Abstract Background and Aims Cardiovascular disease (CVD) is the most common cause of death in patients on hemodialysis (HD). Indeed, left ventricular (LV) dysfunction, symptomatic congestive heart failure and arrhythmia are independent predictors of poor survival and hospitalization in these patients. Among the patients who initiated HD, the proportions of the elderly patients aged >65 years are increasing worldwide. According to the Korean Renal Data System, the percentage of incident HD patients aged>65 years increased from 11% in 1990 to 53% in 2022. In addition, the prevalence of CVD increases with age. Therefore following these trends, investigating the impact of heart function on prognosis in elderly patients on HD is important for predicting prognosis and making appropriate treatment plans. This study aims to investigate the all-cause mortality in elderly incident HD patients according to heart function and presence of arrhythmia. Method 1,002 incident HD patients aged >70 years were recruited from a retrospective cohort of the Korean Society of Geriatric Nephrology. The primary outcome is all-cause mortality (6 months, overall) according to the LV function and presence of arrhythmia. To analyze, patients were divided into 6 groups according to the LV function (LV ejection fraction [LVEF]≥50%, 40≤LVEF<50%, LVEF<40%) and the presence of arrhythmia. The secondary outcome is all-cause mortality (6 months, overall) according to the LV function or presence of arrhythmia. To analyze, patients were divided into 3 groups according to the LV function and separately divided into 2 groups according to the presence of arrhythmia. Cox regression univariate and multivariate analyses was used. Results 1,002 patients were divided into 6 groups (LVEF≥50% without arrhythmia [N = 707, 70.6%], 40≤LVEF<50% without arrhythmia [N = 71, 7.1%], LVEF<40% without arrhythmia [N = 66, 6.6%], LVEF≥50% with arrhythmia [N = 124, 12.4%], 40≤LVEF<50% with arrhythmia [N = 15, 1.5%], LVEF<40% with arrhythmia [N = 19, 1.8%]). There were no differences of 6-month all-cause mortality among the groups. Compared with LVEF≥50% without arrhythmia group, 40≤LVEF<50% without arrhythmia (hazard ratio [HR], 1.59; 95% confidence interval [95% CI], 1.21-2.09; P = 0.001), LVEF≥50% with arrhythmia (HR, 1.39; 95% CI, 1.12-1.74, P = 0.003) and LVEF<40% with arrhythmia groups (HR, 1.95; 95% CI, 1.19-3.17, P = 0.007) showed higher all-cause mortality of overall period. According to the LVEF groups, there were no differences of 6-month all-cause mortality among the groups. Compared with LVEF≥50% group, 40≤LVEF<50% group showed higher all-cause mortality of overall period (HR, 1.51; 95% CI, 1.18-1.94, P = 0.001). According to the arrhythmia groups, there were no differences of 6-month all-cause mortality between the groups. The presence of arrhythmia was associated with higher all-cause mortality of overall period (HR, 1.39; 95% CI, 1.15-1.69, P = 0.001). Conclusion Decreased heart function and presence of arrhythmia have negative impact on all-cause mortality of overall period, but not on 6-month all-cause mortality in elderly patients on incident HD. Therefore, evaluation of the heart function and detection of arrhythmia in elderly patients who start HD and providing the appropriate treatment to these patients would be helpful for improving the prognosis.