International Journal of Hypertension (Jan 2018)

Blood Pressure in relation to 24-Hour Urinary Sodium and Potassium Excretion in a Uruguayan Population Sample

  • Paula Moliterno,
  • Ramón Álvarez-Vaz,
  • Matias Pécora,
  • Leonella Luzardo,
  • Luciana Borgarello,
  • Alicia Olascoaga,
  • Carmen Marino,
  • Oscar Noboa,
  • Jan A. Staessen,
  • José Boggia

DOI
https://doi.org/10.1155/2018/6956078
Journal volume & issue
Vol. 2018

Abstract

Read online

Many public health policies in Latin America target an optimized sodium and potassium intake. The aims of this study were to assess the sodium and potassium intake using 24-hour urinary analysis and to study their association with blood pressure in a Uruguayan population cohort using cluster analysis. A total of 149 participants (aged 20–85 years) were included in the study, and office blood pressure, anthropometric measurements, biochemical parameters in the blood, and 24-hour urine samples were obtained. The overall mean sodium and potassium excretion was 152.9 ± 57.3 mmol/day (8.9 ± 3.4 g/day of salt) and 55.4 ± 19.6 mmol/day, respectively. The average office systolic/diastolic blood pressure was 124.6 ± 16.7/79.3 ± 9.9 mmHg. Three compact spherical clusters were defined in untreated participants based on predetermined attributes, including blood pressure, age, and sodium and potassium excretion. The major characteristics of the three clusters were (1) high systolic blood pressure and moderate sodium excretion, (2) moderate systolic blood pressure and very high sodium excretion, and (3) low systolic blood pressure and low sodium excretion. Participants in cluster three had systolic blood pressure values that were 23.9 mmHg (95% confidence interval: −29.5 to −1.84) lower than those in cluster one. Participants in cluster two had blood pressure levels similar to those in cluster one (P = 0.32) and worse metabolic profiles than those in cluster one and three (P 0.47). An effect of sodium and potassium intake on blood pressure levels was not found at the population level using regression or cluster analysis.