Periventricular nodular heterotopia: A challenge for epilepsy surgery
Article Outline
- Summary
- Case history
- Consequences for surgical decision-making and intraoperative technique
- Discussion
- References
- Copyright
Summary
Pharmacoresistant focal epilepsies due to periventricular nodular heterotopia are a diagnostic and therapeutic challenge because of the need of invasive presurgical diagnostics and the selection of an optimal surgical approach. Invasive investigations in previous studies showed that focal epileptic activity can be correlated predominantly either with one of the nodular heterotopia or with neocortical epileptogenic zones distant to the periventricular nodules. Up to now, invasive recordings were required for localization of epileptic activity and its correlation to heterotopia. The following case presentation reports on a non-invasive approach using magnetic source imaging (MSI) combined with intraoperative ECoG. MSI combines preoperative data from magnetic resonance imaging (MRI) with magnetoencephalography (MEG). The MSI data for definition of the localization of the epileptic activity and functional important areas were coregistered with the intraoperative high-field-MRI and diffusion tensor imaging-based fiber tracking (DTI) of the visual pathway using a neuronavigational system. A neuronavigation-guided surgical resection of the epileptogenic area was performed leaving the heterotopia and the visual tract fibers intact. Postoperatively preservation of the visual fields was documented and the frequency of seizures was markedly reduced.
Keywords: Periventricular heterotopia, Epilepsy surgery, MEG, Neuronavigation
Case history
A 12-year-old female patient suffered from febrile convulsions in her second year of life. Since the age of 10, she suffers from focal seizures with clouding of consciousness (stare gaze, oral automatisms, ictal speech, postictal paleness and cold shiver) with a frequency of five to six seizures per month and additional nocturnal seizures of unknown frequency. Because of drug intractability, the patient was referred to presurgical investigation.
Findings
The patient was right handed, the neurological examination was normal except for a slight hemianopic impairment to the left (central scotoma of the upper visual fields of both eyes). The patient was also known to suffer from panhypopituarism.
MRI showed nodular periventricular heterotopia unilaterally on the right, lying laterally along the temporal horn, the trigone and the occipital horn. A radial band of the same signal intensities as the grey matter extended from the nodules to the basolateral cortex. The posterior temporal neocortex also appeared thinner. Finally, there was a hippocampal structural anomaly, the hippocampus showed a disrupted internal architecture and was atrophic (Fig. 1). The heterotopia extended to the right occipital horn of the ventricle. Spectroscopy showed reduced N-acetyl-aspartate (NAA) concentration and reduced NAA/Cr (Creatine) in the right hippocampus.
Figure 1.
(A and B) T2-weighted (axial) and MPRAGE (sagittal) MR imaging show periventricular heterotopia in the right basal temporo-occipital area (marked by arrow). (C) FLAIR (axial) MR image shows no evidence for cortical malformations or cortical lesions. (D) T2-weigthed (coronar) image: the arrow points to a radial band reaching from the nodular heterotopia to the lateral cortex. (E and F) A typical MEG spike of the patient. Two distinct MSI foci are shown in a sagittal view (Nos. I and II).
Fluoro-deoxy-glucose positron emission tomography (FDG-PET) revealed hypometabolism on the right mainly in the posterior temporo-lateral area more than in the periventricular heterotopic area.
Video-scalp EEG monitoring documented complex partial seizures with ictal onset in the right posterior temporal and parieto-occipital regions.
During simultaneous MEG/EEG recordings only spikes in MEG were recorded (Fig. 1) (Magnes II biomagnetometer; four-dimensional Neuroimaging, San Diego, CA, USA). Source localization using ECD (equivalent dipole) source model and a realistic volume conductor model Boundary Element Method (BEM; Curry 4.6, Neuroscan Compumedics), was performed and coregistered with MRI. Visual inspection of MEG spike localizations revealed two clusters of centers of gravity, which was confirmed using k-means cluster analysis (SPSS, Chicago, USA). One cluster were close to the region of the mesial neuronal heterotopia, adjacent to the occipital horn of the right ventricle (I, Fig. 1) and the second in the posterior neocortical part of the right temporal lobe (II, Fig. 1).
Consequences for surgical decision-making and intraoperative technique
Surgical planning was performed on the basis of coregistration of non-invasive data of MRI, MEG and diffusion tensor imaging (DTI) of the visual pathway (Fig. 2). Intracerebral depth electrodes and subdural strips were implanted by means of functional neuronavigation, and guided by MEG localization into the heterotopia (Fig. 3A). Intraoperative ECoG-recording revealed two spatially separated epileptogenic spike activity zones: the main epileptogenic spike activity was localized in the basal occipito-temporal cortex with 77% of the activity and the second one was found close to the periventricular neuronal heterotopia with 23%. The numbers (77% and 23%) were the results of a template searching algorithm based on the Pearson correlation coefficient for independent spikes occurring at different electrodes (details about the spike distribution at different channels are shown in Fig. 3B). The latter area near the heterotopia was lying very close to and within the visual pathway and therefore could not be surgically removed because of the risk of complete postoperative hemianopsia.
Figure 2.
(A) Streamtube visualization of the right optic radiation based on diffusion tensor imaging. (B) For navigation, a three-dimensional object representing the optic radiation (wrapping the individual fibers) and two distinct MSI foci (red) are generated. (C) Relation of optic radiation (visualized as streamlines) to MSI foci. (D–F) Sagittal/coronal/axial view of T1-weighted images with registered DTI and MSI data. Localization of focal epileptic activity is below the optic tract.
Figure 3.
(A) Intraoperative electroencephalographic recordings with platinum-electrodes close to the two suspected spike foci as suggested by MSI (white arrows: temporobasal strip, thin white arrow: depth electrode to heterotopia). Additional measurements were picked up from the lateral cortical surface (hatched arrow). The electrode position was confirmed by intraoperative T1- and T2-weighted high-field-MR imaging. (B) MSI guided electrode implantation of intracerebral depth and subdural electrodes; spike activity in lateral cortex and periventricular heterotopia, the corresponding spike density distribution is shown (upper right). The neocortex shows predominant spike wave activity and 11–12
s−1 polyspikes during intraoperative ECoG. (C) Intraoperative MR-imaging after cortical resection of MSI-focus No. II (with platinum electrodes still in situ). MSI-focus No. I adjacent to the heterotopia was left intact.
The patient was positioned on a rotating surgical table in an operating room with a 1.5-T-MR imager installed (Magnetom Sonata Maestro Class; Siemens Medical Solutions). After fixation of the patient's head in a MRI-compatible headholder, 1.0-mm isotropic three-dimensional MR image was obtained for use as a navigational reference data set. The functional imaging data (MSI, DTI) were integrated into this baseline MRI with the neuronavigation system (ImageFusion software; BrainLAB, Heimstetten, Germany). From these data, a three-dimensional aspect and a microscope-based navigational view were calculated. After craniotomy and exposal of the temporo-occipital cortical surface a 24-channel-electrocorticography (ECoG) recording was performed to further encircle the suspected epileptogenic focus. Intracerebral platinum depth and subdural strip electrodes with contacts over the two suspected epileptogenic foci were inserted under neuronavigational guidance (VectorVision sky cranial; BrainLAB). The exact three-dimensional position of each electrode contact covering the latero-basal temporo-occipital cortex and the periventricular heterotopia was visualized with T1-MPRAGE and T2-weighted images from the intraoperative high-field-MR (Fig. 3A). Electrocorticographic spike recordings confirmed that the main epileptogenic activity arose from the latero-basal neocortex and not from the periventricular heterotopia zone (Fig. 3B). This was concordant with the preoperative MEG spike localizations.
The tailored microsurgical resection was guided by a microscope-based neuronavigation system displaying the integrated information from the two MEG spike foci and the DTI visual fiber tracking. Temporo-occipital neocortical tissue including the predominant lateral MEG and ECoG-spike focus (No. II) was resected, whereas the periventricular area with the heterotopia, the visual tract fibers and the second minor MEG focus (No. I) as well as temporo-mesial structures were spared. The extent of tissue resection was visualized by the intraoperative T1- and T2-weigthed MR imaging (Fig. 3C). The MRI control documented concordance of the resected with the planned volume avoiding the anatomical course of visual fiber tracts.
Neuropathological examination of the surgical specimen revealed a regular hexalaminar organization of the neocortex. There was no immunohistochemical evidence for dysplastic neurons or balloon cells. Astrogliosis was moderate and predominately affected white matter and deep cortical layers. Although ectopic neurons were detectable in white matter there number was not significantly increased (Hildebrandt et al.8).
The patient had an excellent postoperative outcome with respect to seizure control and preservation of her functional neurological integrity. Under lamotrigine monotherapy, three seizures occurred due to fever and one spontaneous seizure was seen in a stressful situation during the first 6 months postoperatively. The patient afterwards was seizure free with a total follow up period of 18 months. Postoperative perimetry documented unchanged visual field defects as compared to preoperative findings.
Discussion
Temporal resections in patients with periventricular nodular heterotopia and intractable focal seizures often yield poor results. In order to define the role of the heterotopic grey matter tissue in epileptogenesis for selective surgery invasive recordings were necessary up to now. The best predictor of surgical outcome according to the experiences of Aghakhani et al.,1 Scherer et al.9 and Tassi et al.10 is the detection of a focal epileptic generator usually demonstrated by intracerebral recordings. The epileptogenic network may be located in the nodular heterotopia or the neocortex as found in our case. The imaging findings showed here were similar to those described by Tassi et al.10 with, in addition, a radial band of grey matter presumably connecting the heterotopia and overlying lateral cortex. The heterotopia appear to be connected with distant cortical structures, which suggests that the ectopic tissue may also participate in the generation of epileptic discharges. Through abnormalities of cortical architecture, neuronal composition and connectivity, the cortex distant to the periventricular malformation may act as a primary epileptogenic substrate.1, 4, 5
Due to panhypopituitarism and the parents’ decision we were not able to carry out prolonged invasive recordings with a larger number of grids and depth electrodes in order to define more precisely the extent of the epileptogenic zone by recording seizures and interictal spike activity. Therefore, a multimodal approach using MSI guided intraoperative electrocorticography was performed. In addition, we are aware of the limitations of the MEG for deeper sources and the spatial resolution of one depth electrode used in the ECoG recording. This electrode in addition to the subdural electrodes was used for confirmation of the MEG localization of the deeper source and therefore confirmed the second less intensive focal area at the periventricular heterotopia. After identification of the two clusters, the further strategy aimed to remove only the dominant latero-basal focal epileptic activity in order to preserve visual functions, the mesial perilesional cluster could not be removed.
We conclude that an accurate definition of the epileptogenic network including the role of the heterotopia and overlying neocortex, is essential to avoid a poor surgical outcome. Though no invasive ictal onset could be recorded, the good surgical outcome obtained in this case, indicates that a crucial region of the epileptogenic zone was removed, either by the removal of the pacemaker or disconnection. The precise anatomical position of the recording electrodes was visualized by the intraoperative MR-imaging and made a spatial correlation with electrophysiological measurements possible.
Thus, the surgical decision-making was based on the integrated visualization of various functional imaging techniques (MRI, MEG, DTI, ECoG). The transfer of preoperative data into the intraoperative neuronavigational system is crucial for precise microsurgical tissue resection adjacent to functionally eloquent areas, e.g. the visual or the motor fiber tracts.3
Intraoperative high-field-MR-imaging is not only able to exactly localize the position of platinum-ECoG-electrodes but gives the opportunity to coregister pre- and intraoperatively collected functional imaging techniques (MSI, DTI).2, 6, 7 The postresection MR-images may lead to a modification of the surgical strategy as the extent of tissue removal may be extended during repeated surgical inspection.
Since this is a single observation, a larger number of patients have to be analyzed. Though intracerebral recordings seem to be traditionally the best method to localize focal epileptic activity, MSI and MSI guided intraoperative ECoG can also provide relevant information about interictal and ictal epileptic activity. The results obtained here open a window to less invasive approaches for the evaluation and surgical planning of epileptic patients with periventricular nodular heterotopia and more complex malformations.
References
- The role of periventricular nodular heterotopia in epileptogenesis. Brain. 2005;128:641–651
- Magnetic brain source imaging of focal epileptic activity: a synopsis of 455 cases. Brain. 2003;126:1–10
- Preoperative and intraoperative diffusion tensor imaging-based fiber tracking in glioma surgery. Neurosurgery. 2005;56:130–138
- Characterisation of nodular neuronal heterotopia in children. Brain. 1999;122:219–238
- . Cortical dysplasia: developmental effects. In: Williamson PD, Siegel AM, Roberts DW, Thadani VM, Gazzangins MS editor. Advances in neurology. vol. 84:Philadelphia: Lippioncott Williams & Wilkins; 2000;p. 497;Chapter 39
- Functional neuronavigation with magnetoencephalography: outcome in 50 patients with lesions around the motor cortex. J Neurosurg. 1999;91:73–79
- Intraoperative high-field-strength MR imaging: implementation and experience in 200 patients. Radiology. 2004;233:67–78
- Neuropathological spectrum of cortical dysplasia in children with severe focal epilepsies. Acta Neuropathol. 2005;110(1):1–11
- Intrinsic epiletogenicity of an isolated periventricular nodular heterotopia. Neurology. 2005;65:495–496
- Electroclinical, MRI and neuropathological study of patients with nodular heterotopia, with surgical outcomes. Brain. 2005;128:321–337
PII: S1059-1311(06)00195-6
doi:10.1016/j.seizure.2006.10.004
© 2006 British Epilepsy Association. Published by Elsevier Inc. All rights reserved.
