Prediction of 24-hour sodium excretion from spot urine samples in South African adults: a comparison of four equations

Abstract

Repeated 24-hour urine collection is considered to be the gold standard for assessing salt intake. This is often impractical in large-population studies, especially in low-middle-income countries. Equations to estimate 24-hour urinary salt excretion from a spot urine sample have been developed, but have not been widely validated in African populations. This study aimed to systematically assess the validity of four existing equations to predict 24-hour urinary sodium excretion (24UNa) from spot urine samples in a nationally representative sample of South Africans. Spot and 24-hour urine samples were collected in a subsample (n = 438) of participants from the World Health Organisation Study on global AGEing and adult health (SAGE) Wave 2 in South Africa in 2015. Measured 24UNa values were compared with predicted 24UNa values from the Kawasaki, Tanaka, INTERSALT and Mage equations using Bland-Altman plots. In this subsample (mean age 52.8 ± 16.4 years; body mass index 30.2 ± 8.2 kg/m 2 ; 76% female; 73% black African; 42% hypertensive), all four equations produced a significantly different population estimate compared with the measured median value of 6.7 g salt/day (IQR 4.4-10.5). Although INTERSALT underestimated salt intake (−3.77 g/d; −1.64 to −7.09), the other equations overestimated by 1.28 g/d (−3.52; 1.97), 6.24 g/d (2.22; 9.45), and 17.18 g/d (8.42; 31.96) for Tanaka, Kawasaki, and Mage, respectively. Bland-Altman curves indicated unacceptably wide levels of agreement. Use of these equations to estimate population level salt intake from spot urine samples in South Africans is not recommended.

Can methods based on spot urine samples be used to estimate average population 24 h sodium excretion? Results from the Isfahan Salt Study

“…In our study, we observed a weak correlation between estimated 24hUNa based on established equations and measured 24hUNa excretion, which was similar to or lower than in previous studies (6,8,10,(30)(31)(32)(33) . Consistent with our study, a recent systematic review of more than 1•3 million adults and children (14) and several studies from South Africa, the UK, Italy and Denmark (11,33,34) have shown that the correlation between measured and estimated 24hUNa using the Tanaka equation is weak. In a Korean population, the correlation coefficient between measured and estimated daily Na excretion by Kawasaki, Tanaka and INTERSALT equations was relatively high, although there were significant differences between the spot urine-based estimation methods and measured 24hUNa (7) .…”

“…In our study, the mean differences between measured and predicted values based on Tanaka, Kawasaki and INTERSALT equations were greater than in most previous reports (11,35) . These studies showed inconsistent levels of under-and overestimation using these formulas (11,35) ; for example, overestimation in the presence of low level of Na excretion and underestimation in high level of daily Na excretion (22) . We found greater limits of agreement (95 % CI) than Brown et al's and Rhee et al's studies for all equations, but the greatest was for the Kawasaki equation (7,10) .…”

“…The Kawasaki and Tanaka equations were developed in Japanese populations (8,9) and the INTERSALT equation was designed in North American and European populations (10) . No single formula is universally accepted and all have reported biases when assessed by Bland-Altman plots (11) . In addition, there is no consensus on the timing of spot urine sample collection (6,11,12) .…”

“…No single formula is universally accepted and all have reported biases when assessed by Bland-Altman plots (11) . In addition, there is no consensus on the timing of spot urine sample collection (6,11,12) .…”

Correcting for Intra-Individual Variability in Sodium Excretion in Spot Urine Samples Does Not Improve the Ability to Predict 24 h Urinary Sodium Excretion

“…We have previously reported that equations using spot urine collections have an unacceptably wide degree of bias and are not appropriate for use in the South African population with Tanaka, Kawasaki, and Mage equations—all overestimating salt intake (15). The INTERSALT equation systematically underestimated measured 24 h sodium excretion [ 15 ]. Other authors have reported similar findings [ 16 ].…”