Including cardiac MRI parameters improves the predictive power of an American Heart Association model that estimates people's 10-year risk for heart failure, according to a study published June 23 in Radiology.
The finding comes from a secondary analysis of a prospective cohort of nearly 40,000 UK Biobank participants who underwent cardiac MRI, reported lead author Menghan Zhu, of the Peking Union Medical College and the Chinese Academy of Medical Sciences in Beijing, and colleagues.
"Integrating cardiac MRI parameters into the Predicting Risk of Cardiovascular Disease EVENTs equations not only improved heart failure risk prediction but also revealed sex-specific pathophysiologic differences, underscoring the predictive value of multidimensional imaging parameters," the group wrote.
The American Heart Association introduced the Predicting Risk of Cardiovascular Disease EVENTs (PREVENT) equations in 2023 as a tool to estimate 10-year and 30-year risks for heart failure and atherosclerotic disease in adults aged 30 to 79. The model is based on standard clinical variables such as age, blood pressure, cholesterol levels, kidney function, diabetes, and smoking status. While the model demonstrates excellent discrimination, the authors hypothesized that including subclinical imaging biomarkers could improve its value.
To that end, they culled data on 39,069 U.K. Biobank participants (mean age 55 years; 52.5% women) who underwent cardiac MRI between March 2006 and December 2022. Those with prevalent heart failure, coronary heart disease, or stroke were excluded. The cohort was randomly divided 70/30 into training and internal test sets. The researchers used least absolute shrinkage and selection operator (LASSO) regression and selected 16 cardiac MRI parameters from an initial pool of 82 variables and incorporated them into PREVENT equations.
Key results included the following:
The C-index increased from 0.753 to 0.812 (P < .001) in the training set and from 0.760 to 0.821 (P = .007) in the internal test set.
Left atrial ejection fraction contributed most to heart failure risk prediction in men (ΔC-index = 0.022), whereas left ventricular ejection fraction (ΔC-index = 0.011) drove the improvement in women.
“Our results suggest that [heart failure] pathophysiology may involve earlier atrial remodeling in men and earlier ventricular impairment in women, reflecting distinct myocardial remodeling, inflammation, fibrosis, and hormonal regulation patterns,” the group wrote.
The authors noted that the study is the first to systematically evaluate the incremental value of multidimensional cardiac MRI parameters for the PREVENTs model in predicting 10-year heart failure risk within a large prospective cohort, and as such, the results require further validation.
"Future studies should replicate these findings in large, multicenter, and ethnically diverse cohorts to further enhance the clinical translation of this model," the researchers concluded.
The full study is available here.
