Background
This network meta-analysis investigated the effect of various combined regimens of sodium-glucose cotransporter-2 inhibitors (SGLT2is) and renin-angiotensin-aldosterone system inhibitors (RAASis) on the occurrence of hyperkalemia in diabetic kidney disease.
Methods
The risk of hyperkalemia was compared using the random-effects model of network meta-analysis, with results expressed as odds ratios (ORs) with 95% confidence intervals (CIs). The comparative effects of all medications and their combinations with placebo were ranked using the surface under the cumulative ranking probabilities.
Results
In total, 27 eligible studies involving 43,589 participants with diabetic kidney disease were included. Major findings showed that the use of mineralocorticoid receptor antagonists (MRAs) on top of angiotensin-converting enzyme inhibitors (ACEIs) or angiotensin receptor blockers (ARBs) prominently increased hyperkalemia incidence when compared with placebo (OR, 6.08; 95% CI, 2.30 to 16.08), ACEI (OR, 3.07; 95% CI, 1.14 to 8.31), ARB (OR, 2.57; 95% CI, 1.10 to 6.02), SGLT2i (OR, 9.22; 95% CI, 2.99 to 28.46), renin inhibitors+ACEI/ARB (OR, 2.23; 95% CI, 1.14 to 4.36), or SGLT2i+ACEI/ARB (OR, 4.10; 95% CI, 2.32 to 7.26). Subgroup analysis among different generations of MRA found that spironolactone had the strongest effect in combination with ACEI/ARB, even higher than the combined use of ACEI and ARB (OR, 2.89; 95% CI, 1.26 to 6.63). In addition, SGLT2i had a significantly lower incidence of hyperkalemia compared with ACEI (OR, 0.33; 95% CI, 0.12 to 0.91), ARB (OR, 0.28; 95% CI, 0.13 to 0.61), dual RAASi (ACEI combined with ARB; OR, 0.17; 95% CI, 0.06 to 0.47), or MRA or renin inhibitors combined with ACEI/ARB (OR, 0.11; 95% CI, 0.04 to 0.33; OR, 0.24; 95% CI, 0.08 to 0.76, respectively.). Moreover, adding SGLT2i to the combination of MRA and ACEI/ARB, as well as the combinations of different RAASis, markedly reduced the occurrence of hyperkalemia.
Conclusions
Among the therapeutic drugs with the potential risk of increasing serum potassium in patients with diabetic kidney disease, MRA added an extra risk of hyperkalemia while SGLT2i had the opposite effect and could even reverse the elevation of serum potassium caused by the combined regimen, including MRAs.