摘要
目的基于术前MR图像的多参数特征建立多参数MR列线图模型评价低级别胶质瘤(LGG)患者预后。方法回顾性分析2016—2019年间在秦皇岛市第一医院343例LGG患者临床资料,其中男190例,女153例。患者术前均接受多模态MRI检查,高年资医师进行胶质瘤影像特征提取。采用单因素Cox回归分析筛选与胶质瘤预后相关的MRI参数和临床变量,将相关参数特征和临床变量纳入多因素Cox回归分析,得到与胶质瘤预后密切相关的独立危险因素。根据独立危险因素,建立预测低级别胶质瘤预后的列线图。结果年龄、胶质瘤级别、病理类型、术后放疗、肿瘤位置和MRI评分是LGG发生的独立影响因素(均P<0.05)。基于以上变量和MRI评分成功建立列线图模型预测LGG预后的曲线下面积为0.8,明显大于WHO分级模型的0.64,(Z=-2.56,P<0.05)。3、5年生存校正曲线提示在观察值与预测值之间有良好的一致性。结论基于多参数MRI的列线图模型可直观全面地预测LGG患者生存概率,可为神经外科医师提供相对准确的预测工具,有利于临床个性化评估患者的生存及预后。
Objective To establish a nomogram based on multi-parameter MRI for predicting prognosis of low-grade gliomas(LGG)patients.Methods 343 patients with LGG from September 2016 to October 2019 were enrolled from the first hospital of Qinhuangdao retrospectively.There were 190 males and 153 females.All patients underwent multimodal MRI examination before operation,and radiological features extraction was performed by 2 senior physicians.Univariate Cox regression analysis was used to screen MRI parameters and clinical variables related to glioma prognosis.Relevant parameter characteristics and clinical variables were included in multivariate Cox regression analysis to obtain independent risk factors closely related to gliomas prognosis.Based on independent risk factors,a nomogram was established to predict the prognosis of low-grade gliomas.Results Age,grade of gliomas,pathological type,post-operative radiotherapy,tumor location and MRI score were independent factors influencing the prognosis of low-grade gliomas(all P<0.05).Based on the above variables and MRI scores,the area under the curve of LGG prognosis predicted by the nomogram model was 0.8,which was significantly higher than 0.64 of the WHO model(Z=-2.460,P<0.05).The 3-year and 5-year survival correction curves suggested a good consistency between the observed and predicted values.Conclusions Nomogram based on multiparameter MRI can predict the survival probability of patients with low-grade gliomas intuitively and comprehensively.It can provide a relatively accurate prediction tool for neurosurgeons to individualized assessment of survival and prognosis for patients.
作者
李文菲
鲍欣然
梁珊珊
温鑫
顾长聪
徐小明
赵月梅
韩芳
Li Wenfei;Bao Xinran;Liang Shanshan;Wen Xin;Gu Changcong;Xu Xiaoming;Zhao Yuemei;Han Fang(Department of Radiology,First Hospital of Qinhuangdao,Qinhuangdao 066000,China;Department of Neurology,First Hospital of Qinhuangdao,Qinhuangdao 066000,China;Liaoning Prouincial Key Laboratory of high throughput screening and targeted drug transformation of breast and digestive tumor,Shenyang 116001,China;Department of Radiology,First Affiliated Hospital of Xian Jiaotong University,Xian 710061,China;Department of Neurosurgery,Second Hospital of Qinhuangdao,Qinhuangdao 066600,China;Department of imaging,Affiliated Zhongshan Hospital Dalian University,Dalian I1600l,China)
出处
《中华转移性肿瘤杂志》
2021年第2期139-143,共5页
Chinese Journal of Metastatic Cancer
基金
辽宁省自然科学基金项目(2019-ZD-0310)
秦皇岛市级科技计划项目(201805A078)。
关键词
列线图
胶质瘤
磁共振成像
预后
Nomogram
Gliomas
Magnetic resonance image
Prognosis
