Area under receiver operating characteristic curve improved with retinal parameters, outperforming traditional risk factors
THURSDAY, Jan. 16, 2025 (HealthDay News) — Retinal vascular parameters are associated with stroke risk and can improve prediction of stroke, according to a study published online Jan. 13 in Heart.
Mayinuer Yusufu, Ph.D., from The Royal Victorian Eye and Ear Hospital in East Melbourne, Australia, and colleagues used the Retina-based Microvascular Health Assessment System to extract retinal vascular parameters from the U.K. Biobank fundus images and examined the associations between these parameters and incident stroke.
The researchers found that 749 incident strokes occurred among 45,161 participants during a median follow-up of 12.5 years. Twenty-nine significant parameters associated with stroke risk were identified, with a dominance of density parameters. Each standard deviation (SD) change in these parameters was associated with a 9.8 to 19.0 percent increase in stroke risk. Each SD change in identified caliber parameters was associated with an increase in risk ranging from 10.1 to 14.1 percent. Each SD decrease in identified complexity parameters and arterial inflection count tortuosity was associated with an increased risk ranging from 10.4 to 19.5 percent. The area under the receiver operating characteristic curve was improved to 0.752 with the introduction of retinal vascular parameters, significantly outperforming the model using traditional risk factors (0.739).
“Our study showed that this set of comprehensive retinal vascular parameters was of added predictive value for incident stroke, indicating its potential application as a noninvasive screening method for individuals with increased risk,” the authors write.
2005 to 2023 Saw More U.S. Women Never Screened for Cervical Cancer
Onset of Mortality, Morbidity Delayed for Offspring of Centenarians
Machine Learning Model IDs Heart Failure With Reduced Ejection Fraction Using Routine Lab Indicators
Addition of CAC to PREVENT-ASCVD Equations May Improve Predictive Utility in Certain Patients
Warning Symptoms, Clinical History Can Predict Sudden Cardiac Arrest
Micro-, Nanoplastics Detected More Often in Patients With STEMI
AI Can Accurately Detect Left Ventricular Dysfunction From Electrocardiograms
Differences in Cardiovascular Risk Factor Prevalence ID"d for South Asian Adults