BackgroundGeriatric nutritional risk index (GNRI) is a validated nutritional assessment method, and lower GNRI values are closely associated with adverse clinical outcomes in dialysis patients. This study investigated the impact of changes in GNRI during the first year of dialysis on cardiovascular outcomes in incident peritoneal dialysis (PD) patients.MethodsWe reviewed medical records in 133 incident PD patients to determine GNRI at the start of PD and after 12 months. Patients were categorized into improved (delta GNRI > 0) and worsening/stationary (delta GNRI ≤ 0) groups. The primary outcome was major adverse cardiac and cerebrovascular events (MACCEs).ResultsDuring a mean follow-up of 51.1 months, the primary outcome was observed in 42 patients (31.6%). The baseline GNRI at PD initiation was not significantly associated with MACCEs (log-rank test, P = 0.40). However, the cumulative event-free rate was significantly lower in the worsening or stationary GNRI group than in the improved group (log-rank test, P = 0.004). Multivariate Cox analysis revealed that a worsening or stationary GNRI was independently associated with higher risk for MACCEs (hazard ratio, 2.47; 95% confidence interval, 1.15–5.29; P = 0.02). In subgroup analysis, patients with worsening or stationary GNRI were at significantly greater risk for MACCEs in both the lower (P = 0.04) and higher (P = 0.01) baseline GNRI groups.ConclusionBaseline GNRI was not associated with MACCEs, but patients with deteriorating or stationary nutritional status were at significantly greater risk for MACCEs, suggesting that serial monitoring of nutritional status is important to stratify cardiovascular risk in incident PD patients.