Assessing whether a spot urine specimen can predict 24-h urinary sodium excretion accurately: a validation study.

Abstract

Aimed to evaluate the validity of five published equations that estimate 24-h urinary sodium excretion (24-h USE) by timed spot urine specimens. All of the spot urines within 24 h were collected from 99 healthy volunteers aged 21-57 years. The spot urinary sodium and 24-h USE were measured. The 24-h USE was estimated by INTERSALT equation and Tanaka equation with casual spot urine and four timed spot urines (i.e. overnight, second morning, post meridiem, and evening urine), by Kawasaki equation and SunSMU equation with second morning urine, and by SunPM equation with post meridiem urine, respectively. At last, the agreement between measured and estimated 24-h USE was evaluated. The average of the measured 24-h USE was 4558.25 ± 1908.75 mg/day. The proportion of participants with more than ±40% of relative difference between measured 24-h USE and estimated 24-h USE by five equations varied from 21.18 to 42.35%. The absolute differences of 24-h USE among approximately half of the participants were more than ±1170.00 mg/day estimated by Kawasaki equation, INTERSALT equation, and SunPM equation, respectively. In addition, misclassification rates of individual salt intake for five equations were more than 60%. It was found that overestimation occurred at low levels of sodium excretion whereas underestimation occurred at high levels of sodium excretion except Kawasaki equation. The 24-h USE estimated using Tanaka equation with four timed spot urine specimens varied substantially, although Tanaka equation with casual spot urine specimen provided the highest correlation coefficient and the least mean difference. There is still no evidence to support that 24-h USE could be estimated accurately using the current equations, especially for the equations developed by casual spot urine specimens.