Abstract: Objective To investigate the relationship between the expression of growth differentiation factor 15 (GDF-15) in patients with chronic heart failure treated with sacubatril/valsartan and its clinical prognosis. Methods A total of 84 patients with chronic heart failure whose conditions were under stable control during their hospitalization in the Affiliated Hospital of Xuzhou Medical University from February to December 2019 were selected. All patients were given oral sacubatril/valsartan on the basis of routine treatment after admission, and received consecutive treatment for 12 weeks after discharge. According to the presence of heart failure re-admission or all-cause death within 90 d after discharge, the patients were divided into two groups: an event-free group (n=68) and an event group (n=16). The levels of GDF-15 and N-terminal pro-B-type natriuretic peptide (NT-proBNP) were measured when the patients were discharged, during 12-week follow-up or during re-hospitalization for heart failure. The outcomes of heart failure re-hospitalization or all-cause death were recorded within 12 weeks after discharge. The ROC curve was plotted to evaluate the efficacy of GDF-15 and NT-proBNP levels at different time points to predict the adverse prognosis of heart failure patients in a short term. Results Patients in the event group produced remarkably a higher GDF-15 level at discharge, and higher GDF-15 and NT-proBNP levels 12 weeks after discharge than those in the non-event group (P<0.05). There was no statistical difference in the level of NT-proBNP at discharge between the event group and the non-event group (P>0.05). Patients in the event group produced remarkably higher levels of GDF-15 and NT-proBNP 12 weeks after discharge than those after discharge, while the levels of GDF-15 and NT-proBNP in the non-event group were markedly lower 12 weeks after discharge in the non-event group than those after discharge (P<0.05). According to GDF-15 level at discharge, the predicted AUC=0.865 for heart failure re-admission and death, with an optimal cut-off value of 3 357 ng/L, a sensitivity of 81.3%, and a specificity of 82.4%. Conclusions The level of plasma GDF-15 is correlated with early re-hospitalization and mortality of patients with chronic heart failure after discharge, where plasma GDF-15 level at discharge shows certain predictive value for adverse prognosis of patients with chronic heart failure.