METHODS: BMP10 levels were quantified in 2085 chronic HF patients from the European BIOlogy Study to TAilored Treatment in Chronic Heart Failure (BIOSTAT-CHF) cohort and in 1487 patients from the Scottish validation cohort. Multivariable linear regression identified independent associates of BMP10. Proteomic analysis of 6369 proteins with subsequent gene set enrichment analysis was used to explore biological pathways associated with elevated BMP10. Cox proportional hazards models adjusting for established risk factors were used to associate BMP10 levels with clinical outcomes, including all-cause mortality and HF hospitalisation.

RESULTS: In a multivariable model including clinical and echocardiographic parameters, log-transformed and standardised BMP10 levels were significantly associated with a history of atrial fibrillation (Sβ=0.419; p<0.001), and with echocardiographic features reflecting atrial stress, such as increased left atrial diameter (Sβ=0.075; p=0.048). By contrast, these were not among the strongest associates of NT-proBNP levels. Gene set enrichment analysis showed significant overrepresentation in pathways of muscle contraction and extracellular matrix organisation. Higher log-transformed and standardised BMP10 levels predicted a combined outcome of 2-year all-cause mortality and HF rehospitalisation (HR=1.10, 95% CI=1.02-1.19), with the validation cohort yielding comparable results.