Objective: Evaluate the prevalence of hyperkalemia(potassium < 5.5 mmol/l) in hospitalized patientsnot on dialysis, as well as the association ofmedications, impaired renal function and comorbidconditions with hyperkalemia.Design: A retrospective case-control method.Setting: A tertiary care teaching hospital.Patients: Hyperkalemic adults not on dialysis withage and sex matched controls.Interventions: None.Main outcome measures: The use of medicationsassociated with hyperkalemia and renal function usinga calculated creatinine clearance were compared in thehyperkalemic and control groups.Results: 35 adult patients with hyperkalemia who werenot receiving dialysis were identified, with aprevalence in the hospitalized population of 3.3%.The hyperkalemic patients were older than the generalhospital population (p < 0.05). Compared withcontrols, hyperkalemic patients: had a lowercreatinine clearance (p < 0.05), were more likely tobe taking angiotensin-converting enzyme inhibitors (p < 0.05), and had an increased frequency of diabetesmellitus (p < 0.001). All of the control patientssurvived their hospitalization, but the mortality rate in the hyperkalemic group was 17% (p < 0.0001). Noneof the deaths were directly attributable to hyperkalemia.Conclusions: Hyperkalemia is more frequent in olderpatients and is usually mild. Hyperkalemia isassociated with diabetes mellitus, diminished renalfunction and the use of angiotensin-converting enzymeinhibitors. An elevated serum potassium level in ahospitalized patient may be a marker for asignificantly increased risk of death, which is due tounderlying medical problems and is not a consequenceof the hyperkalemia.