connectivity at rest in 12 patients in the acute and postacute (< 14 hours after onset) phases of TGA [60]. The significant reduction in functional connectivity within the episodic memory network was brought to light by comparing the patients’ data with those of 17 control participants. It appeared to be linked to the timing of the MRI examination, being more pronounced in the acute phase than in the postacute one. There was no observable modification 2– 9 months after the episode, attesting to the reversible nature of this functional disturbance. Proton MRS allows us to estimate concentrations of endogenous substances that contain natural paramagnetic nuclei, notably protium (1H). This technique, which makes it possible to study TGA patients’ cerebral metabolism in vivo, has the advantage of being noninvasive, and allows repetitive assessments to be conducted in longitudinal studies. Compounds that can be detected using this technique may help us to identify any brain modifications that occur during an episode of TGA. Although changes in neurometabolite levels are nonspecific, typical modifications have been identified in some neurological diseases. Decreases in the N-acetylaspartate (NAA) signal, a marker of neuronal/axonal injury or dysfunction, associated with an elevated lactate (Lac) signal reflecting anaerobic glycolysis, have been found in both TIA [61] and deep vein thrombosis [62]. If this profile were found in TGA, it would point to a hypoxic-ischaemic origin. Although an increase in the glutamine + glutamate (Glx) peak would suggest spreading depression [63], there has yet to be a study of CSD in

286

revue neurologique 171 (2015) 282–288

humans using 1H-MRS. One study did apply this technique in the acute stage of TGA to investigate metabolic changes in the hippocampal region [28]. The examination was performed 6 hours after onset, while the patient was still amnesic. No changes were found in NAA or Lac, either during the attack or 2 weeks afterwards. A comparison of peak ratio values recorded at these two times showed no significant difference. The authors suggested that their failure to find an ischaemic pattern of brain metabolites in the hippocampal areas might be due to three reasons. First, the ischaemic areas might lie outside the hippocampi and not be detectable. Second, the ischaemia might be so transient that the spectral and metabolic changes were too small to be detected with this technique. Third, the underlying neural dysfunction might be other than ischaemic. The authors concluded that, given a history of migraine and the occurrence of stressful events, CSD and/or an osmotic process was probably the cause, rather than ischaemia. A second study assessed metabolic changes in hippocampal CA1 lesions using a combination of 3-T DWI and MRS in seven patients with TGA [48]. Four of these had a diffusion lesion with a corresponding T2 lesion in the CA1 area. Selective hippocampal MRS showed Lac peaks in three of the four patients with diffusion lesions, but none in patients without a diffusion lesion. The Lac peaks found during the TGA episode were no longer discernible at follow-up (2 to 5 months later) and were associated with a normal NAA level. The authors interpreted the focal hippocampal CA1 diffusion lesions matching the Lac peaks as indicating acute metabolic stress of CA1 neurons. They also confirmed that the lesions are reversible and do not lead to long-term structural sequelae. However, no information about Glx levels or possible ADC changes was supplied. Finally, in these two studies, the long echotimes (136 and 272 ms, and 144 ms, respectively) used for data acquisition, did not allow either myo-inositol or Glx to be measured, and the nature of the pathophysiological mechanism behind TGA could not be determined. In our laboratory (paper in preparation), we used 1H-MRS to examine metabolic changes in the hippocampal region in five patients during the acute phase of TGA and 4 months later. We chose to use short echotimes, in order to assess metabolites such as myo-inositol and Glx. In both hippocampi, measurements of spectra peaks for NAA, creatine-phosphocreatine and Lac failed to reveal changes consistent with cerebral ischaemia, either in the acute phase or at follow-up. We did however observe a trend towards decreased levels of myo-inositol and Glx in both examinations, suggesting a possible release of the neurotoxic glutamate neurotransmitter and a possibly reversible osmotic process.

allow us to determine whether TGA is of vascular or of neurochemical origin, providing we can use short echotimes as well as long ones, in order to measure metabolites such as Glx or myo-inositol. Initial studies in this area have revealed signs of neural stress (osmotic pressure, CSD), but not of major ischaemia. They have failed to find any significant variation in NAA within the hippocampus, once again confirming the absence of structural modifications, although it would be sensible to confirm these results. In sum, given the stereotypical clinical picture of massive anterograde amnesia and variable retrograde amnesia, hippocampal dysfunction is probably a factor in all TGA episodes, but is difficult to pin down, owing to the transient nature of the episode and the doubtless discreet nature of this dysfunction. Other brain regions, such as the neocortical and above all frontal neocortical regions, could well be implicated. The involvement of a more extended network would certainly explain the paradox inherent to TGA, namely, massive episodic memory disturbance but only minor (and sometimes unilateral) anomalies picked up on DWI. This hypothesis is based on preliminary resting-state fMRI results, and looking for correlations between the topography of the affected brain regions (isolated hippocampal or hippocampal-frontal disturbance) and the nature of the episodic memory disorders (anterograde amnesia, retrograde amnesia) should definitely be the priority for future research in this area. Lastly, while we have clearly established that TGA stems from hippocampal dysfunction, researchers are also beginning to report minor morphological modifications in the hippocampus some time after the episode. While these results still need to be confirmed, we cannot afford to ignore them. These structural modifications may be either a direct or an indirect consequence of TGA. For instance, a vascular pathophysiological mechanism could directly lead to brain lesions in areas particularly vulnerable to anoxia, such as the hippocampus. There may also be a more indirect relation, as TGA often occurs in the wake of intense and intensely stressful emotional situations, and can also generate a period of anxiety in the episode’s immediate aftermath [38,39,64], causing excessive and/or prolonged release of neurotoxic glumatate [65], as well as other metabolic disturbances. However, we cannot entirely rule out the possibility that lesions observed in patients were already present before the episode took place, and may even have rendered them more vulnerable to an episode of TGA.

Disclosure of interest 4.

Discussion

The surge in neuroimaging studies of TGA over the past few years has resulted in major advances in our understanding of this syndrome. The main contribution of MRI has been the highlighting of hyperintense signals in the CA1 area of the hippocampus using DWI. These signals appear quite late (24– 72 hours after onset) and subsequently disappear, suggesting that the mechanism behind TGA is indeed functional. Unlike MRI, which cannot tell us exactly which pathophysiological mechanism is at work during the acute phase, MRS could

The authors declare that they have no conflicts of interest concerning this article.

references

[1] Guyotat M, Courjon J. Les ictus amne´siques. J Med Lyon 1956;37:697–701. [2] Bender M. Syndrome of isolated episode of confusion with amnesia. J. Hillside Hosp 1956;5:212–5.

revue neurologique 171 (2015) 282–288

[3] Graff-Radford J, Josephs KA. Primary progressive aphasia and transient global amnesia. Arch Neurol 2012;69(3):401–4. http://dx.doi.org/10.1001/archneurol.2011.1129. [4] Borroni B, Agosti C, Brambilla C, et al. Is transient global amnesia a risk factor for amnestic mild cognitive impairment? J Neurol 2004;251(9):1125–7. http://dx.doi.org/ 10.1007/s00415-004-0497-x. [5] Hodges JR, Warlow CP. Syndromes of transient amnesia: towards a classification. A study of 153 cases. J Neurol Neurosurg Psychiatry 1990;53(10):834–43. [6] Fisher C, Adams R. Transient global amnesia. Acta Neurol Scand 1964;40:46–70. http://dx.doi.org/10.1111/j.16000404.1964.tb07435.x. [7] Bartsch T, Butler C. Transient amnesic syndromes. Nat Rev Neurol 2013;9(2):86–97. http://dx.doi.org/10.1038/nrneurol. 2012.264. [8] Schott B. L’ictus amne´sique. Rev Prat 1969;19:1205–16. [9] Bartsch T, Deuschl G. Transient global amnesia: functional anatomy and clinical implications. Lancet Neurol 2010;9(2):205–14. http://dx.doi.org/10.1016/S1474-4422(09) 70344-8. [10] Winbeck K, Etgen T, von Einsiedel HG, Ro¨ttinger M, Sander D. DWI in transient global amnesia and TIA: proposal for an ischaemic origin of TGA. J Neurol Neurosurg Psychiatry 2005;76(3):438–41. http://dx.doi.org/10.1136/jnnp.2004. 042432. [11] Lauria G, Gentile M, Fassetta G, Casetta I, Caneve G. Transient global amnesia and transient ischemic attack: a community-based case control study. Acta Neurol Scand 1998;97(6):381–5. [12] Melo TP, Ferro JM, Ferro H. Transient global amnesia. A case control study. Brain 1992;115(Pt 1):261–70 [Available at: http://www.ncbi.nlm.nih.gov/pubmed/1559158. Accessed October 23, 2014]. [13] Pantoni L, Bertini E, Lamassa M, Pracucci G, Inzitari D. Clinical features, risk factors, and prognosis in transient global amnesia: a follow-up study. Eur J Neurol 2005;12(5): 350–6. http://dx.doi.org/10.1111/j.1468-1331.2004.00982.x. [14] Uttner I, Weber S, Freund W, Schmitz B, Ramspott M, Huber R. Transient global amnesia – full recovery without persistent cognitive impairment. Eur Neurol 2007;58(3):146– 51. http://dx.doi.org/10.1159/000104715. [15] Mangla A, Navi BB, Layton K, Kamel H. Transient global amnesia and the risk of ischemic stroke. Stroke 2014;45(2):389–93. http://dx.doi.org/10.1161/STROKEAHA. 113.003916. [16] Klo¨tzsch C, Sliwka U, Berlit P, Noth J. An increased frequency of patent foramen ovale in patients with transient global amnesia. Analysis of 53 consecutive patients. Arch Neurol 1996;53(6):504–8. [17] Maalikjy Akkawi N, Agosti C, Anzola GP, et al. Transient global amnesia: a clinical and sonographic study. Eur Neurol 2003;49(2):67–71 [doi:68501]. [18] Lewis SL. Aetiology of transient global amnesia. Lancet 1998;352(9125):397–9. http://dx.doi.org/10.1016/S01406736(98)01442-1. [19] Solheim O, Skeidsvoll T. Transient global amnesia may be caused by cerebral vein thrombosis. Med Hypotheses 2005;65(6):1142–9. http://dx.doi.org/10.1016/ j.mehy.2005.05.042. [20] Quinette P, Guillery-Girard B, Dayan J, et al. What does transient global amnesia really mean? Review of the literature and thorough study of 142 cases. Brain 2006;129(Pt 7):1640–58. http://dx.doi.org/10.1093/brain/ awl105. [21] Sander K, Sander D. New insights into transient global amnesia: recent imaging and clinical findings. Lancet Neurol 2005;4(7):437–44. http://dx.doi.org/10.1016/S14744422(05)70121-6.

287

[22] Modabbernia A, Taslimi S, Ashrafi M, Modabbernia MJ, Hu HH. Internal jugular vein reflux in patients with transient global amnesia: a meta-analysis of case control studies. Acta Neurol Belg 2012;112(3):237–44. http://dx.doi.org/ 10.1007/s13760-012-0072-7. [23] Baracchini C, Tonello S, Farina F, et al. Jugular veins in transient global amnesia: innocent bystanders. Stroke 2012;43(9):2289–92. http://dx.doi.org/10.1161/STROKEAHA. 112.654087. [24] Leao AAP. Spreading depression of activity in the cerebral cortex. J Neurophysiol 1944;7(6):359–90. [25] Latour LL, Hasegawa Y, Formato JE, Fisher M, Sotak CH. Spreading waves of decreased diffusion coefficient after cortical stimulation in the rat brain. Magn Reson Med 1994;32(2):189–98. [26] Frederiks JA. Transient global amnesia. Clin Neurol Neurosurg 1993;95(4):265–83. [27] Schmidtke K, Ehmsen L. Transient global amnesia and migraine. A case control study. Eur Neurol 1998;40(1):9–14. [28] Zorzon M, Longo R, Mase G, Biasutti E, Vitrani B, Cazzato G. Proton magnetic resonance spectroscopy during transient global amnesia. J Neurol Sci 1998;156(1):78–82. [29] Olesen J, Jørgensen MB. Leao’s spreading depression in the hippocampus explains transient global amnesia. A hypothesis. Acta Neurol Scand 1986;73(2):219–20. [30] Gonzalez-Martinez V, Comte F, de Verbizier D, Carlander B. Transient global amnesia: concordant hippocampal abnormalities on positron emission tomography and magnetic resonance imaging. Arch Neurol 2010;67(4):510. http://dx.doi.org/10.1001/archneurol.2010.50. [31] Baron JC, Petit-Taboue´ MC, Le Doze F, Desgranges B, Ravenel N, Marchal G. Right frontal cortex hypometabolism in transient global amnesia. A PET study. Brain 1994;117(Pt 3):545–52. [32] Eustache F, Desgranges B, Petit-Taboue M, et al. Transient global amnesia: implicit/explicit memory dissociation and PET assessment of brain perfusion and oxygen metabolism in the acute stage. J Neurol Neurosurg Psychiatry 1997;63(3):357–67. [33] Guillery B, Desgranges B, de la Sayette V, Landeau B, Eustache F, Baron J-C. Transient global amnesia: concomitant episodic memory and positron emission tomography assessment in two additional patients. Neurosci Lett 2002;325(1):62–6. [34] Fo¨rster A, Griebe M, Gass A, Kern R, Hennerici MG, Szabo K. Diffusion-weighted imaging for the differential diagnosis of disorders affecting the hippocampus. Cerebrovasc Dis 2012;33(2):104–15. http://dx.doi.org/10.1159/000332036. [35] Agosti C, Borroni B, Akkawi NM, De Maria G, Padovani A. Transient global amnesia and brain lesions: new hints into clinical criteria. Eur J Neurol 2008;15(9):981–4. http:// dx.doi.org/10.1111/j.1468-1331.2008.02250.x. [36] Gallardo-Tur A, Romero-Godoy J, de la Cruz Cosme C, Arboix A. Transient global amnesia associated with an acute infarction at the cingulate gyrus. Case Rep Neurol Med 2014;2014:418180. http://dx.doi.org/10.1155/2014/ 418180. [37] Bartsch T, Alfke K, Stingele R, et al. Selective affection of hippocampal CA-1 neurons in patients with transient global amnesia without long-term sequelae. Brain 2006;129(Pt 11):2874–84. http://dx.doi.org/10.1093/brain/ awl248. [38] Kessler J, Markowitsch HJ, Rudolf J, Heiss WD. Continuing cognitive impairment after isolated transient global amnesia. Int J Neurosci 2001;106(3–4):159–68. [39] Noe¨l A, Quinette P, Dayan J, et al. Influence of patients’ emotional state on the recovery processes after a transient global amnesia. Cortex 2011;47(8):981–91. http://dx.doi.org/ 10.1016/j.cortex.2010.10.003.

288

revue neurologique 171 (2015) 282–288

[40] Jia J, Wang L, Yin L, Tang H. Contrast study on cognitive function with MRI and positron emission tomography imaging in transient global amnesia. Chin Med J (Engl) 2002;115(9):1321–3. [41] Nakada T, Kwee IL, Fujii Y, Knight RT. High-field, T2 reversed MRI of the hippocampus in transient global amnesia. Neurology 2005;64(7):1170–4. http://dx.doi.org/ 10.1212/01.WNL.0000156158.48587.EA. [42] Uttner I, Weber S, Freund W, et al. Hippocampal cavities are not associated with cognitive impairment in transient global amnesia. Eur J Neurol 2011;18(6):882–7. http:// dx.doi.org/10.1111/j.1468-1331.2010.03310.x. [43] Strupp M, Bru¨ning 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(2):164–70. http:// dx.doi.org/10.1002/ana.410430206. [44] Choi BS, Kim JH, Jung C, Kim SY. High-resolution diffusionweighted imaging increases lesion detectability in patients with transient global amnesia. AJNR Am J Neuroradiol 2012;33(9):1771–4. http://dx.doi.org/10.3174/ajnr.A3072. [45] Scheel M, Malkowsky C, Klingebiel R, Schreiber SJ, Bohner G. Magnetic resonance imaging in transient global amnesia: lessons learned from 198 cases. Clin Neuroradiol 2012;22(4):335–40. http://dx.doi.org/10.1007/s00062-0120140-7. [46] Kim J, Kwon Y, Yang Y, et al. Clinical experience of modified diffusion-weighted imaging protocol for lesion detection in transient global amnesia: an 8-year large-scale clinical study. J Neuroimaging 2014;24(4):331–7. http:// dx.doi.org/10.1111/jon.12021. [47] Auyeung M, Tsoi TH, Cheung CM, et al. Association of diffusion-weighted imaging abnormalities and recurrence in transient global amnesia. J Clin Neurosci 2011;18(4):531– 4. http://dx.doi.org/10.1016/j.jocn.2010.08.019. [48] Bartsch T, Alfke K, Wolff S, Rohr A, Jansen O, Deuschl G. Focal MR spectroscopy of hippocampal CA-1 lesions in transient global amnesia. Neurology 2008;70(13):1030–5. http://dx.doi.org/10.1212/01.wnl.0000306633.06027.33. [49] Bartsch T, Alfke K, Deuschl G, Jansen O. Evolution of hippocampal CA-1 diffusion lesions in transient global amnesia. Ann Neurol 2007;62(5):475–80. http://dx.doi.org/ 10.1002/ana.21189. [50] Weon YC, Kim JH, Lee JS, Kim SY. Optimal diffusionweighted imaging protocol for lesion detection in transient global amnesia. AJNR Am J Neuroradiol 2008;29(7):1324–8. http://dx.doi.org/10.3174/ajnr.A1105. [51] Yang Y, Kim S, Kim JH. Ischemic evidence of transient global amnesia: location of the lesion in the hippocampus. J Clin Neurol 2008;4(2):59–66. http://dx.doi.org/10.3988/ jcn.2008.4.2.59. [52] Lee HY, Kim JH, Weon Y-C, et al. Diffusion-weighted imaging in transient global amnesia exposes the CA1

[53]

[54]

[55]

[56]

[57]

[58]

[59]

[60]

[61]

[62]

[63] [64]

[65]

region of the hippocampus. Neuroradiology 2007;49(6):481– 7. http://dx.doi.org/10.1007/s00234-007-0213-5. Sedlaczek O, Hirsch JG, Grips E, et al. Detection of delayed focal MR changes in the lateral hippocampus in transient global amnesia. Neurology 2004;62(12):2165–70. Enzinger C, Thimary F, Kapeller P, et al. Transient global amnesia: diffusion-weighted imaging lesions and cerebrovascular disease. Stroke 2008;39(8):2219–25. http:// dx.doi.org/10.1161/STROKEAHA.107.508655. Ahn S, Kim W, Lee Y-S, et al. Transient global amnesia: seven years of experience with diffusion-weighted imaging in an emergency department. Eur Neurol 2011;65(3):123–8. http://dx.doi.org/10.1159/000324154. Bartsch T, Scho¨nfeld R, Mu¨ller FJ, et al. Focal lesions of human hippocampal CA1 neurons in transient global amnesia impair place memory. Science 2010;328(5984): 1412–5. http://dx.doi.org/10.1126/science.1188160. Bartsch T, Do¨hring J, Rohr A, Jansen O, Deuschl G. CA1 neurons in the human hippocampus are critical for autobiographical memory, mental time travel, and autonoetic consciousness. Proc Natl Acad Sci U S A 2011;108(42):17562–7. http://dx.doi.org/10.1073/pnas. 1110266108. LaBar KS, Gitelman DR, Parrish TB, Mesulam MM. Functional changes in temporal lobe activity during transient global amnesia. Neurology 2002;58:638–41. Westmacott R, Silver FL, McAndrews MP. Understanding medial temporal activation in memory tasks: evidence from fMRI of encoding and recognition in a case of transient global amnesia. Hippocampus 2008;18(3):317–25. http://dx.doi.org/10.1002/hipo.20397. Peer M, Nitzan M, Goldberg I, et al. Reversible functional connectivity disturbances during transient global amnesia. Ann Neurol 2014;75(5):634–43. http://dx.doi.org/10.1002/ ana.24137. Giroud M, Walker P, Guy F, et al. Cerebral metabolism after transient ischemic attack. A 1H MR spectroscopy study. Neurol Res 1999;21(6):563–5. Hsu LC, Lirng JF, Fuh JL, Wang SJ, Shyu HY, Liu HC. Proton magnetic resonance spectroscopy in deep cerebral venous thrombosis. Clin Neurol Neurosurg 1998;100(1): 27–30. Gorji A. Spreading depression: a review of the clinical relevance. Brain Res Brain Res Rev 2001;38(1–2):33–60. Do¨hring J, Schmuck A, Bartsch T. Stress-related factors in the emergence of transient global amnesia with hippocampal lesions. Front Behav Neurosci 2014;8:287. http://dx.doi.org/10.3389/fnbeh.2014.00287. Quintana P, Alberi S, Hakkoum D, Muller D. Glutamate receptor changes associated with transient anoxia/ hypoglycaemia in hippocampal slice cultures. Eur J Neurosci 2006;23(4):975–83. http://dx.doi.org/10.1111/j.14609568.2006.04617.x.