Serum uric acid levels show a 'J-shaped' association with all-cause mortality in haemodialysis patients

Background. Although elevated serum levels of uric acid are common in patients with kidney disease or in those receiving maintenance dialysis therapy, the clinical impact of uric acid on mortality in haemodialysis (HD) patients remains unclear. This work was designed to explore the predictive value of serum uric acid levels on all-cause mortality of HD patients. Methods. We retrospectively analysed mortality rates in 146 chronic HD patients that were treated with HD three times per week at our HD unit for a period of one full year. The analysed parameters included demographic characteristics, aetiology of end-stage renal disease, co-morbid conditions, duration (at least 1 year) and delivered dose of HD, normalized protein catabolic rate, serum albumin concentration, haematocrit, serum uric acid (UA) levels and other laboratory parameters. A multivariate Cox proportional hazards model, which included adjustment for the above factors, was applied to identify the predictive value of UA levels on patient mortality. Results. A Cox proportional hazards model revealed that decreased serum albumin, underlying diabetic nephropathy (DMN) and UA groups (less than or equal to20th, 20-80th and greater than or equal to80th percentiles; P=0.016) were all significant, independent predictors of all-cause mortality in HD patients. The hazard ratios of death were: serum albumin (per 0.5g/dl decrease), 3.10 [95% confidence interval (95% CI), 1.80-5.34, P < 0.001]; DMN (vs non-DMN), 3.47 (95% CI, 1.25-9.59, P=0.017); and UA groups (vs 20th to 80th percentile): less than or equal to20th percentile, 2.98 (95% CI, 0.82-10.90, P=0.099); >80th percentile, 5.67 (95% CI, 1.71-18.78, P=0.004). Conclusions. These preliminary observations suggest that HD patients in the lowest and highest quintiles of UA levels would face higher risk of mortality. Further studies with larger sample sizes will be needed to confirm these findings.