Professional Documents
Culture Documents
Review
A R T I C L E I N F O A B S T R A C T
Keywords: In the context of status epilepticus (SE), seizure-induced reversible MRI abnormalities (SRMA) can be difficult to
Epilepsy differentiate from epileptogenic pathologies. To identify patterns and characteristics of SRMA, we conducted a
Transient systematic review in accordance with the Preferred Items Reporting for Systematic Reviews and Meta-Analyses
EEG
(PRISMA) guidelines. We included publications describing patients (a) presenting with status epilepticus, (b)
Imaging
Periictal
exhibiting seizure-induced MRI abnormalities, (c) who demonstrated complete resolution of MRI abnormality at
Postictal follow-up, and (d) who had availability of descriptive MRI results.
A total of 49 cases from 19 publications fulfilled our eligibility criteria. Signal abnormalities were most
frequently reported on T2-weighted sequences followed by diffusion-weighted and fluid-attenuated inversion
recovery imaging. Both unilateral and bilateral SRMA were reported. Unilateral EEG abnormalities were often
associated with ipsilateral SRMA. The signal changes appeared during the ictus itself in some subjects whilst the
median time to SRMA appearance and resolution were 24 h and 96.5 days, respectively.
Based on the distribution of reversible signal alterations, we identified five ‘composite patterns’: (1) pre
dominant cortical (with or without subcortical, leptomeningeal or thalamic involvement), (2) hippocampal (with
or without cortical, subcortical, leptomeningeal, or thalamic involvement), (3) claustrum, (4) predominant
subcortical, and (5) splenium involvement. Amongst treatment-responsive SE patients, the cortical pattern was
the most prevalent whereas hippocampal involvement was most frequently reported in refractory SE.
Cortical atrophy, hippocampal sclerosis, and cortical laminar necrosis were common long-term sequelae after
the resolution of SRMA. In this review, we highlight many limitations of the literature and discuss future di
rections for research.
Abbreviations: ASM, antiseizure medications; SRMA, seizure-induced reversible MRI abnormalities; SE, status epilepticus.
* Corresponding author at: School of Clinical Sciences at Monash Health, Department of Medicine, Monash University, Melbourne, Victoria, Australia
E-mail address: udaya.seneviratne@monash.edu (U. Seneviratne).
https://doi.org/10.1016/j.seizure.2021.09.002
Received 14 July 2021; Received in revised form 3 September 2021; Accepted 5 September 2021
Available online 7 September 2021
1059-1311/© 2021 British Epilepsy Association. Published by Elsevier Ltd. All rights reserved.
F.P. Mariajoseph et al. Seizure: European Journal of Epilepsy 92 (2021) 166–173
167
F.P. Mariajoseph et al. Seizure: European Journal of Epilepsy 92 (2021) 166–173
Table 1 (8%) and one patient (2%) was in a comatose state at the 6-month
Summary of patient demographics, epilepsy classification and seizure follow-up. One patient (2%) died during SE, and one patient (2%) was
characteristics lost to follow-up, and for the remaining 15 cases (31%) long-term clin
Clinical and demographic variables Number of cases ical outcomes were not reported.
(%) The EEG findings were reported with some variability. 25 patients
Median Age 29y (range: 2 to 79) (49%) had only interictal EEG data available, whilst 16 patients (33%)
Sex had only ictal EEG data. Four subjects (8%) had both interictal and ictal
Female 21 (43) EEG information, and a further four (8%) had no EEG data reported. The
Male 24 (49)
most common interictal EEG finding was focal epileptiform discharges
Unspecified 4 (8)
Country observed in 12 patients (85.7%), whilst temporal-onset epileptiform
Italy 11 (22) rhythm was the most frequently reported ictal EEG finding (9 patients).
South Korea 8 (16)
USA 6 (12) 3.2. Bias and quality assessment
Germany 6 (12)
France 4 (8)
Taiwan 4 (8) Of the 49 cases included, 38 (78%) exhibited a satisfactory selection
Portugal 3 (6) of SRMA and 39 cases (80%) appropriately excluded alternative causes.
Austria 1 (2) Follow-up was deemed acceptable in 40 patients (82%) and the quality
Hungary 1 (2)
of reporting was considered sufficient in only 24 cases (49%) reflecting
Japan 3 (6)
Switzerland 1 (2) the variation in reporting standard. Further information relating to the
UK 1 (2) bias and quality assessment of this study can be found in Supplementary
Epilepsy Classification Table 2. Of note, reports of 40 cases (82%) provided images of the
Focal epilepsy 29 (59) abnormal MRI demonstrating SRMA, whilst only 28 (57%) provided
Generalised epilepsy 3 (6)
First unprovoked seizure 10 (20)
figures of the follow-up MR imaging showing complete resolution,
Unspecified 7 (14) although this was not an essential reporting criterion in our quality
Seizure Classification assessment.
NCSE 17 (35)
CSE (focal onset) 13 (27)
3.3. MRI abnormalities
CSE (generalised) 1 (2)
CSE (unknown onset) 3 (6)
FMSE 2 (4) 3.3.1. MRI sequence
Focal SE (not further specified) 8 (16) Table 2 summarises SRMA according to the MRI sequences. The
Myoclonic SE 1 (2) sequence on which abnormality was most commonly reported was T2-
Unspecified 4 (8)
ASMs preceding onset of status epilepticus
weighted imaging in 42 patients (86%). Forty-one subjects (84%)
Phenytoin 2 (4) demonstrated T2 hyperintensity whereas hypointensity with ‘dark white
Phenobarbital 1 (2) matter’ was reported in one patient (4%). Although in 16 (33%) cases
Lamotrigine, phenytoin & levetiracetam 1 (2) field strength of MRI was not reported, all other cases specified that MRI
Not reported 31 (63)
was performed at 1.5T.
Treatment to terminate status epilepticus
PHT 4 (8) Time interval between the seizure and the MRI: Eight patients (16%) had
DZP 1 (2) an MRI conducted less than 24 h postictally, ten (20%) had an MRI
CBZ, CPZ & VPA 1 (2) between 1 and 2 days after the index seizure, and the MRI of five patients
MDZ, PPF, PHT, LEV, LCM, barbiturate coma 1 (2) (10%) was conducted between 2 and 3 days after the seizure. Two pa
MDZ, PPF, PHT, LEV, LCM, VPA, CBZ, TPM, barbiturate 2 (4)
coma
tients (4%) had an MRI between 4 and 7 days post-ictus, and 4 patients
MDZ, PPF, PHT, LEV, TPM, CLB 1 (2) (8%) had an MRI performed after seven days or more. Ten patients
MDZ, PPF, PHT, LEV, VPA, TPM, barbiturate coma 1 (2) (20%) had an MRI during the ictus, whilst the initial MRI timing was not
MDZ, PPF, VPA, CBZ, PHB, barbiturate coma 1 (2) specified in 10 patients (20%). The median time interval to initial MRI
MDZ, VPA, LEV, PHT 1 (2)
was 24 h post-ictal (range: during ictus – 264 h after ictus).
MDZ, VPA, OXC 1 (2)
PHT & CBZ 1 (2) Between the index seizure and follow-up brain MRI showing com
PHT & GBP 1 (2) plete resolution, there was a median time interval of 96.5 days (range:
PHT, CPZ & VPA 1 (2) ictal to 1800 days). Additionally, abnormal serial MRI scans were re
PHT, LEV, GBP, TPM, CPZ 1 (2) ported in 12 patients (24%) ranging from 6 days to 70 days after the
PHT, PHB, CBZ, TPM, LEV, PPF 1 (2)
PHT, PHB, CBZ, TPM, LEV, thiopental 1 (2)
index seizure.
PHT, PHB, CBZ, TPM, LZP, barbiturate coma 1 (2) Anatomical distribution of SRMA: A summary of the anatomical dis
PHT, PPF, CBZ, OXC, LZP, thiopental 1 (2) tribution of SRMA is presented in Table 3. Twenty-seven subjects (55%)
Pharmacologically induced coma 1 (2) exhibited unilateral SRMA whilst the remaining 22 (45%) had bilateral
Not reported 26 (53)
SRMA. There was a widespread of lobar location with cases reporting
Treatment Response
Responsive 10 (20) SRMA in every lobe. With respect to anatomical layers, the cortex and
Refractory 8 (16) hippocampus were the most commonly involved sites of SRMA.
Super-refractory 4 (8) A total of nine patients also had pre-existing structural abnormal
Unspecified 27 (55) ities, such as tumours or vascular pathology, with six specifying a lobar
ASM = antiseizure medication; NCSE = non-convulsive status epilepticus; location. In four patients, the lobe of structural abnormality was at least
CSE = convulsive status epilepticus partially involved in the SRMA lesion, and one of these patients, who
FMSE = focal motor status epilepticus; PHT = phenytoin; DZP = diazepam; had a right parietal cavernoma, exhibited SRMA exclusively in the same
CBZ = carbamazepine; CPZ = clonazepam; VPA = valproate; GBP = gabapentin; lobe. Interestingly, all six subjects demonstrated SRMA ipsilateral to the
PHB = phenobarbital; TPM = topiramate; LEV = levetiracetam; original structural pathology.
LZP = lorazepam; PPF = Propofol; OXC = oxcarbazepine; MDZ = midazolam;
SRMA according to clinical and EEG characteristics: Overall, no specific
LCM = lacosamide; CLV = clobazam
patterns on the lateralisation or localisation of SRMA in relation to
subtypes of SE were evident. Nine (82%) patients presenting with focal
168
F.P. Mariajoseph et al. Seizure: European Journal of Epilepsy 92 (2021) 166–173
SE demonstrated unilateral SRMA (7 were ipsilateral to seizure focus; 2 was the most frequent (6) followed by subcortical (2) and claustrum
were unspecified), although there was not enough data to identify trends patterns (2). Cortical atrophy was evident in 6 patients as long-term
in patients with generalised SE (Table 4). sequelae. Hippocampal involvement was the most frequent stereotypic
In total, 11 patients demonstrated specific unilateral interictal EEG pattern seen in refractory SE (in 5) and four subjects went on to develop
findings, and of these, 10 exhibited SRMA ipsilaterally (and one bilateral hippocampal atrophy on follow-up. Bilateral claustrum
exhibited bilateral SRMA), and 8 demonstrated SRMA involving the lobe involvement was seen in three subjects whilst only one had signal
and location of EEG abnormality. In addition, two patients demon changes in the splenium. Interestingly, all four patients in the super
strated generalized periodic epileptiform discharges (GPEDs) on inter refractory SE group demonstrated bilateral claustrum involvement.
ictal EEG, and both these patients exhibited bilateral SRMA. Unilateral Hyperintensities in T2, FLAIR, and DWI sequences were seen across all
ictal EEG findings specifying location were reported in 14 subjects, and three groups.
of these 11 displayed unilateral abnormalities ipsilateral to the ictal EEG SRMA according to aetiology: It is noteworthy that nine patients in the
finding, and 3 patients had bilateral SRMA. pooled data were diagnosed with new-onset refractory status epilepticus
SRMA characteristics according to the duration of status epilepticus (NORSE) and eight of them had SRMA involving the claustrum bilater
are summarised in Supplementary Table 4. There were no obvious pat ally whilst the remaining patient demonstrated signal changes in the
terns observed. splenium. Outside the NORSE group, only one patient diagnosed as
Response to therapy and SRMA: We also analysed SRMA in relation to possible febrile infection-related epilepsy syndrome (FIRES) had bilat
treatment response. Status epilepticus resolved with benzodiazepines or eral claustrum involvement (see Supplementary Table 1). Though this
standard ASM was defined as ‘responsive SE’ whilst patients who failed data has limitations, these results suggest a possible association between
to respond to ASM but responded to anaesthetic agents were classified as NORSE and SRMA in the claustrum. Alternatively, it could be a reflec
‘refractory SE’. ‘Super-refractory SE’ was defined as the failure of an tion of the severity of SE as all four subjects with super-refractory SE
aesthetics to terminate SE. We found 10 responsive, 8 refractory, and 4 demonstrated signal changes in the claustrum bilaterally.
super-refractory SE patients whereas 27 cases could not be classified due Long term sequelae: Cortical atrophy alone was present in six subjects
to insufficient data. Among ‘responsive SE’ patients, the cortical pattern (12%), whilst hippocampal sclerosis was reported in 9 patients (18%).
169
F.P. Mariajoseph et al. Seizure: European Journal of Epilepsy 92 (2021) 166–173
Table 2 Table 3
Summary of MRI findings according to sequence. Summary of anatomical lobar and layer location of SRMA.
MRI features according to sequence Number of cases (%) Lobar location of MRI abnormalities Number of cases (%)
170
F.P. Mariajoseph et al. Seizure: European Journal of Epilepsy 92 (2021) 166–173
Table 4 the vast majority reporting one MRI scan several months following the
Summary of SRMA according to seizure classification. seizure. As a result, we were only able to report on when the normal MRI
MRI abnormalities according to seizure classification Number of cases (%) scan was performed, rather than the exact timeline for normalisation of
SRMA. Consequently, the timeframes that we report are by no means
NCSE
Unilateral 14 (29) conclusive, although we believe they have provided valuable estimates
Frontotemporal & Parietooccipital 1 (2) for clinicians about the duration of SRMA and its eventual reversal.
Hippocampus 2 (4) Nonetheless, we would suggest that other pathologies be considered
Hippocampus & Pulvinar 3 (6) before confirming SRMA, especially if the lesion persists for extended
Occipital & Parahippocampus & Uncus 1 (2)
Parietal 1 (2)
periods without any evidence of regression.
Parietooccipital & Pulvinar 1 (2) When compared with a recent systematic review of SRMA following
Parietooccipital & Hippocampus & Pulvinar 1 (2) single seizures [28], there are several differences between the charac
Temporoparietal 3 (6) teristics of SRMA in single seizures and status epilepticus. Following a
Temporal & Hippocampus & Uncus 1 (2)
single seizure, abnormalities have been reported to resolve within days
Bilateral 3 (6)
Cingulate gyri 1 (2) to weeks of onset, whilst we observed that SRMA takes longer to reverse
Cingulate cortex & Unilateral frontotemporal 1 (2) in SE and is more likely to take weeks to months before complete nor
Temporal & Unilateral Parietal 1 (2) malisation. Status epilepticus patients also had numerous reports of
CSE (focal onset) long-term sequelae, after resolution of SRMA, such as hippocampal
Unilateral (ipsilateral) 5 (10)
Frontoparietal 1 (2)
sclerosis, cortical laminar necrosis, and cortical atrophy. This was not
Hippocampus 2 (4) characteristic in SRMA following single seizures. Additionally, in single
Parietooccipital 1 (2) seizure patients, the subcortical region was reported to be the most
Temporoparietal 1 (2) frequent location of SRMA, whereas, in the context of SE subcortical
Bilateral 7 (14)
lesions were far less common. Nonetheless, there are some similarities as
Frontotemporal & Hippocampus 1 (2)
Hippocampus 4 (8) well. In both single seizures as well as status epilepticus, SRMA was most
Diffuse 1 (2) commonly reported in T2-weighted imaging. Furthermore, the hippo
Splenium, Hippocampus, Claustrum, Insula & Basal 1 (2) campus was not an infrequent location of manifestation in both contexts,
Ganglia although, single seizures largely demonstrated unilateral abnormalities
CSE (generalised onset)
whilst hippocampal involvement in SE patients was predominantly
Central 1 (2)
Splenium 1 (2) bilateral.
CSE (unknown onset) There was very limited information on seizure characteristics and
Unilateral 1 (2) therefore we were restricted in our exploration of their association with
Frontal 1 (2)
SRMA location. In focal SE, we noted that SRMA were unilateral as well
Bilateral 3 (6)
Hemisphere Cortex 1 (2) as bilateral in nature. Furthermore, in the current study, we were unable
Claustrum & External Capsule 1 (2) to examine the laterality of SRMA in generalised SE due to a paucity of
Cuneus & Precuneus 1 (2) data. Conversely, in single seizure patients, focal seizures are more likely
FMSE to present as unilateral SRMA and patients with generalised seizures are
Unilateral (ipsilateral) 2 (4)
more likely to exhibit bilateral signal changes [28].
Frontal 1 (2)
Temporooccipital 1 (2) The main clinical dilemma posed by SRMA is the differentiation from
Focal SE unspecified underlying cerebral pathologies responsible for SE. There are no
Unilateral (laterality unknown) 2 (4) guidelines on how to approach this question in a systematic manner. It is
Parietal 1 (2)
also unclear how often and how soon the imaging should be repeated.
Unspecified 1 (2)
Bilateral 6 (12)
Though we are unable to provide definitive guidelines due to many
Claustrum 4 (8) limitations of the literature, the stereotypic MRI patterns delineated by
Claustrum & Insular Cortex 1 (2) us may help clinicians recognize this phenomenon early. Further
Claustrum & Unilateral Thalamus 1 (2) research is needed to establish clear guidelines and management
Unspecified SE
pathways.
Unilateral 1 (2)
Parietal 1 (2)
Bilateral 2 (4) 4.2. Pathophysiology
Unspecified 2 (4)
Central 1 (2)
It has been postulated that increased blood flow, metabolic demand
Splenium 1 (2)
Myoclonic SE and neurotransmitter release can be attributed to the transient abnor
Bilateral 1 (2) malities seen in patients exhibiting SRMA, especially when abnormal
Claustrum & External Capsule 1 (2) ities coincide with the location of the seizure focus [16,29]. Epileptic
NCSE = Non-convulsive status epilepticus; CSE = Convulsive status epilepticus; seizure activity causes increased neurotransmitter release as well as
FMSE = Focal motor status epilepticus; SE = status epilepticus raised metabolic requirements, and blood flow to the seizure focus is
markedly increased in an attempt to meet these energy demands. This is
picture. Likewise, serial MRI scans are also recommended as they will ultimately insufficient and leads to dependence on anaerobic respira
provide more accurate information on the time course of reversible tion, which results in the accumulation of lactic acid, carbon dioxide and
signal changes. On another note, we did not observe any association free radicals [30], causing a release of apoptotic enzymes and a break
between the duration of the seizure and SRMA characteristics. None down of the neuronal cell membrane[2].
theless, we found that a pattern did emerge between SRMA and EEG As aforementioned, the hippocampus is commonly involved in SRMA
findings. In both the instances of interictal and ictal EEG, patients with which may be explained by an increased number of glutamate receptors
focal EEG findings were likely to exhibit SRMA ipsilaterally. Addition in the hippocampus, rendering it more susceptible to cytotoxic damage
ally, we observed a possible association between signal alterations [31,32] Additionally, it has been suggested that, especially in the
involving the claustrum bilaterally and NORSE. context of prolonged seizures, the hippocampus can be vulnerable to
In this study, serial imaging was available in a minority of cases, with atrophic changes [33]. Ultimately these conclusions were supported by
our findings, as we found 14 patients (28.5%) with hippocampal SRMA,
171
F.P. Mariajoseph et al. Seizure: European Journal of Epilepsy 92 (2021) 166–173
Fig. 2. Composite patterns of MRI signal abnormalities seen in status epilepticus. Cortical pattern- A: focal diffusion restriction involving cortical gyri, B: ipsilateral
pulvinar signal change on fluid-attenuated inversion recovery imaging; 2. Subcortical pattern-C: hyperintensity in the subcortical left insular region on fluid-
attenuated inversion recovery imaging; 3. Splenial pattern-D: diffusion restriction of the splenium, E: corresponding decreased signal intensity on the apparent
diffusion coefficient image; 4. Hippocampal pattern-F: diffusion restriction in hippocampi bilaterally, G: corresponding decreased signal intensity on the apparent
diffusion coefficient image, H: hyperintensity of hippocampi bilaterally on fluid-attenuated inversion recovery imaging, I: contrast enhancement of hippocampi
bilaterally on T1-weighted imaging (post-contrast); J: gliosis and encephalomalacia in the right fontal region due to previous head trauma. This lesion is the aetiology
of status epilepticus. Note bilateral signal changes in hippocampi due to status epilepticus away from the lesion.
as well as 10 patients (20%) who went on to develop hippocampal resolution of SRMA, as the vast majority of cases only reported the time
sclerosis. point at which SRMA disappeared from MRI. An additional limitation
With regards to MRI sequences specifically, diffusion restriction on inherent to reviewing status epilepticus is the changes in definitions that
DWI/ADC signifies cytotoxic oedema [4], whilst T2-weighted imaging have taken place over the years. It is possible that some relevant cases
and FLAIR abnormalities reflect vasogenic oedema.[2] During epileptic were not included in our study because they were not considered SE at
seizures, changes observed in ADC/DWI occur simultaneously to signal the time of publication, however, we attempted to minimise this by
changes in T2 and FLAIR, which suggest that the pathophysiological carefully considering the length of seizure episodes and applying current
mechanism of these abnormalities cannot be explained by ischaemic definitions to them, although some cases may still have been missed due
causes or hypoxia in which these changes usually appear sequentially. to inadequate information provided in the publication.
Additionally, there was considerable heterogeneity in the reporting
4.3. Strengths of data. Although 82% of cases provided images of the SRMA itself, only
57% presented figures of the resolved MRI scan, which may indeed have
There are a number of strengths inherent to this study. This review provided useful information. Not all MRI sequences were reported.
was conducted in strict accordance with the PRISMA guidelines. Eligi Standard MRI sequences were not mentioned in several studies, with
bility criteria for selection into this review were stringently imple almost two-thirds of cases failing to report T1 weighted imaging results
mented. Complete resolution of SRMA was ensured, as well as the and more than half excluding FLAIR sequence. Similarly, long-term
exclusion of provoked seizures, cluster seizures and single seizures, clinical outcomes of patients were infrequently reported, as was there
minimising the impact of potential confounders in our study. Addi a paucity in the reporting of management, aetiology and comorbidities.
tionally, there is a widespread of patient characteristics represented Further, there was variation in SE classifications, as well as very limited
within this study, rendering our findings applicable in different settings. information regarding seizure focus, which ultimately hindered our
Demographically, patients originated from eleven countries with the analysis of their associations with SRMA characteristics.
youngest and oldest subjects being aged 2 years and 79 years
respectively. 4.5. Gaps in knowledge
172
F.P. Mariajoseph et al. Seizure: European Journal of Epilepsy 92 (2021) 166–173
4.6. Future directions [6] Trinka E, Cock H, Hesdorffer D, Rossetti AO, Scheffer IE, Shinnar S, et al.
A definition and classification of status epilepticus–report of the ILAE task force on
classification of status epilepticus. Epilepsia 2015;56(10):1515–23.
Further investigation on SRMA is warranted given its clinical im [7] Fisher RS, Cross JH, French JA, Higurashi N, Hirsch E, Jansen FE, et al. Operational
plications. Underrecognition of this entity may lead to unnecessary in classification of seizure types by the international league against epilepsy: position
vestigations such as brain biopsy. Continued research into the Paper of the ILAE commission for classification and terminology. Epilepsia 2017;58
(4):522–30.
pathophysiology of this phenomenon would be useful to gain a better [8] Murad MH, Sultan S, Haffar S, Bazerbachi F. Methodological quality and synthesis
clinical understanding. Studies with larger sample sizes would also of case series and case reports. BMJ Evid Based Med 2018;23(2):60–3.
provide additional details and insight into the characteristics of SRMA in [9] Bauer G, Gotwald T, Dobesberger J, Embacher N, Bauer R, Seppi K, et al. Transient
and permanent magnetic resonance imaging abnormalities after complex partial
a variety of settings and presentations. Additionally, serial scanning at status epilepticus. Epilepsy Behav 2006;8(3):666–71.
pre-determined time points would facilitate a more accurate estimate of [10] Canas N, Soares P, Calado S, Pestana R, Ribeiro C, Vale J. Pathophysiology and
the time frames for both the initial occurrence and subsequent resolu long-term outcome of reversible tumor-like lesions induced by presenting status
epilepticus. J Neuroimaging 2010;20(2):169–74.
tion of SRMA. Finally, studies comparing cohorts of patients who [11] Chhetri SK, Mathur S, Soh C, Gosal D. Seizure-induced MRI changes mimicking
develop SRMA with cohorts who do not, may provide information on metastatic brain disease. J Clin Neurosci 2014;21(5):869–72.
predisposing factors associated with the appearance of SRMA. [12] Cianfoni A, Caulo M, Cerase A, Della Marca G, Falcone C, Di Lella GM, et al.
Seizure-induced brain lesions: a wide spectrum of variably reversible MRI
abnormalities. Eur J Radiol 2013;82(11):1964–72.
5. Conclusion [13] Huang YC, Weng HH, Tsai YT, Huang YC, Hsiao MC, Wu CY, et al. Periictal
magnetic resonance imaging in status epilepticus. Epilepsy Res 2009;86(1):72–81.
[14] Ishiguro T, Nitta M, Komori T, Maruyama T, Muragaki Y, Kawamata T. Transient
In the published literature on status epilepticus, SRMA manifested focal magnetic resonance imaging abnormalities after status epilepticus showed
heterogeneously with frequent involvement on T2-weighted sequences. 11C-methionine uptake with positron emission tomography in a patient with
Abnormalities were noted to appear as early as during the ictus and as cerebral cavernous malformation. World Neurosurg 2018;114:43–6.
[15] Juhász C, Scheidl E, Szirmai I. Reversible focal MRI abnormalities due to status
late as 11 days post-ictal, and the median time to resolution was 96.5 epilepticus. An EEG, single photon emission computed tomography, transcranial
days. The cortex, hippocampus, and claustrum were common locations Doppler follow-up study. Electroencephalogr Clin Neurophysiol 1998;107(6):
of SRMA, and long-term sequelae, such as hippocampal sclerosis, 402–7.
[16] Kim JA, Chung JI, Yoon PH, Kim DI, Chung TS, Kim EJ, et al. Transient MR signal
cortical atrophy and cortical laminar necrosis were also noted. Interictal
changes in patients with generalized tonicoclonic seizure or status epilepticus:
and ictal EEG findings were also useful in predicting the lateralisation of periictal diffusion-weighted imaging. Am J Neuroradiol 2001;22(6):1149–60.
abnormalities. However, due to considerable limitations of the litera [17] Kramer R, Lüders H, Lesser R, Weinstein M, Dinner D, Morris H, et al. Transient
ture, no definitive conclusions can be drawn to propose a clear man focal abnormalities of neuroimaging studies during focal status epilepticus.
Epilepsia 1987;28(5):528–32.
agement pathway for those patients presenting with SRMA. This study [18] Lansberg MG, O’Brien M, Norbash A, Moseley M, Morrell M, Albers G. MRI
highlights the need for more research into SRMA, specifically relating to abnormalities associated with partial status epilepticus. Neurology 1999;52(5):
serial scanning and prognostic implications, in order to formulate 1021.
[19] Milligan TA, Zamani A, Bromfield E. Frequency and patterns of MRI abnormalities
practical guidelines to rationally manage this clinical dilemma. due to status epilepticus. Seizure 2009;18(2):104–8.
[20] Senn P, Lövblad KO, Zutter D, Bassetti C, Zeller O, Donati F, et al. Changes on
Funding diffusion-weighted MRI with focal motor status epilepticus: case report.
Neuroradiology 2003;45(4):246–9.
[21] Szabo K, Poepel A, Pohlmann-Eden B, Hirsch J, Back T, Sedlaczek O, et al.
This research did not receive any specific grant from funding Diffusion-weighted and perfusion MRI demonstrates parenchymal changes in
agencies in the public, commercial, or not-for-profit sectors. complex partial status epilepticus. Brain 2005;128(6):1369–76.
[22] Chevret L, Husson B, Nguefack S, Nehlig A, Bouilleret V. Prolonged refractory
status epilepticus with early and persistent restricted hippocampal signal MRI
abnormality. J Neurol 2008;255(1):112–6.
Declaration of Competing Interest
[23] Eguchi K, Tsuzaka K, Yabe I, Sasaki H. Reversible splenial lesion in new-onset
refractory status epilepticus: a case report. Clin Neurol Neurosurg 2019;183:
We know of no conflicts of interest associated with this manuscript, 105392.
and there has been no significant financial support for this work that [24] Meletti S, Slonkova J, Mareckova I, Monti G, Specchio N, Hon P, et al. Claustrum
damage and refractory status epilepticus following febrile illness. Neurology 2015;
could have influenced its outcome. 85(14):1224–32.
[25] Mizutani AU, Shindo A, Arikawa S, Shimada T, Matsuura K, Ikejiri K, et al.
Reversible splenial lesion in a patient with new-onset refractory status epilepticus
Supplementary materials
(NORSE). Eneurologicalsci 2020;18:100220.
[26] Mumoli L, Labate A, Palamara G, Sturniolo M, Gambardella A. Reversible
Supplementary material associated with this article can be found, in symmetrical external capsule hyperintensity as an early finding of autoimmune
the online version, at doi:10.1016/j.seizure.2021.09.002. encephalitis. Neurol Sci 2014;35(7):1147–9.
[27] Sperner J, Sander B, Lau S, Krude H, Scheffner D. Severe transitory encephalopathy
with reversible lesions of the claustrum. Pediatr Radiol 1996;26(11):769–71.
References [28] Mariajoseph FP, Muthusamy S, Amukotuwa S, Seneviratne U. Seizure-induced
reversible MRI abnormalities in patients with single seizures: a systematic review.
Epileptic Disord 2021;23(4):552–62.
[1] Sánchez S, Rincon F. Status epilepticus: epidemiology and public health needs.
[29] Sutula TP. Experimental models of temporal lobe epilepsy: new insights from the
J Clin Med 2016;5(8):71.
study of kindling and synaptic reorganization. Epilepsia 1990;31:S45–54.
[2] Briellmann RS, Wellard RM, Jackson GD. Seizure-associated abnormalities in
[30] Dingledine R, Varvel NH, Dudek FE. When and how do seizures kill neurons, and is
epilepsy: evidence from MR imaging. Epilepsia 2005;46(5):760–6.
cell death relevant to epileptogenesis? Adv Exp Med Biol 2014;813:109–22.
[3] Cole AJ. Status epilepticus and periictal imaging. Epilepsia 2004;45:72–7.
[31] Khaspekov L, Stastný F, Viktorov I, Lisý V. Cytotoxic effect of glutamate and its
[4] Kim SE, Lee BI, Shin KJ, Ha SY, Park J, Park KM, et al. Characteristics of seizure-
agonists on mouse hippocampal neurons. J Brain Res 1990;31(5):635–43.
induced signal changes on MRI in patients with first seizures. Seizure 2017;48:
[32] Margerison J, Corsellis J. Epilepsy and the temporal lobes. Brain 1966;89(3):
62–8.
499–530.
[5] Liberati A, Altman DG, Tetzlaff J, Mulrow C, Gøtzsche PC, Ioannidis JP, et al. The
[33] Pohlmann-Eden B, Gass A, Peters C, Wennberg R, Bluemcke I. Evolution of MRI
PRISMA statement for reporting systematic reviews and meta-analyses of studies
changes and development of bilateral hippocampal sclerosis during long lasting
that evaluate health care interventions: explanation and elaboration. J Clin
generalised status epilepticus. J Neurol Neurosurg Psychiatry 2004;75(6):898–900.
Epidemiol 2009;62(10):e1–34.
173