Diagnosis Efficacy Of FDG-PET/MRI Co-registration In MRI-negative Focal Epilepsy

Background and objective:18F-fluorodeoxyglucose-positron emission tomography(FDG-PET)is one of the important presurgical evaluation in pharmacoresistant focal epilepsy which can improve ratio of the epileptogenic zone detection.While because of its own limitation,the rate of epileptogenic zone detection and positioning accuracy of FDG-PET is not high.But through the use of 18F-fluorodeoxyglucose-positron emission tomography/magnetic resonance imaging coregistration(PM-C),the rate of epileptogenic zone detection and positioning accuracy will be greatly improved.This study aims to validate the diagnosis efficacy of PM-C for MRI-negative focal epilepsy and a guidance of high resolution MRI-reread and clinical strategy.Methods:This study prospective included the patients with pharmacoresistant focal epilepsy who underwent presurgical evaluation in our epliepsy monitoring center since September 1st,2012.The data inclusion was updated until August 31,2016.During admission,presurgical evaluation includes detailed clinical history and neurological examination,neuropsychological assessment,long-term video-EEG monitoring,high-resolution MRI(HR-MRI)and FDG-PET.Before admission,patients usually had performed conventional brain MRI scan(1.5 T or 3.0T)either in our hospital or outside hospitals.The level of HR-MRI abnormality was classified as "suspicious" or "normal".Using SPM8 to make the coregistration of FDG-PET images and HR-MRI image.Under the guidance of PM-C,diagnostic change of HR-MRI imaging re-read was classified as type Ⅰ(include Ⅰa and Ⅰb),type Ⅱ(Ⅱa and Ⅱb)and no change.In a weekly multidisciplinary patient management conference,the patients were recommended to performed resective surgery or an invasive EEG evaluation,or others(i.e.neuromodulation or medial adjustment)based on all obtained information including the result of PM-C.According the clinical data and imaging finds,we diveded the patients into temporal lobe epilepsy and extratemporal lobe epilepsy.Then analyse the imaging changing and decision making of these two groups sepetately.Results:It included 112 patients with 67 patients of extra-temporal and 40 patients of temporal lobe epilepsy.Patients with ETLE and TLE demonstrated distinct patterns of hypometabolism distribution on PET.The hypometablism of TLE patients was often extensive and 50%of them had remote hypometablism distributed in ipsilateral orbital frontal,ipsilateral insular,ipsilateral parsopercularis and contralateral anterior temporal lobe.The hypometabolic pattern of ETLE patients was usually focal.When FDG-PET was interpreted by nuclear imageologist alone,PET report was localizing in 20 patients(29.9%)in ETLE,and 23 patients(57.5%)with TLE.In contrast,PM-C providing further localizing information(Change Ⅰa,Ⅰb and Ⅰa)in 53 patients(79.1%)in ETLE,and 38 patients(95.0%)in TLE.Among the 15 patients of BOSD in our study,PM-C information converted 11 patients from MRI "negative" to MRI "positive"(Change Ⅰa),then change the clinical decision.Conclusion:Our study found that incorporating PM-C in the electroclinical information converted more than half of the patients MRI "negative" to MRI "positive",and subsequently change the decision making of invasive and resection surgical.Moreover,for the MRI strictly negative patients,PM-C provided additional localizing information.So that these bad controlled patients can get better treatment,and then control or reduce the impact of epilepsy.