Exploring the incremental utility of circulating biomarkers for robust risk prediction of incident atrial fibrillation in European cohorts using regressions and modern machine learning methods

• DOI: 10.1093/europace/euac260
• Publication Date: 2023-01-07T21:21:51Z
• Authors: Betül Toprak; Stephanie Brandt; Jan Brederecke; Francesco Gianfagna; Julie K K Vishram...; Francisco M Ojeda; Simona Costanzo; Christin S Börschel; Stefan Söderberg; Ioannis Katsoularis; Stephan Camen; Erkki Vartiainen; Maria Benedetta D...; Jukka Kontto; Martin Bobak; Ellisiv B Mathiesen; Allan Linneberg; Wolfgang Koenig; Maja-Lisa Løchen; Augusto Di Castel...; Stefan Blankenberg; Giovanni de Gaetano; Kari Kuulasmaa; Veikko Salomaa; Licia Iacoviello; Teemu Niiranen; Tanja Zeller; Renate B Schnabel
• Keywords: Interquartile range; Rotterdam Study

To identify robust circulating predictors for incident atrial fibrillation (AF) using classical regressions and machine learning (ML) techniques within a broad spectrum of candidate variables.In pooled European community cohorts (n = 42 280 individuals), 14 routinely available biomarkers mirroring distinct pathophysiological pathways including lipids, inflammation, renal, myocardium-specific markers (N-terminal pro B-type natriuretic peptide [NT-proBNP], high-sensitivity troponin I [hsTnI]) were examined in relation to AF Cox ML methods. Of individuals (21 843 women [51.7%]; median [interquartile range, IQR] age, 52.2 [42.7, 62.0] years), 1496 (3.5%) developed during follow-up time 5.7 years. In multivariable-adjusted Cox-regression analysis, NT-proBNP was the strongest predictor [hazard ratio (HR) per standard deviation (SD), 1.93 (95% CI, 1.82-2.04); P < 0.001]. Further, hsTnI [HR SD, 1.18 1.13-1.22); 0.001], cystatin C 1.16 1.10-1.23); C-reactive protein 1.08 1.02-1.14); 0.012] correlated positively with AF. Applying various techniques, high inter-method consistency selected variables observed. identified as blood-based marker highest predictive value Relevant clinical use antihypertensive medication, body mass index.Using different variable selection procedures methods, consistently remained ranked before cardiovascular risk factors. The benefit these findings identifying at-risk targeted screening needs be elucidated tested prospectively.