基于CT影像学特征及临床数据对涎腺多形性腺瘤复发风险预测的价值分析  

Establishment of a nomographic chart model based on CT imaging characteristics and clinical data in predicting the recurrence risk of pleomorphic adenoma of salivary gland

作　　者：路丽 柳成良 左超 于淑靖[1] 张力[1] Lu Li;Liu Chengliang;Zuo Chao;Yu Shujing;Zhang Li(Department of CT Diagnosis,Cangzhou Central Hospital,Cangzhou 061000,China;The Second Section of Ultrasound Department,Cangzhou Central Hospital,Cangzhou 061000,China;Department of Nuclear Medicine,Cangzhou Central Hospital,Cangzhou 061000,China)

机构地区：[1]沧州市中心医院CT诊断科,沧州061000 [2]沧州市中心医院超声二科,沧州061000 [3]沧州市中心医院核医学科,沧州061000

出　　处：《中国医学装备》2024年第5期54-58,63,共6页China Medical Equipment

基　　金：2022年度河北省医学科学研究课题项目(20220378)。

摘　　要：目的:探讨基于CT影像学特征及临床数据预测涎腺多形性腺瘤(PA)复发风险的列线图模型构建。方法:回顾性分析2019年11月至2022年11月沧州市中心医院收治的99例PA患者的临床及CT影像学资料,根据是否复发将其分为复发组(15例)和未复发组(84例)。采用单因素分析、多因素logistic回归分析PA复发的影响因素,并构建风险预测模型,采用受试者工作特征(ROC)曲线分析该模型的预测价值。结果:多因素logistic回归分析结果显示,肿瘤部位为深叶、包膜不完整及周围组织侵犯3个指标为PA复发的独立危险因素(OR=79.730、67.194、479.801,P<0.05)。基于以上3个指标构建的风险预测模型的ROC曲线下面积(AUC)为0.960,灵敏度为80.0%,特异度为97.6%,均高于模型中各单项指标,且该风险预测模型和实际观测值间无明显差异(x^(2)=0.823,P=0.935)。结论:肿瘤部位为深叶、包膜不完整及周围组织侵犯3个指标为PA复发的独立危险因素,基于以上3个指标构建的风险预测模型对PA复发具有较高预测价值,能为预防PA复发提供理论依据。Objective:To explore the establishment of a nomographic chart model based on computed tomography(CT)imaging characteristics and clinical data in predicting the recurrence risk of pleomorphic adenoma(PA)of salivary gland.Methods:Clinical and CT imaging data of 99 PA patients admitted to Cangzhou central hospital from November 2019 to November 2022 were retrospectively analyzed,and they were divided into recurrence group(15 cases)and non-recurrence group(84 cases)according to whether occurred recurrence.The influence factors of PA recurrence were analyzed by univariate analysis and multivariate logistic regression,and the risk prediction model was constructed.The predictive value of the model was analyzed by receiver operating characteristic(ROC)curve.Results:Multivariate Logistic regression analysis showed that the tumor site was at deep lobe,the capsule of tumor was incomplete and the surrounding tissue of tumor existed invasion were respectively independent risk factors for PA recurrence(OR=79.730,67.194,479.801,P<0.05).The area under curve(AUC)of ROC curve of the risk prediction model based on the above three indexes was 0.960,and the sensitivity and specificity were respectively 80.0%and 97.6%,which were higher than each single index in the model.There was no significant difference between this risk predictive model and actual observation value(x^(2)=0.823,P=0.935).Conclusion:The tumor site is deep lobe,the capsule of tumor is incomplete,and the surrounding tissue of tumor exists invasion are independent risk factors of PA recurrence.The risk prediction model based on the above three indexes has higher predictive value for PA recurrence,which can provide theoretical basis for preventing PA recurrence.

关 键 词：CT影像学特征 临床数据 涎腺多形性腺瘤(PA) 复发风险预测 列线图模型 

分 类 号：R816.98[医药卫生—放射医学]