Table 1.
Summary of associations between potassium and risk of incident diabetes and the highest levels of evidence supporting these associations.
| Study (year) | Potassium-related exposure | Study design | Comparison groups | Magnitude of effect on risk of incident diabetes† | Ref. |
|---|---|---|---|---|---|
| Elliott et al. (2007) | Thiazide diuretic use (associated with decreases in potassium levels) | Network meta-analysis of data from hypertension trials | Diuretic use compared with placebo | Increased risk OR: 1.30 (95% CI: 1.07–1.58)‡ | [25] |
| Tocci et al. (2011) | ACE-I use (associated with increases in potassium levels) | Meta-analysis of data from placebo-controlled hypertension trials | ACE-I compared with placebo | Decreased risk OR: 0.8 (95% CI: 0.7–1.0) | [34] |
| Bosch et al. (2006) | Randomized controlled trial with incident diabetes as primary outcome in participants with IFG or IGT and no cardiovascular disease at baseline | Ramipril (up to 15 mg/day) compared with placebo | Decreased risk HR: 0.91 (95% CI: 0.81–1.03) | [29] | |
| Tocci et al. (2011) | ARB use (associated with increases in potassium levels) | Meta-analysis of data from placebo-controlled hypertension | ARB compared with placebo | Decreased risk OR: 0.8 (95% CI: 0.8–0.9)‡ | [34] |
| McMurray et al. (2010) | Randomized controlled trial with incident diabetes as primary outcome in participants with IGT and with cardiovascular disease or risk factors at baseline | Valsartan (up to 160 mg/day) compared with placebo | Decreased risk HR: 0.86 (95% CI: 0.80–0.92)‡ | [30] | |
| Chatterjee et al. (2010) | Low serum potassium (K+) | Prospective cohort study; African–Americans and whites | Lowest quartile of serum K+ (<4.0 mEq/l) compared with highest quartile | Increased risk HR: 1.64 (95% CI: 1.29–2.08)‡ | [13] |
| Heianza et al. (2011) | Prospective cohort study; Japanese men | Lowest tertile of serum K+ (2.8–3.9 mEq/l) compared with highest tertile | Increased risk HR: 1.57 (95% CI: 1.15-2.15)‡ | [14] | |
| Hu et al. (2005) | Low dietary potassium (K+) | Prospective cohort study; Finnish population | Highest quartile of K+ excretion compared with second lowest quartile | Decreased risk HR: 0.84 (95% CI: 0.51–1.38) | [45] |
| Chatterjee et al. (2010) | Prospective cohort study; American African– Americans and whites | Lowest quartile of K+ intake compared with highest quartile | Increased risk HR: 1.03 (95% CI: 0.81–1.31) | [13] | |
| Colditz et al. (1992) | Prospective cohort study; American women | Highest quintile of K+ intake compared with lowest quintile | Decreased risk RR: 0.76 (95% CI: 0.44–1.30) | [46] | |
| Carter et al. (2010) | Increased intake of green, leafy vegetables (potassium-rich foods) | Meta-analysis of prospective cohort studies with incident diabetes as primary outcome | Highest intake of green leafy vegetables (1.35 servings/day) compared with lowest intake | Decreased risk HR: 0.86 (95% CI: 0.77–0.97)‡ | [49] |
†
Magnitude of effects based on multivariate models except for those from randomized controlled trials.
‡
Statistically significant association.
ACE-I: Angiotensin-converting enzyme inhibitor; ARB: Angiotension II receptor blocker; HR: Hazard ratio; IFG: Impaired fasting glycemia; IGT: Impaired glucose tolerance; OR: Odds ratio; RR: Relative risk.