Study Objective. To identify patient‐specific factors associated with spironolactone‐induced potassium level elevation in patients with heart failure.Design. Prospective cohort study.Setting. Two adult heart failure clinics.Patients. Sixty‐two adult (mean ± SD age 54 ± 16 yrs) aldosterone antagonist‐naïve patients with heart failure.Intervention. Patients received spironolactone 12.5 mg/day, titrated to 25 mg/day if tolerated.Measurements and Main Results. Blood samples were obtained at baseline, 1 week after spironolactone initiation, and 1 week after spironolactone dose titration for assessment of baseline aldosterone level, serum chemistry, and angiotensinogen (AGT) c.‐6G>A and p.M268T and mineralocorticoid receptor (NR3C2) c.215C>G and p.I180V genotypes. Patient characteristics, laboratory values, and genotypes were compared between those whose potassium levels increased by more than 0.5 mEq/L (15 patients) and those with lower potassium level elevations (47 patients) after spironolactone initiation and dose titration. Patients with a greater potassium level elevation had a higher mean ± SD aldosterone concentration (178 ± 92 vs 102 ± 57 pg/ml, p=0.007) and NR3C2 215G allele frequency (50% vs 22%, p<0.01). Aldosterone concentrations positively correlated with diuretic dose (r=0.313, p=0.014) and negatively correlated with serum potassium level (r=‐0.319, p=0.012). On regression analysis, factors predictive of potassium level increases greater than 0.5 mEq/L with spironolactone were aldosterone level greater than 150 pg/ml (odds ratio [OR] 30, 95% confidence interval [CI] 3.2–287] and NR3C2 215G carrier status (OR 17, 95% CI 1.6–167).Conclusion. Our data suggest that potassium should be monitored with particular caution when spironolactone is started in patients with heart failure who have evidence of elevated aldosterone levels, such as high diuretic requirements, or the NR3C2 215G allele.