#738 Low serum Na-Cl difference level before alkali therapy is an independent predictor of rapid renal deterioration in CKDG3b-G5 patients with hyperkalemic metabolic acidosis

Abstract Background and Aims Advanced chronic kidney disease (CKD) is often complicated by serum potassium[K]-elevated metabolic acidosis (high K-MA) and MA can be a cause of CKD progression. High K-MA can occur with decreased mineralocorticoid action (e.g., use of RAAS inhibitors (RAASi), hyporeninemic states such as diabetes and volume expansion) or, in advanced CKD, with decreased nephron mass exceeding compensatory excretion of H+ (NH4+) and K+ in the renal tubules. Metabolic acid-base status is usually assessed by serum HCO3−/total-CO2 concentrations, but respiratory status, especially respiratory compensatory changes in severe metabolic acid-base imbalance, also affect these values. On the other hand, the serum Na-Cl difference ([Na-Cl]) is a simplified strong ion difference and is useful as an indicator of the dominant metabolic component independent of respiratory status. In patients with advanced CKD, especially those with high K-MA requiring alkalizing therapeutic intervention, it is unclear whether pre-intervention [Na-Cl] determines renal prognosis and incidence of cardiovascular events. Method In this single-center retrospective observational study, CKDG3b-G5 outpatients with ([HCO3−] <22 mM) and elevated K (serum [K] ≥5.0 mEq/L) at our center and who had started alkali therapy for high K-MA were included. Total of 59 patients were selected, excluding those who had already been prepared for renal replacement therapy at the start of alkali therapy. Background factors such as age, sex, cause of CKD, and parameters immediately prior to the initiation of alkali therapy (i.e., estimated glomerular filtration rate (eGFR), CKD stage, urinary protein (g/gCr), serum [K]/[HCO3−]/[Na-Cl]/albumin/phosphate, hemoglobin levels, and use of RAASi/K binders/diuretics) were investigated. The endpoints were the incidence of renal events (>30% reduction in eGFR or renal replacement therapy) and cardiovascular events within 2 years of starting alkali therapy. Results In all 59 subjects, the mean age was 72.0 ± 11.7 years, 81% were male, mean eGFR was 19.6 ± 7.2 mL/min/1.73 m2, CKDG4/5 account for 92% (G4 60%/G5 32%), primary disease was diabetic nephropathy 30%, nephrosclerosis 43%, RAASi use 68%, mean [K] 5.5 ± 0.4 mEq/L, mean [HCO3−] 18.1 ± 2.2 mM and mean [Na-Cl] 28 ± 3.2 mEq/L. In a multivariate analysis among potential determinants for metabolic acid-base status (i.e., [Na-Cl], albumin, phosphate, hemoglobin), lower[Na-Cl] (P = 0.002) and lower hemoglobin level (P = 0.005) were significantly associated with poor cumulative renal survival. In a multivariate analysis performed among factors drawn from background factors that were significant in univariate analysis, lower eGFR (P = 0.004), lower [Na-Cl] (P = 0.006), no RAASi use (P = 0.006), and lower hemoglobin level (P = 0.051) were associated with poor renal outcome. Neither [HCO3−] (used instead of [Na-Cl]) nor [K] was significant (P = 0.252 and P = 0.199). In tertile classification of [Na-Cl], the lowest ([Na-Cl] ≤ 26) group had significantly poorer cumulative renal survival (P = 0.001) (Figure). In Cox regression analysis, adjusted hazard ratios for [Na-Cl] ≤ 26 (reference [Na-Cl] ≥ 30) were 4.47–6.84 (P < 0.001- P = 0.022), adjusted for many factors including age, sex, cause of CKD, eGFR, affecting metabolic acid-base status, and type of alkalizing treatment (Table). There were only a few cardiovascular events during the study period, and the relationship between [Na-Cl] and these events was not evident. Conclusion In advanced CKD (G3b-G5) patients complicated by metabolic acidosis with elevated serum potassium requiring alkalizing therapeutic intervention, a low serum Na-Cl difference level can be an independent determinant of rapid renal deterioration.