期刊
WORLD NEUROSURGERY
卷 110, 期 -, 页码 E404-E426出版社
ELSEVIER SCIENCE INC
DOI: 10.1016/j.wneu.2017.11.012
关键词
Craniopharyngioma; Hypothalamus; MRI; Surgical risk; Third ventricle; Tumor adherence
BACKGROUND AND OBJECTIVE: Craniopharyngioma (CP) adherence represents a heterogeneous pathologic feature that critically influences the potentially safe and radical resection. The aim of this study was to define the magnetic resonance imaging (MRI) predictors of CP adherence severity. METHODS: This study retrospectively investigated a cohort of 200 surgically treated CPs with their corresponding preoperative conventional MRI scans. MRI findings related to the distortions of anatomic structures along the sella turcica-third ventricle axis caused by CPs, in addition to the tumor's shape and calcifications, were analyzed and correlated with the definitive type of CP adherence observed during the surgical procedures. RESULTS: CP adherence is defined by 3 components, as follows: 1) the specific structures attached to the tumor, 2) the adhesion's extent, and 3) its strength. Combination of these 3 components determines 5 hierarchical levels of adherence severity with gradually increasing surgical risk of hypothalamic injury. Multivariate analysis identified 4 radiologic variables that allowed a correct overall prediction of the levels of CP adherence severity in 81.5% of cases: 1) the position of the hypothalamus in relation to the tumor-the most discriminant factor; 2) the type of pituitary stalk distortion; 3) the tumor shape; and 4) the presence of calcifications. A binary logistic regression model including the first 3 radiologic variables correctly identified the CPs showing the highest level of adherence severity (severe/critical) in almost 90% of cases. CONCLUSIONS: A position of the hypothalamus around the middle portion of the tumor, an amputated or infiltrated appearance of the pituitary stalk, and the elliptical shape of the tumor are reliable predictors of strong and extensive CP adhesions to the hypothalamus.
作者
我是这篇论文的作者
点击您的名字以认领此论文并将其添加到您的个人资料中。
推荐
暂无数据