Abstract Chronic kidney disease (CKD) is the most frequent comorbidity associated with resistant hypertension (RH). Despite mineralocorticoid-receptor antagonists (MRAs) have been proven to be effective in RH, this drugs are underused in this patients and must be used very carefully in CKD due to the higher risk of hyperkalemia (HK) and significant renal dysfunction (RD), particularly in combination with renin angiotensin inhibitors. The aim was to compare the efficacy and safety of using of MRAs: spironolactone (SP) and eplerenone (EP) in RH and CKD stages 3 and determine predictors of developing HK and worsening of renal function (RF). Methods 54patients with true RH (antihypertensive treatment mean of 3.4±1.4 drugs per patient including diuretic, ACE-I or an ARB) and CKD 3 (baseline eGFR between 30 and 60 ml/min) were included in the study. The potassium (K) and creatinine (C) levels, plasma aldosterone (AS) and active renin concentration (ARC) were estimated at baseline. After chemical evaluation, 28 patients started on SP treatment with a mid-dose 18 mg daily (titrated), 26 patients started on EP with a mid-dose 24 mg (titrated). K and RF were checked at weeks 1, 4, 8, 12. At baseline and after 12 weeks of therapy, patients underwent clinic and 24-hour BP measurement. Results SP and EP both effectively decreased BP (9.5/-3.7 mm Hg; P<0.05 for both; 9.9±3.9 mm Hg; P<0.05 for both; respectively), changes in BP were dose dependent and not significantly different. K and C levels increased significantly after the start of MRAs treatment for both SP and EP patients: mean of K on SP increased from 4.26±0.61 to 5.25±0.69 mEq/l and on EP from 4.29±0.71 to 5.13±0.71 mEq/l (P<0.001; P<0.001; respectively) and was dose dependent. After 12 weeks of treatment the incidence of sever HK (K ≥6.0 mmol/L) was <4% (two on SP 25 mg and two on EP 25 mg). A K 5.5–5.9 mmol/L occurred in 9 patients (four on SP) and was predicted by baseline K ≥5.0 mmol/L, eGFR ≤40 ml/min/1.73m2 and dose of diuretic. Mean of eGFR on SP decreased from 45.26±3.19 to 40.15±2.88 ml/min/1.73m2 and on EP from 46.29±3.56 to 39.13±2.99 ml/min/1.73m2 (P<0.001; P<0.001; respectively). Three patients (one on SP) experienced significant RD result in withdrew MRAs. Two cases of gynecomastia were reported with SP use resulting in a switch to EP. Female, ARC >20 ng/ml and eGFR≤40 ml/min/1.73m2 at baseline had modest discriminative powers for predicting decline RF (P<0.05; P<0.01; P<0.001; respectively). Conclusion SP and EP demonstrated comparable efficacy in RH and CKD 3, but EP was well tolerated. In both groups MRAs treatment resulted in a significant dose dependent rise in K and C levels. Female, higher levels of ARC and eGFR were identified as potential predictors of worsening of RF. The occurrence of HK was predicted by baseline K level and eGFR. So, caution should be advised when using MRAs in RH with CKD 3 and an ARC >20 ng/ml, P of ≥5.0 mmol/L and eGFR ≤40 ml/min/1.73m2 for safety reasons.