# A nomogram for predicting the risk of coronary artery disease in premenopausal women with suspected coronary artery disease

## TL;DR

This study creates a tool to predict coronary artery disease risk in premenopausal women using factors like age, diabetes, and blood markers.

## Contribution

The study introduces a novel nomogram for predicting CAD risk in premenopausal women using LASSO and logistic regression.

## Key findings

- A nomogram with five variables (age, diabetes, AST, ALP, Lp(a)) was developed for CAD risk prediction.
- The nomogram showed good predictive performance with an AUROC of 0.819.
- Calibration curves and decision curve analysis confirmed the model's clinical utility.

## Abstract

Due to the cardioprotective effects of estrogen, premenopausal women have a relatively lower risk of developing coronary artery disease (CAD). However, the incidence of CAD in premenopausal women has been increasing in recent years. Therefore, the aim of this study is to develop a clinical prediction model to estimate the risk of CAD in premenopausal women. This study included premenopausal women who underwent coronary angiography at the First Hospital of Hebei Medical University from September 2018 to December 2021. The Least Absolute Shrinkage and Selection Operator (LASSO) regression method was used to identify the optimal variables for predicting the risk of CAD in premenopausal women. A nomogram was then constructed using multivariate logistic regression analysis. Finally, the predictive performance of the nomogram was evaluated using the area under the receiver operating characteristic curve (AUROC), its calibration performance was assessed using calibration curves, and clinical net benefit was evaluated using Decision Curve Analysis (DCA). A total of 222 premenopausal women were ultimately included for analysis, of whom 86 were diagnosed with CAD. Through LASSO and multivariate logistic regression, five predictive variables were finally selected: age, diabetes mellitus (DM), aspartate transaminase (AST), alkaline phosphatase (ALP), and lipoprotein (a) (Lp(a)). These five variables were used to construct a prediction model, which was presented in the form of a nomogram. The calibration curves of the nomogram showed good fit. The area under the receiver operating characteristic curve (AUROC) for the nomogram was 0.819 (95%CI: 0.760–0.878). Additionally, decision curve analysis (DCA) indicated that the nomogram can achieve good net benefit in clinical applications.