Abstract Background Hypertension is the leading cause for the development of heart failure (HF). Increased hemodynamic load, including mechanical stretch and neurohumoral factors, is able to trigger hypertrophic growth of cardiac myocytes. Although hypertensive HF is prevalent, there is no useful biomarker to identify HF due to chronic hypertension. Aims To identify plasma markers associated with incidence of hypertensive HF. Methods Circulating levels of 149 proteins were measured by proximity extension assay at baseline examination in 4469 individuals from the Malmö Diet and Cancer study. Protein levels were compared to stretch-activated gene expression changes in cultured neonatal rat ventricular myocytes (NRVM) in response to 1, 4, 12, 24 or 48 hours of cyclic mechanical stretch. Association between plasma proteins level and HF incidence and hypertension was studied using respectively Cox proportional hazards model and binary logistic regressions. Results After Bonferroni correction, 44 circulating proteins were significantly differentially expressed in individuals who developed HF during follow-up versus controls (P<3.4E-4). Out of these, 5 proteins (Interleukin-6 (IL-6), Growth Differentiation Factor-15 (GDF15), Interleukin-1 Receptor-Like-1 (ST2), Plasminogen Activator Urokinase Receptor (U-PAR), Transforming Growth Factor-α (TGF-α)), corresponding mRNA levels were upregulated by mechanical stretch in NRVM at all time points (P<0.05). Similar upregulation for the 5 proteins was shown in hypertensive versus normotensive individuals (P≤8.05E-4). In a model with all 5 proteins entered simultaneously, GDF15 and IL-6 were predictive of incident HF after adjustment for age, sex and NT-BNP levels (205 events; hazard ratio [HR] per SD increment of protein: HR=1.29, CI=1.05–1.58, P=0.013 and HR=1.16, CI=1.02–1.33, P=0.028). Using the same model, IL-6 but not GDF15 associated with hypertension (Odds ratio [OR] per SD increment of IL-6: OR=1.18, CI=1.09–1.27, P=3.3E-5). In hypertensive individuals GDF15 and IL-6 were individually predictive of future HF after adjustment for age, sex, NT-BNP levels, smoking, BMI and diabetes (183 events; HR=1.36, CI=1.16–1.60, P=1.64E-4 and HR=1.21, CI=1.05–1.40, P=0.008). Furthermore, in these hypertensive individuals, GDF15 and IL-6 were predictive of HF in a model with IL-6, GDF15, ST2 and TGF-α entered simultaneously after adjustment for age, sex and NT-BNP levels (176 events; HR=1.36, CI=1.13–1.64, P=0.001 and HR=1.16, CI=1.01–1.34, P=0.041). Conclusions Circulating levels of IL-6 and GDF15 might be used as NT-BNP independent biomarkers for HF development in hypertensive patients. Acknowledgement/Funding Påhlsson, Crafoord, Lundström, Åke Wiberg, Royal Physiographic Society and the Swedish Foundation for Strategic Research for IRC15-0067
Read full abstract