Chao-Hung Wang1, Mei-Ling Cheng2, Min-Hui Liu3, Li-Tang Kuo4, Ming-Shi Shiao2. 1. Heart Failure Research Center, Division of Cardiology, Department of Internal Medicine, Chang Gung Memorial Hospital, Keelung 20401, Taiwan; Chang Gung University College of Medicine, Taoyuan 33302, Taiwan. Electronic address: bearty54@gmail.com. 2. Healthy Aging Research Center, Chang Gung University, Tao-Yuan 33302, Taiwan; Metabolomics Core Laboratory, Chang Gung University, Tao-Yuan 33302, Taiwan; Department of Biomedical Sciences, College of Medicine, Chang Gung University, Tao-Yuan 33302, Taiwan. 3. Heart Failure Research Center, Division of Cardiology, Department of Internal Medicine, Chang Gung Memorial Hospital, Keelung 20401, Taiwan; Chang Gung University College of Medicine, Taoyuan 33302, Taiwan; Department of Nursing, National Yang-Ming University, Taipei 11221, Taiwan. 4. Heart Failure Research Center, Division of Cardiology, Department of Internal Medicine, Chang Gung Memorial Hospital, Keelung 20401, Taiwan; Chang Gung University College of Medicine, Taoyuan 33302, Taiwan.
Abstract
BACKGROUND: Metabolic profiles have been shown to provide prognostic information in patients with heart failure (HF). Galectin-3 (Gal-3), indicating cardiac fibrosis, is a documented biomarker of prognosis in HF. It is unknown whether metabolic profiles provide prognostic value better than Gal-3. METHODS AND RESULTS: This study analyzed 212 hospitalized HF patients, measuring metabolic score (composed by butyrylcarnitine, dimethylarginine/arginine ratio, spermidine, and total essential amino acids) and Gal-3. Endpoints were composite events (death/HF-related re-hospitalization). The median of metabolic scores and Gal-3 levels were 3.1 (1.3-5.2) and 17.8ng/mL (4.7-100ng/mL), respectively. Patients with higher metabolic scores had worse functional classes, higher atrial fibrillation incidences, levels of Gal-3 and B-type natriuretic peptide (BNP), but lower albumin levels and glomerular filtration rate. Correlations of metabolic score to Gal-3 and BNP were significant, but weak (r=0.34 and 0.41, respectively, both p<0.001). During a follow-up period of 4.2±1.4 years, there were 91 (42.9%) composite events. In univariate analysis, significant predictors of composite events were age, functional class, atrial fibrillation, levels of hemoglobin, log (Gal-3), log (BNP) and metabolic score. In multivariable analysis, adjusted for above variables, metabolic score remained a strong predictor of combined endpoints (hazard ratio=2.596, 95% confidence interval=1.649-4.087, p<0.001). C-statistics for the prediction of composite events significantly increased when metabolic score was incorporated into the model with established risk factors, BNP and Gal-3 [0.76 (0.70-0.83) vs. 0.66 (0.58-0.74), p=0.032]. CONCLUSIONS: Metabolic profile provides prognostic value for HF patients better than Gal-3.
BACKGROUND: Metabolic profiles have been shown to provide prognostic information in patients with heart failure (HF). Galectin-3 (Gal-3), indicating cardiac fibrosis, is a documented biomarker of prognosis in HF. It is unknown whether metabolic profiles provide prognostic value better than Gal-3. METHODS AND RESULTS: This study analyzed 212 hospitalized HF patients, measuring metabolic score (composed by butyrylcarnitine, dimethylarginine/arginine ratio, spermidine, and total essential amino acids) and Gal-3. Endpoints were composite events (death/HF-related re-hospitalization). The median of metabolic scores and Gal-3 levels were 3.1 (1.3-5.2) and 17.8ng/mL (4.7-100ng/mL), respectively. Patients with higher metabolic scores had worse functional classes, higher atrial fibrillation incidences, levels of Gal-3 and B-type natriuretic peptide (BNP), but lower albumin levels and glomerular filtration rate. Correlations of metabolic score to Gal-3 and BNP were significant, but weak (r=0.34 and 0.41, respectively, both p<0.001). During a follow-up period of 4.2±1.4 years, there were 91 (42.9%) composite events. In univariate analysis, significant predictors of composite events were age, functional class, atrial fibrillation, levels of hemoglobin, log (Gal-3), log (BNP) and metabolic score. In multivariable analysis, adjusted for above variables, metabolic score remained a strong predictor of combined endpoints (hazard ratio=2.596, 95% confidence interval=1.649-4.087, p<0.001). C-statistics for the prediction of composite events significantly increased when metabolic score was incorporated into the model with established risk factors, BNP and Gal-3 [0.76 (0.70-0.83) vs. 0.66 (0.58-0.74), p=0.032]. CONCLUSIONS: Metabolic profile provides prognostic value for HF patients better than Gal-3.