Circulating biomarkers for coronary artery disease and adverse outcomes in atrial fibrillation

J. Kemper (Hamburg)1, F. Seum (Hamburg)1, D. Engler (Hamburg)2, P. Schlieker (Hamburg)1, A. Ohlrogge (Hamburg)3, C. Altmann (Hamburg)1, D. Csengeri (Hamburg)3, L. Fabritz (Hamburg)4, A. Ziegler (Hamburg)3, P. Kirchhof (Hamburg)4, S. Blankenberg (Hamburg)3, T. Zeller (Lübeck)5, R. Schnabel (Hamburg)2
1Universitäres Herz- und Gefäßzentrum Klinik für Kardiologie Hamburg, Deutschland; 2Universitäres Herz- und Gefäßzentrum Hamburg Allgemeine und Interventionelle Kardiologie Hamburg, Deutschland; 3Universitäres Herz- und Gefäßzentrum Hamburg Klinik für Kardiologie Hamburg, Deutschland; 4Universitäres Herz- und Gefäßzentrum Hamburg Klinik für Kardiologie Lübeck, Deutschland; 5Universitätsklinikum Schleswig-Holstein Institut für Kardiogenetik Lübeck, Deutschland

Aims 

Coronary artery disease (CAD) and atrial fibrillation (AF) frequently co-occur and share many risk factors. Circulating biomarkers provide quantifiable proxies for chronic disease processes that may differ between AF patients with and without CAD. We determined associations of emerging biomarkers, bone morphogenetic protein 10 (BMP10), fibroblast growth factor 23 (FGF23), angiopoietin-2 (Ang-2), insulin-like growth factor binding protein 7 (IGFBP7), and the established marker N-terminal pro-B-type natriuretic peptide (NT-proBNP) with CAD prevalence and prognosis in a contemporary cohort of patients with AF.

Methods and Results 

We prospectively enrolled 791 cardiology patients with AF at the University Heart and Vascular Center Hamburg. Baseline plasma concentrations of BMP10, FGF23, Ang-2, IGFBP7, and NT-proBNP were measured using validated assays. Associations with CAD were tested by multivariable logistic regression using log-transformed concentrations. Prediction of MACE (myocardial infarction, stroke, heart failure, and all-cause mortality) was assessed by Cox regression over a median 5-year follow-up. After adjustment for confounders, BMP10 was lower in patients with AF and CAD than in patients with AF but without CAD (odds ratio (OR) 0.823; 95% confidence interval (CI) 0.69–0.99). NT-proBNP was higher with CAD than without (OR 2.11; 95% CI: 1.70–2.63). All biomarkers were significantly associated with cardiovascular events in the CAD subgroup, with the strongest risk increase for NT-proBNP (hazard ratio 2.07, 95% CI 1.50–2.86). 

Conclusions

Low BMP10 and high NT-proBNP could be useful to identify patients with AF and CAD.  Our results suggest that disease processes related to BMP10 could be more active in patients with AF without CAD, while disease processes related to NT-proBNP could be more active in patients with AF and CAD. These findings require validation in larger cohorts and call for further research into the disease processes linked to BMP10 and NT-proBNP in patients with AF.