Synapse
⌘+K
Synapse
PulseExploreClubsResearchersJournals
Instagram
HomeClubsExplore
June 7, 2026PeerJOpen Access

Integrated clinical, inflammatory, and imaging model predicts five-year CVD risk in advanced CKD with ~0.85 AUC.

View Full Paper
Ask AI
Bookmark
Share

Why the study?

Traditional CVD risk factors have diminished predictive utility in advanced CKD, and existing prediction models based solely on traditional risks show limitations and inaccuracies.

Does a risk prediction model integrating clinical, inflammatory, and imaging parameters improve the prediction of 5-year CVD events in patients with CKD stages 3-5 compared to traditional risk factors?

Population

Three hundred and one patients with CKD stage 3-5

Comparison

Model integrating clinical, inflammatory, and imaging parameters vs traditional risk models

Design

Retrospective model development and internal validation study

Follow-up

5 years

Key result

A prediction model integrating clinical, inflammatory, and imaging parameters accurately predicted 5-year CVD risk in patients with CKD stages 3-5 (AUC 0.845; 95% CI 0.802-0.888).

Authors

HLHuixia LiuJXJing Xiong

Discussion

Loading...

Member takes

Overview

A novel risk prediction model incorporating CRP and interventricular septum thickness alongside traditional risk factors accurately predicts 5-year CVD events in patients with advanced CKD.

Key Points

  • The study aimed to develop and validate a five-year cardiovascular disease risk prediction model for patients with chronic kidney disease stages 3–5.
  • Retrospective study of 301 patients with CKD stages 3–5 from January 2010 to January 2022, followed until July 2022.
  • Established risk prediction model using LASSO regression and logistic regression, validated through tenfold cross-validation.
  • C-statistic and Hosmer-Lemeshow test assessed model performance and calibration.
  • 169 patients (56.1%) had a first cardiovascular event within 5 years, median occurrence at 10 months.
  • Identified 11 independent predictors from 29 candidate variables, improving prediction with CRP and IVS markers.
  • The final model showed excellent discrimination (AUC 0.845, 95% CI [0.802–0.888]) and calibration (P=0.332).

Study Design

Type

Cohort (n=301)

Structured PICO

Does a risk prediction model integrating clinical, inflammatory, and imaging parameters improve the prediction of 5-year CVD events in patients with CKD stages 3-5 compared to traditional risk factors?

P
Population
301 patients with chronic kidney disease stage 3-5 followed for up to 5 years to develop and validate a cardiovascular disease risk prediction model.
E
Exposure
5-year CVD risk prediction model integrating clinical, inflammatory (CRP), and echocardiographic (IVS) parameters
C
Comparator
Prediction models based only on traditional CVD risk factors
O
Outcome
First CVD event within 5 yearscomposite

Main Result

Effect estimate: AUC 0.845 (95% CI 0.802-0.888)

A novel risk prediction model incorporating CRP and interventricular septum thickness alongside traditional risk factors accurately predicts 5-year CVD events in patients with advanced CKD.

Cite This Study

Liu et al. (2026) conducted a cohort in Chronic kidney disease stage 3-5 (n=301). Full prediction model integrating clinical, inflammatory (CRP), and imaging (IVS) parameters vs. Traditional risk factors was evaluated on First cardiovascular disease event within 5 years (AUC 0.845, 95% CI 0.802-0.888). A prediction model integrating clinical, inflammatory, and imaging parameters accurately predicted 5-year CVD risk in patients with CKD stages 3-5 (AUC 0.845; 95% CI 0.802-0.888).

synapsesocial.com/papers/6a250c1c7def13d035e1c1a0https://doi.org/10.7717/peerj.21312
View Full Paper
Ask AI
Bookmark
Share