# Analysis of risk factors and construction of nomogram model for cardiac valve calcification of patients undergoing hemodialysis

## TL;DR

This study builds a predictive model to identify patients on hemodialysis at higher risk of developing cardiac valve calcification based on factors like age, gender, and cholesterol levels.

## Contribution

A novel nomogram model is developed and validated for predicting cardiac valve calcification in hemodialysis patients.

## Key findings

- Age, male sex, dialysis duration ≥ 36 months, high cholesterol, and increased fat tissue index are significant risk factors for CVC.
- The nomogram model showed good discrimination (AUC 0.789) and calibration in predicting CVC.
- The model's clinical utility was confirmed through decision curve analysis in both modeling and validation groups.

## Abstract

This study aimed to construct a risk prediction nomogram model of cardiac valve calcification (CVC) in patients undergoing maintenance hemodialysis (MHD) and to verify its evaluation effect.

A total of 398 patients undergoing hemodialysis were randomly divided into a modeling group (n = 274) and a validation group (n = 124). In the modeling group, 92 patients had CVC and 182 did not. Multivariate logistic regression analysis was conducted to determine the risk factors for CVC in patients undergoing hemodialysis. A nomogram prediction model was constructed using R software, and its predictive performance was evaluated in terms of discrimination, calibration, and clinical utility.

This study included 398 patients undergoing MHD with a mean age of 51.17 ± 14.09 years, and the prevalence of CVC was 31.66%. Compared with the non-CVC group, patients in the CVC group were older and had a higher proportion of males, longer dialysis duration, higher prevalence of diabetes, and higher levels of total cholesterol, triglycerides, and fat tissue index, while handgrip strength was significantly lower (all P < 0.05). Multivariate logistic regression analysis identified age (OR = 1.052, 95%CI = 1.028–1.077), male sex (OR = 3.164, 95%CI = 1.679–5.962), dialysis duration ≥ 36 months (OR = 2.096, 95%CI = 1.162–3.781), total cholesterol level (OR = 1.582, 95%CI = 1.191–2.101), and fat tissue index (OR = 1.128, 95%CI = 1.046–1.217) as independent risk factors for CVC (all P < 0.05). The area under the receiver operating characteristic curve (AUC) of the nomogram in the modeling group was 0.789, indicating good discriminative ability. The calibration curve demonstrated good agreement between predicted and observed outcomes. In the validation group, the AUC was 0.751, with calibration curve closely aligned with the ideal reference line. Decision curve analysis (DCA) further confirmed the clinical utility of the nomogram.

Patients undergoing hemodialysis who are older, male, have a dialysis duration ≥ 36 months, elevated total cholesterol levels, and increased fat tissue index are at higher risk of developing CVC. The nomogram model demonstrated good predictive performance for CVC in patients undergoing hemodialysis and may serve as a practical tool for identifying high-risk individuals in clinical practice.