Immunohistochemical Study of Six Cases of Taylor's Type Focal Cortical Dysplasia: Correlation with Electroclinical Data

Authors


Address correspondence and reprint requests to Dr. E.A. Cavalheiro at Rua Botucatu 862, CEP 04023-900, São Paulo/SP, Brazil. E-mail: esper.nexp@epm.br

Abstract

Summary:  Purpose: Cortical specimens from six patients operated on for drug-resistant epilepsy diagnosed as Taylor's type focal cortical dysplasia were submitted to neuropathological and immunohistochemical studies.

Methods: All patients were submitted to presurgical investigations including clinical and neuropsychological evaluations, EEG/video telemetry of ictal and interictal events, magnetic resonance imaging, and ictal and interictal single-photon emission computed tomography (SPECT). Recordings from electrocorticography (ECoG) were obtained in four cases and from subdural electrode implantation in two. Postsurgical follow-up was assessed according to Engel's score. Immunohistochemistry (IHC) was processed for parvalbumin (PV), calbindin D28-K (CB), nonphosphorylated neurofilaments (SMI-311), glial fibrillary acidic protein (GFAP) in all cases.

Results: We found continuous/quasi-continuous spikes and sharp-wave patterns in three cases and frequent repetitive bursting of polyspikes and ECoG seizures in two cases. Every patient showed cortical dyslamination, abnormal and giant neurons, and balloon cells. GFAP immunoreactivity was found in astrocytes and some balloon cells that were less intensely stained. Nonphosphorylated neurofilaments SMI-311 immunoreactivity was found in normal and giant neurons and in some balloon cells, making visible thin neuropils. PV immunoreactivity was present in normal interneurons and in fibers in layers IV–V. PV-negative balloon cells were surrounded by abundant PV-positive fibers. CB immunoreactivity was found mostly in interneurons in layers II–III.

Conclusions: Our research is inconclusive. More cases should be investigated, and we must draw more accurate anatomic correlations between the ECoG recordings and surgical specimens studied with IHC.

Focal cortical dysplasias (FCDs) are estimated to be responsible for half of the drug-resistant epilepsies in the children and 20% in the adult population. Pathologically, FCDs include a wide spectrum of lesions including (a) a cortical dyslamination with heterotopic neurons in the white matter (WM; architectural dysplasia), (b) a cortical dyslamination with heterotopic neurons in the WM and dysmorphic neurons (mostly in layers II–III; cytoarchitectural dysplasia), and (c) a cortical dyslamination with heterotopic neurons in the WM, dysmorphic and giant neurons, and balloon cells (Taylor type 2) or not (Taylor type 1). The pathogenesis of FCDs involves an early disorder of cell proliferation/differentiation related to the presence of neuromegaly and balloon cells, but later developmental processes such as neuronal migration and cortical organization are involved as well (1). In FCD, electroencephalograms (EEGs) usually record interictal rhythmic epileptiform activity that could be related to more continuous epileptiform discharges with electrocorticography (ECoG). It is accepted that the best surgical strategy is to remove not only the visible lesion (magnetic resonance imaging; MRI), but also all the ECoG discharging area.

The aim of this work was to correlate ECoG patterns with immunohistochemical (IHC) findings in six patients with Taylor's type FCD who underwent limited corticectomy for drug-resistant epilepsy.

PATIENTS AND METHODS

Patients

All patients were submitted to presurgical investigations including clinical and neuropsychological evaluations, long-duration EEG/video telemetry, MRI, and ictal/interictal single-photon emission computed tomography (SPECT; ECD).

Clinical electrophysiology

ECoG was carried out immediately before surgery (four cases) or over the long term by using subdural electrodes (two cases). The analysis of ECoG activity took into account the presence of ictal/continuous epileptogenic discharges (I/CEDs) of three types: (a) repetitive ECoG seizures with recruiting/derecruiting; (b) repetitive bursting of rhythmic polyspikes at 10–20 Hz for 5–10 s; and (c) continuous or quasi-continuous pattern of rhythmic spikes or sharp waves at 1–8 Hz for>10 s (2).

Neuropathology and immunohistochemistry

Tissue specimens were fixed in 1% paraformaldehyde, transferred to saccharose 20%, and frozen (–50°C). The 40-μm cryostat sections were processed for Nissl stain and IHC. One set of antibodies was used to recognize glial and/or neuronal cytoskeletal components: glial fibrillary acidic protein (GFAP), vimentin, and SMI-311 for nonphosphorylated neurofilaments. The other one was directed against calcium-binding proteins parvalbumin (PV) and calbindin D-28K (CB), proteins specific for partially overlapping populations of inhibitory γ-aminobutyric acid (GABA)ergic interneurons.

RESULTS

EcoG showed continuous/quasi-continuous spikes and sharp-waves patterns in three cases and frequent repetitive bursting of polyspikes and EcoG seizures in two cases. In all cases reported here, we observed cortical dyslamination and the presence of giant and abnormal neurons and balloon cells. Two cases had fewer balloon cells and also less dyslamination in the latter. The findings of IHC did not show any difference from one case to another and were in accordance with already published data (3). GFAP immunoreactivity was found in astrocytes and some balloon cells that were consistently less intensely stained. Nonphosphorylated neurofilaments SMI-311 immunoreactivity was found in normal and giant neurons and in some balloon cells, making visible thin neuropils. PV immunoreactivity was present in normal interneurons and in fibers in layers IV–V. PV-negative balloon cells were surrounded by abundant PV-positive fibers. CB immunoreactivity was found mostly in interneurons in layers II–III (Fig. 1).

Figure 1.

Results of immunohistochemistry. A: Parvalbumin-positive interneuron (chandelier type). B: Parvalbumin-positive terminals to the balloon cells in the white matter (arrow). C: Calbindin-positive interneurons in layers II–III. D: Nonphosphorylated neurofilaments SMI-311–positive neurons in layers IV–V.

DISCUSSION

ECoG and deep electrode recordings have demonstrated that CDs may have specific epileptiform activities by themselves. In vitro electrophysiologic study of dysplastic cortex showed that, although membrane properties of individual neurons are not altered, the occurrence of network organization disturbances responsible for a stronger recruitment is related to hyperexcitability (4).

Rosenow et al. (5) demonstrated a positive correlation among the seizure frequency, numbers of spiking on ECoG, and the presence of balloon cells in 17 patients with FCD. Another finding observed in that study was a relatively good surgical outcome in the patients with balloon cells. In our study, two patients with apparently fewer balloon cells had ECoG isolated spikes and no continuous/quasi-continuous spikes and sharp waves. In the present study, three patients had a good surgical outcome but without a correlation with the number of balloon cells. Because our results are inconclusive, more cases should be investigated to draw an accurate anatomic correlation between the ECoG recording site and the tissue processed for morphology.

Ancillary