Epilepsia, 52(Suppl. 4):25–27, 2011 doi: 10.1111/j.1528-1167.2011.03147.x
IS THERE SUCH A THING AS NONLESIONAL EPILEPSY?
MRI-negative epilepsy: Protocols to optimize lesion detection Louis-Gilbert Ve´zina Department of Diagnostic Imaging and Radiology, Children’s National Medical Center, Washington, District of Columbia, U.S.A.
can call attention to a lesion that is either very subtle or not evident on conventional sequences. Detection of cortical anomalies is best performed early in infancy, preferably before 6 months of age. If the initial magnetic resonance imaging (MRI) scan is performed between 9 and 18 months of age and is negative, a repeat scan after 2 years of age may be necessary. KEY WORDS: Focal cortical dysplasia, Hippocampal sclerosis, Polymicrogyria, Magnetization transfer imaging, Myelination.
SUMMARY Identification of the structural lesions that underlie pediatric epilepsy can be challenging. Careful evaluation of the gray–white matter interface is crucial, and necessitates multiplanar thin images of high resolution that can differentiate focal lesions from partial volume averaging artifacts created by the innate gyral configuration. Careful evaluation of the hippocampus and of the myelination patterns can further increase the diagnostic yield of the study. Magnetization transfer imaging
In children with new-onset seizures, magnetic resonance imaging (MRI) is the preferred imaging modality to identify epileptogenic lesions: malformations of cortical development, tumors, vascular anomalies, and gliosis/ scarring. Identification of some epileptogenic lesions, however, can be challenging; if not recognized, a negative MRI report may erroneously be generated (Von Oertzen et al., 2002). In surgical series, the structural lesions most commonly found in patients with refractory epilepsy and MRI-negative examinations consist of malformations of cortical development [primarily focal cortical dysplasia (FCD) and polymicrogyria] and hippocampal sclerosis (HS) (Bien et al., 2009). FCD is the most common histopathologic substrate identified in infants and children with epilepsy (Lerner et al., 2009; Hsieh et al., 2010). The MRI findings of FCDs may be subtle. They consist of: mild blurring of the gray– white matter junction, mild increase in T2 signal, mild thickening of the cortex, and abnormal gyral configuration (Colombo et al., 2009). These findings are often not identified using standard MRI sequences (Von Oertzen et al.,
2002). Therefore, children with focal epilepsy (and all children with epilepsy younger than age 2 years) need to be evaluated with high quality studies tailored to the identification of these structural lesions (Gaillard et al. 2009). Most epilepsy centers utilize 1.5 Tesla (1.5T) strength magnets; these scanners, especially with the use of multichannel phase array coils, generate excellent studies that can identify most lesions. Three Tesla (3T) systems can generate images of higher resolution; in a small subset of patients, 3T will likely prove more sensitive to subtle epileptic lesions (Nguyen et al., 2010).
Protocols A scanning protocol that is consistently applied in an imaging center that evaluates a high volume of children with epilepsy will yield the highest return for the identification of a seizure focus (Von Oertzen et al., 2002). Use of such dedicated protocols will maximize the early detection of structural abnormalities in children with new-onset seizures, and lead to more optimal treatment—including earlier surgical resection of lesions discovered in patients who develop refractory seizures. A minimum, time efficient protocol should include the following sequences; the rationale for some of the sequences is explained. 1 A high-resolution T2-weighted sequence, slice thickness 2 mm or less. These are traditionally obtained as
Address correspondence to Louis-Gilbert Vzina, M.D., Department of Diagnostic Imaging and Radiology, Children’s National Medical Center, 111 Michigan Av NW, Washington, DC 20010, U.S.A. E-mail:
[email protected] Wiley Periodicals, Inc. ª 2011 International League Against Epilepsy
25
26 L.-G. Ve´zina two-dimensional (2D) images; images need to be of high resolution, voxel size 0.6–0.7 mm2 (for example, field-of-view 20–22 cm, matrix size 256 · 320 or greater). High spatial resolution, whole brain threedimensional (3D) T2-weighted acquisitions are now commercially available; it is not yet clear if these volume sequences provide the contrast resolution needed to resolve the more subtle FCDs. Rationale: T2-weighted images are crucial in the examination of the cerebral cortex and the identification of subtle cortical dysplasia (Colombo et al., 2009). Because cerebral cortex is thin (2 or 3 mm in thickness), images that are thinner than the cortex proper are necessary to achieve maximal resolution and account for partial volume effects (artifactual cortical thickening) in areas with sulci running in different directions. 2 A heavily T1-weighted 3D gradient-echo sequence, which can be reformatted into orthogonal or curved planes. In infants younger than 1 year of age, when myelination is not fully mature on T1-weighted images, the 3D T1-weighted sequence is of limited value; it should be replaced by a second (orthogonal) T2-weighted sequence. 3 A fluid-attenuated inversion recovery (FLAIR) acquisition, 4–5 mm thick, in either axial or coronal plane. Coronal acquisition is preferred in older children and adults as it provides more opportunity to examine the signal intensity of the hippocampus. High-resolution whole brain 3D T2-weighted FLAIR acquisitions are also commercially available and may prove more sensitive than 2D images. In infants younger than 1 year of age, FLAIR images are not sensitive to many pathologies and should be replaced by proton density images. 4 A very high resolution oblique coronal T2-weighted acquisition of the hippocampal formations. Images are obtained perpendicular to the long axis of the hippocampus, using 3-mm slice thickness, submillimeter voxel resolution (suggest 20 cm field of view [FOV], matrix 256 · 480 – about 0.3–0.4 mm2 voxel size). These images are helpful for analyzing the size, signal, and characteristics of the hippocampal formations and adjacent mesial temporal structures. Rationale: HS is the most frequent pathologic substrate in adults with refractory focal epilepsy (Hashiguchi et al., 2010), and is identified in children as early as 1 year of age. HS is a progressive pathologic process due to repeated seizures that can lead to progressive neuronal damage in the hippocampus. The principal volume loss occurs early in the disease, and it is associated with gliosis of the hippocampus; the sclerotic hippocampus becomes smaller than the contralateral hippocampus and shows bright T2 signal. The atrophy and T2 changes can be subtle and detected only on very high resolution studies of the hippocampus (Hashiguchi et al., 2010). These images also provide excellent resolution of the Epilepsia, 52(Suppl. 4):25–27, 2011 doi: 10.1111/j.1528-1167.2011.03147.x
hypothalamus; the latter should be closely examined to exclude the presence of a hypothalamic hamartoma. Additional acquisitions that should be strongly considered include: 1 An axial magnetization transfer (MT) study (4–5 mm thick images). MT imaging is especially useful in the earlier years, when some FCDs can be difficult to detect on conventional T1, T2, or FLAIR images. Rationale: MT is based on the interaction between mobile free water protons and macromolecular bound protons. With MT imaging, an off-resonance radiofrequency (RF) pulse is applied to saturate protons bound to macromolecules, mainly the myelin sheath covering the axons. Due to spin–spin interactions, there is a transfer of the saturation effect of the macromolecular bound hydrogen molecules to the nearby free protons. This results in a decrease in signal from the mobile proton and suppression of signal from background brain tissue. If a lesion contains abnormal myelination or microcalcifications, the signal suppression will be decreased compared to that observed in the healthy white matter; the conspicuity of the lesion may increase, revealed as a T1 bright focus of abnormal signal. MT images are superior in the detection of white matter lesions in tuberous sclerosis (Pinto Gama et al., 2006), and similarly in some FCDs (Fig. 1) (Rugg-Gunn et al., 2003). 2 A high-resolution T2-weighted fast inversion recovery sequence. These generate images with very high contrast between gray and white matter (‘‘in vivo myelin stain’’) and are useful in diagnosis of FCD (Chan et al., 1998) and HS (Hashiguchi et al., 2010).
A
B
Figure 1. (A) A 2-mm–thick axial T2-weighted image of an FCD (arrow) in the right posterior suprasylvian cortex; the affected cortex is slightly thickened, but not easily differentiated from artifactual cortical thickening, which is created when the MR image is not perpendicular to the cortex. (B) A 5-mm MT image in the same patient clearly reveals a curvilinear band of abnormal increased signal (arrow). Epilepsia ILAE
27 MRI-Negative Epilepsy: Protocols to Optimize Lesion Detection 3 A susceptibility gradient recalled echo (GRE) T2weighted sequence. GRE sequences are helpful to detect the presence of calcified or mineralized lesions, and they should be obtained if a computed tomography (CT) scan is not available. Paramagnetic contrast infusion and postcontrast T1weighted images are not routinely indicated in the child with new-onset epilepsy. Contrast should be given if there is concern for a focal lesion such as tumor, inflammation, or infection, or a vascular malformation; or if there is possibility or evidence of a meningitis or meningeal process. Magnetic resonance spectroscopy can be useful for detecting metabolic derangements that lead to epilepsy, and can also assess the lateralization of focal epilepsy (especially temporal lobe epilepsy) (Willmann et al., 2006). Diffusion imaging detects cytotoxic cortical swelling, which can be seen in some metabolic disorders (including mitochondrial encephalopathy, lactic acidosis, stroke-like episodes [MELAS] syndrome) and in the postictal period. Finally, changes in the signal intensity of cerebral white matter throughout infancy (as a result of ongoing myelination) need to be accounted for, as they can result in a change in appearance (over time) of cortical and subcortical lesions. As axons myelinate, their T2 signal shifts from bright (significantly brighter than adjacent cortex) to dark (darker than cortex). Most of these changes occur between 8 and 18 months of life. During this period, T2 contrast between gray and white matter is poor (high-resolution T2-weighted fast inversion recovery sequences are especially useful in this situation). Lesions that are evident on MRI studies performed prior to 6 months of age, when there is high contrast between the T2 bright white matter and adjacent cortex, may ‘‘disappear’’ as the brain myelinates—only to reappear later on. Conversely, if the initial MRI is performed between 8 and 18 months, a lesion may be present but not evident. Therefore, if the initial scan is performed between 9 and 18 months of age and is negative, a repeat scan after 2 years of age may be necessary (Lerner et al., 2009).
References Bien CG, Szinay M, Wagner J, Clusmann H, Becker AJ, Urbach H. (2009) Characteristics and surgical outcomes of patients with refractory magnetic resonance imaging-negative epilepsies. Arch Neurol 66:1491–1499. Chan S, Chin SS, Nordli DR, Goodman RR, DeLaPaz RL, Pedley TA. (1998) Prospective magnetic resonance imaging identification of focal cortical dysplasia, including the non-balloon cell subtype. Ann Neurol 44:749–757. Colombo N, Salamon N, Raybaud C, Ozkara C, Barkovich AJ. (2009) Imaging of malformations of cortical development. Epileptic Disord 11:194–205. Gaillard WD, Chiron C, Cross JH, Harvey AS, Kuzniecky R, Hertz-Pannier L, Vezina LG; ILAE, Committee for Neuroimaging, Subcommittee for Pediatric. (2009) Guidelines for imaging infants and children with recent-onset epilepsy. Epilepsia 50:2147–2153. Hashiguchi K, Morioka T, Murakami N, Suzuki SO, Hiwatashi A, Yoshiura T, Sasaki T. (2010) Utility of 3-T FLAIR and 3D short tau inversion recovery MR imaging in the preoperative diagnosis of hippocampal sclerosis: direct comparison with 1.5-T FLAIR MR imaging. Epilepsia 51:1820–1828. Hsieh DT, Chang T, Tsuchida TN, Vezina LG, Vanderver A, Siedel J, Brown K, Berl MM, Stephens S, Zeitchick A, Gaillard WD. (2010) New-onset afebrile seizures in infants: role of neuroimaging. Neurology 12:150–156. Lerner JT, Salamon N, Hauptman JS, Velasco TR, Hemb M, Wu JY, Sankar R, Donald Shields W, Engel J Jr, Fried I, Cepeda C, Andre VM, Levine MS, Miyata H, Yong WH, Vinters HV, Mathern GW. (2009) Assessment and surgical outcomes for mild type I and severe type II cortical dysplasia: a critical review and the UCLA experience. Epilepsia 50:1310–1335. Nguyen DK, Rochette E, Leroux JM, Beaudoin G, Cossette P, Lassonde M, Guilbert F. (2010) Value of 3.0 T MR imaging in refractory partial epilepsy and negative 1.5 T MRI. Seizure 19:475–478. Pinto Gama HP, da Rocha AJ, Braga FT, da Silva CJ, Maia AC Jr, de Campos Meirelles RG, MendonÅa do Rego JI, Lederman HM. (2006) Comparative analysis of MR sequences to detect structural brain lesions in tuberous sclerosis. Pediatr Radiol 36:119–125. Rugg-Gunn FJ, Eriksson SH, Boulby PA, Symms MR, Barker GJ, Duncan JS. (2003) Magnetization transfer imaging in focal epilepsy. Neurology 27:1638–1645. Von Oertzen J, Urbach H, Jungbluth S, Kurthen M, Reuber M, Fernndez G, Elger CE. (2002) Standard magnetic resonance imaging is inadequate for patients with refractory focal epilepsy. J Neurol Neurosurg Psychiatry 73:643–647. Willmann O, Wennberg R, May T, Woermann FG, Pohlmann-Eden B. (2006) The role of 1H magnetic resonance spectroscopy in pre-operative evaluation for epilepsy surgery. A meta-analysis. Epilepsy Res 71:149–158.
Disclosure The author has no conflict of interest to disclose. The author confirms that he has read the Journal’s position on issues involved in ethical publication and affirms that this report is consistent with those guidelines.
Epilepsia, 52(Suppl. 4):25–27, 2011 doi: 10.1111/j.1528-1167.2011.03147.x
