FP356SERUM PHOSPHATE AND SOCIAL DEPRIVATION PREDICT LONG-TERM ALL CAUSE AND CARDIOVASCULAR MORTALITY IN CHRONIC KIDNEY DISEASE

Abstract

Introduction and Aims: Hyperphosphataemia is associated with reduced survival in end stage renal disease and chronic kidney disease (CKD). Socioeconomic deprivation is generally associated with poor lifestyle, increased cardiovascular risk and reduced survival. We studied the relationship between serum phosphate, socioeconomic deprivation and long-term outcome in non-dialysis CKD patients. Methods: All adult patients with a first renal clinic attendance (excluding dialysis and transplant) during 2010-13 were studied. Baseline clinical and laboratory data at presentation were retrieved. Follow up was recorded to October 2014. Socioeconomic status was assessed by postcode of residence from the National Index of Multiple Deprivation. Time to death was calculated and stepwise Cox regression used to identify covariates significantly associated with survival. Results: During the study period 2,989 patients (mean age 64.1 years; 49.8% male) had a first renal clinic attendance. The mean estimated glomerular filtration rate (eGFR) at baseline was 46ml/min. Median follow up was 1.7 years. During follow up 386 (12.9%) patients died and 101 (3.4%) commenced renal replacement therapy (RRT). There were 67 cardiovascular deaths (17.4% of all deaths, 26.0% of known causes of death). Patients who died were older (74.8 vs. 62.6 years, p<0.001), had high higher serum phosphate (1.20 vs. 1.12 mmol/L, p<0.001), lower haemoglobin (111 vs. 125g/L, p<0.001), eGFR (33 vs. 48ml/min, p<0.001) and serum albumin (34 vs. 37g/L, p<0.001) than those alive at follow up. Patients who resided in lower socioeconomic areas had higher mean serum phosphate (most deprived 1.14 vs. least deprived 1.10mmol/L, p<0.01) despite similar mean eGFR (47.0 vs. 48ml/min, p=0.5). In a multivariate model significant independent predictors of mortality were increased age, male gender, increased serum phosphate (although low phosphate was also associated with increased risk-see figure), low albumin, eGFR, haemoglobin and lower socioeconomic status. Increased age and phosphate and lower serum albumin and socioeconomic status were significant independent predictors of cardiovascular death. Patients who started RRT were younger (59.1 vs. 64.3 years, p<0.01), had higher systolic blood pressure at baseline (159 vs. 148mmHg, p=0.001), lower eGFR (19 vs. 47ml/min, p<0.001) and higher urinary protein:creatinine (uPCR) (520 vs. 154mg/mmol, p<0.01). In a multivariate model independent predictors of requiring RRT were low eGFR, low serum albumin, male gender, raised systolic blood pressure and uPCR. Socioeconomic status and phosphate did not influence need for RRT. Conclusions: Increased serum phosphate and low socioeconomic status are independent risk factors for reduced survival but not need for RRT in patients with non-dialysis dependent CKD. Elevated phosphate despite similar renal function suggests increased dietary intake of processed foods as a possible mechanism for raised phosphate in lower socioeconomic groups. Further work on the determinants of raised phosphate, target phosphate and optimal method for phosphate lowering in CKD is required. (Figure Presented).