Hypokalaemia and subsequent hyperkalaemia in hospitalized patients.

BACKGROUND: The objective was to study the epidemiology of hypokalaemia [serum potassium concentration (S(K)) <3.5 mmol/l] in a general hospital population, specifically focusing on how often and why patients develop subsequent hyperkalaemia (S(K) > or =5.0 mmol/l).
METHODS: In a 3-month hospital-wide study we analysed factors contributing to hypokalaemia and subsequent hyperkalaemia.
RESULTS: From 1178 patients in whom S(K) was measured, 140 patients (12%) with hypokalaemia were identified (S(K) 3.0 +/- 0.3 mmol/l). One hundred patients (71%) had hospital-acquired hypokalaemia. Common causes of hypokalaemia included gastrointestinal losses (67%), diuretics (36%) and haematological malignancies (9%). In 104 patients (74%), hypokalaemia was multifactorial. Hypokalaemia frequently coexisted with hyponatraemia (24%) and, when measured, hypomagnesaemia (61%). Twenty-three patients (16%) developed hyperkalaemia (highest S(K) 5.7 +/- 0.7 mmol/l) following hypokalaemia. In these patients, potassium suppletion was not more common (70 vs 59%, P = 0.5), but when potassium was given, the total amount administered was significantly higher (median 350 mmol vs 180 mmol, P = 0.02). Furthermore, these patients more often received total parenteral nutrition (17 vs 4%, P = 0.02) and magnesium suppletion (30 vs 9%, P = 0.009), and more often had haematological malignancies (22 vs 6%, P = 0.03).
CONCLUSIONS: Hypokalaemia is a multifactorial and usually hospital-acquired condition associated with hyponatraemia and hypomagnesaemia. One out of every six patients with hypokalaemia developed subsequent hyperkalaemia. Besides potassium suppletion, total parenteral nutrition (source of potassium), magnesium suppletion (may reduce kaliuresis) and haematological malignancy (may cause cell lysis) contribute to hyperkalaemia following hypokalaemia. Caution with potassium suppletion and frequent monitoring of S(K) may prevent iatrogenic hyperkalaemia.