Left atrial (LA) dimension ≥50 mm had approximately four times the risk of developing atrial fibrillation (AF). The aim of this study was to investigate whether the application of clinical and echocardiographic parameters could differentiate between the patients having severely dilated left atrium with and without AF. This retrospective cross-sectional study enrolled consecutive patients with LA dimension ≥50 mm and divided them into three groups: no AF (no-AF), paroxysmal AF (PAF) and non-paroxysmal AF (non-PAF) groups. For PAF and non-PAF groups, all patients underwent radiofrequency ablation, and the echocardiographic parameters were obtained on the next day after ablation. Our study population comprised 160 patients, including 80, 53, and 27 patients in the non-AF, PAF and non-PAF groups, respectively. The no-AF group had a significantly higher body mass index (kg/m2) (29.31±6.27, 27.58±4.12 and 26.57±2.81, P=0.01), and a higher prevalence of diabetes mellitus (DM) [31 (38.80%), 13 (25.00%) and 4 (14.80%), P=0.01] and hypertension [67 (83.80%), 34 (65.40%), and 19 (70.40%), P=0.04], but a lower prevalence of rheumatic heart disease (RHD) [3 (3.80%), 6 (11.50%) and 5 (18.50%), P=0.02] and sick sinus syndrome [0 (0.00%), 6 (11.50%) and 4 (14.80%), P=0.045]. Echocardiographic studies showed that the non-AF group had significantly smaller LA minimal volume index (24.89±9.74, 34.06±19.38 and 42.83±17.44 mL/m2, P<0.01), higher LA emptying fraction (51.99%±13.97%, 38.40%±15.96% and 33.89%±10.73%, P<0.01), longitudinal strain (23.87%±7.72%, 17.11%±8.52% and 12.38%±4.28%, P<0.01) and strain rate than the AF groups. The multivariate analysis showed that the late diastolic component of LA strain rate was the only independent factor associated with the presence of AF (odds ratio, 21.69; 95% CI, 9.77-48.13, P<0.01). LA function plays an important role in the absence of AF in patients with LA dimension ≥50 mm; the late diastolic component of LA strain rate was the only independent variable on multivariate analysis.