Abstract
BACKGROUND: The kidneys maintain acid-base homeostasis through excretion of acid as either ammonium or as titratable acids that primarily use phosphate as a buffer. In chronic kidney disease (CKD), ammoniagenesis is impaired, promoting metabolic acidosis. Metabolic acidosis stimulates phosphaturic hormones, parathyroid hormone (PTH) and fibroblast growth factor 23 (FGF-23) in vitro, possibly to increase urine titratable acid buffers, but this has not been confirmed in humans. We hypothesized that higher acid load and acidosis would associate with altered phosphorus homeostasis, including higher urinary phosphorus excretion and serum PTH and FGF-23. STUDY DESIGN: Cross-sectional. SETTING & PARTICIPANTS: 980 participants with CKD enrolled in the Chronic Renal Insufficiency Cohort (CRIC) Study. PREDICTORS: Net acid excretion as measured in 24-hour urine, potential renal acid load (PRAL) estimated from food frequency questionnaire responses, and serum bicarbonate concentration < 22 mEq/L. OUTCOME & MEASUREMENTS: 24-hour urine phosphorus and calcium excretion and serum phosphorus, FGF-23, and PTH concentrations. RESULTS: Using linear and log-linear regression adjusted for demographics, kidney function, comorbid conditions, body mass index, diuretic use, and 24-hour urine creatinine excretion, we found that 24-hour urine phosphorus excretion was higher at higher net acid excretion, higher PRAL, and lower serum bicarbonate concentration (each P<0.05). Serum phosphorus concentration was also higher with higher net acid excretion and lower serum bicarbonate concentration (each P=0.001). Only higher net acid excretion associated with higher 24-hour urine calcium excretion (P<0.001). Neither net acid excretion nor PRAL was associated with FGF-23 or PTH concentrations. PTH, but not FGF-23, concentration (P=0.2) was 26% (95% CI, 13%-40%) higher in participants with a serum bicarbonate concentration <22 versus ≥22 mEq/L (P<0.001). Primary results were similar if stratified by estimated glomerular filtration rate categories or adjusted for iothalamate glomerular filtration rate (n=359), total energy intake, dietary phosphorus, or urine urea nitrogen excretion, when available. LIMITATIONS: Possible residual confounding by kidney function or nutrition; urine phosphorus excretion was included in calculation of the titratable acid component of net acid excretion. CONCLUSIONS: In CKD, higher acid load and acidosis associate independently with increased circulating phosphorus concentration and augmented phosphaturia, but not consistently with FGF-23 or PTH concentrations. This may be an adaptation that increases titratable acid excretion and thus helps maintain acid-base homeostasis in CKD. Understanding whether administration of base can lower phosphorus concentrations requires testing in interventional trials.