• exercise stress echocardiography
• heart failure
• left ventricular adverse remodeling
• peak cardiac power output-to-mass

Background. It is well known that the left ventricular (LV) remodeling process in heart failure (HF) patients has a significant prognostic value, therefore we wanted to focalize our attention on investigating its possible predictors factors.
Aim. The aim of our study was to test whether exercise stress echocardiography (ESE)-derived variables, including peak cardiac power output-to-mass (CPOM), result useful predictors of LV adverse remodeling in patients with chronic HF with reduced ejection fraction (EF) (HFrEF).
Materials and methods. Our study group included 133 consecutive patients referred to the Cardiovascular Division of Pisa between 2014 and 2015 whit diagnosis of HFrEF, defined according to current guidelines. All patients underwent transthoracic 2D and Doppler echocardiographic examination at baseline, during bicycle semi-supine exercise (peak exercise) and at 6-months follow-up. The LV adverse remodeling was defined as a ≥10% increase in end systolic volume (ESV) at 6-months follow-up (pattern A), whereas no LV remodeling was defined if the volume changes were <10% (pattern B). LV cardiac power output (CPO) was calculated as the product of a constant (K=0.222) with cardiac output (CO) and mean blood pressure (MBP). CPOM was obtained by divided CPO by LV mass to convert the units to Watts/100 g. Patients were followed-up for an average time of 38 months for the primary end point of all-cause mortality and secondary end point of all-cause mortality and hospitalizations for worsening HF.
Results. Twenty-nine patients (21.8%) died and twenty-three patients (17.3%) were hospitalized for worsening HF. In time-indipendent analyses, the increase of ESV ≥10% at 6-months follow-up exhibited the highest accuracy in predicting event-free survival both for primary and secondary end-points (respectively, AUC=0.74, 95% CI 0.65-0.84, p=0.05 and AUC=0.72, 95% CI 0.64-0.80, p=0.04) compared to the other remodeling classification based on EDV value, LV mass index and relative wall thickness. Also in time-dependent analysis, the remodeling pattern defined on the base of ESV value at 6-months follow-up exhibited the best prognostic value, both for primary and secondary end points (respectively, log-rank chi square: 29.6, p<0.0001 and log-rank chi square: 34.3, p<0.0001). At the multivariate logistic regression analysis, peak CPOM and peak LV ESV resulted to be the only independent predictors of LV A pattern of remodeling (respectively, OR 0.03, 95% CI 0.002-0.31, p=0.0037 and 1.01, 95% CI 1.01-1.03, p=0.0398).
Conclusions. The adverce remodeling defined as an increase of LV ESV ≥10% at 6-months follow-up has the best predictive accuracy for long-term survival and hospitalizations for worsening HF. The CPOM at the peak of exercise is a useful predictor of the risk of adverse LV remodeling in patients with HFrEF.