Interleukin-18 is a strong predictor of hospitalization in haemodialysis patients.

BACKGROUND Morbidity and mortality rates are high among patients with end-stage renal disease (ESRD), and recent evidence suggests that this may be linked to inflammation. Current research has also demonstrated the crucial involvement of interleukin-18 (IL-18) in inflammation. In agreement, the activity of IL-18 has been markedly up-regulated in ESRD patients. However, it has not been established whether elevated plasma IL-18 predicts outcome in haemodialysis (HD) patients. METHODS To determine whether plasma IL-18 predicts overall hospitalization, we studied 184 ESRD patients (62% males, 58.5+/-1.0 years of age) undergoing maintenance HD treatment. The patients were followed for 12 months and were stratified by the tertiles of plasma IL-18 levels. Classic factors, such as age, body mass index, duration of HD, nutritional and inflammatory parameters, co-morbidity, dialysis adequacy, and lipid status were entered into a Cox regression model to predict hospitalization. The Kaplan-Meier method was used to analyse the cumulative proportion of hospitalization-free events. RESULTS Significantly different hospitalization days and frequencies (P<0.05) were observed when patients were divided according to tertiles of plasma IL-18 levels. Patients were stratified according to IL-18 tertiles and analysed separately according to the hospitalization-free period. In the Kaplan-Meier model, the upper tertile of IL-18 had the highest probability of a hospitalization event during the entire follow-up period (P log rank = 0.027). In the Cox proportional hazard model, the relative risk for first hospital admission for each increase in Ln IL-18 (pg/ml) concentration was associated with a 1.709 (95% CI, 1.114 to 2.620; P = 0.014) increase in the risk for future hospitalization events. CONCLUSIONS The present study demonstrated a strong predictive value of elevated IL-18 levels for poor outcome in HD patients.