- Click here for explanation of classification scheme.

To evaluate the effects of potassium supplementation on health outcomes and blood pressure in adults with primary hypertension.

• Randomized controlled trials (RCTs) of a parallel or crossover design comparing oral potassium supplements with placebo, no treatment or usual care
• Treatment and follow-up of eight weeks or more (extremely restrictive criterion)
• Oral potassium supplementation, or dietary interventions that manipulated only potassium intake
• Participants over 18 years, with raised systolic blood pressure (SBP) equal or greater than 140mmHg or diastolic blood pressure (DBP) equal or greater than 85mmHg without a known primary cause
• SBP and DBP reported at the end of follow-up
• Crossover trials restricted to designs with two intervention and two treatment periods.

 

Trials that included:

• Pregnant women
• Participants receiving antihypertensive medication that changed during the study
• Participants with potassium supplementation combined with other interventions.

Recruitment:  

• Databases searched included: Cochrane Library, MEDLINE, EMBASE, Science Citation Indes, ISI Proceedings, ClinicalTrials.gov, Current Controlled Trials, CAB abstracts
• The reference lists of systematic reviews and meta-analyses from these databases as well as RCTs were identified
• The search was based in a strategy where terms in capitals are MeSH headings, so the strategy could be developed within MEDLINE and adjusted accordingly for the other databases
• In addition, a general web search using the search engines Google, Zapmeta and Dogpile, as well as websites of the Blood Pressure Association, British Hypertension Society, American Society of Hypertension and Canadian Hypertension Society was carried out
• There was no language restriction
• Methodological quality of included trials was assessed considering these criteria: Blinding (yes, no, unclear); randomization (adequate, inadequate or unclear); allocation concealment from treatment providers and participants (adequate, inadequate or unclear); loss to follow-up (recorded number of participants in each intervention arm whose blood pressure was not reported at the end of the study, or if loss of follow-up was not reported); carryover effects (for crossover trials, assessment of carryover effects at the end of follow-up and whether was reported).

Design

Systematic review/meta-analysis.

Intervention

Trials of potassium supplementation.

Statistical Analysis:

• Separate meta-analysis of parallel and crossover trials were conducted, but they were combined if they did not show heterogeneity
• When standard deviations (SD) of final values were not available, they were imputed. In crossover treatment effect, the SD was imputed, assuming the mean within-person correlation observed in other crossover trials that evaluated the effect on blood pressure of oral supplements of calcium, sodium and magnesium
• Random effects model was performed using the meta-analysis combined with weighted mean differences according to the precision of each trial
• Heterogeneity between trials was assessed using the I²-statistic
• Sub-groups analyses were performed, grouping the trials into those participants in the active arm received higher and lower doses of potassium and participants' mean baseline blood pressure was higher and lower
• Sensitivity analysis were conducted excluding trials that did not report adequate concealment of allocation, blinding of participants, treatment providers and outcome assessors
• A post-hoc sensitivity analysis was performed, excluding one trial in an African population
• Additionally, post-hoc analyses were conducted to evaluate the effect of potassium supplementation on serum potassium levels
• Tolerability of intervention was assessed in the parallel trials by calculating the difference in the rate of withdrawal in the treatment and control arms, and using a random effects model to calculate a pooled risk difference
• Gastrointestinal effects were assessed by calculating the difference in the rate of these effects in treatment and control arms, and using a random effects model to calculate a pooled risk difference.  

Dependent Variables

• SBP at end of follow-up
• DBP at end of follow-up
• Fatal or non-fatal myocardial infarction (MI) 
• Fatal or non-fatal strokes
• Death from all causes.

Independent Variables

Potassium supplementation: Participants who received more than 100mmol per day (active treatment arm) or equal or less than 100mmol per day (control arm). 

• Initial N: Initial screening was 1,282; 48 RCTs were retrieved
• Attrition (final N): Five RCTs; 425 participants. Reasons for exclusion:
  • 42 RCTs excluded at paper screening stage
  • One RCT excluded from meta-analysis had a follow-up less than eight weeks (18 trials)
  • Normotensive participants (11 trials)
  • Antihypertensive medication varied during the trial (four trials)
  • Intervention consisted of multiple supplements (two trials)
  • Survey study design (two trials)
  • Follow-up less than eight weeks and normotensive participants (one trial)
  • Participants not randomly allocated to the sequence of treatment (one trial)
  • Observational study (one trial)
  • Control group on drugs (one trial)
  • Trial of multiple interventions (one trial)
  • Combined increased potassium with decreased sodium intake (one trial)
• Mean age: 50 years (range: 36 to 52 years)
• Ethnicity: Reported from only two trials, which was 61% white
• Other relevant demographics: Overall, 75% of the participants were male. Little information was reported about the normal diet of participants
• Location: US, Australia, Kenya, Germany and Italy.

Key Findings

• Six RCTs (N=483), with eight to 16 weeks follow-up, met the inclusion criteria
• Meta-analysis of five trials (N=425) with adequate data indicated that potassium supplementation compared to control resulted in a large but statistically non-significant (NS) reduction in SBP (mean difference: -11.2, 95% CI: -25.2 to 2.7) and DBP (mean difference: -5.0, 95% CI: -12.5 to 2.4). The substantial heterogeneity between trials (I²=98% and 99% for SBP and DBP respectively) was not explained by potassium dose, quality of trials or baseline blood pressure
• Excluding one trail in an African population with very high baseline blood pressure resulted in smaller overall reductions in blood pressure (SBP mean difference: -3.9, 95%CI: -8.6 to 0.8; DBP mean difference: -1.5, 95% CI:-6.2 to 3.1)
• Two trials administering lower (less than or equal to doses of potassium showed greater reductions in BP than three trials administering 100mmol per day or more, which was significant for DBP (mean differences in the two trials = -17.00 (95% CI: -19.25, -14.75) and -10.50 (95% CI: -16.32, -4.68mmHg)
• Sensitivity analysis of two high-quality trials found overall reduction in blood pressure among participants taking potassium supplementation remained NS for both SBP (mean difference: -7.1, 95% CI: -19.9 to 5.7) and DBP (mean difference: -5.5, 95%CI: -14.5 to 3.5). There was considerable heterogeneity between the trials (I²= 87% for both SBP and DPB).

 Other Findings

• No trials reported deaths or cardiovascular events. Only one trial reported adverse effects in both control and treatment groups,with more stomach pains and flatulence in the potassium-treated group
• Overall mean blood pressure at baseline was 151/95mmHg (SBP range: 145 to 174mmHg; DBP range: 92 to 100mmHg)
• The dose of potassium supplementation varied from 48 and 120mmol per day
• Meta-analysis showed NS difference in the rate of withdrawal between the treatment groups (risk difference= -0.03, 95% CI: -0.07 to 0.02
• Overall participants receiving potassium supplementation had higher serum potassium levels at the end of the trials than the controls (mean difference: 0.20mmol per L, 95% CI: 0.02 to 0.38)
• The funnel plot analysis of estimated treatment effects for SBP and DPB showed little evidence of publication bias.

• Potassium supplementation has no statistically significant effect on blood pressure
• Due to small number of participants in the two high-quality trials, the short duration of follow-up and the unexplained heterogeneity between trials, the evidence about the effect of potassium supplementation on blood pressure is not conclusive
• Further high-quality RCTs of longer duration are required to clarify whether potassium supplementation can reduce blood pressure and improve health outcomes.

Not-for-profit

National Institute for Clinical Excellence, UK. Other non-profit:

• The inclusion criteria for interventions to be at least eight weeks was extremely restrictive, limiting the pool of included studies
• The majority of the included studies were not of good quality, with the exception of two trials
• Substantial heterogeneity was found among the studies that can compromise the outcomes
• There was an over-representation of males in most of the trials included in this review.