Association between sodium intake and change in uric acid, urine albumin excretion, and the risk of developing hypertension

Abstract

Background: A high-sodium diet has little short-term effect on blood pressure in nonhypertensive individuals but, for unclear reasons, is associated with hypertension if consumed long term. We hypothesized that a chronically high sodium intake would be associated with increases in biomarkers of endothelial dysfunction, specifically serum uric acid (SUA) and urine albumin excretion (UAE), and that high sodium intake would be associated with incident hypertension among those with higher SUA and UAE.

Methods and results: We prospectively analyzed the associations between sodium intake and the change in SUA (n=4062) and UAE (n=4146) among participants of the Prevention of Renal and Vascular End Stage Disease (PREVEND) study who were not taking antihypertensive medications. We also examined the association of sodium intake with the incidence of hypertension (n=5556) among nonhypertensive participants. After adjustment for confounders, each 1-g-higher sodium intake was associated with a 1.2-μmol/L increase in SUA (P=0.01) and a 4.6-mg/d increase in UAE (P<0.001). The relation between sodium intake and incident hypertension varied according to SUA and UAE. For each 1-g-higher sodium intake, the adjusted hazard ratio for developing hypertension was 0.98 (95% confidence interval, 0.89-1.08) among those in the lowest tertile of SUA and 1.09 (1.02-1.16) among those in the highest tertile. Corresponding hazard ratios were 0.99 (confidence interval, 0.93-1.06) among participants whose UAE was <10 mg/d and 1.18 (confidence interval, 1.07-1.29) among those whose UAE was >15 mg/d.

Conclusions: Over time, higher sodium intake is associated with increases in SUA and UAE. Among individuals with higher SUA and urine UAE, a higher sodium intake is an independent risk factor for developing hypertension.

Figure 1

Assembly of the three study populations. SBP, systolic blood pressure. DBP, diastolic blood pressure. SUA, serum uric acid. UAE, urine albumin excretion. Relevant data, information on urinary sodium excretion, SUA, or UAE. Analyses of the change in SUA and UAE included individuals who had available data at baseline and 3^rd screening exam, and who were not taking antihypertensives at either time point. Analyses of incident hypertension included non-hypertensive individuals at baseline.

Figure 2

Adjusted change in serum uric acid and urine albumin excretion during follow-up according to sodium intake. The adjusted mean change (± standard error) is shown for each quartile of urinary sodium excretion compared with the lowest quartile. Panel A displays results for change in serum uric acid. Panel B displays results for change in urine albumin excretion.

Figure 3

Adjusted hazard ratio for incident hypertension according to sodium intake and either baseline serum uric acid or urine albumin excretion. Hazard ratios are compared with a common reference group (the lowest quartile of sodium intake and either the lowest tertile of serum uric acid or the lowest category of urine albumin excretion). Hazard ratios are adjusted for age, body mass index, sex, alcohol intake, smoking status, family history of hypertension, estimated glomerular filtration rate, plasma levels of glucose and cholesterol, and urinary levels of potassium, calcium, and creatinine. Panel A displays results for the hazard ratio of hypertension according to quartile of urinary sodium excretion and tertile of serum uric acid; a common reference group was used, specifically those participants in the lowest quartile of urinary sodium excretion and the lowest tertile of serum uric acid. Panel B displays results for the hazard ratio of hypertension according to quartile of urinary sodium excretion and category of urine albumin excretion; a common reference group was used, specifically those participants in the lowest quartile of urinary sodium excretion and the lowest category of urine albumin excretion. A common reference group, which was employed here, was not used in Table 3.