Key result
Retinal AI score linked to ~2.3-fold higher 10-year ASCVD risk beyond SCORE2.
Why the study?
Whether a retinal AI risk score reclassifies SCORE2 cardiovascular risk stratification across the ESC/EAS statin-indication boundary in statin-naive adults was uncertain.
Does adding a retinal artificial intelligence (AI) risk score to SCORE2 improve 10-year ASCVD risk prediction and reclassify risk across the ESC/EAS statin-indication boundary in statin-naive adults without diabetes?
Cohort (n=40,672)
Does adding a retinal artificial intelligence (AI) risk score to SCORE2 improve 10-year ASCVD risk prediction and reclassify risk across the ESC/EAS statin-indication boundary in statin-naive adults without diabetes?
A retinal AI risk score independently predicts 10-year ASCVD and provides modest bidirectional reclassification across the ESC/EAS statin-indication boundary when added to SCORE2, especially in women, though without improving overall C-index or net benefit.
Answers are grounded in this paper and cite their sources.
Retinal AI modestly reclassifies statin eligibility without C-index or net-benefit gains; leaves open clinical utility pending prospective trials.
AIMS: Whether a retinal artificial intelligence (AI) risk score reclassifies SCORE2 cardiovascular risk stratification in statin-naive adults across the ESC/EAS statin-indication boundary is uncertain. We evaluated the retinal AI risk score for 10-year atherosclerotic cardiovascular disease (ASCVD) prediction. METHODS: In 40,672 statin-naive UK Biobank participants (aged 40-69, without diabetes), cardiovascular risk categories and treatment recommendation status (Class I or IIa vs neither) were defined using the 2025 ESC/EAS focused-update framework. Cox models estimated hazard ratios (HRs) adjusted for SCORE2 or for established risk factors. The incremental predictive value of adding the retinal AI risk score to SCORE2 was evaluated using 10-fold cross-validation. RESULTS: Over a 10.0-year median follow-up (914 ASCVD events), the very high retinal AI risk group independently predicted ASCVD (SCORE2-adjusted HR 2.32 vs low risk; p<0.001). Adding the retinal AI risk score to SCORE2 reclassified risk across the ESC/EAS statin-indication boundary (statin-eligibility NRI [net reclassification improvement] 0.012, p<0.001); reclassification was statistically significant but modest (continuous NRI 0.445, p<0.001), without C-index or net-benefit gain. In the statin-indicated subgroup, retinal AI risk groups further stratified risk (SCORE2-adjusted HR 2.16; p<0.001). Statin-not-indicated participants in the high or very high retinal AI risk groups had numerically similar observed 10-year risk to all statin-indicated, untreated participants (p=0.758). There was no evidence of effect modification by clinically detectable retinal disorder (interaction p=0.93). Incremental value was greater in women than men (sex interaction p<0.001). CONCLUSION: Adding the retinal AI risk score to SCORE2 produced modest bidirectional risk reclassification across the ESC/EAS statin-indication boundary; these findings warrant prospective validation.Abstract word count: 250 words.
No takes yet. Share an insight, caveat, or question.
Nam et al. (2026) conducted a cohort in 10-year atherosclerotic cardiovascular disease (ASCVD) risk stratification (n=40,672). Retinal artificial intelligence (AI) risk score added to SCORE2 vs. SCORE2 alone / low retinal AI risk group was evaluated on 10-year atherosclerotic cardiovascular disease (ASCVD) events and risk reclassification across the ESC/EAS statin-indication boundary (HR 2.32, p=<0.001). Adding a retinal AI risk score to SCORE2 independently predicted 10-year ASCVD (adjusted HR 2.32 vs low risk; p<0.001) and modestly reclassified statin eligibility (NRI 0.012; p<0.001).
Synapse has enriched 10 related papers. Consider them for comparative context: