When I first realised that the conventional ideas about heart disease were, to put it kindly, flawed, I decided to try and start again with a blank sheet of paper and see if it was possible to work out what was really going on. Nothing was ruled in, nothing was ruled out. At first, like almost everyone else, I began to look for alternative potential ‘causal’ factors’ e.g. potassium, or stress, or fibrinogen, or other things in the diet.
However – as I have written before – I came to realise that this was a fool’s quest. There was so much noise, so many apparent contradictions, so many possible interactions and confounding variables that you could pick and choose your evidence to support almost any factor, or set of factors, that you wanted. Gradually I began to realise that if I truly wanted to understand heart disease, I had to start looking at the underlying processes.
In order to do this, I had to try and define exactly what ‘heart disease’ might be. I already knew that that heart disease is not really a disease of the heart. It is a disease of the arteries supplying blood to the heart (the coronary arteries). It is also a disease of other arteries around the body, the arteries supplying blood to the brain (coronary arteries), the kidneys (renal arteries) etc.
What we usually call heart disease is really arterial disease, where the arterial walls are thickened with atherosclerotic plaques, which eventually narrows the lumen, or central channel of the artery, causing angina and suchlike. In the arteries around the heart these plaques can finally ‘rupture’ causing a large blood clot to form on top of the plaque leading to a heart attack or, perhaps to be more ‘accurate’, a myocardial infarction.
In the carotid arteries in the neck, blood clots can also form on top of the plaques. The clot can then break off, travels into the brain, and block an artery. This leads to a stroke or a cerebral infarction. [There are other forms of stroke, but this is the most common].
Whilst this is a relatively simplistic model, I shall use it for the purpose of this blog, because it provides a close enough description of what happens. I shall also continue to use the term heart disease to mean the development of atherosclerotic plaques and formation of clots. Using this as the definition of ‘heart disease’, what processes can explain it?
The process of arterial thickening/plaque development
Moving backwards for a moment, I believe that the main reason why the cholesterol hypothesis has proven so resilient to all contradictory evidence is that the process seemed seductively simple. You eat too much cholesterol, the cholesterol level in the blood goes up and this excess cholesterol is then deposited on the artery walls… thickening and narrowing them. I once saw an episode of The Simpsons, demonstrating this exact thing happening in Homer’s blood vessels. Once something appears on The Simpsons you know you have a meme on your hands.
Of course, this initially simple process has altered and adapted and fragmented and shape-shifted so many times that it is now almost impossible to describe what it is. It has become something like this: you eat too much ‘unhealthy food’ (which may or may not include cholesterol and/or saturated fat), this raises LDL/cholesterol levels/or particle numbers and/or size, or other things, or lowers HDL, or raises triglycerides, or all three… which causes inflammation/oxidised LDL levels to go up, leading to development of plaques/thickenings in the arterial wall…
Sorry for the vagueness of it all, but I defy you try to get anyone to give you a more accurate summary of the current bad diet/heart hypothesis. I can’t. However, despite the fact that the cholesterol hypothesis has fragmented into a more and more confused mess, people still cling to the central process of ‘eating something > blood levels of something going up > narrowing of arteries.’ Primarily, I think, because it seems so simple. A leads to B, B leads to C, C leads to D…eath.
More than twenty years ago I realised that this model was bunk. It just could not explain heart disease. But what other process, or processes, could take its place?
In order to try and answer this I decided not to begin at the very beginning. Instead, I started at the very end, with the final event in heart disease. Essentially, this is when a blood clot or thrombus forms over a plaque, fully blocking an artery in the heart. [Most strokes (ischaemic) are also caused by a blood clot blocking an artery in the brain, although the process is not the same, it is very similar.]
Clearly, therefore, blood clotting (thrombus formation) is the terminal event in almost all heart attacks, and most strokes. This is widely accepted. Indeed, almost all forms of treatment for heart attacks, and strokes, involve the use of anticoagulants of different types: clot busters, aspirin, clopidogrel… or inserting stents to prize open the blockage caused by the blood clot etc. Acute cardiology intervention could, in many ways, best be defined as thrombus management.
The importance of blood clotting in heart disease death can be further highlighted if we look at people with Hughes Syndrome. This is a condition where the blood is dangerously more likely to clot – thrombophilia. People with this syndrome are far, far more likely to have strokes and heart attacks – often at a very young age – sometimes before the age of twenty. (Mainly strokes, actually). The condition is managed with various anticoagulants.
I could continue on this theme for some time, but the role of blood clots in causing death from heart disease, and strokes, is not in the slightest controversial. What is somewhat more controversial is to suggest that the earlier process of heart disease, atherosclerotic plaque development, could also be due to abnormalities/ dysfunction, with the system of blood clotting/repair.
At present, although I have never seen it stated clearly, it seems that everyone is happy to accept that atherosclerotic plaque development is due to one set of risk factors. Then the final event, the deadly blood clot, happens…coincidentally? Due to a completely different set of risk factors? This remains unexplained.
I was never comfortable with the idea that the creation of atherosclerotic plaques has one set of ‘causes’ whilst the final event, the blood clot, has another set of, potentially, unrelated causes. This seems clumsy, and always did – two diseases welded together to make one disease? I thought it was much better to see if a blood clot/thrombosis hypothesis could explain the entire process from start to finish. This could just be me trying to make things neat and tidy, but I don’t think so.
Firstly I tried to articulate what the unified ‘clotting’ hypothesis might look like. Whilst it does not have the elegant simplicity of the cholesterol hypothesis, I hope that it is clear:
Step One: Various factors damage the artery wall (endothelial damage)
Step Two: A thrombus or clot forms on top the area of damage
Step Three: Once the thrombus has stopped growing/stabilised, endothelial cells re-grow over the top of it
Step four: As a result of step three the thrombus becomes, effectively, incorporated within the arterial wall
Step five: Various repair processes break it down and clear it up – but often not fully
Step Six: The area of ‘damage/repair’ becomes a focus for further damage/thrombus formation
Step Seven: The thrombus/plaque grows through repeated episodes of thrombus deposition/repair
Step Eight: A final thrombus forms over a large plaque that completely blocks the artery leading to a heart attack
Many parts of this are far from new. For those who have read my previous book, and blog, you will know that Karl Von Rokitansky proposed that plaques in arteries were really thrombi, over one hundred and fifty years ago. The problem that lead to his ideas being dismissed was, essentially, step three.
Whilst he recognised that arterial plaques looked very like thrombi, and contained everything you find in a thrombus, he could not explain how a thrombus could possibly come to be inside the arterial wall, covered by endothelium (the single layer of cells that line arteries). Virchow attacked his hypothesis simply by asking how this could occur – well, obviously, it cannot. Bong!
However, if Rokitansky had known what is now known, his hypothesis may well have won the battle of ideas, and the entire direction of research into heart disease would have gone off in a completely different direction.
The answer to the Rokitansky conundrum is, of course, very simple. If you damage the endothelium, and a thrombus forms over the area of damage (this will always occur), replacement endothelial cells do not come from within the artery wall (as happens if you scratch your skin). They come from the blood itself.
New endothelial cells develop mainly in the bone marrow, they float about in the bloodstream, and they are known in this state as Endothelial Progenitor Cells (EPCs). EPCs are attracted to areas where the endothelium is missing – where a thrombus has formed. Once there, they stick to the top of the thrombus and develop into mature endothelial cells. Hey presto, the thrombus becomes covered by a new layer of fresh endothelial cells and is now, effectively, within the arterial wall itself.
Of course, if you think about it, this has to be what happens. If a thrombus forms on your artery wall, it cannot simply fall off once the artery has ‘healed’ beneath, as would a scab on your skin. If this were to take place, the thrombus would just travel a bit further down the artery until it jammed. As you can imagine, jamming arteries with thrombi is generally pretty catastrophic. See under ischaemic stroke.
Which means that the repair system for thrombi that form on the walls of arteries has to involve covering them up – then clearing the debris away from within the artery wall itself. I would like to say that I hypothesized that EPCs must exist, before I found out that they did. But you only have my word for that.
A whole new process – and potentially causal factors
At this point, I would like you to look afresh at heart disease/plaque development as containing three interconnected processes:
- Endothelial damage
- Thrombus formation
Viewing things in this way, you can see that factors that damage the endothelium e.g. high blood sugar levels, turbulent blood flow, stress, will cause more thrombi to form; the more thrombi that form, the more that plaques will develop. Factors that make the blood more likely to clot – and also create bigger and more difficult to shift thrombi – will accelerate plaque growth, and increase the risk of the final event occurring. Factors that interfere with repair process are likely to make plaques become bigger, and more damaging.
If these are the processes, then ‘factors’ which truly are causes (rather than associations), should fit easily within this model – and indeed they do. At this point you can play a game – if you are as sad as I am! It is one that I play when sitting quietly in a train, or driving, or half watching the television. It is called, think of a risk factor and see if it has a damaging effect on any of these three processes. The other half of this game is to think of something that ‘protects’ against heart disease and see if benefits any of these three process i.e. does it protect the endothelium, reduce blood clotting, or enhance repair.
Now to let you play this game yourself. Hit Google or Pubmed, and see what you come up with. Try endothelial damage, diabetes and CVD. Or smoking, EPCs and thrombus formation. Or try, effects of insulin resistance on EPCs and thrombus formation. Try exercise, nitric oxide and endothelial function. Or yoga and endothelial health, or smoking and blood clotting, EPCs and endothelial health.
Stick in any significant risk factor for heart disease, or stick in any factor known to reduce the risk of heart disease, and you will always find that they have a major impact on one of the three key processes: Endothelial damage, thrombus formation, or repair. Usually all three… This is not a coincidence.
In the light of this, I think it is interesting to review statins. Now I am a great critic of statins, as I believe their downsides greatly outweigh their benefits. However, they do reduce the risk of death from heart disease and strokes – if not by a great amount. At present this is generally attributed to to their impact on lowering cholesterol levels.
But I thought it was interesting to ask another question. Do they also have a significant effect on any of the three processes? Why, yes they do. Firstly, to look at their effect on the key repair system of EPCs. Here is a paper called:
Increase in circulating endothelial progenitor cells by statin therapy in patients with stable coronary artery disease1.
‘Statin treatment of patients with stable CAD was associated with an approximately 1.5-fold increase in the number of circulating EPCs by 1 week after initiation of treatment; this was followed by sustained increased levels to approximately 3-fold throughout the 4-week study period.’
In short, statins increase the number of EPCs which are essential to repair areas of damage to artery walls. Well, who’d a thunk? Well, me, actually.
Now to look at another critically important effect of statins. Before doing this I have an admission to make. It is something I have known about for many, many, years. It is this. Familial Hypercholesterolemia does increase the risk of heart disease. Something that I have tended to gloss over, for obvious reasons.
In my defence I have always known that this increased risk had nothing to do with the LDL/Cholesterol hypothesis. It was something else. The something else is that that Familial Hypercholesterolaemia causes (for a number of reasons, and not in everyone) increased thrombus formation. Or, to put it another way, it makes your blood much more likely to clot.
Here is a paper from the Journal Circulation called ‘Hyperlipidemia and Coronary Disease. Correction of the Increased Thrombogenic Potential With Cholesterol Reduction2.’ It is nearly twenty years old:
‘Background: Hypercholesterolemia is a risk factor for coronary disease, and platelet reactivity is increased with hypercholesterolemia, suggesting a prethrombotic risk. The aim of this study was to measure mural platelet thrombus formation on an injured arterial wall in a model simulating vessel stenosis and plaque rupture in hypercholesterolemic coronary disease patients before and after cholesterol reduction.’
‘Conclusions: Thus, hypercholesterolemia is associated with an enhanced platelet thrombus formation on an injured artery, increasing the propensity for acute thrombosis…cholesterol lowering may therefore reduce the risk of acute coronary events in part by reducing the thrombogenic risk…’
Yes, gentle reader, statins do work to reduce the risk of heart disease, but not directly by lowering LDL/Cholesterol. Instead they work a bit like aspirin, by stopping platelets stick together over areas of damaged endothelium. They also work a bit like Clopidogrel – which does much the same thing. They also work a bit like omega-3 fatty acids (which the Eskimos eat a lot of), and causes them to have nose bleeds.
What statins do not do is to mimic the action of warfarin. Warfarin has little or no impact on platelet thrombus formation, caused by endothelial damage, it works in a very different way. Once you know this, you can, as I promised, understand the conundrum I left in the last article on this topic. Namely, why does warfarin protect against strokes, but does not protect against heart attacks. Whereas aspirin, which is also an anticoagulant, primarily protects against heart disease.
Now you know… possibly?