High and low serum potassium associated with cardiovascular events in diuretic-treated patients

Objective To determine the relationship of moderately high and low concentr ations of serum potassium with cardiovascular disease events among treated hypertensive patients. Design An observational cohort study with prospectively collected data. Setting A worksite treatment program for mild hypertension. Patients All program participants with baseline and at least one annual fol low-up measure of serum potassium; 7653 individuals with 6.7 years mean fol low-up met these criteria. Main outcome measures Outcome events were admissions to hospital because of cardiovascular disease, and deaths. The research question regarding serum potassium categories was formulated after data collection. The serum potass ium concentration (mean +/- 2SD) of the study population was used to define low (3.0-3.5 mmol/l), high (5.1 - 5.9 mmol/l) and middle (3.6 - 5.0 mmol/l ) categories. Results Individuals with low (n = 146) and high (n = 226) serum potassium h ad significantly greater risk for cardiovascular disease events than those in the middle category (n = 7281). Multivariate adjusted hazard ratios from Cox models were 2.6 [95% confidence intervals (CI) 1.5-4.4] for the low po tassium group and 1.7 (95% CI 1.02.7) for the high potassium group, with th e middle group as reference. Among 1679 individuals who regularly took diur etics, hazard ratios were 4.3 (95% CI 2.4-7.9) for the low potassium group and 6.7 (95% CI 2.8-15.9) for the high group. Neither low nor high potassiu m was significantly associated with outcome events for those not regularly using diuretics. Conclusions These data confirm an association of mild hypokalemia with incr eased cardiovascular events among diuretic-treated hypertensive patients. I n addition, we have found a similar increased cardiovascular risk associate d with modest hyperkalemia among these patients. Whether modification of th ese serum potassium concentrations would alter that risk remains to be dete rmined. I Hypertens 19:1315-1323 (C) 2001 Lippincott Williams & Wilkins.