Department of Biomedical, Metabolic and Neural Sciences, University of Modena and Reggio Emilia, Modena, Italy.
Neurology Unit, OCB Hospital, AOU Modena, Modena, Italy.
Epileptic Disord. 2023 Feb;25(1):45-56. doi: 10.1002/epd2.20050. Epub 2023 Apr 17.
To evaluate in a real clinical scenario the impact of the ILAE-recommended "Harmonized neuroimaging of epilepsy structural sequences"- HARNESS protocol in patients affected by focal epilepsy.
We prospectively enrolled focal epilepsy patients who underwent a structural brain MRI between 2020 and 2021 at Modena University Hospital. For all patients, MRIs were: (a) acquired according to the HARNESS-MRI protocol (H-MRI); (b) reviewed by the same neuroradiology team. MRI outcomes measures were: the number of positive (diagnostic) and negative MRI; the type of radiological diagnosis classified in: (1) Hippocampal Sclerosis; (2) Malformations of cortical development (MCD); (3) Vascular malformations; (4) Glial scars; (5) Low-grade epilepsy-associated tumors; (6) Dual pathology. For each patient we verified for previous MRI (without HARNESS protocol, noH-MRI) and the presence of clinical information in the MRI request form. Then the measured outcomes were reviewed and compared as appropriate.
A total of 131 patients with H-MRI were included in the study. 100 patients out from this cohort had at least one previous noH-MRI scan. Of those, 92/100 were acquired at the same Hospital than H-MRI and 71/92 on a 3T scanner. The HARNESS protocol revealed 81 (62%) positive and 50 (38%) negative MRI, and MCD was the most common diagnosis (60%). Among the entire pool of 100 noH-MRI, 36 resulted positive with a significant difference (p < .001) compared to H-MRI. Similar findings were observed when accounting for the expert radiologists (H-MRI = 57 positive; noH-MRI = 33, p < .001) and the scanner field strength (H-MRI 43 = positive, noH-MRI = 23, p < .001), while clinical information were more present in H-MRI (p < .002).
The adoption of a standardized and optimized MRI acquisition protocol together with adequate clinical information contribute to identify a higher number of potentially epileptogenic lesions (especially FCD) thus impacting concretely on the clinical management of patients with focal epilepsy.
在真实临床环境中评估 ILAE 推荐的“癫痫结构性序列的协调神经影像学”-HARNESS 方案在局灶性癫痫患者中的影响。
我们前瞻性招募了 2020 年至 2021 年在摩德纳大学医院接受脑部结构性磁共振成像 (MRI) 的局灶性癫痫患者。对所有患者，MRI 按以下方式进行：(a) 根据 HARNESS-MRI 方案 (H-MRI) 采集；(b) 由同一位神经放射科团队进行审查。MRI 结果测量包括：阳性（诊断）和阴性 MRI 的数量；影像学诊断类型分为：(1) 海马硬化；(2) 皮质发育畸形 (MCD)；(3) 血管畸形；(4) 胶质瘢痕；(5) 低级别癫痫相关肿瘤；(6) 双重病变。对于每位患者，我们验证了之前的 MRI（无 HARNESS 方案，noH-MRI）和 MRI 请求表中的临床信息。然后对测量结果进行了审查和比较。
共纳入了 131 例 H-MRI 患者。从该队列中，有 100 例患者至少有一次之前的 noH-MRI 扫描。其中，92/100 例是在与 H-MRI 相同的医院采集的，71/92 例是在 3T 扫描仪上采集的。HARNESS 方案显示 81 例 (62%) MRI 阳性和 50 例 (38%) MRI 阴性，MCD 是最常见的诊断 (60%)。在整个 100 例 noH-MRI 中，有 36 例结果为阳性，差异具有统计学意义 (p < .001)。当考虑到专家放射科医生 (H-MRI 阳性=57 例，noH-MRI 阳性=33 例，p < .001) 和扫描仪场强 (H-MRI 阳性=43 例，noH-MRI 阳性=23 例，p < .001) 时，也观察到了类似的发现，而临床信息在 H-MRI 中更为常见 (p < .002)。
采用标准化和优化的 MRI 采集方案并结合适当的临床信息有助于识别出更多潜在的致痫病变（特别是 FCD），从而对局灶性癫痫患者的临床管理产生实际影响。
