Trials are best, ignore the rest: safety and efficacy of digoxinBMJ 2015; 351 doi: https://doi.org/10.1136/bmj.h4662 (Published 30 August 2015) Cite this as: BMJ 2015;351:h4662
- Graham D Cole, research fellow,
- Darrel P Francis, professor of cardiology
- 1International Centre for Circulatory Health, National Heart and Lung Institute, Imperial College London, London W2 1LA, UK
- Correspondence: G D Cole
Nearly 20 years after the digoxin in heart failure (DIG) trial showed that digoxin reduced hospital admissions for heart failure by 28% with no effect on mortality,1 many clinicians still see this drug as a last resort—reflected in guidelines that reserve it only for those with severe or worsening heart failure when first and second line treatments have failed.2 One reason might have been the numerous observational studies indicating the potentially worrying trade-off that digoxin is associated with an increased risk of death.
The linked article by Ziff and colleagues (doi:10.1136/bmj.h4451), a meta-analysis of digoxin treatment for heart failure or atrial fibrillation (or both), sheds light on the discordant findings between randomised and observational studies.3 They show that the mechanism of the association with mortality in observational studies is that clinicians preferentially gave digoxin to sicker patients: when these patients died, there was therefore a true but misleading association between death and digoxin. The paper paints …