SP617INDOXYL SULFATE AND MORTALITY IN HAEMODIALYSIS PATIENTS: RESULTS FROM THE JAPAN DIALYSIS OUTCOMESAND PRACTICE PATTERNS STUDY (J-DOPPS)

Abstract

INTRODUCTION AND AIMS: Uraemic toxins are associated with various chronic kidney disease (CKD)-related comorbidities. Indoxyl sulfate (IS), a representative protein-bound uraemic toxin, reacts with vasculature, accelerating atherosclerosis and/or vascular calcification in animal models. Removal of IS by conventional haemodialysis (HD) is often inadequate, owing to its high affinity for protein-binding. Associations of IS with clinical outcomes in CKD and dialysis patients have been reported; however, few studies have examined the relationship in a large cohort of HD patients. In this study, we assessed the relationship between serum IS level and mortality in HD patients in the Japan Dialysis Outcomes and Practice Patterns Study (J-DOPPS). METHODS: We included 1,170 HD patients from the J-DOPPS phase 5 (2012-2015) who had serum IS level measurements collected as part of an ancillary biomarker study. We evaluated the associations of serum IS levels with all-cause mortality and cardiovascular (CV) death or hospitalization using Cox regressions with adjustments for potential confounding by age, sex, vintage, history of diabetes and/or prior CV disease, albumin, normalized protein catabolic rate (nPCR), body mass index, creatinine and spKt/V. We assessed possible effect modification (interaction) of the association between IS and mortality by residual kidney function (RKF; defined as urine output >200 ml/day) and nPCR. Time at risk for each patient began on the date of serum collection and continued until the clinical outcome occurred, the study phase ended, or the patient departed from J-DOPPS (typically due to transfer out of the study site). We used multiple imputation (n=5 datasets) to replace missing values. RESULTS: The median serum IS level at baseline was 31.6 ug/mL (interquartile range, 22.6, 42.0). Serum IS level was positively associated with dialysis vintage, nPCR, and urea nitrogen. We observed 174 deaths (14.9%; crude rate, 0.06/year). Serum IS level was positively associated with all-cause mortality (adjusted hazard ratio [HR] per 10 ug/mL higher, 1.16; 95% confidence interval [CI], 1.04-1.28). The association was stronger in patients with RKF (HR, 1.44; 95% CI, 1.20-1.73; interaction p=0.02 for RKF vs. no RKF) and nPCR<0.85 (HR, 1.29; 95% CI, 1.15-1.44; interaction p=0.01 for nPCR tertiles). We observed a weaker association of serum IS level with CV events (HR per 10 ug/mL higher, 1.08; 95% CI, 0.97-1.20). CONCLUSIONS: In a large cohort study of Japanese HD patients, serum IS level was associated with all-cause mortality, with increased toxicity suggested in patients with RKF and low nPCR. Residual confounding may explain these findings, but strategies to maintain low levels of IS and other uraemic toxins may induce better survival in HD patients.