Department of Molecular and Clinical Pharmacology, Institute of Translational Medicine, University of Liverpool, UK; The Walton Centre NHS Foundation Trust, Liverpool, UK.
The Walton Centre NHS Foundation Trust, Liverpool, UK.
Clin Radiol. 2019 Jan;74(1):78.e1-78.e11. doi: 10.1016/j.crad.2018.08.013. Epub 2018 Sep 28.
To evaluate whether a dedicated epilepsy research protocol with expert image re-evaluation can increase identification of patients with lesions and to attempt to ascertain the potential reasons why lesions were not identified previously on earlier clinical magnetic resonance imaging (MRI).
Forty-three patients (26 female) with focal refractory epilepsy who had failed at least two trials of anti-epileptic drug treatments were studied. Patients were recruited prospectively into the study if previous clinical MRI was deemed to be "non-lesional" by the clinicians involved in the initial assessment. Three-dimensional (3D) T1-weighted (T1W), T2-weighted (T2W), T2 fluid-attenuated inversion recovery (T2-FLAIR) sequences, and two-dimensional (2D) coronal T1-/T2W FLAIR were assessed by a neuroradiologist, including the previous clinical MRI of individual patients.
Twenty-nine or 43 (67%) patients remained MRI-negative after scanning with the epilepsy-dedicated protocol and image reappraisal by expert consultant neuroradiologists; however, 14/43 (33%) patients were found to have potentially epileptogenic brain lesions. The lesion that most frequently escaped the attention of clinicians was hippocampal sclerosis (nine cases, of which two had an additional focal cortical dysplasia, FCD), followed by single FCDs (two cases), and others including gliosis, encephalocoele, and amygdala enlargement (one case each). Eleven of the 14 (79%) previously "non-lesional" patients had electroencephalogram (EEG) imaging-concordant localisation features, rendering them potential candidates for resective surgery.
The primary factors explaining the newly identified lesions were the choice of MRI sequences, imaging parameters, data quality, lesion not reported (human factor), and loss of information through incomplete documentation. It is important for all clinicians to proceed meticulously in the detailed assessment of epilepsy-dedicated in-vivo MRI and discuss difficult patient cases in multidisciplinary team meetings.
评估专门的癫痫研究方案和专家图像重新评估是否可以增加病变患者的检出率,并尝试确定先前临床磁共振成像(MRI)未能识别病变的潜在原因。
43 名(26 名女性)局灶性耐药性癫痫患者入组研究,这些患者至少经历了两种抗癫痫药物治疗的失败。如果参与初始评估的临床医生认为先前的临床 MRI“无病变”,则前瞻性招募患者入组。由神经放射科医生评估三维(3D)T1 加权(T1W)、T2 加权(T2W)、T2 液体衰减反转恢复(T2-FLAIR)序列和二维(2D)冠状 T1/T2W FLAIR,包括个体患者的先前临床 MRI。
29 名或 43 名(67%)患者在癫痫专用方案扫描和专家顾问神经放射科医生重新评估后仍为 MRI 阴性;然而,14/43(33%)名患者发现有潜在致痫性脑病变。最常被临床医生忽视的病变是海马硬化(9 例,其中 2 例还有局灶性皮质发育不良,FCD),其次是单发 FCD(2 例)和其他病变,包括胶质增生、脑膨出和杏仁核增大(各 1 例)。14 名先前“无病变”患者中的 11 名(79%)有脑电图(EEG)成像一致的局灶性特征,使其成为潜在的手术切除候选者。
新发现病变的主要解释因素是 MRI 序列的选择、成像参数、数据质量、未报告病变(人为因素)以及通过不完整的记录丢失信息。所有临床医生都必须在详细评估癫痫专用体内 MRI 时谨慎进行,并在多学科团队会议上讨论困难的患者病例。