Cardiovascular risk prediction in a population with the metabolic syndrome: Framingham vs. UKPDS algorithms

Area of Science:

• Cardiology
• Metabolic Health
• Epidemiology

Background:

• Cardiovascular disease (CVD) risk prediction algorithms are crucial for prevention strategies but often lack population specificity.
• Metabolic syndrome (MetS), a cluster of risk factors, significantly elevates CVD risk but lacks a dedicated prediction algorithm.
• Current algorithms are often designed for general or diabetic populations, not specifically for MetS.

Purpose of the Study:

• To evaluate the predictive performance of two established CVD risk algorithms (one for general, one for diabetic populations) in individuals with MetS.
• To compare the accuracy of the Framingham and UKPDS risk algorithms for predicting 5-year CVD risk in a MetS cohort.

Main Methods:

• Utilized data from 2700 subjects with MetS but no diagnosed CVD from the Australian Diabetes, Obesity and Lifestyle study.
• Estimated 5-year CVD risk using Framingham and UKPDS algorithms, comparing predictions with Wilcoxon-signed rank test.
• Assessed algorithm performance using discrimination (AUC) and calibration (Hosmer-Lemeshow) metrics against actual CVD events.

Main Results:

• The UKPDS algorithm showed age-related inaccuracies, overpredicting risk in younger adults and underpredicting in older adults compared to Framingham.
• Both Framingham and UKPDS algorithms exhibited poor discrimination (AUC ~0.51-0.52) and calibration (Hosmer-Lemeshow >297) in the MetS population.
• A total of 133 CVD events were observed over a median follow-up of 5.0 years.

Conclusions:

• Neither the Framingham nor the UKPDS algorithm is optimal for predicting CVD risk in individuals with metabolic syndrome.
• There is a clear need for the development of novel, population-specific risk-prediction algorithms tailored for the MetS demographic.
• Accurate CVD risk assessment in the growing MetS population requires dedicated predictive tools.