Abstracts

The Extent of Resection of Pre-operative 18F-FDG-PET Hypometabolism and Contralateral Temporal Lobe Hypometabolism Predict Seizure Control Following Anterior Temporal Lobe Resection.

Abstract number : 1.238
Submission category : 5. Neuro Imaging / 5B. Functional Imaging
Year : 2017
Submission ID : 336316
Source : www.aesnet.org
Presentation date : 12/2/2017 5:02:24 PM
Published date : Nov 20, 2017, 11:02 AM

Authors :
Varduhi Cahill, Departments of Medicine, Neurology and Neurosurgery, The Royal Melbourne Hospital, The University of Melbourne, Melbourne, Australia; Benjamin Sinclair, Departments of Medicine, Neurology and Neurosurgery, The Royal Melbourne Hospital, The

Rationale: A significant proportion of patients with mesial temporal lobe epilepsy (mTLE) continue to experience seizures following anterior temporal lobe resection (ATLR). Interictal [18F]-fluorodeoxyglucose (FDG)-PET is commonly employed in the evaluation of surgical candidates and may provide a metabolic marker of epileptogenic neural network dysfunction. The aim of this study was to investigate the relationship between the extent of resection of the FDG-PET hypometabolism and seizure outcomes in patients with non-lesional mTLE.  Methods: Eighty-two patients who underwent ATLR and who had ≥2 years of post-operative follow-up were studied. FDG-PET and MRI processing was performed using the SPM12 software.  The hypometabolic region in each patient was identified with reference to 20 healthy controls (p<0.005). The resected temporal lobe (TL) volume was calculated from the pre-and post-operative MRIs. The volume of pre-operative FDG-PET hypometabolism was calculated using the SPM software, as well as the proportion of the resected TL hypometabolism. The presence of contralateral TL hypometabolism was ascertained by visual inspection of the thresholded SPM images. Results: Patients with excellent seizure outcomes (Engel’s class I; 70.7%) had a significantly greater proportion of the total FDG-PET hypometabolism resected (p=0.01, odds ratio [OR]=0.91, 95% confidence interval [CI]: 0.86-0.97), independent to that of the MRI findings (p=0.02, OR=5.46, 95% CI: 1.26-23.7) as well as the volume of temporal lobe resected (p=0.16). The proportion of the temporal lobe hypometabolism included in the resection was significant on its own in predicting seizure outcomes (p=0.004, OR=0.96, 95% CI: 0.93-0.99). In addition, the presence of contralateral temporal lobe hypometabolism was an independent predictor of an unfavourable seizure outcome. There was a crossover interaction between the presence of contralateral temporal lobe hypometabolism and the proportion of extra-temporal hypometabolism identified preoperatively (p=0.003, OR=1.20, 95% CI: 1.07-1.36).  Conclusions: The amount of pre-operative FDG-PET hypometabolism included in the resection was an independent predictor of seizure outcome, with the proportion of the ipsilateral TL hypometabolism resected being a key factor. The presence of contralateral TL hypometabolism was an independent predictor of unfavourable outcome. Computational analysis of FDG-PET images is observer independent and its integration in multimodal presurgical evaluation would be beneficial for presurgical planning and prognostication for patients with drug resistant mTLE.  Funding: n/a
Neuroimaging