ORIGINAL RESEARCH ARTICLEBACKGROUND: Poor fitness in middle age is a risk factor for heart failure, particularly heart failure with a preserved ejection fraction. The development of heart failure with a preserved ejection fraction is likely mediated through increased left ventricular (LV) stiffness, a consequence of sedentary aging. In a prospective, parallel group, randomized controlled trial, we examined the effect of 2 years of supervised highintensity exercise training on LV stiffness. METHODS:Sixty-one (48% male) healthy, sedentary, middle-aged participants (53±5 years) were randomly assigned to either 2 years of exercise training (n=34) or attention control (control; n=27). Right heart catheterization and 3-dimensional echocardiography were performed with preload manipulations to define LV end-diastolic pressure-volume relationships and Frank-Starling curves. LV stiffness was calculated by curve fit of the diastolic pressure-volume curve. Maximal oxygen uptake (Vo . 2 max) was measured to quantify changes in fitness. RESULTS: Fifty-three participants completed the study. Adherence to prescribed exercise sessions was 88±11%. Vo CONCLUSIONS:In previously sedentary healthy middle-aged adults, 2 years of exercise training improved maximal oxygen uptake and decreased cardiac stiffness. Regular exercise training may provide protection against the future risk of heart failure with a preserved ejection fraction by preventing the increase in cardiac stiffness attributable to sedentary aging.

Background Predicting mortality is important in patients with heart failure (HF). However, current strategies for predicting risk are only modestly successful, likely because they are derived from statistical analysis methods that fail to capture prognostic information in large data sets containing multi‐dimensional interactions. Methods and results We used a machine learning algorithm to capture correlations between patient characteristics and mortality. A model was built by training a boosted decision tree algorithm to relate a subset of the patient data with a very high or very low mortality risk in a cohort of 5822 hospitalized and ambulatory patients with HF. From this model we derived a risk score that accurately discriminated between low and high‐risk of death by identifying eight variables (diastolic blood pressure, creatinine, blood urea nitrogen, haemoglobin, white blood cell count, platelets, albumin, and red blood cell distribution width). This risk score had an area under the curve (AUC) of 0.88 and was predictive across the full spectrum of risk. External validation in two separate HF populations gave AUCs of 0.84 and 0.81, which were superior to those obtained with two available risk scores in these same populations. Conclusions Using machine learning and readily available variables, we generated and validated a mortality risk score in patients with HF that was more accurate than other risk scores to which it was compared. These results support the use of this machine learning approach for the evaluation of patients with HF and in other settings where predicting risk has been challenging.

No abstract