Synapse
⌘+K
Synapse
PulseExploreClubsResearchersJournals
Instagram
HomeClubsExplore
June 27, 2026EP EuropaceOpen Access

Novel 18-protein score independently predicts ~69% greater incident AF risk per SD in HF patients.

View Full Paper
Ask AI
Bookmark
Share

Why the study?

Does an 18-protein plasma risk score improve the prediction of incident atrial fibrillation over 4 years compared to established clinical and genomic models in patients with heart failure?

Population

3,289 patients with heart failure in the derivation cohort and 41,049 in the external validation cohort…

Comparison

18-protein plasma risk score derived via machine… vs Established clinical model and genomic model

Design

Cohort

Follow-up

4 years

Key result

A novel 18-protein plasma risk score predicted 4-year incident atrial fibrillation in heart failure patients independent of clinical profile (PRSD 1.69; 95% CI 1.40-2.1; p<0.0001).

Authors

MHM T HuttelmaierSZS ZeidAGA Gieswinkel

Discussion

Loading...

Member takes

Overview

AF protein score may refine incident AF prediction; leaves open clinical utility pending prospective validation.

Key Points

  • This study aimed to develop and validate a plasma protein risk score to predict atrial fibrillation in heart failure patients.
  • Analyzed data from the MyoVasc HF cohort (n=3,289) to identify incident AF over 4 years.
  • Utilized proximity extension assay to analyze 536 proteins and derive an AF-protein score using elastic net-regularized logistic regression.
  • Conducted external validation with UK Biobank Resource for comprehensive analysis.
  • Incident AF occurred in 6% of participants (n=126; 69% men; mean age 68.6 years).
  • The AF protein score showed significant predictive capability (PRSD 1.69, 95% CI: 1.40; 2.1, p<0.0001, AUC 0.80).
  • In external validation, the protein score had the strongest association with AF risk (PRSD 1.42, 95% CI 1.33 - 1.51) compared to traditional models.

Study Design

Type

Cohort (n=3,289)

Multicenter

Yes

Structured PICO

Does an 18-protein plasma risk score improve the prediction of incident atrial fibrillation over 4 years compared to established clinical and genomic models in patients with heart failure?

P
Population
3,289 patients with heart failure (mean age 68.6 years, 31% female) in the derivation cohort and 41,049 in the validation cohort, followed for 4 years for incident atrial fibrillation.
E
Exposure
18-protein plasma risk score (AF-protein-score) derived via machine learning
C
Comparator
Established clinical model (CHARGE-AF) and genomic model (AF polygenic risk score [AF-PRS])
O
Outcome
Incident atrial fibrillation over 4 yearshard clinical

Main Result

Effect estimate: PRSD 1.69 (95% CI 1.40-2.1)

p-value: p=<0.0001

A novel 18-protein plasma risk score improves the prediction of 4-year incident atrial fibrillation in heart failure patients compared to established clinical and genetic models.

Cite This Study

Huttelmaier et al. (2026) conducted a cohort in Heart failure (n=3,289). AF protein score vs. CHARGE-AF and AF-PRS models was evaluated on Incident atrial fibrillation over 4 years (PRSD 1.69, 95% CI 1.40-2.1, p=<0.0001). A novel 18-protein plasma risk score predicted 4-year incident atrial fibrillation in heart failure patients independent of clinical profile (PRSD 1.69; 95% CI 1.40-2.1; p<0.0001).

synapsesocial.com/papers/6a3f97bb125782b61d865934https://doi.org/10.1093/europace/euag105.376
View Full Paper
Ask AI
Bookmark
Share