LONDON – The best-quality available evidence demonstrates that digoxin reduces hospital admissions and has no effect upon all-cause mortality in patients with heart failure with or without atrial fibrillation, Dr. Dipak Kotecha reported at the annual congress of the European Society of Cardiology.
His meta-analysis of all observational and randomized controlled studies published since 1960 included 41 studies with 260,335 digoxin-treated patients, roughly 1 million controls, and more than 4 million person-years of follow-up.
Bruce Jancin/Frontline Medical News
Dr. Dipak Kotecha
What was unique about this meta-analysis is Dr. Kotecha and coworkers assessed each study in terms of its degree of bias due to confounding by indication and other limitations using two standardized bias scoring systems: the Cochrane Collaboration’s risk of bias tool for randomized controlled trials and the Risk of Bias Assessment tool for Non-randomized Studies (RoBANS).
The level of bias varied greatly among the observational studies. Moreover, the higher an observational study’s bias score, the greater the reported association between digoxin and mortality. In contrast, the seven randomized controlled trials all scored very low on bias and carried a consistent message that digoxin had a neutral effect on mortality, according to Dr. Kotecha of the University of Birmingham (England).
He noted that digoxin was first introduced into clinical cardiology back in 1785 in the form of digitalis. And although digoxin has seen widespread use for heart rate control and symptom reduction in patients with heart failure and/or atrial fibrillation over the years, prescriptions for the venerable drug have markedly declined recently in response to observational studies reporting a link with increased mortality. That purported association, he asserted, is highly dubious.
“Our comprehensive systematic review would suggest that confounding is the main reason why observational studies continue to show increases in mortality associated with digoxin. And this suggests that digoxin should continue to be considered as a treatment option to avoid hospital admissions in patients with heart failure and to achieve heart rate control in those with atrial fibrillation until better randomized data become available,” the cardiologist explained.
“I think that there’s one further important point to make here, and that’s that observational data, particularly when there are systematic differences in the two groups, should not be used to determine clinical efficacy. Propensity matching does not replace a randomized controlled trial in the assessment of efficacy,” he emphasized. “We’ve shown that observational data should be taken with extreme caution because digoxin is a second-line therapy and, like most of us, I tend to give digoxin to patients who are sicker, so we’re bound to see an increase in mortality and hospitalizations in those patients.”
Across all 41 studies, digoxin was associated with a small but clinically meaningful 8% reduction in the rate of all-cause hospitalization.
Asked what data are available regarding optimal dosing of digoxin, the cardiologist replied that the data are limited. “What data there are would suggest lower digoxin doses are better and higher doses tend to cause mortality. So our suggestion would be to continue using low digoxin doses, in combination if necessary with other drugs,” he said.
All of the randomized controlled trials have focused on patients with heart failure, alone or with comorbid atrial fibrillation. None have evaluated the impact of digoxin in patients with atrial fibrillation without heart failure. Given that atrial fibrillation is an emerging modern epidemic and a large chunk of digoxin-prescribing today is for rate control in such patients, this lack of high-quality data constitutes a major unmet need, Dr. Kotecha observed.
The RATE-AF trial, designed to advance knowledge in this area, is due to begin next year. The study, based at the University of Birmingham, will enroll patients with atrial fibrillation without heart failure who need rate control, randomizing them to a beta-blocker or digoxin.
Dr. Kotecha reported having no financial conflicts regarding his meta-analysis, which was funded by a grant from the university’s Arthur Thompson Trust.
Simultaneously with his presentation at the ESC congress, the meta-analysis was published online (BMJ 2105;351:h4451 doi:10.1136/bmj.h4451).