MRI海马结构体积分割分析对颞叶癫痫的诊断价值及临床相关性研究

目的　利用MRI体积分割分析的方法研究海马萎缩的不同形式 ,探讨颞叶癫痫的病因和发病机制 ,提高颞叶癫痫诊断的敏感性。方法　采用GE 1 5TSignaHorizonLX超导型磁共振机 ,行垂直于海马长轴的倾斜冠状面T1WI。将颞叶内侧结构分为杏仁核和海马头、体、尾部 4段 ,采用体积测量、逐层比较的方法对 5 0例病理诊断为海马硬化的颞叶癫痫患者行颞叶内侧结构体积分割分析 ,并比较各组间的临床特征及手术预后的差异。结果　 5 0例海马硬化患者中 ,海马弥漫萎缩者 2 2例(44 % ) ,同时伴杏仁核萎缩者 5例 (10 % )。海马局限性萎缩者 2 0例 (40 % ) ,无萎缩者 8例 (16 % )。 38例 (76 % )海马硬化萎缩累及海马体部 ,其次累及海马头部者 2 9例 (5 8% ) ,累及海马尾部者 2 4例(48% ) ,累及杏仁核者 8例 (16 % )。 10例患者MRI体积分割分析提示海马局限性萎缩而海马绝对体积正常。海马不同萎缩形式各组间的癫痫病程长短、发作频率及手术预后的差异具有显著性意义 (P<0 0 5 ) ;各组间的有无高热惊厥史、起病年龄及有无大发作史的差异不具有显著性意义 (P >0 0 5 )。结论　颞叶内侧结构体积分割分析能提高颞叶癫痫诊断的敏感性 ,对探讨颞叶癫痫的病因和发病机制提供帮助。

MRI segmentation in the diagnosis and clinical correlations of temporal lobe epilepsy

Objective To study the different patterns of hippocampal atrophy by MRI segmental analysis and to investigate the etiology and pathogenesis of temporal lobe epilepsy Methods GE 1 5 T Signa Horizon LX MRI scanner was used Oblique coronal T 1 weighted images perpendicular to the long axis of the hippocampus were obtained The mesial temporal structures were divided into four parts: the amygdala, hippocampal head, body and tail MRI patterns of atrophy in 50 patients with histologically confirmed hippocampal sclerosis were investigated by MRI volumetric measurement and segmental analysis, and the differences of clinical features and surgical outcome in different groups were compared Results Diffuse hippocampal atrophy was found in 22 of 50 patients (44%), 5 of the 50 patients (10%) showed diffuse atrophy involving both the amygdala and hippocampus 20 of the 50 patients (40%) had hippocampal focal atrophy and 8 of 50 patients (16%) had no obvious atrophy 38 of 50 (76%) hippocampal sclerosis had atrophy in the hippocampal body, 29 of 50 (58%) had hippocampal head atrophy, 24 of 50 (48%) had hippocampal tail atrophy, and the least involved part was the amygdala (16%, 8/50) 10 patients who had normal hippocampal volume showed focal hippocampal atrophy by segmental analysis Various patterns of hippocampal atrophy were found to be statistically related to the duration of epilepsy, the frequency of seizure and the outcome of surgery, respectively ( P <0 05), but they were not related to the history of febrile convulsions, age of onset and the history of secondary generalized tonic clonic seizures ( P >0 05) Conclusion MRI segmental analysis can improve the diagnostic sensitivity of temporal lobe epilepsy and help to investigate its etiology and pathogenesis