PURPOSE: Leg cramps are common in patients with heart failure. Quinine is frequently prescribed in low doses to these patients, but safety of this practice is unknown. We studied the outcomes associated with use of quinine in a nationwide cohort of patients with heart failure. METHODS: Through individual-level-linkage of Danish national registries, we identified patients discharged from first-time hospitalization for heart failure in 1997-2010. We estimated the risk of mortality associated with quinine treatment by time-dependent Poisson regression models. RESULTS: A total of 135 529 patients were included, with 14 510 patients (11%) using quinine at some point. During a median time of follow-up of 989 days (interquartile range 350-2004) 88 878 patients (66%) died. Patients receiving quinine had slightly increased mortality risk, adjusted incidence rate ratio (IRR) 1.04 (95% confidence interval [CI] 1.01 to 1.07). The risks differed according to concomitant β-blocker treatment. For patients treated with both quinine and β-blockers IRR was 1.15 (95% CI 1.09 to 1.21) vs. 0.99 (95% CI 0.96 to 1.03) for patients treated with quinine but not β-blockers. The risks were highest shortly after initiation of therapy: for the first 14 days of treatment IRR was 2.12 (95% CI 1.54 to 2.93) for patients in treatment with β-blockers and 1.17 (95% CI 0.86 to 1.59) for patients not treated with β-blockers. CONCLUSIONS: Use of quinine was common and associated with increased mortality in heart failure, especially if administered together with β-blockers and shortly after treatment initiation. Mechanisms underlying the findings remain to be established.
PURPOSE:Leg cramps are common in patients with heart failure. Quinine is frequently prescribed in low doses to these patients, but safety of this practice is unknown. We studied the outcomes associated with use of quinine in a nationwide cohort of patients with heart failure. METHODS: Through individual-level-linkage of Danish national registries, we identified patients discharged from first-time hospitalization for heart failure in 1997-2010. We estimated the risk of mortality associated with quinine treatment by time-dependent Poisson regression models. RESULTS: A total of 135 529 patients were included, with 14 510 patients (11%) using quinine at some point. During a median time of follow-up of 989 days (interquartile range 350-2004) 88 878 patients (66%) died. Patients receiving quinine had slightly increased mortality risk, adjusted incidence rate ratio (IRR) 1.04 (95% confidence interval [CI] 1.01 to 1.07). The risks differed according to concomitant β-blocker treatment. For patients treated with both quinine and β-blockers IRR was 1.15 (95% CI 1.09 to 1.21) vs. 0.99 (95% CI 0.96 to 1.03) for patients treated with quinine but not β-blockers. The risks were highest shortly after initiation of therapy: for the first 14 days of treatment IRR was 2.12 (95% CI 1.54 to 2.93) for patients in treatment with β-blockers and 1.17 (95% CI 0.86 to 1.59) for patients not treated with β-blockers. CONCLUSIONS: Use of quinine was common and associated with increased mortality in heart failure, especially if administered together with β-blockers and shortly after treatment initiation. Mechanisms underlying the findings remain to be established.