Secondary Outcomes Meta-analysis: Effect of placebo and diuretic on serum potassium, plasma renin activity and blood pressure
Changes in serum potassium were: thiazide/thiazide-like -0.275 (-0.62, -0.78) mEq/L, loop -0.617 (-0.983, -0.251) mEq/L, K+ sparing/MRA 0.248 (0.151, 0.345) mEq/L and combination 0.048 (-0.327, 0.422) mEq/L, with significant between-class heterogeneity (Q = 49.4, P < 0.001, Figure 3). The analysis of PRA and SBP can be found in the supplementary material. Of the 45 included studies, 26 had differences in SBP, 26 had differences in serum potassium and 23 had differences in PRA both before and after diuretic. All diuretics decreased blood pressure significantly, apart from the loop diuretics, with combination diuretics lowering SBP by the largest amount (-23.4 [-35.33, -11.48] mmHg). In studies where there was no background antihypertensive use/a washout, the average decrease in SBP was -13.05 (-17.7, -8.40) mmHg and similar those in whom diuretic was added to previous antihypertensives: -14.76 (-18.78, -10.75) mmHg.
The overall average increase standardised mean PRA difference was 0.560 (0.275, 0.846) with a similar trend in both previously untreated subjects and those receiving another antihypertensive medication.