Increased risk for cardiovascular mortality among malnourished end-stage renal disease patients

Background: Studies have shown that protein-energy malnutrition (PEM) is a strong predictor of total mortality among patients with end-stage renal disease (ESRD). The aim of this study is to assess the relationship between nutritional indices and cardiovascular (CV) mortality among patients with ESRD by using data from the US Renal Data System (USRDS) Dialysis Morbidity and Mortality Study Wave I (DMMS-1).

Methods: Demographic and medical data were abstracted from 5,058 patients who participated in the USRDS DMMS-1. Nutritional measurements of interest included subjective assessment of malnutrition, serum albumin level, body mass index (BMI), and cholesterol level. The USRDS provided follow-up data on mortality through July 1998. The Cox proportional hazard model was used to estimate the risk for CV death associated with nutritional markers.

Results: The risk for CV death was 39% greater for each 1-g/dL (10-g/L) decrement in serum albumin level (95% confidence interval [CI], 1.20 to 1.60; P < 0.001). A care provider's assessment of malnutrition was associated with a 27% greater risk for CV mortality (95% CI, 1.08 to 1.50; P < 0.004). For each one-unit decrement in BMI, the risk for CV disease (CVD) was 6% greater (95% CI, 1.00 to 1.13; P < 0.046). Among patients without CVD at the study start, serum albumin level remained a significant risk factor for CV death (adjusted relative risk = 1.39 per 1-g/dL (10-g/L) increment; P = 0.026). In addition, change in albumin levels over time was significantly associated with CV mortality. For each 0.1-g/dL (1-g/L) decrement in albumin level per month, the risk for CV death was 2.24-fold greater (95% CI, 1.65 to 3.02; P < 0.001) among the entire cohort and 3.86-fold greater (95% CI, 1.96 to 7.60; P < 0.010) among those without a known history of CVD at the study start.

Conclusion: Both PEM at baseline and worsening PEM over time are associated with a greater risk for CV death. This finding persists among dialysis patients without preexisting CVD at baseline.