February 11, 2026Clinical Journal of the American Society of Nephrology1 citations
View Full PaperCardiovascular Risk Assessment Tools in Chronic Kidney Disease
Key Result
CKD-specific CV risk models like QRISK, CKD-PC, and PREVENT better predict cardiovascular risk in CKD patients and are recommended by 2024 KDIGO guidelines.
Key Points
- This narrative review aims to assess the cardiovascular risk calculators tailored for chronic kidney disease populations.
- Reviewed existing cardiovascular risk tools and their applicability to chronic kidney disease.
- Analyzed strengths and limitations of models like QRISK, CKD-PC, and PREVENT.
- Considered international guidelines and the necessity for individualized management strategies.
- Current models often underestimate cardiovascular risk in chronic kidney disease patients.
- Enhanced models integrating kidney function have been developed and recommended.
- Artificial intelligence approaches show promise for better risk prediction but require further validation.
Structured PICO
Do CKD-specific cardiovascular risk assessment tools improve cardiovascular risk prediction in patients with chronic kidney disease compared to standard models?
P
Population
Patients with chronic kidney disease (CKD)
I
Intervention
CKD-specific and enhanced cardiovascular risk models (e.g., SCORE2 Add-ons, QRISK, CKD-PC, PREVENT)
C
Comparator
Standard cardiovascular risk models (e.g., Framingham Risk Score, SCORE2, ASCVD Pooled Cohort Equations)
O
Outcome
Cardiovascular risk prediction accuracy (discrimination and calibration)
CKD-specific cardiovascular risk calculators, such as QRISK, CKD-PC, and PREVENT, provide better risk estimation for CKD patients than standard models and are recommended by current guidelines.
Main Result
Absolute Event Rate: 0% vs 0%
Abstract
Cardiovascular (CV) risk calculators estimate the likelihood of CV events by integrating factors such as age, sex, blood pressure, lipids, smoking, and diabetes. Commonly used tools in the general population include the Framingham Risk Score, SCORE2, Atherosclerotic Cardiovascular Disease (ASCVD )Pooled Cohort Equations, and QResearch Cardiovascular Risk Algorithm (QRISK), the latter being regularly updated using large-scale United Kigdom (UK) health records and including a wider range of variables such as chronic kidney disease (CKD). CKD is a strong, independent risk factor for CV disease, but most standard models omit kidney function and CKD-specific factors (e.g., inflammation, vascular calcification, mineral metabolism disorders). Consequently, they often underestimate CV risk in CKD, particularly in advanced stages and among dialysis patients, where competing risks (e.g., non-CV death) further complicate risk estimation. To overcome these limitations, enhanced and CKD-specific models have been developed. Systematic Coronary Risk Evaluation (SCORE2 ) Add-ons incorporate estimated glomerular filtration rate (eGFR) and albuminuria to refine prediction in CKD. QRISK includes CKD as a variable. The CKD Prognosis Consortium (CKD-PC) and Predicting Risk of Cardiovascular Disease EVENTs (PREVENT) models are tailored to CKD populations and leverage large, diverse datasets to improve discrimination and calibration. Reflecting this evidence, the 2024 Kidney Disease Global Outcomes ( KDIGO) guidelines recommend QRISK, CKD-PC, and PREVENT for CV risk prediction in CKD. Artificial intelligence and machine learning approaches may further enhance risk prediction by exploiting complex clinical and laboratory data, but they require external validation, transparency, and assessment of clinical utility before broad implementation. CV risk calculators should support, not replace, clinical judgment and must be embedded within individualized, multifactorial management strategies. This narrative review summarizes the main characteristics, strengths, and limitations of currently available CV risk tools for CKD, focusing on those endorsed by major international guidelines and used in routine practice. We adopt a global perspective, while acknowledging that most evidence derives from high-income settings, which limits generalizability.