Higher sodium, lower potassium urine excretion tied to increased CV risk

Roshini Claire Anthony
29 Nov 2021
Higher sodium, lower potassium urine excretion tied to increased CV risk

Higher sodium and lower potassium intake, measured through repeat 24-hour urine samples, are tied to an elevated risk of cardiovascular (CV) events, according to a study presented at AHA 2021.

“Our study showed a significant linear association between sodium intake, as measured with the use of multiple 24-hour urine samples, and CV risk in a dose–response manner with a daily sodium intake of approximately 2,000 to 6,000 mg. Higher potassium intake was associated with a lower CV risk,” said the authors.

“These findings may support reducing sodium intake and increasing potassium intake from current levels,” they said.

For this study, the researchers used individual-participant data from six prospective cohorts of generally healthy adults in the US and Europe. Each of the 10,709 participants (mean age 51.5 years, 54.2 percent female) included had 2 24-hour urine samples, the average of which determined potassium and sodium excretion.

Twenty-four urinary sodium excretion, based on 37,896 samples, was a median 3,270 mg. Daily intake of sodium and potassium was estimated at 3,516 and 3,292 mg, respectively.

The participants were followed up for a median 8.8 years during which time 571 CV events (a composite of coronary revascularization or fatal or nonfatal myocardial infarction [MI] or stroke) were documented, corresponding to an incidence rate of 5.9 per 1,000 person-years. These included 232 MIs, 213 coronary revascularizations, 136 strokes, and 12 additional deaths from CV causes. Over a median follow-up of 14.8 years, 1,100 deaths were documented.

Based on quartiles of urinary biomarkers, there was a stepwise increasing risk of CV events with increasing levels of sodium excretion (adjusted hazard ratio [HR], 1.25, 95 percent confidence interval [CI], 0.96–1.63 [quartile 2], HR, 1.44, 95 percent CI, 0.93–2.23 [quartile 3], and HR, 1.60, 95 percent CI, 1.19–2.14 [quartile 4] vs quartile 1). [AHA 2021, abstract 10005; N Engl J Med 2021;doi:10.1056/NEJMoa2109794]

Increasing levels of potassium excretion were tied to a reduced risk of CV events (HR, 1.00, 95 percent CI, 0.78–1.26 [quartile 2], HR, 0.86, 95 percent CI, 0.67–1.11 [quartile 3], and HR, 0.69, 95 percent CI, 0.51–0.91 [quartile 4] vs quartile 1).

A higher sodium-to-potassium ratio was also tied to an increased risk of CV events (HR, 1.02, 95 percent CI, 0.78–1.33 [quartile 2], HR, 1.40, 95 percent CI, 0.99–1.99 [quartile 3], and HR, 1.62, 95 percent CI, 1.25–2.10 [quartile 4] vs quartile 1), with each unit increase tied to a 24 percent increased risk of CV events (HR, 1.24, 95 percent CI, 1.12–1.37).

Each 1,000 mg daily increase in sodium excretion was associated with an increased risk of CV events (HR, 1.18, 95 percent CI, 1.08–1.29), while each 1,000 mg daily increase in potassium excretion was tied to a decreased risk of CV events (HR, 0.82, 95 percent CI, 0.72–0.94).

Secondary analyses showed no association between sodium excretion and any-cause death (HR, 1.02 per 1,000 mg increase), while higher potassium excretion and lower sodium-to-potassium ratio were tied to a lower risk of any-cause death.

“[W]e speculate [that the lack of association between sodium intake and any-cause death] may be due to the occurrence of deaths from non-CV causes diluting the relation between sodium intake and mortality,” the authors suggested.

According to the authors, previous studies assessing the impact of sodium intake on CV risk have been hampered by methodological issues including the use of spot or single 24-hour urine tests, questionnaires, and populations with chronic diseases.

“[O]ur study constitutes a large study on CV risk and sodium intake as assessed by a very reliable method. By studying generally healthy populations, we also reduced confounding and potential reverse-causation bias.” Nonetheless, residual confounding is still a possibility and between-study heterogeneity may have affected the findings, they acknowledged.


Editor's Recommendations