If you don't remember your password, you can reset it by entering your email address and clicking the Reset Password button. You will then receive an email that contains a secure link for resetting your password
If the address matches a valid account an email will be sent to __email__ with instructions for resetting your password
Divison of Child Neurology, Epilepsy, and Neurophysiology, Children’s National Medical Center, George Washington University School of Medicine and Health Sciences, Washington, DC, United States
MRI adds diagnostic yield for inflammation and remote injury.
•
MRI changes may include: increased T2/FLAIR signal and/or diffusion restriction.
•
There is increased cerebral perfusion during SE and decreased post-ictal.
Abstract
Neuroimaging, including computed tomography (CT) and magnetic resonance imaging (MRI) are frequently performed in patients who present with status epilepticus (SE). Here we discuss the role of these neuroimaging modalities in clinical evaluation and seizure localization. Additionally, translational neuroimaging research and advanced imaging technologies, such as perfusion and radionucleotide imaging may also contribute to localization of the seizure onset zone and yield opportunities to better understand the pathophysiology of SE and aid in prognostication.
Practice parameter: diagnostic assessment of the child with status epilepticus (an evidence-based review): report of the Quality Standards Subcommittee of the American Academy of Neurology and the Practice Committee of the Child Neurology Society.
Neurology.2006; 67 (Lippincott Williams & Wilkins): 1542-1550
]. Computed tomography (CT) and magnetic resonance imaging (MRI) are the most commonly employed methods to evaluate patients with SE. The aim and utility of imaging studies in SE falls into four overlapping categories: diagnosis, localization, evaluating pathophysiology, and prognostication.
In this review, we will discuss the role of neuroimaging following SE in children and adults. We will discuss the diagnostic yield of CT, typically performed in emergency settings, and MRI.
Typical MRI findings will be reviewed as well as differences between focal and generalized SE. Additionally, localization of focal onset epilepsy is a critical element of diagnosis, management and prognostication. Improved resolution of traditional MRI sequences, as well as advanced MRI techniques and other modalities, are playing an increasing part in localization. Furthermore, advanced neuroimaging may add value to clinical and preclinical studies that seek to understand the pathophysiology of these conditions. Finally, we will review the evidence that neuroimaging may contribute to prognostication beyond identifying an underlying cause of SE.
2. Clinical diagnosis
The most critical role for neuroimaging in clinical practice is to aid in diagnosis and identifying the etiology of SE. Neuroimaging is recommended for all patients presenting with new onset status epilepticus and should be considered in those with known epilepsy with a first episode of SE [
Practice parameter: diagnostic assessment of the child with status epilepticus (an evidence-based review): report of the Quality Standards Subcommittee of the American Academy of Neurology and the Practice Committee of the Child Neurology Society.
Neurology.2006; 67 (Lippincott Williams & Wilkins): 1542-1550
] (see Table 1). CT scans are the most readily available in emergency department setting and better than structural MRI at detecting blood. CT scans detect abnormalities in 15–33% patients [
Practice parameter: diagnostic assessment of the child with status epilepticus (an evidence-based review): report of the Quality Standards Subcommittee of the American Academy of Neurology and the Practice Committee of the Child Neurology Society.
Neurology.2006; 67 (Lippincott Williams & Wilkins): 1542-1550
]. Combining CT and MRI there is a slightly higher diagnostic yield, identifying abnormalities in 30–36%. A large portion of this improvement is due to MRI being superior for detecting inflammatory changes or subtle, remote structural abnormalities [
]. One prospective study, combining CT and MRI identified CNS infections and cerebral dysgenesis as the most common acute and remote symptomatic causes of SE, respectively [
]. Additionally, obtaining a head CT or MRI may change management. Neuroimaging lead to urgent or emergent interventions in 8.5% of children with new onset SE, while 27% of those were identified by MRI only, including stroke, acute demyelinating encephalomyelitis (ADEM) and menigoencephalitis [
Practice parameter: diagnostic assessment of the child with status epilepticus (an evidence-based review): report of the Quality Standards Subcommittee of the American Academy of Neurology and the Practice Committee of the Child Neurology Society.
Neurology.2006; 67 (Lippincott Williams & Wilkins): 1542-1550
]. Acute management changes may be initiated following the discovery of acute symptomatic causes of SE, such as hemorrhage, neoplasm, and cerebral edema. Conversely, common remote symptomatic causes of epilepsy include longstanding structural abnormalities, such as neonatal or other remote brain injuries, and malformations of cortical development including focal cortical dysplasias (see Table 2).
Table 1Indications for imaging and recommended MRI sequences for the evaluation of new onset seizures and status epilepticus. Abbreviations: 3D: 3-dimensions, FLAIR: fluid attenuated inversion recovery, DTI: diffusion tensor imaging.
Practice parameter: diagnostic assessment of the child with status epilepticus (an evidence-based review): report of the Quality Standards Subcommittee of the American Academy of Neurology and the Practice Committee of the Child Neurology Society.
Neurology.2006; 67 (Lippincott Williams & Wilkins): 1542-1550
], although the exact incidence remains unclear as many studies rely on case series or descriptive approaches. Furthermore, inflammatory, autoimmune, or parainfectious causes of SE are difficult to diagnose, as they may have MRI findings similar to those described with SE alone [
MRI changes most commonly associated with SE are increased T2 or fluid-attenuated inversion recovery (FLAIR) signal and/or diffusion restriction. Diffusion weighted imaging (DWI) and the corresponding apparent diffusion coefficient (ADC) are imaging modalities that reveal local edema that may be induced by SE. Generalized tonic-clonic seizures may lead to decreased ADC values in cortical and/or subcortical regions. These changes are typically more focal, even when seizures are generalized and frequently correlate with T2 or FLAIR hyperintensities [
]. Similarly, focal or unilateral SE, as well as lateralized periodic discharges detected by EEG are associated with focal diffusion restriction and increased T2 or FLAIR signal that may persists for days to weeks [
Fig. 1Common MRI findings due to status epilepticus. DWI (A) sequence reveals increased intensity in the left insular and frontal cortices (white arrows) with corresponding decreased apparent diffusion coefficient (ADC, B). There are similar regions of increased intensity on FLAIR imaging (C) and contrast enhancement (D) implying a component of vasogenic edema. These changes resolved on repeat MRI one week later.
]. Serial MRIs have demonstrated hippocampal signal abnormalities following febrile SE that may be related to the development of hippocampal sclerosis [
Imaging studies of refractory SE requiring a continuous infusion, demonstrate similar patterns of changes with T2 hyperintensities seen in cortex and subcortical nuclei [
]. Case reports and small series of patients following refractory and super refractory status epilepticus, defined as seizures persisting 24 h after initiation of a continuous anesthetic infusion [
]. Studies that correlate neuroimaging with pathologic findings have found widespread areas of abnormalities, including the cortex, thalamus and deep grey nuclei as well as hemispheric and crossed cerebellar atrophy [
]. This atrophy may be secondary to energy failure from excitotoxicity or ischemia or a combination of these.
3. Localization
MRI and other imaging modalities frequently reveal focal abnormalities following SE. Seizure localization may even be appreciated in the setting of seemingly generalized seizures, which likely implies focal onset with secondary generalization. This lends itself to further reliance on neuroimaging as a tool to localize seizure onset. Improved seizure localization increases the number of therapeutic options, including surgery. Traditional MRI and more recently, high resolution epilepsy sequences may identify structural etiologies, including chronic lesions (e.g. remote stroke, calcified cysticercosis) and dysplasias [
Practice parameter: diagnostic assessment of the child with status epilepticus (an evidence-based review): report of the Quality Standards Subcommittee of the American Academy of Neurology and the Practice Committee of the Child Neurology Society.
Neurology.2006; 67 (Lippincott Williams & Wilkins): 1542-1550
Beyond MR imaging, focal epilepsy is associated with perturbations in physiologic measures of brain function, especially metabolism and blood flow, which are typically tightly coupled and may be measured with radionucleotide imaging [
] FDG-PET) and measures of cerebral blood flow (CBF) by [99] ECD or HMPAO (hexamethylpropyleneamine oxime) single-photon emission computed tomography (SPECT). Similarly, advanced MR imaging techniques, such arterial spin labeling (ASL) may be beneficial in localizing a more discrete seizure onset [
]. ASL imaging is a form of MRI that utilizes a radiofrequency pulse to label arterial water allowing the ability to image cerebral blood flow, thus yielding an image with the intensity of the signal corresponding to local perfusion. These imaging techniques may provide value added in the interictal, ictal and post-ictal periods, particularly when comparing the regional metabolism and blood flow between these time points.
In the interictal state the epileptogenic zone is invariably hypometabolic [
]. Seizures have been long known to be associated with regional increases in cerebral blood to the origin of the ictal event or to regions representing propagation of ictal activity [
]. These observations are enhanced when interictal CBF is subtracted form an ictal study. After the ictal event CBF becomes reduced before normalizing [
]. In the setting of prolonged seizures, or SE (where it is not practical to obtain an interictal study), SPECT studies in children and adults show increases in CBF that are generally confined to the seizure focus based on concomitant ictal EEG and MRI findings that may show underlying structural abnormalities [
A case of nonconvulsive status epilepticus with a reversible contralateral cerebellar lesion: temporal changes in magnetic resonance imaging and single-photon emission computed tomography finding.
A recent ASL study of 21 patients with epilepsy found that 71% of subjects had post-ictal hypoperfusion within 90 min following a seizure. Eighty percent of those with hypoperfusion had regional concordance with the suspected seizure onset zone [
]. These findings recapitulate earlier studies by Rowe and colleagues who used SPECT to demonstrate that combining ictal and interictal studies of CBF yielded a correct localization in 72% of patients [
Neuroimaging may provide an opportunity to investigate and elucidate the pathophysiology of SE and brain injury. SE results in both vasogenic and cytotoxic cerebral edema. Vasogenic edema occurs with blood brain barrier breakdown and results in T2 hyperintensity without diffusion restriction [
], as well as contrast enhancement (see Fig. 1). Conversely, cytotoxic edema describes the intracellular accumulation of fluid that occurs with a failure of cellular metabolism. Brain injury, such as ischemia, leads to ATP-dependent sodium/potassium (Na+/K+) membrane pump failure and is visualized as diffusion restriction on MRI [
The timeline of cellular changes following seizure initiation has been studied extensively in animal models. These preclinical studies suggest a reduction in inhibitory signal and/or persistence of excitatory signal as the cause of prolonged seizures [
]. The etiology of seizure-induced injury is likely multifactorial, but the relative contributions of intrinsic neuronal factors versus global issues of cerebral perfusion and oxygen delivery remains an important topic of research with neuroimaging playing a key role.
The relative contribution of and timing of these types edema in SE may play a role in understanding the pathophysiology and localization of seizure onset. In studies of super refractory SE, there is increased T2 signal without restricted diffusion, indicative of vasogenic edema [
]. This may add evidence to the role of blood brain barrier breakdown and the immunopathophysiology that has been associated with many cases of super refractory SE. The resolution of similar MRI findings has been seen in more typical, non-super refractory cases of SE as well, again supporting the role of vasogenic edema [
Diffusion restriction that occurs following seizures may be related to postictal ischemia. This has been supported by animal models of SE. In these studies chemically induced seizures lead to an ADC decrease that peaked between 1–2 days after prolonged seizures, with resolution by day 7 [
Magnetic resonance imaging of changes elicited by status epilepticus in the rat brain: diffusion-weighted and T2-weighted images, regional blood volume maps, and direct correlation with tissue and cell damage.
]. Several hypotheses have been proposed to explain pathophysiology of these imaging changes in the postictal state. These theories include decreased excitatory neurotransmitter availability or desensitization, a relative increase in inhibitory transmitters and receptors, and decreased cerebral blood flow [
]. Thus these imaging findings and the presence of cytotoxic edema may be a more ominous sign, implying longer lasting or permanent cellular injury (see Fig. 2).
Fig. 2Timeline of imaging changes in status epilepticus. Imaging findings that may occur during and following SE include: 1. PET hypermetabolism and SPECT or ASL measured hyperperfusion while a seizure is ongoing; 2. an increase in diffusion restriction acutely, that remains elevated for 1–7 days following seizure and gradually decreases; 3. a gradual increase in FLAIR/T2 signal that rises through the post-ictal period and then, typically resolves over days to weeks. Note: timing of imaging changes and relative signal increases vary. PET = positron emission tomography, SPECT = single-photon emission computed tomography, ASL = arterial spin labeling, DWI = diffusion weighted imaging, FLAIR = fluid attenuated inversion recovery.
There is the most evidence for alterations in cerebral perfusion during SE. These studies, utilizing, CT and MRI perfusion and more recently ASL imaging have demonstrated increased perfusion during SE and hypoperfusion in the post-ictal state [
]. Perfusion based modalities may play a role in investigating the pathophysiology and the relative contribution of blood flow and oxygen delivery to cellular injury in SE and the post-ictal state (see Table 3).
Table 3Summary of Imaging modalities, findings and interpretation during and following resolution of status epilepticus. Abbreviations: FLAIR: fluid-attenuated inversion recovery, DWI/ADC: diffusion weight imaging/apparent diffusion coefficient, CTP: computed tomography perfusion, SPECT: single-photon emission computed tomography, ASL: arterial spin labeling, PET: positron emission tomography.
Imaging Modality
Imaging Findings
Interpretation
T2/FLAIR MRI
Ictal or Post-ictal: Hyperintensity
Cytotoxic or vasogenic edema
DWI/ADC MRI
Ictal: Diffusion restriction
Cytotoxic edema
Perfusion Imaging (e.g. CTP, SPECT, ASL)
Ictal: Hyperperfusion Post-ictal: Hypoperfusion
Changes in blood flow according to demand
FDG-PET
Ictal: Hypermetabolism Interictal: Hypometabolism
Changes in cellular metabolism during and following SE
Finally, neuroimaging has a place in prognostication following repeated seizures or SE. In SE, neuroimaging may play a prognostic role by helping identify the underlying etiology of the seizures and/or by understanding the relative contribution of the seizures themselves on neurologic outcome. Identifying an underlying structural cause of the SE has implications for acute management, mortality and neurologic outcomes, depending on the abnormality identified and its acuity [
To address the independent contribution of SE to neuroimaging findings and prognosis, prospective neuroimaging studies would be ideal. These are logistically challenging and it is rarely feasible to obtain baseline images on individuals before they present with SE. Clinical and preclinical studies have generally found transient signal changes on conventional MRI sequences [
Magnetic resonance imaging of changes elicited by status epilepticus in the rat brain: diffusion-weighted and T2-weighted images, regional blood volume maps, and direct correlation with tissue and cell damage.
]. As an exception, some studies in animals and humans have identified longer lasting structural abnormalities in the setting of presumed cytotoxic edema with T2 hyperintensities and diffusion restriction. Volumetric analysis of the hippocampus 21 days after pilocarpine induced seizures in a rat model correlated with the degree of T2 and ADC change on day 2 following SE [
] . Similarly, increased T2 signal and diffusion restriction in humans may lead to permanent cellular injury based on the lasting changes visualized on follow-up MRI [
]. Thus, certain early MRI changes may predict decreased structural volumes later on.
The role of SE in long term structural changes has been investigated the most with regards to the relationship between febrile SE and mesial temporal sclerosis (MTS) [
]. Serial MRIs of 11 toddlers presenting with febrile SE found 5 of them with evidence of hippocampal sclerosis when scanned an average of 9 months later [
]. In addition, it was appreciated in a group of immunosuppressed patients that human herpesvirus 6 was associated with increased T2 signal and increased FDG-PET uptake in the hippocampus who had a clinical syndrome including epilepsy and follow-up MRI or autopsy in a subset revealed MTS [
] and found a correlation between amygdala and hippocampal volumes with a history of prolonged febrile seizures in patients with existing temporal lobe epilepsy [
More recently, the Consequences of Prolonged Febrile Seizures in Childhood Study (FEBSTAT) a multicenter collaboration was developed to identify the relationship between febrile SE and subsequent MTS and temporal lobe epilepsy. This group sought to compare imaging findings of children with febrile SE with simple febrile seizures. They compared the brain MRIs of 199 children with febrile SE to 96 children with their first simple febrile seizure. The group identified more acute hippocampal abnormalities, including increased T2 signal in febrile SE compared with controls. Additionally, there were more developmental abnormalities of the hippocampus, including malrotation, occurring in 10.5% of SE patients versus 2.2% of controls. Extratemporal imaging abnormalities were not different between the two groups [
]. With continued follow-up the FEBSTAT study will be able to further address the contribution of developmental and acute hippocampal abnormalities on subsequent evolution of hippocampal atrophy and on the development of temporal lobe epilepsy.
6. Conclusion
Neuroimaging is a critical component of the evaluation of status epilepticus. It provides diagnostic value in many cases and in some prompts emergent interventions that would not have been appreciated otherwise. Serial neuroimaging may allow for better assessments of brain injury over time and improve our understanding of the pathophysiology, including the timing and relative contributions of impairments in cellular metabolism, changes in cerebral blood flow and blood brain barrier breakdown. With better understanding of the underlying process we gain insights into the pathogenesis of SE in different disease states while also aiding in targeted interventions and prognostication. Finally, of equal importance is the ability of MRI and other advanced imaging techniques to provide seizure onset localization that is the first step in consideration for surgical approaches to treat underlying epilepsy and in rare cases persistent status.
Existing recommendations on the indications and timing for neuroimaging in SE rely heavily on expert opinion. We have described here data supporting the diagnostic value of neuroimaging that has established it as a standard of care in new onset SE. However, knowledge gaps remain. The impact of neuroimaging findings on the type and escalation of anti-epileptic therapies has not been described. Additionally, there is limited prospective data about the prognostic value of neuroimaging, which is in part related to the heterogeneity of etiologies. Further investigation of SE based on etiology and neuroimaging findings may facilitate future targeted therapies and a better understanding of prognosis. The FEBSTAT study is an important step forward to fill this void.
Disclosures
RMG and WDG: Have no relevant disclosures.
References
Raspall-Chaure M.
Chin R.F.
Neville B.G.
Scott R.C.
Outcome of paediatric convulsive status epilepticus: a systematic review.
Practice parameter: diagnostic assessment of the child with status epilepticus (an evidence-based review): report of the Quality Standards Subcommittee of the American Academy of Neurology and the Practice Committee of the Child Neurology Society.
Neurology.2006; 67 (Lippincott Williams & Wilkins): 1542-1550
A case of nonconvulsive status epilepticus with a reversible contralateral cerebellar lesion: temporal changes in magnetic resonance imaging and single-photon emission computed tomography finding.
Magnetic resonance imaging of changes elicited by status epilepticus in the rat brain: diffusion-weighted and T2-weighted images, regional blood volume maps, and direct correlation with tissue and cell damage.
30033-5&id=gr1.jpg){kind=link}
30033-5&id=gr2.jpg){kind=link}