Construction of a visual model for predicting the risk of recurrence of thyroid cancer after radical surgery via areola endoscopy
10.3969/j.issn.1009-9905.2025.10.003
- VernacularTitle:甲状腺癌经乳晕腔镜下根治术术后复发风险预测模型的构建
- Author:
Qing-feng SHI
1
;
Bu-yong ZHANG
;
Xuan ZHANG
;
Yang BAI
;
Ling-bo XUE
;
Jie LI
Author Information
1. 沧州市中心医院 甲状腺乳腺外科(河北 沧州 061000)
- Publication Type:Journal Article
- Keywords:
Thyroid cancer;
Radical surgery via areola endoscopy;
Recurrence;
Influencing factors;
Nomogram pre-diction model
- From:
Chinese Journal of Current Advances in General Surgery
2025;28(10):769-775
- CountryChina
- Language:Chinese
-
Abstract:
Objective:To explore the risk factors for recurrence of thyroid cancer after radical resection via areola endoscopy,and to construct a visual risk prediction model.Methods:The clinical data of 350 thyroid cancer patients who underwent radical surgery via areola endoscopy in our hospital from January 2016 to October 2018 were retro-spectively analyzed,and they were randomly divided into the modeling group(233 cases)and the internal validation group(117 cases)in a 2:1 ratio.All patients were followed up for 3 years after surgery,and the patients of modeling group were further divided into recurrent group(51)and non recurrent group(182)according to whether they with or not recurrence.Another 163 patients with thyroid cancer who underwent laparoscopic radical mastectomy at our hos-pital from January 2019 to May 2020 were selected as the external validation group.The risk factors for recurrence of thyroid cancer after radical surgery via areola endoscopy was analyzed by using Cox regression method,and a risk prediction nomogram model was established based on this.Internal validation of the nomogram model was conducted by using the Bootstrap method,and the calibration,predictive efficacy and clinical net benefit of the nomogram model were evaluated by the calibration curve,receiver operating characteristic(ROC)curve and decision curve analysis(DCA).The external validation group data was used for external validation.Results:The recurrence rate of thyroid cancer patients after 5 years of radical surgery via areola endoscopy was 21.64％(111/513).The proportions of multiple le-sions,preoperative lymph node metastasis,TNM stages Ⅲ-Ⅳ and maximum tumor diameter,the levels of thyro-globulin(TG),triiodothyronine(T3),thyroxine(T4),free triiodothyronine(FT3),free thyroxine(FT4)and thyroid stimulating hormone(TSH)in the recurrence group were higher than those in the non recurrence group(P<0.05).The Cox regres-sion analysis results showed that the maximum tumor diameter,multiple lesions,preoperative lymph node metasta-sis,TNM stage Ⅲ-Ⅳ and TG,T3,T4,FT3,FT4 and TSH levels were all risk factors for recurrence of thyroid cancer after radical surgery via areola endoscopy(P<0.05).The risk prediction nomogram model of recurrence of thyroid cancer af-ter radical surgery under areola endoscopy was constructed based on the above influencing factors.After internal and external validation,the consistency indices of the modeling group,internal verification group and external verification group were 0.832,0.825 and 0.41 respectively,and the calibration curves of three groups were close to the standard curve.The ROC curve analysis and verification showed that the area under the curve predicted by the nomogram model of the modeling group,internal verification group and external verification group were 0.859,0.847 and 0.853 respectively.The DCA curve showed that the nomogram model had good clinical net benefits when the threshold probability of the modeling group,internal verification group and external verification group were 0.03-0.82,0.02-0.78 and 0.06-0.88 respectively.Conclusion:The maximum tumor diameter,multiple lesions,preoperative lymph node metastasis,TNM staging stage Ⅲ-Ⅳ and levels of TG,T3,T4,FT3,FT4 and TSH are all risk factors for recurrence of thy-roid cancer after radical surgery via areola endoscopy,and the risk prediction visualization nomogram model con-structed based on this is helpful for clinical screening of high-risk patients to guide early intervention and reduce the risk of recurrence.
