Reply to the letter "Prognostic value of late gadolinium enhancement in cardiomyopathy: Causative risk factor or surrogate marker?"

Cardiac magnetic resonance and galectin-3 level as predictors of prognostic outcomes for non-ischemic cardiomyopathy patients

This study was aimed at determining whether late gadolinium enhancement (LGE) in conjunction with Galectin-3 (Gal-3) level offered more precise prognosis of non-ischemic cardiomyopathy (NICM) in comparison to LGE alone. Results of LGE and Gal-3 expression in 192 patients with NICM, including 85 subjects with dilated cardiomyopathy (DCM) and 107 with hypertrophic cardiomyopathy (HCM), were examined. As suggested by the characteristics of LGE and Gal-3 levels, patients were divided into four groups: LGE positive + low Gal-3 (n = 10 for DCM, n = 15 for HCM), LGE positive + high Gal-3 (n = 25 for DCM, n = 51 for HCM), LGE negative + low Gal-3 (n = 32 for DCM, n = 29 for HCM), LGE negative + high Gal-3 (n = 18 for DCM, n = 12 for HCM). Primary endpoints over the follow-up period included major adverse cardiac events (MACEs). Kaplan-Meier survival analysis and univariate Cox proportional hazard models were used to analyze the survival status of patients with NICM. The optimal cut-off value of Gal-3 level for two types of NICM was determined by receiver operating characteristic analysis (13.38 U/L for DCM and 14.40 U/L for HCM). The combination of LGE and Gal-3 levels offered a more significant prognostic value than using LGE alone for both DCM and HCM (DCM P = 0.001 < 0.012; HCM P = 0.037 < 0.040). Moreover, the Cox proportional hazard model suggested that both LGE status [Hazard ratio (HR) = 2.62, P = 0.017] and Gal-3 level (HR = 1.16, P = 0.013) were significant predictors of MACEs in DCM, while they did not appear to have significant prognostic values for HCM (P = 0.06 and 0.64). Furthermore, the multivariate analysis only confirmed LGE as an independent element in predicting prognosis of DCM (HR = 12.19, P = 0.026). In conclusion, LGE status was an independent indicator of DCM prognosis, yet the insignificant role of LGE in HCM prognosis could be limited by sample size.