Role of the Brock, Mayo, and PKUPH combined model in risk stratification of solitary pulmonary nodules in a health check-up cohort
10.3760/cma.j.cn115624-20241231-01088
- VernacularTitle:Brock、Mayo和PKUPH联合模型对健康体检人群孤立性肺结节的风险分层作用
- Author:
Xuejiao LIU
1
;
Bin LI
1
;
Yan LI
1
;
Xiangfeng KONG
1
;
Juan SUN
1
;
Xuelin LI
1
;
Xichun WANG
1
;
Biqiang LI
1
Author Information
1. 重庆市人民医院（重庆大学附属人民医院）健康管理中心，重庆401121
- Publication Type:Journal Article
- Keywords:
Solitary pulmonary nodules;
Risk stratification;
Predictive models;
Combined model;
Physical examination;
Cohort study
- From:
Chinese Journal of Health Management
2025;19(7):550-556
- CountryChina
- Language:Chinese
-
Abstract:
Objective:To investigate the role of the Brock, Mayo, and PKUPH combined model in risk stratification of solitary pulmonary nodules (SPNs) in health check-up population.Methods:An ambispective cohort study was conducted on 668 eligible SPNs cases from the health management center in Chongqing General Hospital from June 2018 to June 2019. The exposure condition was prospectively followed or historically retrospected, and the clinical outcomes were prospectively followed. SPNs were classified into benign and malignant groups. Descriptive statistics and univariate analysis were performed to assess the differences in risk characteristics between two groups. Receiver operating characteristic (ROC) curve, clinical decision curve, and Hosmer-Lemeshow goodness-of-fit test were used to evaluate and compare the predictive performance, clinical utility, and calibration of the combined model versus individual Brock, Mayo, and PKUPH models.Results:Among the 668 SPNs cases, 82 (12.28%) were diagnosed as malignant. Age, sex, smoking history, extrapulmonary tumor history, diameter, upper lobe, clear border and spicule sign in the malignant group were significantly different from those in the benign group (all P<0.05). The combined model demonstrated superior predictive performance, clinical utility, and calibration compared to the best-performing individual Brock model [Area under the curve (AUC): 0.88 (95% CI: 0.84-0.92) vs 0.86 (95% CI: 0.82-0.91)]. Besides, multi-grade risk stratification enabled by the combined model was better than binary classification, with the malignant rate of the four risk levels were 0.60%, 4.62%, 14.58% and 56.07%, respectively. Conclusion:The combined model addresses the limitations of individual models in SPNs risk stratification for health examination populations, improving predictive performance, clinical utility, and calibration, while proposing a superior multi-grade risk stratification system.
