699 HYPERKALIEMIA, USE OF DRUGS TARGETING THE RENIN-ANGIOTENSIN-ALDOSTERONE SYSTEM AND MORTALITY IN PATIENTS FOLLOWED AT A TERTIARY CENTER

Abstract Background Hyperkalemia (HK) has been related to increased mortality in patients with heart failure (HF), at least in part because it leads to down-titration or interruption of renin angiotensin system inhibitors (RASi) and mineralocorticoid receptor antagonists (MRA). Methods We reviewed the records of the HF outpatients with ≥2 visits at our clinic between 2014 and 2021 and complete data on therapy, and identified those who had HK at any time after the first visit. The characteristics between these subjects and those without HK were compared by chi-square test, T-test, or Mann-Whitney test. The use of RASi, MRA and beta-blockers at multiple time points was examined by chi-square or Fisher exact test, and the mean dose for each drug class, expressed as percentage of the guideline-recommended dose, by Wilcoxon rank sum test. The trend of RASi dose over time in HK vs no-HK was assessed by two-way ordinal regression analysis, and the association of HK with all-cause death by Kaplan-Meier method and Cox regression analysis. Results 110 of 690 patients (15.9%) had HK, with the mean potassium concentration being 5.6 mEq/L. Subjects with HK were older [73 (66-79) vs 68 (58-78) years, p<0.001], had more often diabetes (42% vs 27%, p=0.002) and LVEF <40% (77% vs 67%, p=0.03), and had higher creatinine [1.3 (1-1.8) vs 1.1 (0.9-1.3) mg/dL, p<0.001], potassium [4.6 (4.2-5.2) vs 4.2 (3.9-4.5) mEq/L, p<0.001], and NT-proBNP levels [6,480 (2,110-11,000) vs 2,900 (1,115-6,070)], p=0.003] than those without HK. The distribution of RASi, MRA, and beta-blocker was comparable between the two groups at baseline and during follow-up, with the exception of a transient reduction in an early phase (panel A). The mean dose of RASi did not vary significantly between groups and over time (group/timepoint interaction p=0.27; panel B). HK was not associated with all-cause death (panel C; hazard ratio 1.05, 95% confidence interval 0.72–1.52). A sensitivity analysis limited to patients with LVEF <40% yielded similar results (not shown). Conclusions The negative impact of HK on RASi and MRA prescription and on prognosis may be mitigated in tertiary centers with expertise in HF treatment.