BACKGROUND AND AIM: Many scoring systems are used to evaluate malnutrition, but there is no consensus on which scoring system would be more appropriate. We aimed to investigate the effect of malnutrition in patients with non-valvular atrial fibrillation (NVAF) and to compare three scoring systems. METHODS AND RESULTS: A total of 2592 patients with non-valvular AF from 35 different centers in Turkey were included in this prospective study. All participants were divided into two groups: 761 patients who died and 1831 patients who were alive. The malnutrition status of all participants was evaluated with three scoring systems. The primary outcome was all-cause mortality. The mean age of the population was 68.7 ± 11.1 years, and 55.5 % were female. According to Cox regression analysis, the geriatric nutritional risk index (GNRI) (HR = 0.989, 95 % CI: 0.982-0.997, p = 0.007), controlling nutritional status (CONUT) score (HR = 1.121, 95 % CI: 1.060-1.185, p <
0.002), and prognostic nutritional index (PNI) (HR = 0.980, 95 % CI: 0.962-0.999, p = 0.036) were found to be significant mortality predictors. ROC curve analysis indicated GNRI (AUC = 0.568), CONUT (AUC = 0.572), and PNI (AUC = 0.547) had moderate predictive values. Kaplan-Meier analysis showed that increasing the risk class based on GNRI (p <
0.002) and CONUT (p <
0.002) was associated with decreased survival, while PNI staging had no statistically significant effect (p = 0.266). CONCLUSIONS: Malnutrition, determined by three scoring systems, was found to be an independent predictor of all-cause mortality in NVAF patients. Nutritional examination may provide useful information for prognosis and risk stratification in patients with NVAF.