Invited Commentary: Quantifying Salt in Urine—A Complex Solution

Abstract

Reduction of dietary sodium intake has been identified as a priority to reduce the worldwide burden of hypertension and cardiovascular disease. Dietary sodium intake is most precisely ascertained by using timed urine collection. Casual urine sodium measurements are relatively easy to perform, but their relationship to timed urine sodium measurements is unclear. In this issue of the Journal, Brown et al. (Am J Epidemiol. 2013;177(11):1180–1192) report the development and validation of equations to estimate 24-hour urine sodium excretion from casual urine samples. Their study included a large number of participants on 2 continents, a well-collected gold standard, separate discovery and validation samples, and relevant covariates. The resulting equations represent the best available methods to estimate dietary sodium intake from casual urine samples. However, the study is limited by evidence of a suboptimal model fit, restriction to people 20–59 years of age in North America and Europe, and exclusion and adjustment that further limit external validity. In addition, individual-level correlations of estimated and measured 24-hour urine sodium excretion were modest. Properly applied, the results will facilitate tracking of dietary sodium intake within populations over time and identification of communities for which dietary sodium restriction is most likely to be beneficial. Further work is needed to extend estimation to additional populations and improve individual-level assessment.

Reduction of dietary sodium intake has been identified by the World Health Organization and other leading health authorities as a priority in the effort to reduce the worldwide burden of hypertension and cardiovascular disease (CVD). Convincing evidence has demonstrated that an excessive intake of dietary sodium increases blood pressure by expanding extracellular volume and cardiac output. Moreover, dietary sodium restriction has been shown to reduce blood pressure in randomized clinical trials (1). Hypertension is a preeminent risk factor for CVD, particularly stroke, and population-level reductions in blood pressure, attained mostly via pharmacological treatment, account for a substantial portion of the reduced incidence of CVD observed in industrialized nations over the last 3 decades (2). As a result, controlling blood pressure by reducing sodium consumption, in addition to or instead of using pharmacological antihypertensive therapy, is an attractive approach to limiting CVD.

However, the relationship between dietary sodium intake and CVD remains incompletely defined in human populations, raising concerns about the utility of universal dietary sodium restriction and curtailing its implementation. Observational studies have shown inconsistent associations between dietary sodium intake and CVD events. Most recently, in the largest cohort study to date (4,729 CVD events), investigators found an association between higher urine sodium levels and CVD; however, they also observed a higher risk of CVD among individuals who had extremely low urine sodium excretion (3). This J-shaped relationship, which has been observed in other populations (4), may represent confounding by chronic illnesses. Alternatively, a very low intake of dietary sodium may truly increase the risk of CVD. Dietary sodium restriction can lead to adverse biologic responses, including activation of the renin-angiotensin system, stimulation of the sympathetic nervous system, and unfavorable alterations to serum lipids (5). In a meta-analysis of clinical trials, dietary sodium restriction was found to result in a nonsignificant reduction in CVD events overall, but the incidence of CVD events was markedly increased in 1 study in which very strict sodium restriction was applied (6, 7). Given these considerations, dietary sodium restriction may be most beneficial in populations or individuals who consume the greatest quantities of sodium.

Dietary sodium intake and urinary sodium excretion are balanced in steady state, and dietary sodium intake is most precisely ascertained through measurement of urine sodium excretion. Food questionnaires, dietary recall, and even prospective food diaries are limited by imprecise quantification of added salt and sodium-rich preservatives. Urine collection for 24 hours (or more) is considered the gold standard for assessing urine sodium excretion, but it is impractical in population-based studies. Casual or “spot” urine sodium measurements are relatively easy to perform, but their relationship to timed urine sodium measurements is unclear. Casual urine sodium values are affected by characteristics other than sodium intake (e.g., urinary volume). To account for these differences, other characteristics are used to transform urine sodium concentration. This approach risks the introduction of bias related to the added characteristics (8, 9), but it can substantially improve precision. The approach is now regularly applied to quantify urine albumin excretion in research and clinical practice (10), but it has not been well developed for urine sodium excretion.

To address the this important issue, in this issue of the Journal, Brown and et al. (11) reported the development and validation of equations to estimate 24-hour urine sodium excretion from casual urine samples. Their study included 5,693 participants in the International Cooperative Study on Salt, Other Factors, and Blood Pressure (INTERSALT), a multinational study of blood pressure and its determinants. Both 24-hour and casual urine samples were collected from each participant in close temporal proximity. In a discovery sample, linear combinations of casual urine sodium concentration and other relevant participant characteristics were developed to estimate 24-hour urine sodium excretion. In a validation sample, individual-level correlation coefficients for estimated and measured 24-hour urine sodium excretion were 0.50 and 0.51 for men and women, respectively, whereas correlation coefficients for mean values of the 29 communities included were 0.79 and 0.71, respectively.

To our knowledge, their study is the most extensive effort thus far to develop methods to estimate urine sodium excretion from casual urine samples. It has a number of important strengths, including its inclusion of participants from 2 continents, the large sample size, the use of separate discovery and validation samples, the availability of a well-collected gold standard 24-hour urine sample, the inclusion in models of a number of relevant participant characteristics, and the independent modeling of urine components (as opposed to generation of an inflexible urine sodium-creatinine ratio, for example). The resulting equations now represent the best available method to estimate urine sodium excretion from casual urine samples.

The study also has limitations pertaining to both internal and external validity. First, the plot of observed versus estimated 24-hour urine sodium values among individual subjects still demonstrated noticeable bias; the fitted equation systematically overestimated 24-hour urine sodium for low casual urine sodium measurements and systematically underestimated 24-hour urine sodium for high casual measurements. This finding indicates nonlinearity between casual and timed urine sodium measurements. Simple transformations may have improved model fit, reduced bias, and maintained interpretation. Second, measurement of urine potassium concentration, which may not be routinely available, did not meaningfully improve prediction despite reaching statistical significance in the model. Third, dummy variables for 5 regions were included in models, which could have artificially reduced variability among the regions studied and inflated community-level performance of the equation. Finally, although the study is the broadest to date, external validity remains limited. INTERSALT included communities in Africa, Asia, South America, and the Pacific Islands, but the study by Brown et al. (11) was restricted to 29 communities in North America and Europe. In addition, 4 communities were excluded because of evidence of improper 24-hour urine collection (n = 2) or outlying mean values of urine sodium excretion (n = 2), and the inclusion of dummy variables for included regions makes it unclear how to estimate 24-hour urine sodium for regions not included in the study. Consequently, the resulting equations should be applied with caution to populations outside of North America and Europe, to children and older adults, and to people with intakes of urinary sodium at the extremes of the range.

Nonetheless, there are a number of potential applications for the new equations for estimating urine sodium excretion. As Brown et al. suggested, longitudinal analysis of estimated urine sodium excretion could be used to track changes in dietary sodium intakes within populations, monitoring the results of public health interventions or trends in lifestyle and behavior. In addition, estimated urine sodium excretion could be used to identify communities with particularly high dietary sodium intakes for targeted interventions. These communities could be defined at the level of countries, states/provinces, cities, or neighborhoods. An increasing number of studies point to neighborhood characteristics as determinants of health (12). Diet, perhaps determined by access to healthy foods, is an important potential mechanism that connects neighborhood to health.

As Brown et al. noted, the individual-level correlation of estimated and measured 24-hour urine sodium excretions is modest. Thus, the use of single casual urine sodium measurements to further define individual-level associations of dietary sodium intake with health outcomes, test interactions of genetic polymorphisms and dietary sodium intake with regard to blood pressure and cardiovascular disease, study the effects of dietary sodium intake on response to antihypertensive therapies, and pursue related compelling patient-level questions remains somewhat limited. Application to clinical care also requires further study. Diurnal variation in urine sodium excretion may explain part of the modest performance of casual urine sodium ascertainment. However, much is probably related to large within-individual variation in urine sodium excretion over time (13). The use of multiple casual urine samples may improve the classification of individual-level dietary sodium intake substantially.

In summary, Brown et al. (11) have taken an important step forward in the assessment of population-level dietary sodium intake. Properly applied, their results will facilitate the tracking of dietary sodium intake within populations over time and help identify communities in which dietary sodium restriction is most likely to be beneficial. Further work is needed to extend estimation to additional populations and to improve individual-level assessment of urine sodium excretion.