Evaluation of Focal Cortical Dysplasia and Mixed Neuronal and Glial Tumors in Pediatric Epilepsy Patients Using (18)F-FDG and (11)C-Methionine PET

Abstract

Focal cortical dysplasia (FCD) and mixed neuronal and glial tumors share many clinical characteristics; therefore, the presurgical differential diagnosis of these diseases using MRI is difficult in some cases. The aim of this study was to determine whether (11)C-methionine PET, compared with (18)F-FDG PET, was useful for the evaluation of these diseases. Methods: The clinical and imaging data of 30 pediatric lesional epilepsy patients pathologically diagnosed with FCD, dysembryoplastic neuroepithelial tumor (DNT), or ganglioglioma were reviewed. Eleven patients had FCD, 8 patients had a DNT, and 11 patients had a ganglioglioma. (18)F-FDG and (11)C-methinine PET scans were obtained from 25 patients and 15 patients, respectively. Visual grading analysis and quantitative assessment of (18)F-FDG and (11)C-methionine PET, represented as a lesion-to-gray matter ratio (LGR), were performed. Results: In the visual grading analysis, both (18)F-FDG PET and (11)C-methionine PET detected a significant difference among the 3 disease groups (P < 0.033 and P < 0.016, respectively), but discrimination of FCD from mixed neuronal and glial tumors was possible only with (11)C-methionine PET. The mean LGR of (18)F-FDG PET was 0.502 +/- 0.119 for FCD, 0.631 +/- 0.107 for DNTs, and 0.620 +/- 0.196 for gangliogliomas; there was no significant difference in LGR among the groups (P = 0.111). However, the mean LGR of (11)C-methionine PET was 1.078 +/- 0.182 for FCD, 1.564 +/- 0.368 for DNT, and 2.114 +/- 0.723 for gangliogliomas; there was a significant difference in LGR among the groups (P < 0.014). Post hoc analysis revealed that the LGR of FCD was significantly different from that of DNTs and gangliogliomas. The mean LGR value of DNTs fell between those of FCD and gangliogliomas. Conclusion: Although (18)F-FDG plays a major role in the preoperative work-up of epilepsy surgery patients, it appears from this study that (18)F-FDG does not contribute to the differential diagnosis and that another tracer such as (11)C-methinine is required. (11)C-methinine PET results correlated well with the pathologic spectrum in pediatric lesional epilepsy patients.