Diagnostic value of FNDC5 in patients with subclinical diabetic cardiomyopathy

Objective: To estimate the diagnostic value of fibronectin type Ⅲ-domain containing protein 5 (FNDC5) in subclinical diabetic cardiomyopathy. Methods: A total of 94 patients with type 2 diabetes (T2DM), who were hospitalized from April 2018 to June 2019 in the Third Affiliated Hospital of Soochow University, were enrolled in this study. Patients were divided into T2DM with cardiac dysfunction (subclinical DCM) group (n=47) and T2DM without cardiac dysfunction (non-DCM) group (n=47) according to echocardiography and gated myocardial perfusion imaging results. Basic clinical data and serum FNDC5 level were compared between the two groups. Logistic regression analysis was used to establish predicting models and the diagnostic efficiency of established models was compared by ROC curve analysis. Results: Compared to non-DCM group, patients in subclinical DCM group were older, with longer duration of diabetes, and had higher levels of glycosylated hemoglobin A1c (HbA1c), total cholesterol (TC), triglyceride (TG) and low-density lipoprotein cholesterol (LDL-C) (all P<0.05). Serum FNDC5 level was significantly lower in subclinical DCM group than in non-DCM group (P<0.001). FNDC5 level was positively correlated with ventricular septal e'(r=0.451,P=0.005), mitral valve e'(r=0.291,P<0.001), the ratio of peak early diastolic trans-mitral flow velocity (E) to peak late diastolic trans-mitral flow velocity (A)(r=0.490,P=0.002), while negatively correlated with A(r=-0.399,P<0.001), the average ratio of E/e'(r=-0.490,P<0.001), tricuspid regurgitation velocity(r=-0.567,P<0.001), left atrial volume index(r=-0.491,P<0.001). Univariate ROC analysis showed that the diagnostic efficacy of FNDC5(AUC=0.940,95%CI 0.897-0.982)was superior to age(AUC=0.639,95%CI 0.523-0.752), diabetic duration(AUC=0.663,95%CI 0.555-0.772), HbA1c(AUC=0.740,95%CI 0.638-0.839), TG(AUC=0.661,95%CI 0.547-0.776), TC(AUC=0.675,95%CI 0.563-0.788)and LDL-C(AUC=0.644,95%CI 0.532-0.756). Model 1 was established with subclinical DCM as dependent variable, age, diabetic duration, TG, TC, LDL-C and HbA1c as independent variables. Model 2 was established by adding FNDC5 as independent variable on the basis of model 1. Diagnostic efficacy for subclinical DCM was compared between the two models by ROC analysis. The diagnostic efficiency was better with model 2 (AUC=0.980) than with model 1 (AUC=0.879, P<0.001). When sensitivity was set at 0.617, the specificity of model 2 was higher than that of model 1(0.979 vs. 0.936). When sensitivity was set at 0.532, the sensitivity of model 2 was higher than that of model 1 (1.000 vs. 0.915). Conclusions: Our findings suggest that serum FNDC5 could be used as a novel biomarker for the diagnosis of subclinical DCM.