Meta-analysisindependent cross-checkfull text checked I-VEG-13 in-depth review
Potassium and blood pressure
Effect of increased potassium intake on cardiovascular risk factors and disease: systematic review and meta-analyses.
Overview
Aburto NJ et al.. BMJ 2013 · PMID 23558164 · DOI 10.1136/bmj.f1378
More potassium lowered blood pressure in people with hypertension and was linked to 24% lower stroke risk.
Does the study support the specific claim? No · Methodological quality: high
Plain-language summary
Study explained simply
- What did the researchers want to find out?
Potassium effects
- How was the study done?
Meta-analyses
- How many people, animals or samples were studied?
RCTs and cohorts
- What was actually found?
Lower BP, less stroke
- How large was the observed effect?
SBP -3.49 mm Hg
- What do the results mean in everyday life?
Potassium-rich foods are beneficial.
- What are the study's limitations?
Indirect for vegetables
- What does it say about Elawady's specific claim?
Supports vegetable intake
- Do the results apply to Germany and Europe?
High
- What can explicitly not be concluded from this study?
Not specific to vegetables.
Scientific in-depth review
1. Full reference
- Authors
- Aburto NJ et al.
- Journal
- BMJ
- Year
- 2013
- Identifier
- PMID 23558164 DOI 10.1136/bmj.f1378
- Verification
- verified – PubMed efetch 2026-10-25
- Retracted
- no
- Corrections
- none
- Extent checked
- full text
- Conflicts of interest and funding
Funded by WHO, the Kidney Evaluation Association Japan, and the governments of Japan and the Republic of Korea; one author WHO staff; authors unpaid members of salt-reduction advocacy groups (CASH, WASH) and unpaid WHO/PAHO advisers; one author NIHR-supported.
2. Research question
Effects of potassium intake on BP and CVD.
3. Study design and population
- Study design
- Meta-analysis – WHO-commissioned systematic review and meta-analyses of RCTs (blood pressure, lipids, catecholamines, renal function) and prospective cohorts (CVD, stroke, CHD, mortality), Cochrane methods, PRISMA, GRADE
- Population
Apparently healthy adults and children without impaired renal potassium handling; acutely ill, hospitalised and HIV-positive people excluded
- Country
Multiple (RCTs from 14 countries incl. Europe, Asia, Africa, Americas; cohorts from Finland, Japan, Netherlands, Scotland, Taiwan, USA, one 40-country study)
- Period
Searches to August-September 2011 (CENTRAL, Medline, Embase, WHO ICTRP, LILACS)
- Sample size
Adults: 22 RCTs (1606 participants; 12-353 per trial; 4 weeks-12 months) and 11 cohorts in meta-analysis (127,038 participants); children: 4 studies (2 RCTs, 1 non-randomised trial, 1 cohort)
- Inclusion and exclusion criteria
RCTs >=4 weeks with 24-h urinary potassium and no co-interventions; cohorts with >=1 year follow-up; non-randomised designs only if <3 RCTs for an outcome
4. Methods and endpoints
- Exposure / intervention
Higher potassium intake; in RCTs 20 of 22 used potassium supplements, 2-3 used dietary advice; cohorts assessed dietary potassium
- Comparator
Lower potassium intake / placebo
- Endpoints
Systolic/diastolic BP; blood lipids; catecholamines; serum creatinine; incident stroke, CVD, CHD; all-cause mortality
- Statistical methods
Inverse-variance random-effects meta-analysis; I2; prespecified subgroups; sensitivity analysis excluding high-RoB studies; funnel plots; GRADE
5. Actual results
- Results
Adults: SBP -3.49 mm Hg (95% CI -5.15 to -1.82) and DBP -1.96 (-3.06 to -0.86) after removing outlier trials (all trials: SBP -5.93, -10.15 to -1.70; I2 96%). Hypertensive: SBP -5.32 (-7.20 to -3.43); normotensive: SBP +0.09 (-0.77 to 0.95). Achieved intake 90-120 mmol/day: SBP -7.16 (-12.41 to -1.91). Dietary advice subgroup: SBP -4.19 (-6.46 to -1.92). No adverse effects on lipids, catecholamines or creatinine (-4.86 umol/L, NS). Cohorts: stroke RR 0.76 (0.66-0.89); CVD 0.88 (0.70-1.10); CHD 0.96 (0.78-1.19); all-cause mortality not poolable. Children: SBP -0.28 (-0.49 to 1.05), NS.
- Absolute risks
not reported in full text
- Authors' conclusion
High-quality evidence shows increased potassium lowers blood pressure in people with hypertension without adverse effects on lipids, catecholamines or renal function; higher intake is associated with 24% lower stroke risk (moderate quality by authors' overall rating; direct cohort evidence graded low); increased potassium from foods such as fruits, vegetables and pulses is likely beneficial for people without impaired renal potassium handling.
6. Strengths
- Protocol developed with WHO guideline group; comprehensive multi-database search without language limits
- Duplicate screening and extraction; Cochrane RoB; GRADE
- RCT evidence for blood pressure plus cohort evidence for hard outcomes
- Sensitivity analyses excluding high-RoB studies
7. Weaknesses
- Most RCTs tested supplements, so food-based evidence for vegetables is indirect
- High heterogeneity in overall BP analysis (reduced by excluding outliers)
- Stroke evidence observational (GRADE low for direct evidence); CVD/CHD imprecise
- Few data in children
- Possible small-study bias; few trials reported randomisation/allocation methods
- Excludes people with renal impairment, in whom high potassium can be harmful
8. Risk of bias
- Orienting tool
- AMSTAR 2
- Type of assessment
- structured assessment based on the full text by a single reviewer (no formal double review)
- Full formal assessment
- no
- Overall risk of bias
- low
- Rationale
Single-reviewer structured AMSTAR 2 appraisal from full text; not formal (needs two independent reviewers). Removal of outlier trials to reduce heterogeneity is a post-hoc choice, but conclusions hold in all-trial analysis.
| Domain | Judgement | Rationale |
|---|---|---|
| Item 1 PICO components | low | Questions defined by WHO guideline group. |
| Item 2 Protocol a priori (critical) | low | Protocol developed in March 2011 with prespecified subgroups (not stated as publicly registered). |
| Item 3 Explanation of included designs | low | Rule for including non-randomised designs set a priori. |
| Item 4 Comprehensive search (critical) | low | Five databases incl. trial registry and LILACS; no language limits. |
| Item 5 Duplicate selection | low | Two independent reviewers. |
| Item 6 Duplicate extraction | low | Two extractors plus third-reviewer check. |
| Item 7 Excluded studies justified (critical) | some concerns | Flow diagram with counts; full excluded list not in main text. |
| Item 8 Description of included studies | low | Supplementary characteristics tables. |
| Item 9 RoB assessment (critical) | low | Cochrane domains for RCTs; additional domains for cohorts. |
| Item 10 Funding of included studies | high | Not reported. |
| Item 11 Appropriate meta-analytic methods (critical) | low | Random-effects inverse variance; heterogeneity explored. |
| Item 12 Impact of RoB on results | low | Sensitivity analysis excluding high-RoB studies. |
| Item 13 RoB considered in interpretation (critical) | low | GRADE applied. |
| Item 14 Heterogeneity explained | low | Hypertension status explained most heterogeneity. |
| Item 15 Publication bias (critical) | some concerns | Funnel plots unclear; possible small-study bias acknowledged. |
| Item 16 Conflicts of interest | low | ICMJE declarations. |
Full-text review (25 October 2026): Added that 20 of 22 RCTs used supplements (dietary-advice subgroup SBP -4.19), all-trial vs outlier-excluded estimates, hypertension subgroups, 90-120 mmol/day subgroup, GRADE per outcome, search dates, funding (WHO, Japan, Korea) and advocacy roles; AMSTAR 2 item-level appraisal (overall unchanged: low); directness lowered to very indirect.
9. Assessment of the statistical interpretation
Moderate BP reductions in hypertensive adults with high-certainty RCT evidence (mostly supplements); stroke association observational. Bears on vegetables only indirectly via potassium intake, which can also come from other foods.
10. Relevance to the specific claim
Supports nutritional importance of vegetables/legumes (C-VEG-16).
11. Independent contrary findings
None.
12. Transferability to Germany and Europe
High.
13. Overall methodological judgement
Key nutrient evidence.
| Multidimensional assessment | |
|---|---|
| Authenticity of the source | verified |
| Methodological quality | high |
| Risk of bias | low |
| Precision of results | moderate |
| Directness | very indirect |
| Reproducibility | replicated |
| Consistency with the overall evidence | consistent |
| Regional transferability | high |
| Adequacy of the interpretation (by the archive) | not applicable |
14. Open questions
Food-based (vegetable) potassium trials with hard outcomes; potassium intake achieved on diets that exclude vegetables and legumes.
15. Linked fact checks
- Avoiding (nearly) all vegetables, salads and legumes is compatible with good health. – Studies contradicting or limiting the claim