Protein energy wasting, malnutrition-inflammation complex syndrome, and frailty are thought to affect mortality in chronic dialysis patients. We performed a historical cohort study to investigate effects of the nutritional status on cardiovascular （CV） and non-CV mortalities as well as CV events using Geriatric Nutritional Risk Index （GNRI）. We registered patients undergoing maintenance hemodialysis （HD）therapy from January to March in 2006 to analyze mortality and causes of death in 7-years observation. A Kaplan-Meyer curve showed that the survival rate tends to be lower in 102 patients with GNRI<92, compared to 171 patients with GNRI ≧92. In a multivariate Cox proportional hazard analysis, the GNRI was an independent predictor for all-cause mortality with hazard ratio of 0.960 （95% confident interval: 0.928-0.993） after adjustment with age, gender, dialysis duration, blood access, the presence of diabetes mellitus, serum levels of calcium, phosphate, alkaline phosphatase, parathyroid hormone and C-reactive protein, and drug use of active vitamin D analog and non-calcium containing phosphate binders. The GNRI was not associated with CV death but with non-CV death after adjustment with the covariates. In parallel with these findings, there was no significant association between the GNRI and CV events. Nutritional status was significantly associated with non-CV death but not with CV death and CV events at least in our maintenance hemodialysis patients.