Randomized trials of salt restriction have consistently demonstrated that decreasing salt consumption lowers blood pressure, but results of observational studies of salt intake and cardiovascular disease have been conflicting. After excluding individuals with prevalent cardiovascular or kidney disease in the prospective UK Biobank study, we examined the within-person variability in spot urinary sodium excretion and its impact on associations with systolic blood pressure and risk of incident cardiovascular disease. Spearman correlation coefficients were used to assess within-person variability in spot urinary sodium, and associations between sodium and blood pressure were assessed using linear regression in participants with measurements at baseline (N=355 134) and after 9 years (N=33 915). Cox regression was used to assess associations with the risk of cardiovascular disease over the same follow-up period (N=5566 events). The within-person variability in urinary sodium was extreme, with a self-correlation coefficient of 0.35 over 4 years. Each 100 mmol/L higher usual urinary sodium was associated with 3.09 mm Hg higher systolic blood pressure (95% CI, 2.7-3.48) at baseline, but had no association at 9 years (0.97 [-0.44 to 2.37]). Likewise, there was no association between urinary sodium and risk of cardiovascular disease over the same follow-up period (hazard ratio, 1.05, [0.87-1.26]). While spot urinary sodium measurements were associated with immediate effects on blood pressure at baseline, the extreme within-person variability in urinary sodium precluded detection of associations with future blood pressure at resurvey or risk of cardiovascular disease. The limitations of observational studies, irrespective of study size, should be recognized when assessing public policy on salt restriction.
blood pressure, cardiovascular disease, cohort study, hypertension, sodium