#4412 URINARY SODIUM-TO-POTASSIUM RATIO DOES NOT ACCURATELY REFLECT DIETARY SODIUM-TO-POTASSIUM RATIO

Abstract

Abstract Background and Aims Low sodium intake is associated with a lower blood pressure and less proteinuria, which are important therapeutic targets in chronic kidney disease. In contrast, a low potassium intake has been associated with higher blood pressure and a higher incidence of chronic kidney disease and cardiovascular events. Counselling of patients regarding their sodium and potassium intake requires accurate estimation of their intake, but 24-hour urinary sodium and potassium excretion can deviate substantially from actual intake. Urinary sodium-to-potassium (Na/K) ratio is a promising alternative as it is less affected by incomplete urine collections and additionally captures both the effect of sodium and potassium. Our study aimed to assess whether the Na/K ratio in 24-hour urine reflects dietary intake more accurately than separate measurement of sodium or potassium in 24-hour urine. Method We performed a post-hoc analysis on data from the long-term sodium balance studies Mars105 and Mars520. Ten healthy participants consumed a diet with a known sodium and potassium content and collected 24-hour urine samples for 105 or 205 days. We calculated the log fold difference between dietary intake and urinary excretion of sodium, potassium and Na/K ratio. A mixed-effects model with a random intercept per subject was used to compare these estimates of accuracy. Subsequently, we performed a subgroup analysis per salt intake level (i.e. 6, 9 or 12 grams per day) and assessed the effect of increasing the number of 24-hour urine collections. Results Overall, the urinary Na/K ratio underestimated dietary Na/K ratio with a median difference of -0.21 (IQR -0.47 to 0.09). Estimation of dietary Na/K ratio intake using the urinary Na/K ratio was significantly less accurate compared to using urinary sodium or potassium measurement for estimation of sodium or potassium intake (Figure 1A). Only during a salt intake of 6 grams per day, the urinary Na/K ratio did not perform significantly worse than sodium measurements. Although increasing the number of 24-hour urine measurements to three or seven improved the accuracy of the urinary Na/K ratio, it remained inferior to separate assessment of urinary sodium and potassium excretion (Figure 1B). Conclusion The 24-hour urinary Na/K ratio is less accurate than 24-hour urinary sodium or potassium excretion for estimation of dietary intake in a controlled setting in healthy volunteers.