Abstract P170: The Association of Serum Potassium with Mortality in Older Community-Dwelling Individuals: The Cardiovascular Health Study (CHS)

Background: High serum potassium (K) is associated with death in chronic kidney disease (CKD) patients, and in acute illness. Associations in other settings are uncertain. We determined associations between K concentrations and total mortality, coronary heart disease (CHD) death, and sudden cardiac death (SCD) in an older, community-dwelling population. Methods: Among 5137 CHS participants aged ≥ 65 years at baseline, we evaluated associations between serum K categories [< 4.0, 4.0-4.5, 4.5-5.0, & ≥ 5.0 mMol/dL] with CHD death, SCD, and all-cause mortality using Cox proportional hazards models. We also evaluated whether associations differed by angiotensin converting enzyme (ACE) inhibitor / angiotensin II receptor blocker (ARB) use, diuretic use, and CKD status [eGFR < 60 vs. higher]. All CHD and SCD events were adjudicated by committee. Results: Mean age was 72 years, 39% were male, and 17% were Black. Individuals in the ≥ 5.0 mMol/dL category were older, more frequently men, diabetic, to have CKD, and to use ACE/ARBs. They were also less likely to use diuretics. Mean follow-up was 14 ± 6 years during which there were 4122 total deaths including 971 CHD deaths. Follow-up for SCD was 12 ± 5 years during which there were 162 SCD events. In models adjusted for demographics, CVD risk factors, eGFR, and use of ACE/ARBs, diuretics and K supplements, those with K ≥ 5.0 mMol/dL had 32% higher risk of all-cause mortality (HR 1.32; 95% CI: 1.07-1.63) than the 4.0-4.5mMol/dL reference category. The association was similar irrespective of diuretic or ACE/ARB use or by CKD status (p interaction all > 0.18). Those with K < 4.0 mMol/dL had 14% higher risk of all-cause mortality (HR 1.14; 95% CI: 1.05-1.23) than the reference category; and this association was limited to those with K < 4.0 mMol/dL and were not on diuretics and did not have CKD (p interaction both < 0.02). No association of K < 4.0 mMol/dL with mortality was observed in those on diuretics or with CKD. There was no significant association of either high or low K with either CHD death (HR 1.08; 95% CI: 0.71-1.65; and HR 0.96; 95% CI: 0.81-1.13, respectively) or SCD (HR 1.13; 95% CI: 0.41-3.11; and HR 1.01; 95% CI: 0.68-1.50, respectively) in adjusted models. Conclusions: Higher and lower serum K are independently associated with all-cause mortality but not CHD death or SCD in older community-dwelling individuals. Mechanisms linking high and low K with mortality from diseases other than CHD and SCD require future study.