【摘要】 目的：分析腔镜甲状腺癌术后癌因性疲乏(CRF)的影响因素，建立列线图模型并评价其效能。方法：选取2018年7月—2020年12月在我院行腔镜甲状腺癌切除术治疗的389例患者作为训练集，并将2021年1—8月95例患者作为验证集。采用单因素分析和二元Logistic回归分析筛选训练集中CRF的独立预测因素，应用R软件构建列线图模型，采用受试者操作特征曲线(ROC)和校准曲线评价模型的预测效能。结果：训练集中，共有185例(47.6%)出现CRF,多因素分析表明年龄、术前抑郁、术后放射性碘治疗以及促甲状腺激素(TSH)抑制水平是CRF的独立危险因素(P<0.05)。对模型进行内部验证和外部验证，训练集和验证集的ROC曲线下面积分别为0.744(95%CI:0.696～0.792)和0.837(95%CI:0.781～0.894);两组患者的校准曲线显示预测结果和实际结果的一致性良好。结论：本研究建立的列线图模型具有较好的区分度和校准度，有利于医护人员预测甲状腺癌术后出现CRF的风险，并制定相应的预防措施。更多还原

【Abstract】 Objective:To analyze the influencing factors of cancer-related fatigue(CRF) after endoscopic thyroid cancer surgery, establish a nomogram model and evaluate its efficacy. Methods:A total of 389 patients who underwent laparoscopic thyroid cancer resection in our hospital from July 2018 to December 2020 were selected as the training set, and 95 patients from January 2021 to August 2021 were selected as the validation set. Univariate analysis and binary logistic regression analysis were used to screen the independent predictors of CRF in the training set.R software was used to build a nomogram model, and receiver operator characteristic curve(ROC) and calibration curve were used to evaluate the prediction efficacy of the model. Results:In the training set, a total of 185 cases(47.6%) developed CRF. Multivariate analysis showed that age, preoperative depression, postoperative radioactive iodine therapy and thyroid stimulating hormone(TSH) inhibition were independent risk factors for CRF(P<0.05).The models were validated internally and externally, and the areas under the ROC curves for the training and validation sets were 0.744(95%CI:0.696～0.792) and 0.837(95%CI:0.781～0.894), respectively.The calibration curves for the two groups of patients show the predicted results are in good agreement with the actual results.Conclusion:The nomogram model established in this study has good discrimination and calibration, which is helpful for medical staff to predict the risk of CRF after thyroid cancer surgery and formulate corresponding preventive measures.更多还原