Does coffee cause arrhythmia?
Many patients ask me whether coffee and other caffeine sources are beneficial or harmful to the health of their heart and circulation. Typically, these questions focus on whether caffeine causes irregular heart rhythms such as atrial fibrillation and ventricular ectopic beats. A typical patient raising such a question might have come to see me to find out what is causing their feeling of palpitations; perhaps we’ve done some tests and identified short periods of disturbed heart rhythm that explains their symptoms. The next question is inevitably, “What can I do to reduce the problem?” and sooner or later we will get onto discussing the question of caffeine and whether they should cut down or avoid it all together.
This is a nuanced area of medicine and one where the “evidence” (i.e. the results of scientific studies) do not necessarily fit with patients’ lived experience. Both the individual and the scientific perspectives on the issue are valid, and both deserve some sort of explanation.
Scientific literature in the area falls into two categories:
First: “basic science” studies in which the effects of pure caffeine applied to cardiac and circulatory tissue are studied in a laboratory setting. Typically these studies are looking for a correlation between the amount of caffeine squirted onto cells in a lab, versus the likelihood that those cells will behave oddly.
Second: “clinical” studies in which groups of people are asked to quantify their caffeine intake and then watched closely to see whether they develop cardiac problems. Broadly speaking these studies seek to establish a correlation between caffeine intake and likelihood of developing arrhythmia.
The results of the basic science studies is clear: in a laboratory setting working on cells outside a living body, caffeine has a direct effect on heart, nerve, brain and other tissue types, including a significant effect on cells taken from animals’ (and humans’) cardiac conduction system. These are the cells that are responsible for coordinating the heartbeat and sustaining the electrical meter and rhythm of the heart and it is certainly interesting that caffeine has such a clear stimulatory effect on them.
However, the dose of caffeine applied to cells in basic science studies is usually very high, and it is quite a leap to say that the effect of purified caffeine directly applied to cells in a laboratory is the same as drinking a beverage with caffeine in it. The two processes are very different. These basic science studies are therefore the starting point for a scientific answer to our question, not an answer in themselves.
Clinical studies involving real patients drinking real caffeine and looking for real heart rhythm abnormalities are more directly relevant to my discussions with patients in my clinic. However, studies in this area are very “heterogeneous”, or - in plain language - contradictory. Expert opinion articles on this subject tend to categorise studies as either “positive” (suggestive of a link between caffeine and arrhythmia), “negative” (suggesting that caffeine reduces arrhythmia) or “neutral” (suggesting that caffeine intake has no effect on arrhythmia). One approach to dealing with contradictory results like these is to group all the studies together and look for an overall trend. Advanced statistical methods are employed to adjust for slight differences between the studies and produce an overall answer to the question “does caffeine cause arrhythmia?”
You can read one such review article - a rare one written in plain language - here. The overall message is that if you take all clinical studies together the sum total result is that caffeine does not seem to influence arrhythmia occurrance.
However this approach of grouping studies together and looking for an overall message risks glossing over the detail of individual studies.
One of the most interesting and relevant findings is from a small study conducted in the 1980s (scientific abstract available here) where a group of patients with frequent ventricular ectopic beats (a common form of heart beat irregularity) were asked to abstain from caffeine for 24hrs and then take a moderate dose of caffeine the following day. The number of ectopic beats the patients suffered on each day was recorded via a continuous ECG monitor. The frequency of ectopic beats increased significantly during the caffeinated period compared to the caffeine free period. Interestingly, this study also included a group patients without frequent ventricular ectopic beats when not drinking caffeine; these “control” patients experienced no difference in their heart rhythm with caffeine consumption. The message from this study is that some patients with an underlyin predisposition to ectopic beats will experience and increase in ectopics after consuming caffeine. You can see that this message, focussed on one small group of patients with one particular heart rhythm abnormality, would be drowned out if the study was pooled with many others asking slightly different questions in different patient populations.
This study is particularly interesting to me because it highlights two factors that I recognise from my everyday practice:
First, not all people are the same in their response to caffeine. Some patients who come to see me with problems unrelated to their heart rhythm drink large quantities of caffeine daily with seemingly no ill effect and specifically without causing problematic heart rhythm issues. This contrasts quite markedly with some other people I see with heart rhythm symptoms who tell me very clearly that coffee and other sources of caffeine trigger their symptoms and that avoiding caffeine has been an effective treatment for them. Presumably the first group have more in common with the “control” patients in the study described above, and the second group have more in common with the ectopic beat patients in the study. What separates these groups of people and how we can predict whether you are in group 1 or group 2 is a really tricky question and I don’t feel that we have a good answer to this currently.
The second message that I take from the study above is that there clearly is a group of patients for whom caffeine has an effect on their arrhythmia. This message is lost in much of the discussion around the issue and in some of the larger studies that fail to drill down to identify differences between groups of patients and different arrhythmia types. An example of this approach is this study here which included a huge number of patients but failed to distinguish different arrhythmias or disease states, and unsurprisingly found no link between caffeine and “all arrhythmia in everyone”. Differences between patients and diseases are real and they are important; glossing over the details helps no-one understand.
Given the complexity and contradiction of much of the evidence in this area, it is easy to become confused and give up trying to apply the evidence to your own condition. It is the role of cardiologists to interpret the evidence and to distill the relevant parts into useful advice that patients can follow.
My approach is generally quite straightforward: if you’re troubled by arrhythmia and you drink more than a minimal amount of caffeine then we can discuss the pros and cons of cutting down or cutting it out. For some people, no matter what the studies say, it does seem to work. For others, it doesn’t. I’ve not yet found a reliable way of predicting which group you will be in.
---------------------------------
You can read more about atrial fibrillation and other cardiac problems on the patient information pages of my Kent Heart Clinic website at https://www.kentlondoncardio.com.