Risk factors of endometrial lesions after breast cancer surgery and construction of a graph model
10.3760/cma.j.cn.115807-20221201-00348
- VernacularTitle:乳腺癌术后子宫内膜病变的危险因素和列线图模型构建
- Author:
Bo LI
1
;
Tianmei ZHANG
;
Shuang LIN
;
Yanmei SHI
Author Information
1. 烟台市烟台山医院妇科，烟台 264000
- Keywords:
Endometrial lesions;
Breast cancer;
Nomogram model;
Logistic risk regression model
- From:
Chinese Journal of Endocrine Surgery
2023;17(2):190-193
- CountryChina
- Language:Chinese
-
Abstract:
Objective:To explore the risk factors affecting endometrial lesions after breast cancer surgery, and build a nomogram prediction model.Methods:From Oct. 2019 to Nov. 2021, 103 patients with abnormal bleeding after breast cancer surgery were selected, the clinical data of the patients were collected, and they were divided into the non-lesion group and the lesion group according to whether the endometrial lesion occurred. A Logistic risk regression model was established to analyze the risk factors affecting endometrial lesions in postoperative patients with breast cancer, a nomogram prediction model was constructed and verified, and receiver operating characteristic curve (ROC) analysis was performed to analyze the nomogram model for predicting sensitivityof endometrial lesions.Results:Childbirth history ( OR=37.100, 95% CI: 3.777-527.7, P=0.004), endometrial thickness ( OR=2.489, 95% CI: 1.699-4.007, P<0.001), menopause ( OR=0.099, 95% CI: 0.015-0.499, P=0.009), abnormal bleeding time ( OR=6.922, 95% CI: 2.221-24.800, P=0.002), and types of treatment drugs ( OR=3.738, 95% CI: 1.187-13.200, P=0.030) had statistical significance in predicting endometrial lesions in postoperative patients with breast cancer. Using the above five variables to construct a nomogram model, the consistency of the nomogram in predicting endometrial lesions in postoperative patients with breast cancer was 0.739, and the discrimination was good. The calibration curve showed that the average absolute error between the predicted probability and the actual probability was 0.041,and ROC curve showed that the AUC value of the nomogram model for predicting endometrial lesions was 0.800. Conclusion:Establishing a nomogram model for predicting the risk of endometrial lesions in postoperative patients with breast cancer has good accuracy and high clinical value.
