The Impact of Normal Range of Serum Phosphorus on the Incidence of End-Stage Renal Disease by A Propensity Score Analysis

BACKGROUND Although hyperphosphatemia is deemed a risk factor of the progression of chronic kidney disease (CKD), it remains unclear whether the normal range of serum phosphorus likewise deteriorates CKD. A propensity score analysis was applied to examine the causal effect of the normal range of serum phosphorus on the incidence of end-stage renal disease (ESRD). METHODS A retrospective CKD cohort of 803 participants in a single institution was analyzed. Propensity score was estimated using 22 baseline covariates by multivariate binary logistic regression for the different thresholds of time-averaged phosphorus (TA-P) in the normal range of serum phosphorus incremented by 0.1 mg/dL from 3.3 to 4.5 mg/dL. RESULTS The incidence rate of ESRD was 33.9 per 1,000 person-years over median follow-up of 4.3 years. Total patients showed the mean baseline phosphorus of 3.37 mg/dL and were divided to quartile. The higher quartile was associated with the parameters consistent with the advancement of CKD. A stratified Cox regression showed the highest hazard ratio (HR) at TA-P 3.4 mg/dL (HR 17.60, 95% CI 3.92-78.98) adjusted for baseline covariates such as sex, age, diabetic nephropathy, estimated GFR, serum albumin, Na-Cl, phosphorus, LDL-C and proteinuria. Adjusted HRs remained high up to TA-P 4.2 mg/dL (HR 2.22, 95% CI 1.33-3.71). After propensity score matching conducted at the thresholds of TA-P 3.4, 3.6, 3.8 and 4.0 mg/dL, the higher levels of TA-P showed the higher HRs by Kaplan-Meier analysis (p < 0.05 by stratified log-rank test). The numbers needed to treat were calculated as 3.9 to 5.3 over 5 years. CONCLUSIONS The propensity score analysis shows that even the normal range of serum phosphorus clearly accelerates CKD progression to ESRD. Our results encourage clinicians to target serum phosphorus to inhibit CKD progression in the manner of 'the lower the better.'