Elevated Lipoprotein(a) Independently Increases Risk for Short-Term Atherosclerotic Cardiovascular Events in Machine Learning Predictive Models
Authors' conclusion
Lp(a) is a significant predictor of short-term ASCVD risk.
Study context
- Population
- Individuals with Lp(a) testing in a claims databasevalidation cohort n = 53,930
- Participants
- 731,983 individuals
- Comparison
- Lp per 50 nmol/L increase
- Outcome
- Incident ASCVD events and all-cause mortality
- Follow-up
- 1, 2 and 3 years after Lp(a) testing
Methodological detailsclaims database, independent validation cohort
- Claims database
- Independent validation cohort
Performance
As reported in the abstract
Association of Lp(a) (per 50 nmol/L increase) with outcomes
Lp(a) +50 nmol/LHazard ratio (95% CI)
Incident ASCVD events
1.072(1.059–1.084)
All-cause mortality
1.041(1.015–1.068)
Per 50 nmol/L increase, as printed
Further findings
Machine learning (ML) models featuring Lp(a): discrimination for incident atherosclerotic cardiovascular disease (ASCVD) at 1, 2 and 3 years
- C-statistic, derivation and validation cohorts
- 0.83-0.84
Calibration of 1-year model, validation cohort
- Modest underestimation of risk; calibration slope
- 1.25
Added value of Lp(a) in 1-year model
- Integrated discrimination improvement
- 0.03
- Optimal net reclassification improvement at risk threshold of 26%
- 10%
- Lp(a) contributed more to 1-year ASCVD prediction than smoking, DM and other lipid parameters
Original publication
Elevated Lipoprotein(a) Independently Increases Risk for Short-Term Atherosclerotic Cardiovascular Events in Machine Learning Predictive Models. JACC Adv. Published 24 October 2025. doi:10.
Composed by AI from the published abstract and checked against it. How Briefs are made →