Shared Primacy of Sodium and Potassium on Cardiovascular Risk

What Does this Important Study Show? 

TOHP I and TOHP II4, 5 evaluated nonpharmacological interventions aimed at reducing blood pressure in participants aged 30 to 54 years with prehypertension (diastolic blood pressure of 80 to 89 mm Hg in TOHP I and diastolic blood pressure of 83 to 89 mm Hg with systolic blood pressure < 140 mm Hg in TOHP II; not using antihypertensive medication in either trial). A mean of 3 to 7 twenty-four-hour urinary excretion measurements of sodium and potassium were performed in each participant during 18 months in TOHP I and 36 months in TOHP II. Urinary excretion values for participants assigned to a sodium reduction intervention or a short-term potassium supplement intervention were not included.

The TOHP Follow-up Study3 examined the relationship between these previously established mean intakes of sodium and potassium with subsequent CVD (including stroke, myocardial infarction, coronary revascularization, and cardiovascular mortality) in 2,275 participants during 10 to 15 years after termination of the trials. Mean urinary sodium excretion was 176 mEq/24 h in men and 138 mEq/24 h in women, and mean urinary potassium excretion was 67 mEq/24 h in men and 51 mEq/24 h in women. Mean sodium-potassium excretion ratios were 2.88 and 2.97 in men and women, respectively. Cox proportional hazards regression models were used to assess the hazard ratio for CVD from the end of the trial period to the end of the follow-up period for quartile categories of urinary excretion. Models were adjusted for age, sex, race/ethnicity, treatment assignment, education status, family history, weight, alcohol use, smoking, and exercise. A non–statistically significant trend toward increasing CVD risk with increasing urinary sodium excretion was detected and was attenuated in fully adjusted analyses. An inverse relationship between urinary potassium excretion and CVD risk was detected, but also did not reach statistical significance in fully adjusted models. However, there was a statistically significant increased risk of CVD and greater sodium-potassium excretion ratio, with a 50% increase across extreme quartiles detected in fully adjusted analyses (P = 0.04). When urinary sodium and potassium excretion were included in the same model, their relationship with CVD risk increased. Examining sodium and potassium excretion as linear terms, there was a 42% estimated increase in CVD risk per 100-mEq/24 h increase in urinary sodium excretion (P = 0.05) and a 33% risk reduction in CVD per 50-mEq/24 h increase in urinary potassium excretion (P = 0.12). However, sodium-potassium excretion ratio showed a statistically significant linear relationship with CVD risk, with a 24% increase in risk per unit of the ratio (P = 0.01), which was consistent across subgroups. The estimated effects of the urinary cationic ratio were similar for coronary artery disease and stroke. The investigators interpreted the improved fit of the cationic ratio with CVD as reflecting either an inherent biological synergism between sodium and potassium2 or correction for urinary collection inaccuracies. We consider the latter possibility unlikely because urinary sodium-potassium ratio is not constant throughout the day because of variable cationic content of ingested foods; inaccuracies of urinary collection should alter both the total excretion of cations and their ratio. Nonetheless, the evidence presented provides some support for the hypothesis that lower sodium consumption and higher dietary potassium intake may reduce the incidence of CVD.

How Does this Study Compare With Prior Studies? 

Observational studies and randomized controlled trials have shown that individuals with lower sodium intake or higher potassium intake have lower blood pressure and reduced risk of hypertension.2, 5, 6, 7, 8 Additionally, limited evidence from long-term interventions suggests that decreased dietary sodium or use of potassium substituted for some dietary sodium might reduce CVD risk.9, 10 However, as noted, those studies generally have used a single dietary recall or single urinary excretion as measures of sodium and potassium intake. In contrast, Cook et al3 assessed the usual dietary intake of sodium and potassium in each participant as the mean of approximately 4 measurements of urinary excretion, thereby enhancing the level of certainty. Corrections for within-person variability and correlation between measurements were performed by using multivariate regression calibration. Additional strengths of this study include its large number of participants and long follow-up. Among the limitations of the study are the lack of measurement of urinary sodium and potassium excretion during the long follow-up and absence of data for certain cardiovascular risk modifiers, such as lipid profile, diabetes, renal function, and medications (eg, angiotensin-converting enzyme inhibitors and statins). Also, identification of nonfatal events was based on communication with participants by mail and telephone, followed by confirmation through review of medical records.

Notably, the urinary sodium-potassium ratio in the International Study of Salt and Blood Pressure (INTERSALT) with 10,079 participants from 32 countries had a significant direct relationship with blood pressure that was stronger than that of either sodium or potassium excretion alone.6 In support of the TOHP Follow-up Study, a randomized trial in Taiwan10 involving 1,981 male, largely elderly, veterans showed that a combined increase in potassium intake and decrease in sodium intake compared with regular salt consumption was associated with a significant 41% reduction in CVD mortality and significant reduction in medical expenses during an average 31-month follow-up.

What Should Clinicians and Researchers Do? 

The TOHP Follow-up Study documents in middle-aged prehypertensive Americans that mean urinary excretion of sodium and potassium, reliable measures of intake, are approximately 3-fold and one-half the levels recommended by the Institute of Medicine in 2005, respectively.11 As a corollary, the sodium-potassium excretion ratio is approximately 2.9, contrasting with the recommended ratio of approximately 0.5. These data are in general agreement with other measurements in American adults.11 Consequently, drastic modifications in the diet are needed to close this large gap.

Human kidneys are poised to conserve sodium and excrete potassium, a physiological platform that served prehistoric humans well, but is counterproductive for the sodium-rich and potassium-poor modern diet. We have recently emphasized2 the interaction of the modern Western diet and the kidneys in the pathogenesis of primary hypertension (Fig 1). The study of Cook et al3 provides valuable data supporting the interaction of sodium and potassium as the dominant environmental factor in the pathogenesis of primary hypertension and its associated CVD risk. Prospective randomized long-term trials are required to establish the cardiovascular effects of this dual dietary strategy.

View Large Image Download to PowerPoint

Figure 1. Interaction of the modern Western diet and the kidneys in the pathogenesis of primary hypertension. Reproduced from Adrogué and Madias2 with permission of the Massachusetts Medical Society, ©2007, all rights reserved.

Implementation of the Institute of Medicine's recommendations would require a systematic campaign targeting both the general public and health care professionals. One cannot overemphasize the importance of gaining the cooperation of the food industry to limit the deviation in sodium and potassium content of processed foods from their natural counterparts. Making this dietary transition is a tall order. However, failure to do so would represent a missed opportunity to decrease markedly the CVD risk. Extrapolating the data from Cook et al3 to full compliance with the Institute of Medicine's guidelines projects a more than 50% reduction in CVD risk.