Patients with end-stage renal disease have a higher risk of cardiovascular morbidity and mortality. In this study, we investigated the predictive ability of a combination of cardiothoracic ratio (CTR) and aortic arch calcification (AoAC) for overall and cardiovascular mortality in patients receiving hemodialysis. We also evaluated the predictive power of AoAC and CTR for clinical outcomes. A total of 365 maintenance hemodialysis patients were included, and AoAC and CTR were measured using chest radiography at enrollment. We stratified the patients into four groups according to a median AoAC score of three and CTR of 50%. Multivariable Cox proportional hazards analysis was used to identify the risk factors of mortality. The predictive performance of the model for clinical outcomes was assessed using the χ2 test. Multivariable analysis showed that, compared to the AoAC < 3 and CTR < 50% group, the AoAC ≥ 3 and CTR < 50% group (hazard ratio [HR], 4.576; p < 0.001), and AoAC ≥ 3 and CTR ≥ 50% group (HR, 5.912; p < 0.001) were significantly associated with increased overall mortality. In addition, the AoAC < 3 and CTR ≥ 50% (HR, 3.806; p = 0.017), AoAC ≥ 3 and CTR < 50% (HR, 4.993; p = 0.002), and AoAC ≥ 3 and CTR ≥ 50% (HR, 8.614; p < 0.001) groups were significantly associated with increased cardiovascular mortality. Furthermore, adding AoAC and CTR to the basic model improved the predictive ability for overall and cardiovascular mortality. The patients who had a high AoAC score and cardiomegaly had the highest overall and cardiovascular mortality among the four groups. Furthermore, adding AoAC and CTR improved the predictive ability for overall and cardiovascular mortality in the hemodialysis patients.