Temporal Lobe Epilepsy Workup

Electroencephalography (EEG) and neuroimaging are basic workup of seizures and epilepsy. MRI is the neuroimaging modality of choice for patients with temporal lobe epilepsy. Other imaging modalities that can be used in the diagnosis of temporal lobe epilepsy include computed tomography (CT) scanning, positron emission tomography (PET) scanning, single-photon emission CT (SPECT) scanning, MR spectroscopy, and magnetoencephalography (MEG).

CT scanning of the head is often obtained in the emergency department, as it is ubiquitous and is adequate for assessing blood and large lesions, but the resolution is not that of MRI.

As mentioned, MRI is the neuroimaging modality of choice for patients with temporal lobe epilepsy. Most brain MRI scans do not include coronal images, but for temporal lobe epilepsy this sequence is more informative than are the axial and sagittal cuts.

Thin, coronal, oblique slices of 1.5–2 mm with no gap, using spoiled gradient recall images (SPGR) are recommended.

All patients with newly diagnosed temporal lobe epilepsy should have a high-resolution MRI scan with at least a 1.5-Tesla MRI, although the availability of a stronger magnet like 3-Tesla is increasing resolution.

High-resolution MRI shows hippocampal atrophy in many patients with temporal lobe epilepsy by visual analysis alone, and, although volumetric studies can be performed, they are labor intensive. Hippocampal atrophy is bilateral in 10–15% of cases. An increase in the T2-weighted signal intensity in the hippocampus may be seen on fluid-attenuated inversion recovery (FLAIR) MRI; this finding is also consistent with hippocampal sclerosis.

PET with 18-fluorodeoxyglucose (PET-FDG) is a useful tool for interictal seizure localization in surgical candidates when the MRI result is normal.

PET-FDG scans usually are performed as an adjunctive measure to delineate the epileptogenic zone.

Interictal deficits include reduced glucose metabolism in the medial and lateral temporal lobe. PET scans can be fused with either CT or MRI and are useful in the presurgical evaluation.

Ictal PET scan recordings are rare, and EEG should be obtained during PET scan to determine if the the study is incterictal or ictal.

SPECT scanning is also an adjunctive imaging modality useful only for surgical candidates; the accuracy of seizure localization is about 80–90%.

Ictal SPECT scans done with hexamethylpropyleneamine oxime (HMPAO) show hyperperfusion in the region of seizure onset. The characteristic pattern is hyperperfusion of the medial and lateral temporal lobe. This requires ictal injection within 30 seconds of seizure onset. The ictal SPECT scan subtracted from the interictal scan (SISCOM) can be very useful in the presurgical evaluation.

Interictal SPECT testing is less sensitive than are PET-FDG and ictal SPECT scanning and is not used routinely for localization of the epileptogenic zone.

Magnetic resonance spectroscopy (MRS) may be clinically useful in selected patients with possbile neoplstic process. It has great research applications.

Electroencephalography should be performed in all patients with suspected temporal lobe epilepsy.

Epileptiform discharges: Interictal abnormalities, consisting of spike/sharp and slow complexes, usually are located in the anterior temporal region (F7/F8 and T3/T4 electrodes) or basal temporal electrodes (most commonly T1/T2 and in research settings, T9/T10 and F9/F10). During video-EEG monitoring, sphenoidal electrodes can be useful.

One third of patients with temporal lobe epilepsy have bilaterally independent, temporal interictal epileptiform abnormalities.

Ictal recordings from patients with typical temporal lobe epilepsy usually exhibit 5-7 Hz, rhythmic, sharp theta activity, maximal in the sphenoidal and the basal temporal electrodes on the side of seizure origin.

In documented temporal lobe seizures, lateralized postictal slowing, when present, is a reliable lateralizing finding.

A patient with temporal lobe epilepsy can have a normal EEG. The yield of the EEG can be increased on a repeat study with prolonged recordings, and, in certain patients, activation with sleep deprivation can be useful.

Video-EEG telemetry is used as part of the presurgical evaluation. It also is used if the diagnosis of temporal lobe epilepsy is suspected but still in question and in patients suspected of having psychogenic seizures.

Intracranial EEG with placement of intracranial subdural electrodes is done only if the patient is a surgical candidate and MRI and other non-invasive EEG data are not sufficiently localizing.

Another complementary method to assess cerebral physiologic activity similar to EEG is magnetoencephalography (MEG), which measures the magnetic fields generated by the epileptic spikes. The main use of MEG is the co-registration with the MRI to give magnetic source imaging (MSI) in 3-dimensional space. The spikes that are analyzed for MSI are interictal spikes and this is not as informative as ictal EEG recordings in surgical evaluations.  

In the past, a prolactin level could be obtained after a prolonged temporal lobe seizure, or a seizure that spread to become bilateral tonic clonic, but there have been some issues of sensitivity and specifcity.