Porto Luciana, Schöning Stefan, Hattingen Elke, Sörensen Jan, Jurcoane Alina, Lehrnbecher Thomas
Neuroradiology Department, Johann Wolfgang Goethe University, Frankfurt/Main, Germany.
Pediatric Hematology and Oncology, Hospital for Children and Adolescents, Johann Wolfgang Goethe University, Frankfurt/Main, Germany.
Radiol Oncol. 2015 Aug 21;49(3):242-9. doi: 10.1515/raon-2015-0024. eCollection 2015 Sep.
The aim of our study was (1) to describe central nervous system (CNS) manifestations in children with Langerhans cell histiocytosis (LCH) based on images sent to a reference center and meeting minimum requirements and (2) to assess the inter-rater agreement of CNS-MRI results, which represents the overall reproducibility of this investigation.
We retrospectively reviewed brain MRI examinations in children with LCH, for which MRI minimum requirements were met. Abnormalities were rated by two experienced neuroradiologists, and the inter-rater agreement was assessed.
Out of a total of 94 imaging studies, only 31 MRIs met the minimum criteria, which included T2w, FLAIR, T1w images before/after contrast in at least two different section planes, and thin post contrast sagittal slices T1w through the sella. The most common changes were osseous abnormalities, followed by solid enlargement of the pineal gland, thickened enhancing stalk and signal changes of the dentate nucleus. Whereas inter-rater agreement in assessing most of the CNS lesions was relatively high (κ > 0.61), the application of minimum criteria often did not allow to evaluate the posterior pituitary.
The diversity of radiological protocols from different institutions leads to difficulties in the diagnosis of CNS abnormalities in children with LCH. Although the inter-rater agreement between neuroradiologists was high, not all the LCH manifestations could be completely ruled out when using the minimum criteria. Brain MRIs should therefore follow LCH guideline protocols and include T1 pre-gadolinium sagittal images, and be centrally reviewed in order to improve the comparison of clinical trials.
我们研究的目的是（1）基于发送至一个参考中心并符合最低要求的影像，描述朗格汉斯细胞组织细胞增多症（LCH）患儿的中枢神经系统（CNS）表现；（2）评估CNS-MRI结果的阅片者间一致性，这代表了本研究的总体可重复性。
我们回顾性分析了符合MRI最低要求的LCH患儿的脑部MRI检查。由两名经验丰富的神经放射科医生对异常情况进行评分，并评估阅片者间的一致性。
在总共94项影像学研究中，只有31例MRI符合最低标准，其中包括至少两个不同截面平面的T2加权、液体衰减反转恢复序列（FLAIR）、对比剂前后的T1加权图像，以及通过蝶鞍的对比剂后矢状位T1加权薄层图像。最常见的变化是骨质异常，其次是松果体实性增大、强化的垂体柄增粗和齿状核信号改变。虽然在评估大多数CNS病变时阅片者间一致性相对较高（κ>0.61），但应用最低标准往往无法评估垂体后叶。
不同机构放射学检查方案的多样性导致LCH患儿CNS异常的诊断存在困难。尽管神经放射科医生之间的阅片者间一致性较高，但使用最低标准时并非所有LCH表现都能完全排除。因此，脑部MRI应遵循LCH指南方案，包括钆喷酸葡胺前矢状位T1图像，并进行集中审查，以改善临床试验的比较。
