摘要
目的构建基于术前临床及超声特征的预测模型,评估术中冷冻切片病理学提示“滤泡性肿瘤但良恶性不确定”患者的甲状腺滤泡癌风险。方法回顾性分析2019年1月至2024年12月在宁波市医疗中心李惠利医院接受甲状腺结节切除手术且术中冷冻病理学检查提示“滤泡性肿瘤但良恶性不确定”693例患者人口学特征、实验室检测结果及超声声像学特征,采用多因素logistic回归分析筛选预测甲状腺滤泡状癌的独立危险因素。数据集按7∶3分为训练集和验证集,在训练集中采用逐步回归法构建预测模型,并通过ROC曲线、校准曲线及决策曲线分析评估模型预测甲状腺滤泡状癌的效能。结果多因素logistic回归分析显示,甲状腺球蛋白阳性(OR=4.38,95%CI:2.96~6.57,P<0.001)、甲状腺球蛋白抗体阳性(OR=2.01,95%CI:1.17~3.48,P=0.012)、低回声(OR=6.18,95%CI:2.97~13.73,P<0.001)、分叶状或不规则边缘(OR=12.27,95%CI:5.87~27.95,P<0.001)和厚晕环(OR=7.83,95%CI:4.76~13.14,P<0.001)为甲状腺滤泡癌的独立预测因子。模型在训练集和验证集中的AUC分别为0.846和0.791。决策曲线分析表明,该模型在不同风险阈值下均具较高的临床净获益。结论基于临床及超声特征的风险预测模型可较好评估术中病理不确定患者的甲状腺滤泡癌风险,为术中决策提供依据。
Objective To develop a predictive model based on preoperative clinical and ultrasonographic features for assessing the risk of follicular thyroid carcinoma(FTC)in patients with intraoperative frozen pathology indicating"follicular neoplasm with indeterminate malignancy".Methods A retrospective analysis was conducted on 693 patients who underwent thyroid surgery at Ningbo Medical Center Lihuili Hospital between January 2019 and December 2024.All of the patients had intraoperative frozen pathology suggesting"follicular neoplasm with indeterminate malignancy".A comprehensive data set encompassing demographic characteristics,laboratory findings,and ultrasonographic features was systematically collected.Multivariate logistic regression analysis was performed to identify independent risk factors for FTC.The dataset was randomly partitioned into a training set and a validation set at a ratio of 7∶3.A stepwise logistic regression approach was applied in the training set to establish the predictive model,which was evaluated using ROC curves,calibration curves,and decision curve analysis(DCA).Results Multivariate logistic regression analysis identified thyroglobulin positivity(OR=4.38,95%CI:2.96-6.57,P<0.001),thyroglobulin antibody positivity(OR=2.01,95%CI:1.17-3.48,P=0.012),hypoecho genicity(OR=6.18,95%CI:2.97-13.73,P<0.001),lobulated or irregular margins(OR=12.27,95%CI:5.87-27.95,P<0.001),and thick hypoechoic halo(OR=7.83,95%CI:4.76-13.14,P<0.001)as independent risk factors for FTC.The model achieved an AUC of 0.846 in the training set and 0.791 in the validation set.DCA indicated substantial clinical net benefit across a wide range of threshold probabilities.Conclusion A predictive model incorporating clinical pathogens and ultrasonographic features can effectively evaluate the risk of FTC in patients with indeterminate intraoperative frozen pathology,providing valuable support for intraoperative surgical decision-making.
作者
蒋燕
罗勇
彭媛媛
周超瑜
甘咏莉
司徒明珠
JIANG Yan;LUO Yong;PENG Yuanyuan;ZHOU Chaoyu;GAN Yongli;SITU Mingzhu(Department of Ultrasound,Ningbo Medical Center Lihuili Hospital,Ningbo 315000,China;不详)
出处
《浙江医学》
2025年第24期2604-2609,共6页
Zhejiang Medical Journal
基金
浙江省医药卫生科技计划项目(2024KY1484)。
关键词
甲状腺滤泡癌
预测模型
临床特征
超声特征
列线图
Follicular thyroid carcinoma
Predictive model
Clinical features
Ultrasonographic features
Nomogram
