MRI volumetric measurement of amygdala and hippocampus in temporal lobe epilepsy

Bilateral hippocampal volume loss in patients with a history of encephalitis or meningitis

Volumetric analysis of high-quality magnetic resonance imaging (MRI) scans identifies asymmetric hippocampal atrophy in most patients with temporal lobe epilepsy. However, bilateral hippocampal atrophy can be missed by unnormalized volume measures. We considered two patient groups with temporal lobe epilepsy, one with a history of febrile convulsions (FC, n = 14) and one with a history of encephalitis or meningitis (E/M, N = 12), to compare the prevalence of bilateral volume loss between the groups. A volume normalization process defines a normal range of hippocampal volumes in control subjects (n = 32). Normalized volumes indicated that 11 of 14 subjects with a history of FC had unilateral hippocampal atrophy and 9 of 12 subjects with a history of E/M had bilateral hippocampal atrophy as compared with the controls. Visual assessments of unilateral hippocampal atrophy (n = 17) correlated well with measured unilateral volume loss (n = 14 ), but visual assessment of bilateral hippocampal atrophy (n = 3) correlated poorly with measured bilateral volume loss (n = 12). Mean age at seizure onset was lower in the FC group (7 years) than in the E/M group (13 years), but other clinical features were similar between the two groups. Hippocampal volume normalization is necessary to detect bilateral volume loss, which is common in patients with a history of encephalitis or meningitis.

Segmentation and feature extraction techniques, with applications to MRI head studies

To obtain a three-dimensional reconstruction of the hippocampus from a volumetric MRI head study, it is necessary to separate that structure not only from the surrounding white matter, but also from contiguous areas of gray matter--the amygdala and cerebral cortex. At present it is necessary for a physician to manually segment the hippocampus on each slice of the volume to obtain such a reconstruction. This process is time consuming, and is subject to inter- and intraoperator variation as well as large discontinuities between slices. We propose a novel technique, making use of a combination of gray scale and edge-detection algorithms and some a priori knowledge, by which a computer may make an unsupervised identification of a given structure through a series of contiguous images. This technique is applicable even if the structure includes so-called false contours or missing contours. Applications include three-dimensional reconstruction of difficult-to-segment regions of the brain, and volumetric measurements of structures from series of two-dimensional images.

Focal intermittent delta activity in patients with mesiotemporal atrophy: a reliable marker of the epileptogenic focus

We attempted to determine the significance of background abnormalities and their relation with spikes and location of atrophy in 56 patients with temporal lobe epilepsy (TLE) and mesiotemporal atrophy (MTA) assessed by volumetric magnetic resonance imaging (MRI): 35 patients had unilateral (group I) and 21 had bilateral atrophy with lateralized predominance (group II). Trains of delta waves over temporal regions were observed in > or = 90% of patients. They lateralized with accuracy equal to that of the spikes to the site of atrophy (delta 92% or 29 of 32 patients in group I and 63% or 12 of 19 of group II; spikes 85% or 28 of 33 of group I and 65% or 13 of 20 of group II). Delta waves and spikes occurred together in > 85% of cases. With respect to their location, a striking concordance was observed: delta activity almost always occurred ipsilateral with unilateral spiking (90% or 19 of 21 of patients with unilateral and 88% or 7 of 8 of patients with bilateral atrophy) and bilaterally independently with bilateral spiking (67% or 6 of 9 of patients in group I and 100% or 10 of 10 in group II). Furthermore, spiking and delta activity were never in disagreement with respect to lateralization. In TLE related to MTA, delta transients are a reliable indicator of the epileptogenic focus and presumably reflect the epileptogenic process rather than the underlying structural pathology.

MRI-based hippocampal volumetrics: data acquisition, normal ranges, and optimal protocol

The process of producing magnetic resonance (MR) volume measurements can be divided into considerations of acquisition and postprocessing of the MR data. With careful attention to both of these, precise and reproducible measurements can be achieved. A statistical description of hippocampal measurements in normal volunteers must be available for comparison if volumetrics are employed either for clinical or research purposes. A wide range in "normal" hippocampal volume is present in the studies of normal young adults that have been reported to date. This variability is most probably due to interinstitutional differences in hippocampal boundary criteria, and in the software employed for counting pixels in a defined region of interest (ROI). Because the numeric output from the volume measurement procedure is highly technique-dependent, the statistical description of "normal" should be determined or calibrated at each institution wishing to use these techniques.

Proton magnetic resonance spectroscopic images and MRI volumetric studies for lateralization of temporal lobe epilepsy

We obtained 2D magnetic resonance (MR) spectroscopic images (MRSI) and MRI volumetric measurements (MRIV) of amygdala and hippocampus in 30 consecutive patients with temporal lobe epilepsy (TLE) being evaluated for surgical treatment. Both MRSI and MRIV lateralization showed good agreement with the current gold standard of clinical-EEG lateralization. Each exam separately correctly lateralized 25 out of 30 patients with no false lateralization. Combining both exams, lateralization could be achieved in 28 out of 30 patients. The two patients with no significant asymmetry had bitemporal EEG abnormalities, and bilateral damage on both MRIV and MRSI. There was a good correlation between the magnitude of the MRSI and MRIV asymmetry (Pearson coefficient = 0.83; p < .0001). Both MRSI and MRIV were normal in our patients with seizures originating outside the temporal lobes. Both MRIV and MRSI can lateralize TLE in 83% of patients. Combination of the two modalities allows lateralization in 93% of patients. Patients who cannot be lateralized generally have symmetrical bitemporal abnormalities; they are not incorrectly lateralized. The structural and chemical pathologic abnormalities seen in TLE seem to be associated with the seizure focus, and may be as, or even more, reliable than a few recorded seizures in predicting the side from which most seizures originate.

FDG-PET and volumetric MRI in the evaluation of patients with partial epilepsy

We performed interictal FDG-PET- and MRI-based hippocampal volumetric measurements on 18 adult patients with complex partial epilepsy of temporal lobe origin in whom we had identified their ictal focus by video-telemetry EEG. Sixteen patients (89%) had regional hypometabolism, 11 (61%) had focal 1.5-tesla T2-weighted MRI (two structural abnormalities, nine hippocampal formation [HF] increased T2 signal), and nine (50%) had absolute HF atrophy ipsilateral to the temporal ictal focus. Ten (55%) had abnormal L/R HF ratios, nine ipsilateral to the EEG focus. All patients with abnormal MRI volumetric studies had focal PET abnormalities. Only seven had both abnormal HF volume ratios and T2 MRI (all increased HF T2 signal). There was a significant correlation between hippocampal volume and inferior mesial and lateral temporal lobe cerebral metabolic rate of glucose asymmetry index (p < 0.01), suggesting that hypometabolism may reflect hippocampal atrophy. PET is more sensitive than MRI volumetry in identifying the ictal focus but does not provide additional information when HF atrophy is present.

MRI-based hippocampal volume measurements in epilepsy

In the study of brain morphometry, it is accepted that a relationship exists between brain structure and function, both normal and abnormal. One descriptor of morphometric structure is volume. Abnormalities in hippocampal morphology, including unilateral or bilateral volume loss, are known to occur in epilepsy, Alzheimer's disease, and in certain amnestic syndromes. Precise quantitation should improve understanding of the role of any biologic system in normal function and in disease. The objectives of magnetic resonance (MR)-based hippocampal volume measurements are precise quantitation, identification of a normal range, and identification of the association between biologic variables and aberrations in this volumetric parameter. Volumetric measures introduce a level of precision in the estimation of hippocampal size that is not available simply by visually inspecting a set of MR images, thus enabling statistically based hypothesis testing. To produce accurate hippocampal volume measurements with magnetic resonance imaging (MRI), attention must be directed to the two major components of the operation as a whole, MR image acquisition and image processing.

Lateralization of temporal lobe epilepsy based on regional metabolic abnormalities in proton magnetic resonance spectroscopic images

Magnetic resonance spectroscopic imaging (MRSI) is capable of determining the spatial distribution in vivo of cerebral metabolites, including N-acetylaspartate (NAA), a compound found only in neurons. We used this technique in 10 patients with temporal lobe epilepsy (TLE) to determine the location of maximal neuronal/axonal loss or damage and to evaluate the potential of MRSI for presurgical lateralization. Asymmetry of the relative resonance intensity of NAA to creatine was determined for mid and posterior regions of the temporal lobes defined anatomically and also for "metabolic lesions" defined as the regions of maximal abnormality on MRSI. MRSI revealed decreased relative signal intensity in at least one temporal lobe of all patients. Two patients had a widespread reduction in NAA in both temporal lobes. The region of maximal abnormality was usually in the posterior temporal lobe but sometimes in the mid temporal lobe. The side of lowest NAA was ipsilateral to the clinical electroencephalographic lateralization in all patients. Lateralization based on NAA to creatine correlated with the atrophy of amygdala and hippocampus in 8 patients who showed this on magnetic resonance imaging volumetric measurements. MRSI can demonstrate regional neuronal loss or damage that correlates with clinical electroencephalographic and structural lateralization in temporal lobe epilepsy. The ability to identify a region of maximal metabolic abnormality on spectroscopic images may confer greater sensitivity than that available from single voxel methods. The maximal metabolic abnormality may not be located in a voxel defined a priori, and based on anatomical considerations, without knowledge of the distribution of the metabolic abnormality.

Atrophy of mesial structures in patients with temporal lobe epilepsy: cause or consequence of repeated seizures?

We studied 70 epileptic patients by using magnetic resonance imaging volumetric measurements of amygdala (AM) and hippocampal formation (HF). Fifty patients presented with intractable temporal lobe epilepsy (TLE), 10 patients had focal extratemporal lobe epilepsy, and 10 had generalized epilepsy. In 91% of the 45 TLE patients without foreign tissue lesions, there was significant smallness of the AM and/or HF coinciding with the side of electroencephalographic seizure onset. No significant smallness or asymmetry was demonstrated in patients with focal extratemporal or generalized epilepsy. We performed a linear regression analysis, plotting the number of years of recurrent seizures and the estimated seizure frequency against the volumes of the AM and HF. There was no correlation between either of these two parameters and AM or HF volume (p > 0.9). There was also no correlation between the patient's age and volumetric measurements of AM or HF, nor did these measurements correlate with the occurrence of generalized seizures. On the other hand, patients with antecedent prolonged febrile convulsions in early childhood had significantly smaller AM and HF, compared with those without such a history (p < 0.001). The findings indicate that repeated seizures or longer duration of epilepsy do not cause increased atrophy of AM or HF that is measurable by volumetric magnetic resonance imaging.

Increased neocortical spiking and surgical outcome after selective amygdalo-hippocampectomy

We studied the electrocorticogram (ECoG) before and immediately after transcortical selective amygdalo-hippocampectomy, prospectively in 13 consecutive patients and retrospectively in three others. ECoG was performed with surface and two depth electrodes inserted through T2 aimed at the amygdala and anterior hippocampus. Before resection the ECoG showed a variable amount of interictal spiking, recorded either independently from the depth and surface, or synchronously. A small cortical incision (2-3 cm) was made in T2. The hippocampus, amygdala and parahippocampal gyrus were removed subpially. After the resection, increased epileptiform abnormality was observed in all 16 patients and a different ECoG pattern emerged. It consisted of repetitive, high amplitude spikes and polyspikes, separated by attenuated background, recorded from the most anterior temporal area. Similar observations were reported by Niemeyer in 1958. The outcome was comparable to that of standard anterior temporal resection: 62.5% class I and 25% class II (Engel's scale). ECoG is often used to tailor the amount of resection, and the persistence of epileptic abnormalities correlates with worse outcome. This is not the case in selective amygdalo-hippocampectomy, suggesting that a different underlying mechanism is responsible for the increased interictal spiking following this procedure.