Estimating the population distribution of usual 24-hour sodium excretion from timed urine void specimens using a statistical approach accounting for correlated measurement errors

24-hour sodium and potassium excretion in the Americas: a systematic review and meta-analysis

Objective

To determine the 24-hour urinary sodium and potassium excretions in the Americas.

Methods

A systematic review and meta-analysis were performed seeking for studies conducted between 1990 and 2021 in adults living in any sovereign state of the Americas in Medline, Embase, Scopus, SciELO, and Lilacs. The search was first run on October 26th, 2020 and was updated on December 15th, 2021. Of 3 941 abstracts reviewed, 74 studies were included from 14 countries, 72 studies reporting urinary sodium (27 387 adults), and 42 studies reporting urinary potassium (19 610 adults) carried out between 1990 and 2020. Data were pooled using a random-effects meta-analysis model.

Results

Mean excretion was 157.29 mmol/24h (95% CI, 151.42-163.16) for sodium and 57.69 mmol/24h (95% CI, 53.35-62.03) for potassium. When only women were considered, mean excretion was 135.81 mmol/24h (95% CI, 130.37-141.25) for sodium and 51.73 mmol/24h (95% CI, 48.77-54.70) for potassium. In men, mean excretion was 169.39 mmol/24h (95% CI, 162.14-176.64) for sodium and 62.67 mmol/24h (95% CI, 55.41-69.93) for potassium. Mean sodium excretion was 150.09 mmol/24h (95% CI, 137.87-162.30) in the 1990s and 159.79 mmol/24h (95% CI, 151.63-167.95) in the 2010s. Mean potassium excretion was 58.64 mmol/24h (95% CI, 52.73-64.55) in the 1990s and 56.33 mmol/24/h (95% CI, 48.65-64.00) in the 2010s.

Conclusions

These findings suggest that sodium excretions are almost double the maximum level recommended by the World Health Organization and potassium excretions are 35% lower than the minimum requirement; therefore, major efforts to reduce sodium and to increase potassium intakes should be implemented.

The Use and Interpretation of Sodium Concentrations in Casual (Spot) Urine Collections for Population Surveillance and Partitioning of Dietary Iodine Intake Sources

In 2013, the World Health Organization (WHO) called for joint surveillance of population salt and iodine intakes using urinary analysis. 24-h urine collection is considered the gold standard for salt intake assessment, but there is an emerging consensus that casual urine sampling can provide comparable information for population-level surveillance. Our review covers the use of the urinary sodium concentration (UNaC) and the urinary iodine concentration (UIC) from casual urine samples to estimate salt intakes and to partition the sources of iodine intakes. We reviewed literature on 24-h urinary sodium excretion (UNaE) and UNaC and documented the use of UNaC for national salt intake monitoring. We combined information from our review of urinary sodium with evidence on urinary iodine to assess the appropriateness of partitioning methods currently being adapted for cross-sectional survey analyses. At least nine countries are using casual urine collection for surveillance of population salt intakes; all these countries used single samples. Time trend analyses indicate that single UNaC can be used for monitoring changes in mean salt intakes. However; single UNaC suffers the same limitation as single UNaE; i.e., an estimate of the proportion excess salt intake can be biased due to high individual variability. There is evidence, albeit limited, that repeat UNaC sampling has good agreement at the population level with repeat UNaE collections; thus permitting an unbiased estimate of the proportion of excess salt intake. High variability of UIC and UNaC in single urine samples may also bias the estimates of dietary iodine intake sources. Our review concludes that repeated collection, in a sub-sample of individuals, of casual UNaC data would provide an immediate practical approach for routine monitoring of salt intake, because it overcomes the bias in estimates of excess salt intake. Thus we recommend more survey research to expand the evidence-base on predicted-UNaE from repeat casual UNaC sampling. We also conclude that the methodology for partitioning the sources of iodine intake based on the combination of UIC and UNaC measurements in casual urine samples can be improved by repeat collections of casual data; which helps to reduce regression dilution bias. We recommend more survey research to determine the effect of regression dilution bias and circadian rhythms on the partitioning of dietary iodine intake sources.

Evaluation of Equations for Predicting 24-Hour Urinary Sodium Excretion from Casual Urine Samples in Asian Adults

BACKGROUND

The collection of 24-h urine samples for the estimation of sodium intake is burdensome, and the utility of spot urine samples in Southeast Asian populations is unclear.

OBJECTIVE

We aimed to assess the validity of prediction equations with the use of spot urine concentrations.

METHODS

A sample of 144 Singapore residents of Chinese, Malay, and Indian ethnicity aged 18-79 y were recruited from the Singapore Health 2 Study conducted in 2014. Participants collected urine for 24 h in multiple small bottles on a single day. To determine the optimal collection time for a spot urine sample, a 1-mL sample was taken from a random bottle collected in the morning, afternoon, and evening. Published equations and a newly derived equation were used to predict 24-h sodium excretion from spot urine samples.

RESULTS

The mean ± SD concentration of sodium from the 24-h urine sample was 125 ± 53.4 mmol/d, which is equivalent to 7.2 ± 3.1 g salt. Bland-Altman plots showed good agreement at the group level between estimated and actual 24-h sodium excretion, with biases for the morning period of -3.5 mmol (95% CI: -14.8, 7.8 mmol; new equation) and 1.46 mmol (95% CI: -10.0, 13.0 mmol; Intersalt equation). A larger bias of 25.7 mmol (95% CI: 12.2, 39.3 mmol) was observed for the Tanaka equation in the morning period. The prediction accuracy did not differ significantly for spot urine samples collected at different times of the day or at a random time of day (P = 0.11-0.76).

CONCLUSION

This study suggests that the application of both our own newly derived equation and the Intersalt equation to spot urine concentrations may be useful in predicting group means for 24-h sodium excretion in urban Asian populations.