Evaluation of diagnostic efficacy of artery elasticity and endothelial function indexes for coronary heart disease with blood stasis syndrome in postmenopausal women by Logistic regression combined with ROC curve model.
10.19540/j.cnki.cjcmm.20211201.501
- Author:
Shi-Yi TAO
1
;
Lin-Tong YU
1
;
Zi-Han WANG
1
;
Ying CHEN
1
;
Jin ZHANG
1
;
De-Shuang YANG
1
;
Li HUANG
2
Author Information
1. Beijing University of Chinese Medicine Beijing 100029, China.
2. China-Japan Friendship Hospital Beijing 100029, China.
- Publication Type:Journal Article
- Keywords:
Logistic regression;
ROC curve;
artery elasticity;
coronary heart disease(CHD);
postmenopausal woman;
prediction model;
vascular endothelial function
- MeSH:
Ankle Brachial Index;
Brachial Artery;
Coronary Disease/diagnosis*;
Elasticity;
Female;
Humans;
Logistic Models;
Postmenopause;
Pulse Wave Analysis;
ROC Curve;
Retrospective Studies
- From:
China Journal of Chinese Materia Medica
2022;47(8):2244-2250
- CountryChina
- Language:Chinese
-
Abstract:
The present study explored the correlation of coronary heart disease(CHD) with blood stasis syndrome in postmenopausal women with artery elasticity and endothelial function indexes and evaluated the diagnostic efficacy of the prediction model via logistic regression and receiver operating characteristic(ROC) curve model. A retrospective comparison was made between 366 postmenopausal CHD patients from August 1, 2020, to September 30, 2021, in the Department of Cardiology of Integrated Traditional Chinese and Western Medicine of China-Japan Friendship Hospital, who were divided into the blood stasis syndrome group(n=196) and the non-blood stasis syndrome group(n=170). General clinical characteristics of the two groups were compared. Multivariate logistic regression analysis was used to probe the correlation of CHD with blood stasis syndrome in postmenopausal women with brachial-ankle pulse wave velocity(baPWV), ankle-brachial index(ABI), and flow-mediated dilatation(FMD), and the ROC curve was drawn to evaluate the diagnostic efficiency of the prediction model. Multivariate logistic regression analysis showed that the correlation coefficients of CHD with blood stasis syndrome in postmenopausal women with baPWV, ABI, and FMD were 1.123, 0.109, and 0.719, respectively(P=0.004, P=0.005, P<0.001),and the regression equation for predicting probability P was P=1/[1+e~(-(3.131+0.116×baPWV-2.217×ABI-0.330×FMD))]. ROC curve analysis suggested that in the context of baPWV≥19.19 m·s~(-1) or ABI≤1.22 or FMD≤9.7%, it was of great significance to predict the diagnosis of CHD with blood stasis syndrome in postmenopausal women. The AUC of baPWV, ABI, FMD, and prediction probability P was 0.763, 0.607, 0.705, and 0.836, respectively. The AUC of prediction probability P was higher than that of each index alone(P<0.001), and the sensitivity and specificity were 0.888 and 0.647, respectively. The results demonstrate that baPWV, ABI, and FMD are independently correlated with CHD with blood stasis syndrome in postmenopausal women, and show certain independent predictive abilities(P<0.05). The combined evaluation of the three possesses the best diagnostic efficiency.
