A seizure is a transient occurrence of abnormal excessive or synchronous neuronal activity in the brain. Seizures manifest in different ways based on the anatomic regions of hyperactive neuronal activity. For example, patients may develop focal symptoms due to abnormal activity in the temporal lobe, whereas global signs represent widespread aberrant neuronal activity. Seizures may initially manifest as focal symptoms with subsequent generalization to the remaining cortex. Furthermore, patients may or may not lose consciousness during a seizure, depending on whether or not the limbic structures and brainstem are involved.
Seizure activity in the brain can be caused by numerous anatomic abnormalities such as tumors, infection, inflammatory/autoimmune processes, vascular malformations, stroke, trauma, cortical malformations/dysplasias, gray matter heterotopias, mesial temporal sclerosis, encephaloceles or other acquired or developmental abnormalities. Patients may have seizures due to medical factors such as metabolic derangement, drug/alcohol withdrawal, hyperthermia, or toxins as well. However, patients may also suffer from recurrent seizures without known underlying etiology. Patients with at least two unprovoked seizures separated by at least 24 hours may be diagnosed with epilepsy.
Seizure management relies on the treatment of the underlying etiology and/or anti-seizure drug therapy, and, for most patients, part of the evaluation for the underlying cause requires diagnostic workup with imaging. Various diagnostic imaging modalities may be used for patients with recurrent seizures, many adding complementary information for the care of these patients. Furthermore, diagnostic imaging can provide information that localizes epileptogenic lesions in patients with refractory epilepsy that require surgical intervention, potentially obviating the need for invasive electroencephalography (EEG). As such, understanding the uses and limitations of each modality is of critical importance for the treatment of these patients.
Seizures can manifest as a result of a wide variety of anatomic abnormalities within the brain as well as toxic or metabolic derangements. Anatomic abnormalities that result in seizures can be located nearly anywhere within the brain, though usually involve the neocortex or mesial temporal region, particularly the hippocampi. Because of this, imaging is typically employed to adequately scrutinize all structures of the brain with careful attention directed towards the hippocampi.
The hippocampi are situated within the medial aspect of the temporal lobes bilaterally and occupy the medial floor of the lateral ventricles. They are a core limbic structure, responsible for learning and memory formation. The hippocampus is composed of two distinct gray matter structures known as the cornu ammonis and the dentate gyrus. Anterior to posterior it is divided into the head, body, and tail segments. On coronal imaging, the hippocampal head is characterized by digitations which give its superior surface an undulating contour. The hippocampal body can be recognized by its "jelly roll" or "swiss roll" appearance of the interlocking dentate gyrus, Ammon horn, and intervening strata. White matter tracts from the hippocampus traverse its superior surface, forming the alveus, which condenses into bundles called fimbria, which continue posteriorly as the fornix. The fornix terminates just off of midline within the mamillary body; this white matter tract plays a vital role in the Papez circuit. Adjacent to the head of the hippocampus lies the amygdala and entorhinal cortex.
Hippocampal sulcus remnant cysts and incomplete hippocampal inversion are developmental variants in the hippocampus, which should not be confused with pathology.
The bilateral and symmetric nature of the hippocampi allows for direct comparison during imaging. Unilateral abnormalities may shed light on the underlying etiology of a patient's refractory epilepsy. However, gross abnormalities of the hippocampi may be bilateral in up to 10% of cases. Additionally, hippocampal lesions can be associated with extra hippocampal epileptogenic lesions. Proper identification of hippocampal abnormalities is critical for patients with medically refractory epilepsy, as surgical resection of the epileptogenic focus is the standard treatment for these patients.
Plain radiography represents the earliest form of diagnostic imaging. X-rays are used to generate image contrast based on differences in tissue attenuation. Because the soft tissues of the brain exhibit similar attenuation characteristics, the use of plain radiographs to evaluate for structural lesions within the brain is extremely limited. As such, plain radiographs play no role in the diagnostic workup for patients suffering from seizures.
Computed tomography (CT) utilizes helically acquired x-rays and postprocessing techniques to generate cross-sectional images. Modern CT scanners receive x-ray attenuation data in a nearly isotropic manner, which allows the generation of voxels that can be reconstructed in coronal, sagittal, and three-dimensional formats. As with the limitations of plain radiographs, the brain soft tissues are poorly evaluated on routine CT examinations because the attenuation characteristics of the brain soft tissues and many pathologies are similar. Iodinated intravenous contrast can highlight the vascular structures of the brain or areas of disruption of the blood-brain barrier. Seizure activity may result in increased cortical enhancement due to increased cortical perfusion but is typically an unanticipated observation in patients that are not suspected of seizures rather than a sought out diagnostic finding. As such, CT plays a limited role in the imaging workup of patients considering epilepsy surgery.
Magnetic resonance imaging (MRI) is the preferred diagnostic modality for patients with seizures. MRI offers excellent signal-to-noise and contrast within the brain. Seizures that are attributed to known metabolic arrangements may not necessarily require further diagnostic studies; however, nearly every patient that suffers from an unexplained seizure should undergo an MRI to evaluate for underlying structural brain abnormalities. Virtually any MRI can be used to assess for mass lesions within the brain, but high field strength scanners, more than 1.5 Tesla, should be used for evaluating patients with epilepsy when possible. Specialized imaging protocols have been developed which optimize subtle signal intensity alterations and anatomic abnormalities within the hippocampi. This is particularly critical for patients undergoing evaluation for surgical management of intractable epilepsy, as small abnormalities within the hippocampi may be undetectable without specialized techniques.
Patients with medically intractable partial complex epilepsy are most commonly affected by mesial temporal sclerosis (MTS). MRI is essential in identifying MTS, as it has characteristic findings of volume loss and increased T2/FLAIR signal intensity due to hippocampal neuronal cell death and gliosis. There may also be associated atrophy within the ipsilateral amygdala, entorhinal cortex, fornix, or mammillary body. Identifying these subtle differences requires the acquisition of a 1 mm isotropic series with T1 weighting and FLAIR. Reconstructions must be performed perpendicular to the plane of the hippocampi to allow adequate side-to-side comparison. A coronal T2-weighted series should also be obtained with 2 mm slices and sub-mm in-plane resolution to allow both side-to-side comparisons of the hippocampi as well as to delineate the typical internal architecture. These sequences are also useful in detecting focal cortical dysplasias, gray matter heterotopias, and small encephaloceles.
Intravenous contrast may improve the utility of MRI depending on the clinical circumstances. Gadolinium-based contrast agents act to increase the T1 signal, highlighting vascular structures and blood-brain barrier abnormalities. A common approach to patients with seizures is to perform non-enhanced MRI sequences initially and only to administer contrast if the nonenhanced study requires further investigation. That said, patients with intractable epilepsy undergoing evaluation for possible surgical treatment do not routinely require intravenous contrast. Finally, it should be noted that MRI does not require the use of ionizing radiation, where this is a necessary consequence of CT imaging.
Although MRI can be useful for the detection of underlying structural lesions, MRI can also be used to evaluate brain physiology. In patients being evaluated for surgical resection, functional MRI (fMRI) is useful in identifying the language laterality and can, in many instances, replace the invasive Wada test.
Ultrasound utilizes high-frequency sound waves to generate diagnostic images. The advantage of ultrasound is that no ionizing radiation is required for its use. Unfortunately, calcified structures such as the bones of the calvarium preclude adequate sound transmission for an ultrasound to be useful in diagnostic imaging of the brain. As such, ultrasound plays no role in the evaluation of patients with seizures.
Nuclear imaging plays an adjunctive role in seizure imaging. There are circumstances where nuclear imaging studies add complementary information to that of traditional cross-sectional imaging such as MRI.
Positron emission tomography with fluorodeoxyglucose (FDG-PET) allows for metabolic imaging within the brain. The fluorodeoxyglucose is actively taken up by neuronal cells, in an activation-dependent distribution. Thus, FDG uptake is increased in parts of the brain during a seizure, and conversely, uptake is decreased within the seizure focus interictally. These temporal factors contribute significant limitations to FDG-PET imaging, making it technically challenging to obtain the images either during a seizure or immediately after. Furthermore, PET imaging has low resolution compared to CT and MRI, with a resolution limit of approximately 1 cm. Because of this limitation, FDG-PET imaging is often co-registered with either CT or MRI data to provide useful colocalization between the foci of metabolic abnormality and anatomic structures. However, in patients with suspected temporal lobe epilepsy, interictal FDG-PET is frequently useful in seizure localization, especially in patients with normal MRI scans.
Single-photon emission CT (SPECT) produces images through the use of radioisotope production of gamma rays. These radioisotopes are linked to parent molecules, known as radiopharmaceuticals. Radiopharmaceuticals such as Tc99m-HMPAO do not cross the blood-brain barrier and act as perfusion agents within the brain. Through rapid intravenous administration of a radiopharmaceutical within 90 seconds of seizure onset, regions of increased perfusion within the brain can be identified, which correspond to the seizure focus. Similarly, postictal administration results in decreased cerebral blood flow in the epileptogenic center. Subtracting ictal and interictal SPECT studies with coregistration to an MRI (SISCOM) improves the utility of SPECT imaging. However, radiopharmaceutical administration within 90 seconds of seizure onset is technically challenging, limiting the utility of SPECT imaging. This modality is most commonly used when conventional MRI imaging, electroencephalograms, and other adjunct tests are equivocal in seizure focus localization.
As described previously, both CT and magnetic resonance angiography are uncommonly performed in patients with seizure disorders. A notable exception includes patients who are suspected of having underlying ischemic or vascular disease within the head or neck. Outside of these narrow indications, cross-sectional angiography is seldomly performed during the routine seizure workup. Additionally, more invasive tests such as catheter-based angiography may be used for further delineation of vascular pathology but are rarely required. Before surgical interventions, vascular imaging may be warranted, but this is dependent on the individual clinical scenario.
Before the advent of modern CT and MRI scanners, patient positioning was of critical importance to obtain accurate and useful diagnostic images. However, modern scanners can acquire data in an isotropic fashion, which permits post imaging processing and reconstruction. Prior to this technical development, adjusting for differences in patient positioning was not easily performed. Nearly all studies are performed supine with the patient lying in a comfortable position. This is particularly relevant for MRI studies since image quality is significantly degraded with even small patient movements; as such, ensuring that a patient can maintain a particular position long enough for image acquisition is of considerable technical importance. Similar principles also apply to FDG-PET and SPECT imaging studies.
Gaining a full understanding of the various imaging modalities for patients who suffer from seizures is of critical importance. Nearly one-third of patients with epilepsy will not achieve remission with antiepileptic medications alone, and many patients will have underlying anatomic abnormalities that may offer a surgical cure. Specialized MRI protocols are necessary to thoroughly scrutinize the hippocampi, as many of these patients will develop or demonstrate mesial temporal sclerosis. More subtle abnormalities, such as focal cortical dysplasias, may require advanced postprocessing techniques and expert neuroradiologists for a full evaluation. Patients with structural lesions concordant with seizure localization on EEG should receive a preoperative fMRI to map regions of eloquent neocortex. In cases where no anatomic lesion is identified on MRI, FDG-PET may offer complementary information, revealing areas of hypometabolism or help guide the placement of intracranial EEG electrodes. Finally, if PET fails to aid localization, ictal, and interictal SPECT imaging can be performed to evaluate dynamic changes in cerebral perfusion, helping direct the placement of intracranial EEG electrodes for definitive localization of an epileptogenic focus. Further advanced imaging equipment and techniques are under development and available at a limited number of institutions, including the use of 7T MRI. This increased field strength offers an opportunity to detect even more subtle lesions, potentially increasing overall imaging sensitivity and providing curative surgery to a greater proportion of patients. Advanced neuroimaging will continue to play an evolving role in the surgical management of patients with medication-refractory epilepsy.
|||So EL, Classifications and epidemiologic considerations of epileptic seizures and epilepsy. Neuroimaging clinics of North America. 1995 Nov; [PubMed PMID: 8564281]|
|||Friedman E, Epilepsy imaging in adults: getting it right. AJR. American journal of roentgenology. 2014 Nov; [PubMed PMID: 25341150]|
|||Dekeyzer S,De Kock I,Nikoubashman O,Vanden Bossche S,Van Eetvelde R,De Groote J,Acou M,Wiesmann M,Deblaere K,Achten E, [PubMed PMID: 28108955]|
|||Tanenbaum L,Drayer BP,Anderson RE,Braffman B,Davis PC,Deck MD,Hasso AN,Johnson BA,Masaryk T,Pomeranz SJ,Seidenwurm D,Masdeu JC, Epilepsy. American College of Radiology. ACR Appropriateness Criteria. Radiology. 2000 Jun; [PubMed PMID: 11037457]|
|||Duncan JS, Brain imaging in epilepsy. Practical neurology. 2019 Oct; [PubMed PMID: 31420416]|
|||Bernasconi A,Cendes F,Theodore WH,Gill RS,Koepp MJ,Hogan RE,Jackson GD,Federico P,Labate A,Vaudano AE,Blümcke I,Ryvlin P,Bernasconi N, Recommendations for the use of structural magnetic resonance imaging in the care of patients with epilepsy: A consensus report from the International League Against Epilepsy Neuroimaging Task Force. Epilepsia. 2019 Jun [PubMed PMID: 31135062]|
|||Deblaere K,Achten E, Structural magnetic resonance imaging in epilepsy. European radiology. 2008 Jan; [PubMed PMID: 17899110]|
|||Duncan JS,Winston GP,Koepp MJ,Ourselin S, Brain imaging in the assessment for epilepsy surgery. The Lancet. Neurology. 2016 Apr; [PubMed PMID: 26925532]|
|||Rathore C,Dickson JC,Teotónio R,Ell P,Duncan JS, The utility of 18F-fluorodeoxyglucose PET (FDG PET) in epilepsy surgery. Epilepsy research. 2014 Oct; [PubMed PMID: 25043753]|
|||LoPinto-Khoury C,Sperling MR,Skidmore C,Nei M,Evans J,Sharan A,Mintzer S, Surgical outcome in PET-positive, MRI-negative patients with temporal lobe epilepsy. Epilepsia. 2012 Feb [PubMed PMID: 22192050]|
|||O'Brien TJ,So EL,Mullan BP,Hauser MF,Brinkmann BH,Bohnen NI,Hanson D,Cascino GD,Jack CR Jr,Sharbrough FW, Subtraction ictal SPECT co-registered to MRI improves clinical usefulness of SPECT in localizing the surgical seizure focus. Neurology. 1998 Feb [PubMed PMID: 9484370]|
|||la Fougère C,Rominger A,Förster S,Geisler J,Bartenstein P, PET and SPECT in epilepsy: a critical review. Epilepsy [PubMed PMID: 19236949]|
|||Martin P,Winston GP,Bartlett P,de Tisi J,Duncan JS,Focke NK, Voxel-based magnetic resonance image postprocessing in epilepsy. Epilepsia. 2017 Sep; [PubMed PMID: 28745400]|
|||Von Oertzen J,Urbach H,Jungbluth S,Kurthen M,Reuber M,Fernández G,Elger CE, Standard magnetic resonance imaging is inadequate for patients with refractory focal epilepsy. Journal of neurology, neurosurgery, and psychiatry. 2002 Dec [PubMed PMID: 12438463]|