A prediction model of thoracic aortic calcification in chronic kidney disease based on serum nidogen-2
10.3760/cma.j.cn441217-20240716-00727
- VernacularTitle:基于血清巢蛋白-2的慢性肾脏病胸主动脉钙化预测模型构建
- Author:
Yongqi LI
1
;
Jing LU
1
;
Yan DI
1
;
Yinan ZHAO
1
;
Yuxia ZHANG
1
;
Yujia WANG
1
;
Ziyu LIANG
1
;
Rining TANG
1
;
Bicheng LIU
1
Author Information
1. 东南大学附属中大医院肾内科 东南大学医学院，南京 210009
- Publication Type:Journal Article
- Keywords:
Renal insufficiency, chronic;
Vascular calcification;
Models, statistical;
Chronic kidney disease;
Nidogen-2
- From:
Chinese Journal of Nephrology
2025;41(8):605-614
- CountryChina
- Language:Chinese
-
Abstract:
Objective:To explore the correlation between serum nidogen-2 (NID-2) and thoracic aortic calcification in patients with chronic kidney disease (CKD), and construct a risk prediction model based on NID-2 to evaluate its value in predicting the risk of the severe thoracic aortic calcification and cardiovascular and cerebrovascular events in CKD patients.Methods:It was a prospective cohort study. Patients with CKD at stage 3 to 5D in the Zhongda Hospital Affiliated to Southeast University from January 2022 to January 2023 were enrolled. Syngo.via software was used to evaluate the volume of thoracic aortic calcification, and enzyme-linked immunosorbent assay was employed to determine the level of serum NID-2. According to the volume of thoracic aortic calcification, the patients were divided into three groups: no calcification group, mild calcification group and severe calcification group. The top 25% of the patients were defined as no or mild calcification group, and the latter 75% were defined as severe calcification group. The follow-up period was one year. During the follow-up period, cardiovascular and cerebrovascular events, as well as all-cause death among the enrolled patients were recorded. Logistic regression analysis was used to screen the influencing factors of thoracic aortic calcification. Based on the results of logistic regression analysis, a nomogram prediction model was constructed. The receiver operating characteristic curve (ROC curve), calibration curve, and decision curve were employed to evaluate the discrimination, calibration and clinical practicality of the nomogram model.Results:A total of 132 patients were included, with 91 males (68.94%) and age of (56.51±16.37) years. There were 60 CKD 3-5 stage patients (non-dialysis, 45.45%) and 72 CKD 5D patients (dialysis, 54.55%). Serum ND-2 levels differed significantly among healthy individuals, dialysis patients and non-dialysis patients ( H=70.651, P<0.001). There was no statistically significant difference in serum NID-2 level between the no or mild calcification group and the severe calcification group in dialysis patients ( Z=350.00, P=0.426). The serum NID-2 level in the severe calcification group was significantly higher than that in the no or mild calcification group in non-dialysis patients ( Z=242.00, P=0.019). In non-dialysis patients, there was a statistically significant correlation between serum NID-2 level and volume of thoracic aortic calcification ( r=0.40, P<0.001). In dialysis patients, there was no statistically significant correlation between serum NID-2 level and volume of each segment of thoracic aortic calcification (all P>0.05). The univariate logistic regression analysis showed that, age, hemoglobin, serum albumin, estimated glomerular filtration rate, NID-2, hypertension, type 2 diabetes mellitus and cerebral infarction were correlated factors of thoracic aortic calcification in non-dialysis patients (all P<0.05). Multivariate logistic regression analysis showed that age ( OR=1.22, 95% CI 1.08-1.50, P=0.010) was an independent correlated factor of thoracic aortic calcification in non-dialysis patients. The above related variables of univariate logistic regression analysis were incorporated into a nomogram to construct a predictive model for severe vascular calcification in non-dialysis patients, yielding an AUC of 0.94 (95% CI 0.89-0.99) in ROC curve, with a sensitivity of 83% and a specificity of 95%. A nomogram model based on above variables for predicting cardiovascular and cerebrovascular events in non-dialysis patients demonstrated an AUC of 0.95 (95% CI 0.90-1.00) in ROC curve, with a sensitivity of 95% and a specificity of 87%. Conclusions:In non-dialysis patients, serum NID-2 level in the severe calcification group is significantly higher than that in the no or mild calcification group. The serum NID-2 is a related factor of thoracic aortic calcification and cardiovascular and cerebrovascular events in non-dialysis patients. The nomogram prediction model constructed by combining NID-2 with age, hemoglobin, serum albumin, estimated glomerular filtration rate, hypertension, type 2 diabetes mellitus and cerebral infarction has a high predictive value for the risk of thoracic aortic calcification as well as cardiovascular and cerebrovascular events in non-dialysis patients.
