I often get asked this question. After all, there are a lot of therapeutic approaches – complementary or mainstream – that might be effective for atrial fibrillation. Why these two in particular?
It all began with my curiosity. Being an acupuncturist, I’ve treated a few people with atrial fibrillation (AF) over the years and it does seem to respond well (in my experience). And I know a few nutritional therapists who have also recorded good results with AF patients, so I was curious about nutritional therapy’s effect on AF too. By the way, if you’d like to know more about what acupuncture and nutritional therapy actually are, take a look at this page.
But while curiosity got me started, it wasn’t the reason why these two therapies got chosen for the trial. My own limited experience, plus the experience of my nutritional therapist colleagues, is very definitely “anecdotal evidence” – i.e., it’s not scientific in any way. In trials research, there has to be a good rationale for every decision involved in designing a trial. So I carried out a systematic review of the scientific literature on atrial fibrillation to find out which therapies (or “interventions” as they’re known in medical research) would be appropriate ones to investigate. At that point, I was prepared to find out that the appropriate therapies to include in a trial were things I’d never even thought about – like maybe pilates, or reflexology, or osteopathy?
Before I could start on the systematic literature review, I had to define appropriateness. What’s appropriate to investigate in a trial, paradoxically enough, depends on the state of the current evidence – and by evidence, I mean published literature showing that something actually works. It’s a Goldilocks thing: if there’s not enough evidence you can’t trial an intervention, and if there’s too much evidence, you can’t trial it either. You need an intervention that’s at just the right stage of its evidence journey.
What do I mean by too little and too much evidence? Good question. Too little evidence would be where there was no published literature at all on a given intervention – in that case it would be unethical to sign up participants for a trial of that intervention, because it wouldn’t be known whether it was basically safe or not. Also, research into a given intervention generally starts with studies that are lower in both risk and quality of evidence – typically, case studies of single patients, case series of several patients, or observational studies where patients are simply observed so the investigators can understand more about the effect of an intervention (in this case known as an “exposure”) that they were having anyway. It’s generally thought not to be good research practice to start out a field of investigation with a full-scale trial of an unknown intervention.
Equally, “too much” evidence is also bad. If a particular intervention (such as a drug, or surgery, or sometimes a more complex intervention) has a lot of high-quality evidence to show it works for a particular condition, it’s considered unethical to investigate it further because it wastes time and money demonstrating what’s already been proved beyond reasonable doubt.
So in my review of the published literature, I included many different kinds of complementary therapies for atrial fibrillation, to see what therapies might be at just the right stage of their evidence journey: not too little and not too much evidence. Perhaps unsurprisingly, I found there was very little research at all into most complementary therapies for atrial fibrillation. But what I did confirm was a small, low-quality evidence-base for acupuncture and nutrition. In acupuncture there were two case studies, and a smallish trial whose quality was difficult to verify due to reporting omissions (i.e., the authors didn’t give enough information to allow readers to judge the quality of the trial). In nutritional therapy there were many studies of different dietary interventions of the type that nutritional therapists might use to support AF clients.
I also found some fairly good evidence for yoga. I chose not to include that in the trial, though, because I wanted to compare two interventions that were broadly similar in how they are delivered. Acupuncture and nutritional therapy are both delivered by practitioners during one-to-one appointments, during which the practitioner and the patient/client form a therapeutic relationship, or “healing partnership” as it’s sometimes known. That gave us the opportunity to study the possible association between that healing partnership and the outcome (in other words, does a participant who rates their healing partnership highly experience a greater reduction in symptoms and a better quality of life than someone who doesn’t rate their healing partnership so highly?). So yoga, which is typically delivered in a class-based environment without opportunities for a healing partnership, wasn’t suitable for that.
In many ways, acupuncture and nutritional therapy are birds of a feather. Not only are they delivered in a similar way to each other, they also have a similar model: despite the fact that one is based in ancient Chinese medicine and the other in modern Western medicine, they both take a holistic approach to the patient. By that much-maligned word, I mean no New Age babble, but simply the fact that both therapies treat the patient as a whole, with a personal history, lifestyle, habits, living circumstances and emotions – all of which can and do affect the patient’s physical, mental and emotional health. And all of these factors are taken into account when formulating a treatment plan for the patient.
Of course, the question of how these two therapies work for AF is not the business of Santé-AF – that’s an entirely different study (the famous “mechanism of action” question, as it’s known in research). But I also get asked that question a lot! So I’ll hazard one (hugely unscientific) opinion about how acupuncture might work for AF. Basically, almost anything that activates the parasympathetic nervous system (i.e., deactivates the fight-or-flight stress response) is going to ratchet down the level of stress hormones in the average body. This particular effect of acupuncture is well-documented [1, 2, 3] and it’s broadly similar to the overall effect of beta-blockers and rhythm control drugs, which are a popular first-line treatment for AF; so if they’re having the same kind of effect as those medications, that’s presumably going to have some therapeutic effect. An equally unscientific opinion about now nutritional therapy works for AF would be that certain dietary strategies have a positive effect on the physiological mechanisms that give rise to AF. For instance, a low glycaemic index (GI) diet can have anti-inflammatory effects [4], and a heightened inflammatory response is known to be associated with AF [5] – so it would make sense that if inflammatory effects were dampened down by following a low-GI diet, AF may improve.
So that’s why these two particular therapies appear in Santé-AF, and despite using plenty of unscientific language I hope it’s given a good account of how that choice was made!
[1] Haker, Egekvist and Bjerring (2000) Effect of sensory stimulation (acupuncture) on sympathetic and parasympathetic activities in healthy subjects. Journal of the Autonomic Nervous System, 79: 52-59.
[2] Uchida, et al. (2019) Effects of Acupuncture Sensations on Transient Heart Rate Reduction and Autonomic Nervous System Function During Acupuncture Stimulation. Medical Acupuncture, 31(3):176-184. Abstract available at: doi: 10.1089/acu.2019.1350
[3] Li, et al. (2013) Acupuncture Effect and Central Autonomic Regulation. Evidence-Based Complementary and Alternative Medicine, 267959:6. Available at: doi: 10.1155/2013/267959
[4] Kim, et al. (2018) Lower dietary inflammation index scores are associated with lower glycemic index scores among college students. Nutrients, 10(2):182. Available at: https://www.ncbi.nlm.nih.gov/pmc/articles/PMC5852758/
[5] Scott, Li and Dobrev (2019) Role of Inflammatory Signaling in Atrial Fibrillation. International Journal of Cardiology, 287:195. Available at: doi: 10.1016/j.ijcard.2018.10.020