Spot Urine Samples to Estimate Na and K Intake in Patients With Chronic Kidney Disease and Healthy Adults: A Secondary Analysis From a Controlled Feeding Study

Abstract

The objective of this study was to assess the agreement between estimated 24-hour urinary sodium excretion (e24hUNa) and estimated 24-hour urinary potassium excretion (e24hUK), calculated from a spot urine sample using several available equations and actual sodium and potassium intake from a controlled diet in both healthy participants and those with chronic kidney disease (CKD). This study is a secondary analysis of a controlled feeding study in CKD patients matched to healthy controls. Participants (n=16) consumed the controlled diet, which provided ∼2400mg Na/day and ∼3000mgK/day, for 8days. On days 7 and 8, participants consumed all meals and collected all urine in an inpatient research setting, and they were discharged on day 9. The day 7 morning spot urine sample was used to calculate e24hUNa and e24hUK, which was compared with known sodium and potassium intake, respectively. Average e24hUNa from the INTERSALT and Tanaka-Na equations were higher than actual sodium intake by 373mg and 559mg, respectively, though the differences were not significant. e24hUNa from the Nerbass-SALTED equation in CKD participants was significantly higher than actual sodium intake by ∼2000mg (P<.001), though e24hUNa from the Nerbass-RRID equation was not different from intake. e24hUK from the Tanaka-K equation was significantly lower than actual potassium intake (P<.001). For both e24hUNa and e24hUK for all participants, agreement with actual intake was poor, and e24hUNa and e24hUK were not correlated with actual sodium or potassium intake, respectively. e24hUNa and e24hUK are poor indicators of true sodium and potassium intake, respectively, in both healthy and CKD participants. Findings should be confirmed in larger sample sizes with varying levels of dietary sodium and potassium.