With patients referred to emergency departments (EDs) for acute dyspnea, emergency physicians should consider all possible diagnoses and assess patients' risk stratification. Copeptin has been shown to have prognostic power for subsequent events, such as death and rehospitalization in patients admitted for dyspnea. The aim of this study was to investigate prognostic role of copeptin variations during hospitalization in patients admitted for dyspnea.We conducted a prospective, multicentric, observational study in acute dyspneic patients in three ED centers in Italy. Clinical data and copeptin assessments were performed at admission, and at discharge. A 90-day follow-up was performed.A total of 336 patients were enrolled, and on the basis of final diagnosis distinguished into two groups: acute heart failure and no acute heart failure. Compared to a control group, in all studied population copeptin values at admission resulted in a significantly (p<0.001) higher median (maximum-minimum): 31 (0-905) versus 8 (0-13) pmol/L. Median copeptin value at admission was 42 (0-905) pmol/L in acute heart failure patients and 20 (0-887) pmol/L in no acute heart failure, respectively (p<0.001). In all studied patients and in each group copeptin at admission and discharge showed significant predictive value for 90-day events (p<0.001). Furthermore, in all patients population and in both groups Δ copeptin values from admission to discharge also showed significant predictive value for 90-day events (p<0.001).In patients admitted for acute dyspnea, admission, discharge and Δ copeptin variations have significant prognostic value from subsequent 90-day death and rehospitalization.