TY - JOUR T1 - Localized Quantitative Analysis of Positron Emission Tomography (PET) for Temporal Lobe Epilepsy Lateralization and Surgical Intervention JF - Journal of Nuclear Medicine JO - J Nucl Med SP - 1927 LP - 1927 VL - 57 IS - supplement 2 AU - Hassan Bagher-Ebadian AU - Jason Schwalb AU - Fariborz Mahmoudi AU - Ellen Air AU - Saeed Shokri AU - Mohammad Nazem-Zadeh AU - Marianna Spanaki-Varelas AU - Vibhangini Wasade AU - Hamid Soltanian-Zadeh Y1 - 2016/05/01 UR - http://jnm.snmjournals.org/content/57/supplement_2/1927.abstract N2 - 1927Objectives We hypothesized that localized quantitative analysis of hypometabolism level in hippocampi using interictal PET scans, in combination with qualitative and quantitative analyses of MRI, can increase the confidence in lateralization and selection of patients for surgical intervention.Methods In this study, 11 patients with refractory epilepsy (ages 18 to 58 years, mean age of 40 years, 8 females, 3 males) were studied. All patients had symptomatic temporal lobe epilepsy and underwent MRI, EEG, and PET/CT scans. MR Imaging: All studies were performed using a 3T GE Excite HD MR system (GE Healthcare, Waukesha, WI) using a standard eight-channel phased-array RF coil and receiver. All patients underwent coronal T2-Fluid attenuated inversion recovery (T2-FLAIR) and T1-weighted (T1WI) MR studies using a spoiled gradient-echo (SPGR) sequence with adequate spatial resolutions appropriate for hippocampal T1 volumetry and FLAIR intensity analysis. PET Scan: All patients were asked to fast for 5 to 6 hours on the day of their scan. The blood glucose level of the patient was checked before the 18F-FDG administration. For each patient, 18F-FDG (8 ~ 12 mCi) was injected via an IV. All patient were scanned using a CT/PET (GE) scanner with a wait time of 60 min. For each patient, using FSL software, FLAIR and PET modalities were skull stripped (BET) and co-registered (rigid-FLIRT) to the skull-stripped T1WI as the reference. The regions of interest (ROIs) enclosing the hippocampi were delineated manually using coronal T1WI. A single investigator (AH) outlined all coronal hippocampal contours using an in-house software. Then, for each subject, the 3D Standard Uptake Value (SUV) was calculated (Figures 1 and 2, note that increased hypomethabolism is red in figure-2) from the co-registered PET modality for the entire brain and for the hippocampi using the hippocampi's ROIs. From Anterior to Posterior, SUV profiles for the left and right hippocampi were calculated and the two profiles were spatially normalized to each other (Figure 3). Then, the normalized profiles (L and R) along with a test of significance (Welch test - unpaired and unequal variances - see Figure 4) were used to estimate the hypometabolism probability profile (HPP) from each individual slice of the hippocampi at the confidence level of 95%.Results As shown in Figure 5, the slice based Welch test along with the proposed sampling method (Figures 3-4) generated a standard HPP profile consisting of three possible conditions for the hippocampi: left hypometabolism area (green), right hypometabolism area (blue), and the area of uncertainty (red - no judgment can be made). In addition to the HPP method, MRI data of the subjects were processed using quantitative lateralization techniques of hippocampal volumetry and T2-FLAIR intensity analysis. The HPP and the scatter plots generated by the MRI methods were used for lateralization and decision making for ECoG. The patient’s medical history, laboratory studies, long-term video-electro-encephalographic (EEG) data, and Neuropsychological tests such as Wada were also used in the decision making process. Table 1 shows that, compared to MRI lateralization techniques (27% and 36% for FLAIR intensity and T1 volume, respectively), the judgments of the HPP technique are closer (64%) to the decisions made by the clinicians for the ECoG monitoring study.Conclusions Despite the facts that the final decision made for ECoG may not be the optimal decision and also the population size in this study was small and skewed towards ‘right’, the results imply that the HPP method could be beneficial for the TLE patients to make a decision regarding the ECoG study and to do presurgical evaluations. We believe that an integrated decision making system that benefits from HPP as well as MRI measures would work best for the decision making process. ER -