Quantification of the risk and predictors of hyperkalemia in patients with left ventricular dysfunction: a retrospective analysis of the Studies of Left Ventricular Dysfunction (SOLVD) trials.

BACKGROUND: Limited data are available to predict the occurrence of hyperkalemia. Risk assessment is complicated by the lack of consistency of definition between trials.

METHODS: We conducted a retrospective analysis of the SOLVD to evaluate the incidence of hyperkalemia and the value of several baseline characteristics as predictors of hyperkalemia in patients with left ventricular dysfunction.

RESULTS: The incidence of hyperkalemia was 6.0% and 1.1% using a definition of > or = 5.5 and > or = 6.0 mmol/L, respectively. Independent predictors of hyperkalemia (> or = 5.5 mmol/L) were randomization to enalapril, baseline serum creatinine, serum potassium, New York Heart Association functional class III or IV, a history of diabetes, and atrial fibrillation (all P < .05). The use of loop diuretics was also associated with an increased risk of hyperkalemia but only in patients included in the SOLVD prevention trial. Similar results were obtained when renal function was evaluated using the estimated creatinine clearance.

CONCLUSIONS: The definition of hyperkalemia is important when evaluating its incidence in clinical trials. Renal dysfunction, baseline serum potassium, diabetes, atrial fibrillation, New York Heart Association functional class, and treatment with an angiotensin-converting enzyme inhibitor are factors associated with the development of hyperkalemia in patients with left ventricular dysfunction. More specifically, our results suggests that before initiating drugs that can cause hyperkalemia in patients with heart failure, a strong consideration should be given to calculate creatinine clearance and that patients with a creatinine clearance < 60 mL/min should undergo a close monitoring of their serum potassium to prevent the development of hyperkalemia.