The Association Between Digoxin Use and Long-Term Mortality After Acute Coronary Syndrome

Abstract

Digoxin is used to treat atrial fibrillation and heart failure. Previous studies have reported an association between digoxin and higher mortality, but the results have been conflicting. This study assessed the association between digoxin use and all-cause mortality using comprehensive health data of patients treated for acute coronary syndrome (ACS). This was a retrospective analysis of 8,388 consecutive ACS patients treated in Tays Heart Hospital between 2007 and 2017, with a follow-up until the end of 2018. The adjusted Cox regression model was used to analyze the association between digoxin treatment and all-cause mortality with and without the inverse probability of treatment weighting (IPTW) method. IPTW was applied to estimate the residual confounding by the treatment selection. Clinical phenotype data were collected from various sources, including a prospectively updated online database maintained by physicians. The median follow-up time was 6.0years (interquartile range 3.5 to 9.0years). During the follow-up, 30.8% (n=2,580) of the patients died. Altogether, 4.0% (n=333) of the patients were treated with digoxin during hospitalization. In the Cox regression model, digoxin associated with increased mortality (age- and sex-adjusted hazard ratio [HR] 1.76 [1.51 to 2.05], p <0.001 and in the full risk factor-adjusted HR 1.23 [1.04 to 1.45], p=0.016). The IPTW Cox analysis average treatment effect HR was 1.71 (1.12 to 2.62, p=0.013), standardized average treatment effect HR was 1.35 (0.96 to 1.90, p=0.082), and treatment effect among the treated HR was 1.32 (1.09 to 1.59, p=0.004). In conclusion, digoxin treatment during ACS associates with increased mortality, despite adjusting for other risk factors and after accounting for factors explaining the residual confounding by selection bias.