Objective To investigate the changes of growth differentiation factor-15 (GDF-15) level and echocardiographic parameters to evaluate the left ventricular remodeling in the patients with chronic heart failure (CHF), and determine the effect of GDF-15 levels on the severity of the remodeling. Methods A total of 75 CHF patients admitted in our department from June 2015 to June 2016 were enrolled in this study, and their clinical data were collected and retrospectively analyzed. According to the causes of heart failure, they were divided into 4 subgroups, that is, hypertensive heart disease subgroup (n=22), coronary heart disease subgroup (n=28), valvular disease subgroup (n=17) and dilated cardiomyopathy subgroup (n=8). At the same time, another 80 healthy individuals were recruited as control group. The serum GDF-15 and other related laboratory indicators were measured. The left ventricular end diastolic diameter (LVEDD), left atrial diameter (LAD) and left ventricular ejection fraction (LVEF) were measured by type M ultrasound examination. SPSS statistics 13.0 was used to analyze the data. The measurement data were expressed as mean±standard deviation ([AKx-D]±s), and the Student’s t test was used for the comparison between the 2 groups. The enumeration data were expressed as percentages, and Chi-square test was used for the inter-group comparison. The correlation between the indicators of the 2 groups was analyzed by Pearson linear correlation analysis. Results The level of GDF-15 was significantly higher in the CHF group than in the control group[(1195.48±43.26) vs (345.28±42.25) ng/L, P<0.05]. But there were no significant differences in GDF-15 level in the 4 subgroups (P>0.05). The GDF-15 level was significantly positively correlated with LVEDD (r=0.63, P<0.05) and LAD (r=0.68, P<0.05), and negatively correlated with LVEF (r=-0.47, P<0.05) and total cholesterol (r=-0.36, P<0.05). Conclusion GDF-15 level is obviously higher in the CHF patients. GDF-15 level may be used as an indirect indicator of ventricular remodeling in the patients.