Csaba Juhász
PET Center, Children's Hospital of Michigan, Wayne State University School of Medicine, Detroit, MI, USA, 2002.
Ideggyogy Sz. 2003 Jul 20;56(7-8):249-54.
The success of cortical resection for intractable epilepsy of neocortical origin is highly dependent on the accurate presurgical delineation of the regions responsible for generating seizures. In addition to EEG and structural imaging studies, functional neuroimaging such as positron emission tomography (PET) can assist lateralization and localization of epileptogenic cortical areas. In the presented studies, objectively delineated focal PET abnormalities have been analyzed in patients (mostly children) with intractable epilepsy, using two different tracers: 2-deoxy-2-[18F]fluoro-D-glucose (FDG), that measures regional brain glucose metabolism, and [11C]flumazenil (FMZ), that binds to GABAA receptors. The PET abnormalities were correlated with scalp and intracranial EEG findings, structural brain abnormalities, as well as surgical outcome data. In patients with extratemporal foci and no lesion on MRI, FMZ PET was more sensitive than FDG PET for identification of the seizure onset zone defined by intracranial EEG monitoring. In contrast, seizures commonly originated from the border of hypometabolic cortex detected by FDG PET suggesting that such areas are most likely epileptogenic, and should be addressed if subdural EEG is applied to delineate epileptic cortex. In patients with cortical lesions, perilesional cortex with decreased FMZ binding was significantly smaller than corresponding areas of glucose hypometabolism, and correlated well with spiking cortex. Extent of perilesional hypometabolism, on the other hand, showed a correlation with the life-time number of seizures suggesting a seizure-related progression of brain dysfunction. FMZ PET proved to be also very sensitive for detection of dual pathology (coexistence of an epileptogenic cortical lesion and hippocampal sclerosis). This has a major clinical importance since resection of both the cortical lesion and the atrophic hippocampus is required to achieve optimal surgical results. Finally, the author demonstrated that in patients with neocortical epilepsy, FDG PET abnormalities correctly regionalize the epileptogenic area, but their size is not related to the extent of epileptogenic tissue to be removed. In contrast, complete resection of cortex with decreased FMZ binding predicts good surgical outcome suggesting that application of FMZ PET can improve surgical results in selected patients with intractable epilepsy of neocortical origin.
新皮质起源的顽固性癫痫皮质切除术的成功高度依赖于对引发癫痫发作区域的准确术前划定。除了脑电图（EEG）和结构成像研究外，正电子发射断层扫描（PET）等功能神经成像可辅助癫痫皮质区域的侧别和定位。在本研究中，使用两种不同的示踪剂，对顽固性癫痫患者（大多为儿童）客观划定的局灶性PET异常进行了分析：测量局部脑葡萄糖代谢的2-脱氧-2-[18F]氟-D-葡萄糖（FDG），以及与GABAA受体结合的[11C]氟马西尼（FMZ）。PET异常与头皮和颅内EEG结果、脑结构异常以及手术结果数据相关。在颞外病灶且MRI无病变的患者中，FMZ PET在识别由颅内EEG监测定义的癫痫发作起始区方面比FDG PET更敏感。相反，癫痫发作通常起源于FDG PET检测到的低代谢皮质边界，这表明这些区域最有可能是致痫性的，如果应用硬膜下EEG来划定癫痫皮质，则应处理这些区域。在有皮质病变的患者中，FMZ结合减少的病变周围皮质明显小于相应的葡萄糖低代谢区域，且与棘波皮质密切相关。另一方面，病变周围低代谢的范围与癫痫发作的终生次数相关，提示脑功能障碍与癫痫发作相关的进展。FMZ PET在检测双重病理（致痫性皮质病变与海马硬化并存）方面也非常敏感。这具有重要的临床意义，因为需要切除皮质病变和萎缩的海马体才能获得最佳手术效果。最后，作者证明，在新皮质癫痫患者中，FDG PET异常可正确定位致痫区域，但其大小与待切除的致痫组织范围无关。相反，FMZ结合减少的皮质完全切除可预测良好的手术结果，这表明应用FMZ PET可改善部分新皮质起源的顽固性癫痫患者的手术效果。
