Abstract 14621: Deviation of Three Methods to Estimate 24-h Urinary Sodium Excretion From Spot Urine Samples in Individuals.
Introduction: Twenty-four-hour (24-h) urine collection is widely regarded as the gold standard for evaluating sodium (Na) intake. As a convenient alternative, spot urine method has been proposed to estimate 24-h urinary Na excretion, but its reliability remains highly controversial. The aim of the study was to assess the accuracy and deviation of three commonly used methods--- Kawasaki (K), INTERSALT (I) and Tanaka (T) for estimating 24-h urinary Na excretion using spot urine samples against measured 24-h urinary Na excretion in Chinese female adults.
Methods: A total of 144 community housewives were enrolled in baseline investigation of a salt reduction program. Both a spot and a 24-h urine sample were collected for each participant and Na was measured in all samples. K, I and T equations were respectively used to estimate 24-h urinary Na excretion from spot urine samples. The correlation and difference between the measured and estimated 24-h Na excretions were then explored.
Results: The analysis included 139 participants aged 51±8 years after exclusion of reported incomplete collections (n=2) and 24-h urine volume<500ml (n=3). The predicted and measured 24-h Na excretions were 5652±1529 (K), 3267±529 (I), 4231±889 (T) and 5051±1794 (measured) mg, respectively. The correlation coefficients between measured and estimated values were 0.36 (K), 0.22 (I) and 0.39 (T), respectively (all P values<0.01). When using the measured 24-h Na excretion as the reference, the proportion of relative deviation within ±10% in K, I and T methods was only 26.6%, 12.2%, 16.5%,and the proportion over 40% was 40.9%(K), 41%(I), 20.8%(T). The proportion of difference within ±400mg (1g salt) in K, I and T methods was only 19.4%, 15.1%, 16.5%, respectively. Moreover, the absolute difference for the three methods was larger than 1200mg (3g salt) in approximately half of the participants, i.e., 51.1%, 63.2% and 47.5% for K, I and T methods.
Conclusions: Although the correlation between the measured and estimated 24-h Na excretions was moderate, estimate of 24-h urinary Na excretion using spot urine samples showed large bias and might lead to misclassification in individuals.
Author Disclosures: H. Liu: None. Y. Tian: None. L. Zhou: None. X. Wen: None. M. Guo: None. Z. Yang: None. L. Zhao: None.
- © 2016 by American Heart Association, Inc.