BACKGROUND: Discovering metabolomic markers of dietary potassium may help improve dietary assessment of potassium and trace the impact of dietary potassium on chronic kidney disease (CKD) development. METHODS: We included adults from the Atherosclerosis Risk in Communities (ARIC) study without CKD at visit 1 (N = 3,812). Cross-sectional associations between dietary potassium and serum metabolites were assessed using multivariable linear regression models. Cox regression models estimated hazard ratios for potassium-related metabolites and incident CKD. Incident CKD was defined as estimated glomerular filtration rate (<
60 mL/min/1.73 m2 and ≥25% decline), CKD-related hospitalization or death, or kidney replacement therapy identified via United States Renal Data System registry from visit 1 (1987-1989) through December 31, 2020. RESULTS: There were 33 significant associations between dietary potassium and serum metabolites, including pyridoxate, N-methylproline, stachydrine, pantothenate, and scyllo-inositol. During more than two decades of follow-up (median: 23 years, 25th-75th percentile: 14-30), 1,616 (42%) of participants developed incident CKD. Ten of the 33 potassium-related metabolites were significantly associated with incident CKD. Metabolites involved in phenylalanine and tyrosine metabolism, 3-(4-hydroxyphenyl)lactate and 3-phenylpropionate, were significantly associated with dietary potassium and CKD. Additionally, glycerate, involved in glucose metabolism, was positively associated with dietary potassium (β=0.09, p=4.01 x 10-17) and inversely associated with CKD (HR 0.77, 95% CI: 0.69-0.85, p=8.57 x 10-7). There was a significant trend for CKD risk across quartiles of 3-(4-hydroxyphenyl)lactate, 3-phenylpropionate, and glycerate. CONCLUSIONS: Dietary potassium was associated with 33 serum metabolites. 3-(4-hydroxyphenyl)lactate 3-phenylpropionate, and glycerate are candidate markers of dietary potassium's impact on chronic kidney disease.