Abstract P450: Sodium and Potassium Intake and Risk of Cardiovascular Events: Updated Meta-analyses
Introduction. Recently a few prospective population studies provided additional information concerning the reported statistical associations between sodium or potassium intake and risk of cardiovascular (CV) accidents. Therefore, we updated our previous meta-analyses of sodium and potassium intake and risk of CV events and assessed in addition the relationship between sodium/potassium ratio and CV outcomes.
Methods. We performed a systematic search of publications (1966-2012) in online databases. Criteria for inclusion were: population studies with assessment of sodium and potassium intake as baseline exposure and CV disease determined prospectively as outcome. Relative risks (RR) and 95%CI were extracted and pooled using a random effect model. Sensitivity analysis was carried out, heterogeneity and publication bias were assessed.
Results. Ten studies for sodium intake were included in the analyses (192,973 participants;6,218 CV events; 5,379 stroke). Higher compared with lower sodium intake (2.1 g/day difference or ~ 5 g of salt) was associated with 20% significantly greater risk of stroke (95%CI:+4 to +39%) and with a non significant trend toward higher risk of total CV events (+13%; CI= -2 to +31%), that achieved statistical significance after exclusion of a single cohort upon sensitivity analysis (+17%; CI=+1 to 34%). In the analysis of 10 studies (268,276 participants; 8,746 events) a 1.6 g/day difference in potassium intake was significantly and inversely associated with stroke risk (-20%; CI= -29 to -9%). The analysis of the relationship between the urinary sodium/potassium ratio and total CV risk (5 studies; 78,401 participants; 3,554 events) showed a non significant positive association (RR:1.14;0.97-1.35), that after exclusion of a single outlier led to a statistically significant pooled estimate of 1.21 (1.02-1.43). In all analyses, there was significant heterogeneity among studies, but no evidence of publication bias.
Discussion. These updated meta-analyses confirm that both high sodium and low potassium intake are associated with greater risk of stroke. The analyses also indicate weaker positive associations of urinary sodium excretion and urinary sodium/potassium ratio with total CV event rate. These results once again confirm the appropriateness of worldwide recommendations for a population decrease in salt intake and for increased consumption of potassium-rich foods to prevent CV disorders.
- © 2013 by American Heart Association, Inc.