Estimation of potassium intake: single versus repeated measurements and the associated cardiorenal risk

Background: High potassium intake has been associated with lower blood pressure and a lower incidence of chronic kidney disease and cardiovascular events. In cohort studies, potassium intake is often estimated with a single 24-h urine collection. However, this may not represent actual long-term individual intake. We assessed whether a single baseline versus multiple follow-up measurements of 24-h urine potassium excretion results in different estimates of individual potassium intake and different associations between potassium intake and long-term outcome. Methods: We performed a retrospective cohort study in outpatient subjects with an estimated glomerular filtration rate >60 mL/min/1.73 m2 who had sampled a 24-h urine collection at baseline and had ≥1 collection during a 17-year follow-up. Potassium intake was estimated with a single baseline 24-h urine collection but also during 1-year, 5-year, and 15-year follow-up. We used cox regression analysis to assess the association between cardiorenal outcome and estimated potassium intake. Results: Average population (n = 541) 24-h potassium excretion was similar at baseline and follow-up but significant individual changes in potassium intake between baseline and follow-up were observed. Forty-four percent of the subjects switched between tertiles of estimated potassium intake when follow-up measurements were used instead of baseline measurements. Hazard ratios for renal and cardiovascular outcomes were 6.9 and 1.7 times higher when follow-up estimates of potassium intake were replaced by baseline estimates. Conclusions: Estimated potassium intake and its association with long-term outcome change significantly when multiple follow-ups 24-h urine collections are used for estimation of potassium intake instead of a single baseline measurement.