Statistical parametric mapping of interictal 123I-iomazenil SPECT in temporal lobe epilepsy surgery

Abstract

Brain single photon emission computed tomography (SPECT) for epilepsy is divided into two types (using three radionuclide tracers)-perfusion SPECT (123I-IMP or 99 mTc-ECD), identifying epileptogenic foci by detecting abnormality in regional cerebral blood flow, and 123I-iomazenil SPECT, identifying epileptogenic foci based on distribution of central benzodiazepine receptors. This study aimed to statistically evaluate and compare the SPECT effectiveness for the three tracers. Statistical parametric mapping (SPM) analysis was performed on 30 mesial temporal lobe epilepsy (mTLE) patients. The radionuclide and patient data were categorized as follows: abnormality in the medial temporal lobe on the operated hemisphere (AAA), in the entire temporal lobe on the operated hemisphere (AA), in the dominantly affected temporal lobe on the operated hemisphere (A), in bilateral temporal lobes (B), with no abnormalities in bilateral temporal lobes (C), and with abnormality in the temporal lobe on the nonoperated hemisphere (D). For analyses of (AAA), (AA), and (A), examining the hemisphere containing epileptogenic foci, IMP-SPECT was significantly superior to ECD-SPECT (P<0.05). For (AAA), indicating localization, IMZ-SPECT was significantly superior to the other two (P<0.05). IMP-SPECT was superior for lateralizing and IMZ-SPECT was useful for localizing epileptogenic foci in mTLE patients though the applicability of the results in extratemporal lobe epilepsy is unknown.