Population Epidemiology of Hyperkalemia: Cardiac and Kidney Long Term Health Outcomes

Rationale & Objective
The population burden and long-term implications of hyperkalemia have not been comprehensively studied. We studied how often and where hyperkalemia occurs as well as its independent association with survival and long-term cardiac and kidney health.
Study Design
Population-based cohort study of adults residents of Grampian, UK (adult population 468,594).
Setting and Participants 
Among the 468,594 adult residents, 2012-2014, 302,630 people with at least one blood test were followed until2019.Exposures
Hyperkalemia was defined as serum potassium ≥5.5 mmol/L. Adjustment for comorbidities,demographics, KDIGO measures of acute and chronic kidney function, and medications prescribed prior to measurement of serum potassium.
Outcomes
All-cause mortality, cardiac events and kidney failure.
Analytical Approach
Description of the annual incidence of hyperkalemia and the characteristics associated with itsoccurrence. Adjusted Cox proportional hazards (PH) analysis to evaluate the independent longterm association of hyperkalemia with all-cause mortality among people who survived >90-days after blood testing. Cause-specific PH models were fit to evaluate the association of HK with cardiac events/death, non-cardiac death, and kidney failure. Effect modification by level of eGFR at the time of blood testing was explored.
Results
The annual population incidence of HK was 0.96 per 100 person-years. This represented 2.3%, 2.1%, and 1.9% of people with at least one blood test in 2012, 2013 and 2014, respectively. Two thirds of episodes of hyperkalemia occurred in the community. The HK rate was two-fold higher for each 10-year greater age. Those with HK were 20 times more likely to have concurrent AKI,and 17 times more likely to have eGFR <30 ml/min/1.73m2. Throughout five years of follow-up(2,483,452 person-years), hyperkalemia was associated with poorer health outcomes. This association held across all levels of kidney function and was irrespective of concurrent AKI, but was stronger among those with baseline eGFR ≥60 ml/min/1.73m2 (p for interaction <0.001). The adjusted hazard ratios (HR) (hyperkalemia vs no hyperkalemia) and 95% confidence intervals for people with eGFR ≥60 ml/min/1.73m2 and eGFR<30 ml/min/1.73m2 were 2.3 (2.2-2.5) and 1.5 (1.3-1.6) for mortality; 1.8 (1.6-1.9) and 1.4 (1.2-1.6) for cardiac events; and 17.0 (9.3-31.1) and 2.0 (1.5-2.8) for kidney failure, respectively.
Limitations
The observational nature of this study limits evaluation of causal relationships.
Conclusions
There is a substantial burden of hyperkalemia in the general population. Hyperkalemia isassociated with poorer long-term health outcomes, especially kidney outcomes, that are independent of other established risk factors.