This past December, my colleagues and I published the findings from our latest clinical study – providing randomized controlled trial data that suggests digoxin is a viable alternative to beta blockers when it comes to treating older patients with permanent atrial fibrillation and symptoms of heart failure.
In the RATE-AF trial, patients randomized to digoxin had no difference in heart rate, physical quality of life, or ejection fraction compared to beta-blockers, but had significantly better symptom control, lower natriuretic peptides and substantially less adverse events.
For some, the news that digoxin – widely believed to be less effective than beta blockers in observational studies – held up so well has inspired talk of a “comeback” for the drug. Yet while the results certainly show promise, it’s important to also note its limitations: this was an open label trial involving 160 outpatients with established symptomatic atrial fibrillation using controlled and low-dose digoxin.
Rather than see this as a watershed moment for digoxin, we should instead view it as one in a series of research advancements in this critical area of medicine. By understanding how we got here, we can put these new findings in context and get a better sense of what’s next for this therapy.
How we got here
For much of the modern era, digoxin was considered a go-to treatment for heart failure. In 2003 the DIG trial identified no mortality benefit (but a small reduction in hospitalization) when digoxin was given to heart failure patients in sinus rhythm. By 2006, studies found the use of digoxin steeply declining.
Safety concerns began to gain steam due to a series of observational studies. In the 1980s, for instance, two studies found digoxin use associated with increased mortality in survivors of myocardial infarction. Another suggested that digoxin may contribute to higher mortality among women. The observational nature of these studies, however, makes them unable to account for prescription biases – notable with digoxin, as this therapy is usually reserved for sicker patients, additive therapy, or for those who cannot tolerate beta-blockers.
At the same time, a number of new therapies for heart failure were starting to be developed and promoted by the pharmaceutical industry. In the past decade or so, these drugs have effectively replaced digoxin as the primary treatment for heart failure, at least in sinus rhythm.
My colleagues and I wanted to know more, particularly for what is now the most common group of patients given digoxin in clinical practice: those with atrial fibrillation. Our research programme stemmed from the seminal observation in double-blind trials (published in Lancet in 2014) that beta-blockers did not reduce mortality in patients with heart failure and concomitant atrial fibrillation.
Given the negative perceptions of digoxin, we couldn’t simply start with a clinical trial against beta-blockers in this population. We began instead by comprehensively studying the safety and efficacy of digoxin with a systemic review and meta-analysis of observational and controlled trial data. By taking account of all published data on digoxin and over 4 million patient-years of follow-up – something prior analyses had failed to do – we were able to show the danger of assessing digoxin through observational datasets.
Published in the BMJ in 2015, we found digoxin to be associated with a neutral effect on mortality in randomized trials (all in sinus rhythm). What’s more, across all types of studies, one of the strongest predictors of death associated with digoxin use was actually the bias of the study!
RATE-AF
These two meta-analyses gave us the ethical imperative to seek funding for a healthcare-embedded, randomized controlled trial: the first head-to-head long-term comparison of digoxin and beta-blockers in patients with atrial fibrillation and heart failure.
Funded by the UK National Institute for Health Research (NIHR), RATE-AF was published in JAMA late last year after a late-breaking ‘hotline’ at the European Society of Cardiology conference. Contrary to popular belief, the study indicated that, among patients with permanent atrial fibrillation and symptoms of heart failure, there was no statistically significant difference in physical-related quality of life or heart rate at six months with digoxin versus beta-blockers. By 12 months, 8 of 20 outcomes showed a significant difference and all favoured digoxin. There was significantly better symptom control with digoxin for both AF and heart failure-related symptoms, a reduction in B-type natriuretic peptide, no compromise in ejection fraction and significantly less adverse events than with beta-blockers.
RATE-AF also represented a model for patient-centred design that offered added layers of insight. As recounted in our European Heart Journal article on the topic, the RATE-AF patient and public involvement team were involved every step of the way: helping design the trial with a focus on quality of life and symptoms; participating as part of the study team that ran the trial; developing patient-facing materials to enhance recruitment, adherence and retention; and running their own qualitative research on the extremely low quality of life seen in patients with atrial fibrillation.
Next steps for digoxin research
If this process tells us anything, it’s that medical research builds on itself over time. Next we plan on updating our meta-analysis of digoxin studies to include RATE-AF and other studies, and we look forward to results from two other trials assessing digoxin. Unfortunately, we continue to see journal editors make questionable decisions in accepting flawed observational research on digoxin that, regardless of sample size, are unable to account for the systematic bias inherent in digoxin prescription. Only new and larger randomized trials can quantify the true safety and efficacy of digoxin therapy.
Towards that end, new opportunities are needed that can innovate clinical trial processes and facilitate the generation of real-world healthcare evidence on a greater scale than achieved in RATE-AF. DaRe2THINK is a national trial programme in collaboration with the UK Clinical Practice Research Datalink that we hope (with public funding) will one day be tasked with the goal of clarifying the hospitalization and mortality effects of digoxin in patients with atrial fibrillation.
Already, DaRe2THINK’s ability to screen over 13 million patients with completely remote enrolment and no study visits is enabling the programme to test if younger patients with atrial fibrillation should be treated to prevent stroke, cognitive dysfunction and vascular dementia. Planned to enrol 3000 patients across England, DaRe2THINK is randomizing those with low or intermediate risk of stroke to direct oral anticoagulation early, versus usual care of starting anticoagulation when risk factors develop.
The forthcoming batch of clinical trials on digoxin and those planned in the future will clearly take some time to complete – meaning that when it comes to digoxin, much remains to be proven. But that’s as it should be: patience, scientific rigor, and methodological progress are all just part of the research journey. Over 230 years after William Withering first used digitalis here in this city of Birmingham, England, it’s reassuring to know that clinicians and patients still need to make that joint decision about what is best for each individual circumstance.
Dr. Kotecha is a paid consultant for BTG International, Ltd.
Dipak Kotecha MBChB PhD (Clinical Cardiology) MSc (Clinical Trials) MRCP FHEA FESC is a Professor of Cardiology at the University of Birmingham, a Consultant Cardiologist specialising in cardiac imaging, and has completed a prestigious National Institute for Health Research (NIHR) Fellowship. He works in education, trials and guidelines, and his research focuses on the interaction of heart failure and atrial fibrillation, supported by grants from the NIHR, British Heart Foundation and European Union. Prof. Kotecha collaborates widely, with interests in patient involvement, big data approaches, the application of artificial intelligence, and the use of novel technology such as machine learning and wearable devices. He was awarded the British Cardiovascular Society Michael Davies Medal in 2018 for his contribution to cardiovascular medicine.
