Predictive value of galectin-3 for incident cardiovascular disease and heart failure in the population-based FINRISK 1997 cohort

Objectives: Galectin-3 is an emerging biomarker playing an important, complex role in intracellular pathways of cardiovascular diseases and heart failure. We aimed therefore to investigate the predictive value of galectin-3 for incident cardiovascular disease and heart failure. Methods: Galectin-3 levels were measured in 8444 participants of the general population-based FINRISK 1997 cohort. Cox proportional hazards regression analyses, adjusting for traditional Framingham risk factors, prevalent valvular heart disease, eGFR (estimated glomerular filtration rate) as well as NT-proBNP, were used to examine the predictive power of galectin-3. Measurements of discrimination and reclassification using 10-fold cross-validation were performed to control for over-optimism. Cardiovascular death (CD), all-causemortality, myocardial infarction (MI), ischemic stroke (hemorrhagic strokes were excluded) and heart failure (HF) were used as endpoints. Results: During the follow-up of up to 15 years there were in total 1136 deaths from any cause, 383 cardiac deaths, 359 myocardial infarctions, 401 ischemic strokes and 641 cases of incident heart failure. Hazard ratios (HR) were statistically significant for all-cause mortality (1.12, p < 0.001), cardiac death (1.15, p = 0.033) and heart failure (1.10, p = 0.049). Statistical significance was lost when analyzing by gender except for all-cause mortality. No significant improvements were observed in model discrimination or overall reclassification upon inclusion of galectin-3. Compared to NT-proBNP, the predictive power of galectin-3 was weaker but both remained significant, independently of each other. Conclusion: Galectin-3 levels were predictive for future cardiovascular events but improvements in discrimination and reclassifications were modest. (C) 2015 Elsevier Ireland Ltd. All rights reserved.