Serum C-reactive protein (CRP) and risk of death in chronic dialysis patients.

BACKGROUND The prognosis of chronic dialysis patients is poor, in part due to the high incidence of cardiovascular disease. Malnutrition, such as hypoalbuminaemia, has been shown to be a predictor of death in this group of patients, while serum C-reactive protein (CRP) is a predictor of myocardial infarction and sudden death. Thus, the aim of the present study was to determine of the relationship between CRP and serum albumin concentration, and the value of baseline CRP data in the prediction of death. METHODS In one of the dialysis units in Okinawa, Japan, baseline CRP data was available (n=163, 95 men and 68 women) in January 1991. These patients were divided into two groups according to their baseline CRP levels, with group 1 consisting of CRP<10 mg/l (n=128) and group 2 of CRP> or =10 mg/l (n=35), and then followed up until the end of 1997. Survival curves were calculated using the Kaplan-Meier method. The statistical significance of the relationship between CRP levels and the risk of death was evaluated by multiple logistic analysis with covariables such as age, sex, diabetes mellitus, serum albumin, and blood pressure. RESULTS The mean (SD) level of serum albumin was 38 (3) g/l in group 1 and 36 (3) g/l in group 2 (P<0.00001). The 5-year survival rate was significantly poorer in group 2 (44.4%) than in group 1 (82.5%) (P<0.0001). Furthermore, the risk of death was significantly higher in group 2 (relative risk 3.48 (95% confidence interval 1.76-6.89), P<0.0003) by multivariate Cox proportional hazard analysis. CONCLUSIONS CRP is a significant predictor of death in chronic dialysis patients, independent of serum albumin and other possible confounders. Dialysis patients with high CRP levels should be carefully evaluated and monitored regardless of serum albumin concentrations in the normal range.