Abnormal serum potassium levels and 6-month all-cause mortality in patients co-treated with antipsychotic and diuretic drugs - A Danish register-based cohort study

Psychiatric patients have excess mortality compared to the general population, and several potential mechanisms may explain this increased risk. This study examined the relationship between serum potassium levels and risk of all-cause mortality in patients co-treated with antipsychotic and diuretic drugs. Using a register-based cohort design, we identified patients between 1995 and 2012 who received a combination of an antipsychotic and a diuretic drug and who further had a serum potassium measurement within 90 days. During the study period, we included the most frequently redeemed antipsychotic drugs with regard to the propensity of corrected QT (QTc) prolongation: zuclopenthixol (unknown/mild), flupentixol (mild), levomepromazine (moderate), and quetiapine (moderate/severe). Patients co-treated with antidepressant drugs, lithium, and other antipsychotic drugs were excluded. Outcome was 6-month all-cause mortality, estimated with multivariable Cox regression. Patients were divided into seven serum potassium levels using restricted cubic splines (reference: 4.2-4.4 mmol/L) and stratified according to the included antipsychotic drugs. Of 6729 patients (median age: 74.0 years; women: 65.3%), 10.8% had hypokalemia and 4.9% had hyperkalemia. Hyperkalemia (>5.0 mmol/L, HR 2.82 [95% CI 2.25-3.54]), hypokalemia (<3.5 mmol/L, HR 1.59 [95% CI 1.29-1.95]), and high normal potassium levels (4.5-4.7 mmol/L, HR 1.44 [95% CI 1.19-1.75]; 4.8-5.0 mmol/L, HR 1.60 [95% CI 1.26-2.04]) were associated with an increased risk of 6-month all-cause mortality. This risk was independent of antipsychotic drugs (interaction: P = 0.06). Our findings imply that excess mortality in patients co-treated with antipsychotic and diuretic drugs is related to serum potassium levels and independent of antipsychotic drugs.