Circadian variation in response to potassium infusion

The response of 5 normal men to an intravenous infusion of potassium chloride was compared at midday and midnight. Each subject was maintained on strict supine bedrest with oral intake limited to 100 ml distilled waterlhr for the 9 hr before and after each infusion. Potassium chloride, 37 mEq, (with an added label of 200 µCi 42 KCl) in iso‐osmolar solution was administered via a central venous catheter over 1 hr starting either at midday or midnight. Plasma potassium concentration was elevated by 40% more at midnight than at midday, and plasma 42 K activity also rose to a higher level at midnight. These differences were reflected by greater T wave elevations of the electrocardiogram at midnight than at midday, although urinary potassium excretion (total and 42 K labeled) was higher at midday than at midnight, indicating that there was reduced renal excretory responsiveness to elevations in plasma potassium concentration at midnight compared with midday. Plasma aldosterone concentration rose during the potassium infusions at both midday and midnight by a similar amount, which suggests that the induced increments in aldosterone secretion were not a major determinant of the differing renal response. These findings indicate that circadian modulations in the physiologic mechanisms which regulate potassium homeostasis profoundly influence the response to exogenous potassium loads. Special caution must therefore be taken in administering potassium infusions at night.