We read the letter by Kornej et al and thank them for their comments regarding our studies evaluating the relation between serum galectin-3 levels and atrial fibrillation.

As Kornej et al have mentioned in their comments, our study population is a highly selected sample of atrial fibrillation patients who do not have co-morbidities and therefore does not actually reflect the patient population encountered in the routine clinical practice. However, our primary aim in this study was to demonstrate the possible relation between serum galectin-3 levels and atrial fibrillation in a relatively small sample of patients devoid of possible confounders. Our findings may provide an insight for future studies in a larger group of patients with atrial fibrillation.

Obesity has been shown to be significantly associated with nonalcoholic fatty liver disease (NAFLD) and known to be an independent predictor for clinically significant liver fibrosis in patients with NAFLD. It has been previously reported that serum galectin-3 levels were elevated in obesity, and the main source of galectin-3 synthesis was the splanchnic region in obese patients with liver fibrosis. These data suggest that galectin-3 levels in patients with higher body mass index (BMI) may obscure the possible elevation related to atrial fibrillation. Most of the patients in our study had BMI values in the normal range because these patients were relatively younger patients without co-morbidities, who were possibly free of NAFLD and associated liver fibrosis. The nonsignificant correlation between BMI and serum galectin-3 levels in our study may be explained with this fact.

Establishing the exact causative or contributory role of galectin-3 in initiation and/or progression of atrial fibrillation merits further basic science studies to provide a direct link between galectin-3 and atrial pathophysiological changes in atrial fibrillation.