周安悦,侯晋轩,袁芊芊,等.甲状腺癌术后保护甲状旁腺功能的联合预测模型建立[J].肿瘤学杂志,2022,28(8):644-650.
甲状腺癌术后保护甲状旁腺功能的联合预测模型建立
Establishment of Prediction Model for Postoperative Parathyroid Function in Patients with Thyroid Cancer
投稿时间:2022-03-08  
DOI:10.11735/j.issn.1671-170X.2022.08.B004
中文关键词:  甲状旁腺  甲状腺癌  影响因素  列线图
英文关键词:parathyroid function  thyroidectomy  influence factors  nomogram
基金项目:武汉大学中南医院科技创新培育基金临床研发项目(lcyf202007);武汉大学中南医院科技创新培育基金临床研究(产业化)项目(WJ2018H0014)
作者单位
周安悦 武汉大学中南医院 
侯晋轩 武汉大学中南医院 
袁芊芊 武汉大学中南医院 
曹家兴 武汉大学中南医院 
摘要点击次数: 509
全文下载次数: 221
中文摘要:
      摘 要:[目的] 分析甲状腺恶性肿瘤手术后影响甲状旁腺功能的因素并建立列线图预测模型。 [方法] 收集890例甲状腺手术患者临床资料,使用SPSS(23.0.0)软件进行单因素检验及多因素Logistic回归分析。使用R(4.0.4)及Rstudio(1.3.1093) rms 程序包建立列线图预测模型,同时应用Caret 程序包进行 Bootstrap 法做内部验证,拟合影响因素联合预测因子,并绘制受试者工作特征曲线,计算曲线下面积及 95%CI双侧检验,得出列线图模型的预测效率。 [结果] 术后并发甲状旁腺功能减退症182例,占20.45%。Logistic回归分析结果表明:女性(OR=1.888,95%CI:1.061~3.360)、低BMI(OR=0.879,95%CI:0.825~0.936)、多发癌灶(OR=1.584,95%CI:1.057~2.372)、腺外侵犯(OR=1.837,95%CI:1.020~3.307)、甲状腺全切及中央区淋巴结清扫(OR=1.709,95%CI:1.073~2.722)、旁腺切除移植(OR=4.450,95%CI:2.540~7.797)与术后1 d甲状旁腺激素水平降低显著相关(P<0.05)。重复抽样偏差曲线接近于标准曲线,拟合度良好。联合预测因子AUC=0.829(95%CI:0.780~0.868),有较高意义。[结论] 女性、低体质指数、多发癌灶、有甲状腺外侵犯、行甲状腺全切及中央区淋巴结清扫术、有旁腺切除是甲状腺癌术后并发暂时性甲状旁腺功能减退的独立危险因素,联合预测因子区分度良好,据此建立的列线图预测模型具有可靠的预测能力。
英文摘要:
      Abstract: [Objective] To analyze the factors affecting parathyroid function after thyroid cancer surgery and establish a nomogram prediction model. [Methods] Clinical data of 890 patients undergoing thyroid surgery were collected. The risk factors for postoperative hypoparathyroidism were determined with Logistic regression analysis by SPSS(23.0.0). The R(4.0.4) and Rstudio(1.3.1093) rms program packages were used to establish a nomogram prediction model. The Caret program package was used to perform the Bootstrap method for internal verification, fitting the Combined predictor of influencing factors. The receiver operating characteristic curve was used to evaluate the prediction efficiency of the established nomogram model. [Results] Among 890 patients, 182 cases(20.45%) were complicated with hypoparathyroidism. Logistic regression analysis showed that female gender(OR=1.888, 95%CI: 1.061~3.360), low BMI(OR=0.879, 95%CI: 0.825~0.936), multiple cancer(OR=1.584, 95%CI: 1.057~2.372), extrathyroid invasion(OR=1.837, 95%CI: 1.020~3.307), total thyroidectomy and central lymph node dissection(OR=1.709, 95%CI: 1.073~2.722), parathyroid gland resection and transplantation(OR=4.450, 95%CI: 2.540~7.797) were significantly associated with the decrease of PTH 1 d after operation(P<0.05). The repeated sampling deviation curve was close to the standard curve with good fitting degree. The area under ROC curve(AUC) of the established nomogram for predicting risk of postoperative hopyparathyroidism was 0.829(95%CI: 0.780~0.868). [Conclusion] Female gender, low BMI, multiple cancer, presence of external thyroid invasion, total thyroidectomy and central lymph node dissection, and parathyroid gland resection and transplantation are independent risk factors for temporary hypoparathyroidism in patients with thyroid cancer after surgical treatment. The established nomogram prediction model has a good prediction ability for postoperative hypoparathyroidism.
在线阅读   查看全文  查看/发表评论  下载PDF阅读器