Evidence base
This guideline is based on a review of the epidemiologic literature, including three new systematic reviews conducted by WHO (44-46) to summarize the evidence regarding potassium intake and health outcomes. Specific health outcomes considered were:
in adults – blood pressure, all-cause mortality, cardiovascular disease, stroke, coronary heart disease, renal function, blood lipids, catecholamine levels and other potential adverse effects;
in children – blood pressure, blood lipids, catecholamine levels and other
potential adverse effects.
The specific research questions guiding these systematic reviews were:
intake on health outcomes in adults and children?
in adults and children of increased potassium intake to:
- –
less than approximately 90 mmol/day;
- –
approximately 90–120 mmol/day;
- –
approximately 120–155 mmol/day;
- –
more than approximately 155 mmol/day?
Evidence was considered conclusive of either a benefit or a harm from increased potassium intake if the point estimate suggested a benefit or harm and the 95%CI did not overlap a threshold of relevance. That is to say, if the real value were the high or the low 95%CI and that value was still of clinical relevance, the evidence was considered conclusive. If the point estimate were near the null value and the 95%CI did not overlap a threshold of relevance (e.g. if the real value were the high or the low 95%CI value and that value was not of clinical relevance) the evidence was considered conclusive of no effect. In such cases, the point estimates were considered precise.
Conversely, evidence was considered inconclusive if the point estimate suggested a benefit or a harm but the 95%CI crossed a threshold of relevance (e.g. if the real value were the high or the low 95%CI value and that value was not of clinical relevance). In such cases, the point estimates were considered imprecise.
Adults
Blood pressure in adults
WHO conducted a systematic review to explore the relationship between potassium and blood pressure in adults. The review identified 22 RCTs that met the inclusion criteria (44). Of these RCTs, 17 were conducted in individuals with hypertension (defined as a blood pressure ≥140/90 mmHg (47)), three in individuals without hypertension, and two in populations of individuals with and without hypertension. Two studies were conducted only in women, and the other 20 in populations of men and women. Studies were conducted in all regions of the world and all measured 24-hour urinary potassium excretion. WHO estimated potassium intake levels from 24-hour urinary potassium excretion using the conversion factor of 1.3 (48).
The lower potassium intake level varied among studies and ranged from 45 to 100 mmol/ day, with a median value of 73 mmol/day. The increased potassium intake ranged from 70 to 247 mmol/day, with a median value of 127 mmol/day. There was a median difference in increased potassium relative to lower potassium of 57 mmol/day (74%).
The meta-analysis of 21 studies with 21 comparisons found that increased potassium resulted in a decrease in resting systolic blood pressure of 3.49 mmHg (95% confidence interval [CI]: 1.82, 5.15) (quality of evidence high1) and a decrease in resting diastolic blood pressure of 3.02 mmHg (95%CI: 1.17, 4.86) (quality of evidence high). The meta-analysis of four studies with four comparisons reporting ambulatory blood pressure found that increased potassium intake decreased ambulatory systolic blood pressure by 3.04 mmHg (95%CI: 0.66, 5.42) (quality of evidence moderate), and ambulatory diastolic blood pressure by 1.24 mmHg (95%CI: –0.66, 3.13) (quality of evidence moderate). The findings demonstrate that across a wide range of baseline intakes, increasing potassium intake is beneficial in terms of blood pressure and they concur with three earlier systematic reviews and meta-analyses (3, 4, 31), but not with a fourth (41).
The results suggest that the greatest impact on blood pressure was achieved when the increased potassium intake was approximately 90–120mmol/day (44). The meta-analysis of five studies (with five comparisons) that achieved an increased potassium intake of 90–120mmol/ day demonstrated a reduction in systolic blood pressure of 7.16 mmHg (95%CI: 1.91, 12.41) (quality of evidence high), and a reduction in diastolic blood pressure of 4.01 mmHg (95%CI: –0.42, 8.44) (quality of evidence moderate). Only one study with one comparison, in which the increased potassium intake was 90–120 mmol/day, reported ambulatory systolic and diastolic blood pressure; increased potassium intake resulted in a non-significant decrease of 1.80 mmHg (95%CI: -2.42, 7.02) in ambulatory systolic blood pressure and 1.40mmHg (95%CI: -2.34, 5.14) in ambulatory diastolic blood pressure (quality of evidence moderate).
All-cause mortality, cardiovascular disease, stroke, and coronary heart disease in adults
WHO conducted a systematic review on the relationship between potassium consumption and cardiovascular disease, stroke, coronary heart disease and all-cause mortality (45). The review updated and reanalysed data from the recent systematic review of D’Elia and colleagues (30). Only one study that met the inclusion criteria reported all-cause mortality. The results of this study were inconclusive (risk ratio [RR] 1.082 ; 95%CI: 0.91, 1.29) (quality of evidence very low). Twelve prospective cohort studies with more than 127,000 participants measured potassium intake through urinary potassium excretion, dietary records or some combination of these methods, and compared the incidence of cardiovascular disease, stroke or coronary heart disease between the lowest and highest potassium-consuming groups. Populations had wide ranges of potassium intake: some consumed approximately 35 mmol/ day in the lowest group and 65 mmol/day in the highest group, whereas others consumed approximately 65 mmol/day or more in the lowest group and 110–150 mmol/day in the highest group. The follow-up period ranged from 2 to 19 years. Studies were conducted in Asia, Europe and the USA, and one study used data from individuals in 40 different countries. Two studies were conducted exclusively in individuals without hypertension, two were specifically in a heterogeneous group of individuals both with and without hypertension, and the remaining studies did not specify the blood pressure status of the study population.
The meta-analysis of four studies with four comparisons with cardiovascular disease as an outcome was inconclusive regarding the association between potassium and cardiovascular disease (RR 0.88; 95%CI: 0.70, 1.11) (quality of evidence very low). The meta-analysis of nine studies with 10 comparisons with stroke as an outcome was supportive of a reduction in risk of stroke with increased potassium (RR 0.79; 95%CI: 0.68, 0.93) (quality of evidence low). The meta-analysis of four studies with four comparisons between a potassium intake of ≥90 mmol/day and <90 mmol/day demonstrated a reduction in stroke associated with consuming ≥90 mmol/day (RR 0.70; 95%CI: 0.56, 0.88) (quality of evidence low). The outcome of the meta-analysis of three studies with three comparisons reporting coronary heart disease as the outcome was inconclusive (RR 0.97; 95%CI: 0.77, 1.24) (quality of evidence very low).
There is a well-established relationship between increasing blood pressure and increasing risk of cardiovascular diseases, particularly coronary heart disease and stroke (12, 49). Blood pressure is thus considered a reliable biomarker for estimating risk of cardiovascular disease (50, 51). Recognizing the limitations of any biomarker, it was nonetheless determined that blood pressure could be a suitable proxy indicator for risk of cardiovascular disease, stroke, and coronary heart disease. Therefore, in addition to the direct evidence from observational cohort studies on the relationship between potassium intake and cardiovascular disease, stroke, and coronary heart disease, the data from RCTs on the effect of increased potassium consumption on blood pressure (44) were used as part of the evidence base for considering the effect of increased potassium intake on these outcomes.
Potential adverse effects in adults
Potential adverse effects from increased potassium intake included changes in blood lipids (e.g. increased total cholesterol, low-density lipoprotein [LDL] cholesterol and triglyceride; decreased high-density lipoprotein [HDL] cholesterol); changes in renal function; increases in catecholamine levels; and any other adverse effects or minor side-effects (i.e. dizziness, headache, muscle pain) reported by study authors. The decrease in blood volume caused by increased potassium activates the sympathetic nervous system, resulting in the release of adrenaline and noradrenaline. Decreased blood volume may also be responsible for changes in blood lipid concentrations. The effect of 4 or more weeks of increased potassium consumption on these outcomes was addressed in the systematic review of the literature that considered the effect of potassium on blood pressure (44). The meta-analysis of three trials reporting total cholesterol concentration showed that increased potassium intake relative to lower potassium intake resulted in a non-significant decrease of 0.12mmol/L (95%CI: –0.33, 0.09) in total cholesterol (quality of evidence high). Only one study in the literature reported LDL concentration, and the results indicated a non-significant decrease in LDL of 0.10 mmol/L (95%CI: –0.18, 0.38) (quality of evidence high). Although the data were limited, the meta-analysis of two studies with two comparisons reporting HDL and triglyceride concentrations also showed non-significant decreases in those indicators with increased potassium consumption (HDL 0.01 mmol/L; 95%CI: –0.1, 0.13; triglyceride 0.11 mmol/L; 95%CI: –0.26, 0.48) (quality of evidence high). The meta-analysis of three trials showed that increased potassium intake relative to lower potassium intake resulted in a non-significant decrease of 4.32 pg/mL (95%CI: –15.13, 23.78)in plasma noradrenaline (and
3.94 pg/mL (95%CI: –1.34, 9.22) plasma adrenaline ( (quality of evidence high). No studies that met the inclusion criteria reported urinary catecholamine levels. As an indicator of renal function, three studies quantified serum creatinine concentration. The meta-analysis of these studies suggested that increased potassium intake had no effect on renal function with a non-significant increase of 4.86 µmol/L (95%CI: –3.87, 13.59) in these samples of individuals with apparently normal renal function (quality of evidence high). Although the evidence was limited, the data from RCTs were conclusive of no adverse effect of increased potassium intake in terms of blood lipids, catecholamine levels or renal function.
No minor side-effects as a result of increased potassium intake were reported in any of the RCTs. Though these studies were all of relatively short duration, the absence of any complaints of adverse effects with increased potassium intake is consistent with the literature (32, 33). The body is able to efficiently adapt and excrete excess potassium via the urine when consumption exceeds needs (18, 32, 33), and there have been no reports of toxicity of potassium from consumption in food (34).
Final considerations of the evidence
WHO attempted to discern differences in the effect of increased potassium on outcomes according to type of intervention (i.e. supplements, fortification or food), type of potassium supplement (i.e. potassium citrate, potassium chloride or other) and gender. In the systematic review and meta-analysis of RCTs in adults reporting blood pressure as an outcome (44), the subgroup analysis of 19 studies using potassium supplements showed a decrease in systolic blood pressure of 3.31 mmHg (95%CI: 1.55, 5.07) (quality of evidence high), and the subgroup analysis of three studies using dietary changes showed a decrease in systolic blood pressure of 4.19 mmHg (95%CI: 1.92, 6.46) (quality of evidence high). The results suggest that an increase in potassium intake from either supplement or food has a beneficial effect on blood pressure.
In assessing the results of supplementation studies, it was possible to isolate the effect of potassium because it was the only variable manipulated between increased potassium and usual or lower potassium groups. The consistency in results from studies with increased potassium through dietary change supports the health benefit of potassium specifically, and not the conjugate anion found in the supplements used in the supplementation studies. Additionally, all cohort studies compared groups consuming different levels of potassium from foods (44). The cohort studies suggested a positive effect of increased potassium on stroke, further strengthening the conclusion that specifically increasing potassium has beneficial effects on health. No studies that met the inclusion criteria looked specifically at potassium fortification of food, mainly because such studies also manipulated sodium levels. One study used potassium citrate, one used potassium bicarbonate and one used a combination of potassium citrate and bicarbonate, whereas all other supplementation studies used potassium chloride; thus, it was not possible to compare different supplement types. Twenty of the 22 RCTs and nine of the 12 cohort studies were in mixed populations of men and women. Although differences by gender could not be compared, the overall positive effect of increased potassium found in these studies supports a beneficial effect in both men and women.
Addressing the optimal ratio of sodium to potassium was outside the scope of this guideline; however, we undertook subgroup analysis of the RCTs to explore whether different levels of sodium intake influence the effect of potassium on blood pressure. Only one study had a mean sodium intake level of <2 g/day, and it found a non-significant decrease of 2.00 mmHg (95%CI: –7.70, 11.70) on systolic blood pressure with increased potassium intake (quality of evidence moderate), but conclusions should not be drawn from such limited evidence. In the 15 studies with a mean sodium intake of 2–4 g/day, increased potassium intake decreased systolic blood pressure by 1.97 mmHg (95%CI: 0.52, 3.41) (quality of evidence high). In the five studies with a mean sodium intake of
>4 g/day, increased potassium intake decreased systolic blood pressure by 6.91 mmHg (95%CI: 2.29, 11.53) (quality of evidence high). Although the difference in the effect estimates was not statistically significant, the results suggest that potassium may be more effective in reducing blood pressure at higher sodium consumption levels, which is consistent with previous findings (4). There was still a significant benefit of increased potassium intake on blood pressure when populations consumed 2–4 g/day of sodium; hence, with most populations around the world consuming more than 2–4 g/day of sodium (40), increased potassium intake should benefit most countries.
The RCTs were also grouped by baseline potassium intake. In the two studies in which baseline intake was <50mmol/day, increased potassium intake decreased systolic blood pressure by 3.89 mmHg (95%CI: 0.74, 7.03). In the 14 studies with a baseline potassium intake of 50-80mmol/day, increased potassium intake decreased systolic blood pressure by 3.39mmHg (95%CI: 1.26, 5.51) and diastolic blood pressure by 1.53 mmHg (95%CI: 0.25, 2.80). In the five studies with a baseline potassium intake of >80mmol/day, increased potassium intake decreased systolic blood pressure by 4.11 mmHg (95%CI: 1.97, 6.26) and diastolic blood pressure by 3.38 mmHg (95%CI: 2.02, 4.74). Thus, increased potassium intake had a beneficial effect on blood pressure regardless of baseline potassium intake.
The RCTs were grouped by blood pressure status at baseline. In the three studies conducted exclusively in individuals with normal blood pressure, increased potassium intake resulted in a non-significant increase in systolic blood pressure of 0.09 mmHg (95%CI: –0.95,0.77) (quality of evidence moderate). In the 16 studies conducted in individuals with hypertension, increased potassium intake decreased systolic blood pressure by 5.32 mmHg (95%CI: 3.43, 7.20 (quality of evidence high). Although it appears that potassium may only reduce blood pressure in individuals with hypertension, the studies in individuals without hypertension were of relatively short duration, and the effect over time on the prevention of elevated blood pressure is not known. Given the high prevalence of hypertension in adult populations globally (2), the relatively low potassium intake in most populations (9, 10, 53), and the clear benefit of increased potassium intake in individuals with high blood pressure, increasing potassium intake is likely to be broadly beneficial to populations around the world.
Finally, the modest reduction in systolic blood pressure (3.49 mmHg) and in diastolic blood pressure (3.02 mmHg) would have important public health benefits. Elevated blood pressure is the leading risk factor for mortality, accounting for almost 13% of death globally (2). In the USA, a relatively small decrease of 2 mmHg in diastolic blood pressure in the population could result in an estimated 17% decrease in the prevalence of hypertension, 6% decrease in risk of coronary heart disease, and 15% decrease in risk of stroke; it could also prevent an estimated 67,000 coronary heart disease events and 34,000 stroke events every year (54). In the United Kingdom, researchers estimate that a 5 mmHg reduction in systolic blood pressure could reduce the prevalence of hypertension by 50% (55). Additionally, the relationship between blood pressure and risk of vascular mortality is positive, strong and linear down to a systolic blood pressure of 115 mmHg, below which there is no evidence (49). Thus, almost all reduction in blood pressure is beneficial for health, and modest population-wide reductions in blood pressure result in important reductions in mortality, substantial health benefits and meaningful savings in health-care costs (2, 12, 13).
