‘Statins lower the LDL level and protect against heart disease.’ This is the only fact that really matters; all else is so much hot air, do I hear you say? On the face of it, the data certainly seems to represent the so-called ‘reversibility of effect’, which is considered one of the strongest forms of scientific proof – for good reason.
If you think factor x causes disease y, then you remove factor x and disease y disappears. ‘Well… HOW MUCH MORE PROOF DO YOU WANT?’ A bit more, actually. Firstly, with statins, even if you get the LDL level down to about 2mmol/l, you reduce the relative risk of dying of heart disease by about 30 per cent, absolute max. Which means that you can completely remove the risk factor – high LDL levels – yet people still die of heart disease.
Statin data are one thing. The interpretation and presentation of data on statins is another. Here, for example, is a press release from the British Heart Foundation about the Heart Protection Study, published in 2004 – the last really big statin study to report:
Heart experts call for urgent action to implement new findings on cholesterol-lowering treatment
Tens of thousands of lives could be saved each year by changing prescribing guidelines for statins, say UK researchers
Research reported in tomorrow’s [Saturday 6 July] Lancet is set to revolutionise the way cholesterol-lowering drugs are prescribed. It shows that using ‘statin’ drugs to lower blood cholesterol levels protects a far wider range of people at risk of heart attacks and strokes than had previously been thought to benefit. These findings should lead to major changes in treatment guidelines, preventing tens of thousands of deaths each year, a London news briefing has been told.
At present, statins are often restricted to people who have heart disease and elevated cholesterol levels. But, new findings from the UK’s 20,000-patient Heart Protection Study show that statins also cut the risks of heart attacks and strokes in people who have diabetes, or have narrowing of arteries in their legs, or have had a stroke. Most remarkably, the study found substantial benefits even among those high-risk patients considered to have ‘normal’ or ‘low’ cholesterol levels. It provides definite evidence that guidelines should be changed so that – irrespective of the blood cholesterol level – a statin is considered for anybody at increased risk of either heart attacks or strokes.
‘The clear message from this study is: “Treat risk – not cholesterol level”,’ said Professor Sir Charles George, Medical Director of the British Heart Foundation – the UK’s leading heart charity. He called for an urgent review of national and international guidelines on statin use by government organisations, such as the National Institute of Clinical Excellence (NICE) in the UK and the National Institutes of Health (NIH) in the USA, as well as by professional bodies, such as the European Society of Cardiology (ESC) and the American Heart Association (AHA).
The study overturns conventional wisdom in a number of other areas. For example, current guidelines say that there is little evidence that statins help older individuals. By deliberately studying large numbers of older people, the researchers were able to show that cholesterol-lowering with statins was just as effective for the over 70s as for those in middle age.
Likewise, not many women had been included in previous studies so there was little direct evidence on the benefits and safety of statins in women. With more than 5,000 women included in the Heart Protection Study, it has been able to show that statins work just as well for women as for men.
HPS lead investigator Professor Rory Collins said: ‘HPS shows unequivocally that statins can produce substantial benefit in a very much wider range of high-risk people than had been thought. These new findings are relevant to the treatment of some hundreds of millions of people worldwide. If now, as a result, an extra 10 million high-risk people were to go onto statin treatment, this would save about 50,000 lives a year – that’s a thousand each week. In addition, this would prevent similar numbers of people from suffering non-fatal heart attacks or strokes.
The HPS team estimates that implementing these new findings fully would more than triple the numbers of people benefiting from statins. In the UK, the numbers treated with statins would increase from a current figure of less than 1 in 20 of the population aged over 40 (or about 1 million people) to about 1 in 8 (about 3 million people). This would save an extra 10,000 lives each year.
Yee-haaa! How fantastic is that, we can save tens of thousands of lives by prescribing a few million more statins a day. A thousand lives a week in the UK alone! And this gushing praise does not come from a paid PR agency in full bullshit mode. It comes from ultra-respectable sources. The Lancet, Oxford University, Professor Sir Charles George, Medical Director of the British Heart Foundation no less. If you could win an argument by the sheer number of gongs and letters after your name, I would have been beaten into a senseless mush by now, a white towel fluttering despondently on to the canvas.
However I hope that, by now, you may recognise that, despite the glow of eminence, this press release is actually riven with gaping holes. Here are a few statements that are worth a bit more analysis:
‘It [the study] provides definite evidence that guidelines should be changed so that – irrespective of the blood cholesterol level [my emphasis] – a statin is considered for anybody at increased risk of either heart attacks or strokes.
So we should be treating everyone at risk of heart disease with statins, no matter what their cholesterol level. Why? Because statins reduce the risk at all cholesterol levels. High, average, or low. Just ponder that statement for a moment or two. Do you find anything strange about it?
While you do that. Here’s another statement:
By deliberately studying large numbers of older people, the researchers were able to show that cholesterol-lowering with statins was just as effective for the over 70s as for those in middle age.
But a high cholesterol level is not a risk factor in the over 70s. If anything, raised cholesterol protects against heart disease in the over 70s, especially in women – for whom, in fact, a raised cholesterol level isn’t a risk factor for heart disease at any age.
Another of the main findings of the HPS was that it protected against stroke – much more so than against heart disease. Yet, a raised cholesterol level is NOT a risk factor for stroke. Probably the exact opposite, if truth be told.
Were any of these issues raised, or even hinted at? In the press release? Not at all. Statins are marvellous, everyone should take them, end of discussion.
However, the issues I have just raised are as trifles compared to the elephant that is sitting in the middle of the room here. It sits quietly cleaning its tusks, unremarked upon, but it is there nonetheless, no matter how much mainstream researchers would like it to pack up its trunk and trumpety, trump, trump, off to the circus. This elephant even has a name. It is called ‘Nellie the total mortality data’. Not a very snappy name, I agree, but it is fully accurate.
STATINS AND TOTAL MORTALITY
With more than 5,000 women included in the Heart Protection Study, it has been able to show that statins work just as well for women as for men.
That rather depends on what you mean by work, I suppose. What I mean by work is that statins saved lives. The HPS team, however, decided not to release the total mortality data on women.
I was not the only one to notice this. Arnold Jenkins, a UK GP, picked up on this not insignificant point, and wrote about it in the BMJ:
Imagine my delight when I heard of the large heart protection study showing clear benefits in the use of statins for women. On reading this study I was therefore disappointed to find the total mortality data for women missing. I now understand that the total mortality benefit for women did not reach significance and therefore was not published.
Louise Bowman, personal communication, 2002 http://bmj.bmjjournals.com/cgi/content/full/327/7420/933-b
I too, have attempted to get hold of the mortality data for women from the HPS study. No such luck. Overall mortality is the most important end point in any clinical trial. It is also the easiest to measure. The HPS researchers are not actually alone in failing to publish overall mortality data in women. This is also true of most of the statin trials, although on those occasions when it has been published, it has shown nothing at all. The 4S, one of the earliest, and by far the most positive of all the statin trials ever, showed no difference at all in female mortality. (Actually, two more women in the trial who were taking statins died, but this was a ‘non-significant’ difference.) In short, statins do not save lives in women.
Statins do not save lives in women.
Statins do not save lives in women.
Statins do not save lives in women.
Is it possible to highlight how important this fact actually is?
STATINS DO NOT SAVE LIVES IN WOMEN!
But you would never know that from reading anything that is written on the subject. You have to dig and dig. This information is buried very deeply. Actually, that is not quite true. The information isn’t buried, it just plain doesn’t exist. And how can you find something that doesn’t exist?
Firstly, you have to ask yourself the most difficult question of all: ‘What’s missing?’ It’s a bit like ‘the dog that didn’t bark’. It’s not what happened that’s important, it’s what didn’t happen. While our attention is distracted by 50,000 lives being saved from heart disease every year, we fail to spot the total mortality data escaping through a hole in the floor.
As a man, I can’t truly speak for women. But ladies, I really think that you should be outraged by the complete silence on this issue. When your doctor is badgering you to take a statin, do you not think that it might be of some importance for them to mention that, while this pill may slightly reduce your risk of stroke and heart disease, it will not on average increase your life expectancy by one single day? Or, to put it another way, taking a statin may change what is written on your death certificate, but it will not change the date.
Despite the complete and utter lack of evidence of any mortality benefit, GPs in the UK are actively encouraged to check cholesterol levels in women, and further encouraged to get the cholesterol level below 5.0mmol/l. If they achieve this in a high enough percentage of their practice population, they are then paid large sums of money.
If this were not so serious it would be laughable. But frankly, at this point, I do not feel like laughing. I feel like grabbing a few people by the lapels and shaking them with great vigour. How can you justify putting millions upon millions of women on powerful and potentially very damaging drugs, when they will not save one single life? This question requires an answer.
If this book achieves nothing else but to start a debate on this issue, then I will be perfectly content – as it is a debate that can have only one conclusion. Perhaps you think statins are harmless, so it doesn’t really matter all that much? Well, if you are a foetus, statins are not harmless at all.
While it is certainly true that not many women of childbearing age take statins, it is becoming more and more common. And with statins now available over the counter in the UK, there is an increasing danger that warnings about taking statins in pregnancy will go unheeded. Or perhaps someone will forget to mention it as they hand over a pack of statins on a busy afternoon in the supermarket pharmacy. Or maybe a husband will pick them up for his wife, without telling anyone that they are not for him.
I have been told that this can’t possibly happen, because statins are contraindicated in pregnancy. Well, so are about ten thousand other drugs, most of which have never actually been shown to do harm. If a drug hasn’t been tested in pregnant women – and which company would now risk doing this – it is often ‘contraindicated’, just to be on the safe side.
However, there are degrees of contraindication. Roaccutane, for example is hyper-contraindicated. This drug is used for resistant acne, and can only be prescribed by a dermatologist with severe warnings handed out about getting pregnant, and instant termination recommended if this happens. Why? Because of the terrible birth defects that this drug can cause.
Roaccutane represents a full-on contraindication, one that is taken very seriously indeed. This is a long way from the ‘It’s probably not a good idea, but it probably won’t do any harm’ type of contraindication. Or the contraindication against certain forms of antibiotic because they can cause permanent tooth discolouration – which is unfortunate, but hardly life threatening.
In short, to say that a drug is ‘contraindicated’ in pregnancy does not necessarily ring many alarm bells. And where, in doctor’s minds, do statins sit on this spectrum of contraindication? Pretty low down, to judge by my discussions with fellow doctors.
Making things even more likely to go wrong, statins have been presented as a universal panacea, with no side effects worth mentioning. Taking a statin is now viewed, among doctors, as akin to taking a multivitamin or low-dose aspirin.
You may remember a quote from Dr John Reckless about the use of statins:
So maybe people should be able to have their statin, perhaps if not in their drinking water, with their drinking water.
Statins can now be bought in the UK without a prescription. In this climate women will take statins over the counter, and will become pregnant while on statins. This is inevitable – it has almost certainly already happened.
But be afraid, be very afraid. In April 2004 an article appeared in the New England Journal of Medicine entitled ‘Central nervous system and limb anomalies in case reports of first-trimester statin exposure’. As statins are ‘contraindicated’ in pregnancy, there wasn’t much data to go on. But they still managed to find 178 cases. This number was whittled down through first-trimester elective and spontaneous abortions, loss to follow-up, and suchlike, to an eventual figure of 52 confirmed cases.
Out of these 52 there were 20 reports of malformation, including severe defects of the nervous system, unilateral limb deficiencies, complex lower-limb abnormalities and much more.
Here are just three of the reports:
• Holoprosencephly (defective septum separating lateral cerebral ventricles, with cerebral dysfunction), atrial septal defect, aortic hypoplasia, death at one month of age.
• Cervicothoracic-to-lumbar neural tube defect, myelocele, duplication of spinal cord, cerebellar herniation with hydrocephalus; apparent agenesis of palate.
• Left leg: femur 16% shorter than right side; foot: aplasia of metatarsals and phalanges 3, 4 and 5; additional VACTERL defects8: left renal dysplasia reversed laterality of aorta, disorganised lumbosacral vertebrae, single umbilical artery; additional findings: clitoral hypertrophy, vaginal and uterine agenesis.
As the authors point out, data from case series cannot be used to test hypotheses of teratogenicity (substances causing birth defects). But 20 severe birth defects out of 52 children is an extremely disturbing figure. As high as anything found with thalidomide, and with more serious defects.
I believe this study constitutes proof that statins are extremely dangerous in pregnancy, and cause terrible birth defects, a belief that is reinforced by the fact that the defects fit within the known effects of inhibiting cholesterol synthesis in the foetus. Oh yes, statins are truly wonder drugs. We should be putting them into the water supply. Just don’t expect too many healthy babies to emerge if we do. A few hundred thousand with ‘duplication of the spinal cord’, perhaps.
Before getting pulled down too far into the damaging effects of statins – an issue that I will return to – I want to get back to the overall mortality data. This time in men.
TOTAL MORTALITY IN MEN
At this point in the discussion I need to split men into two distinct types:
Type A: Those who already have diagnosed heart disease (previous heart attack, or angina).
Type B: Those who don’t.
This distinction has become blurred more recently – I would say quite deliberately – into high risk and low risk. However, it is a very important distinction to make. Because, according to the clinical trials, if you give statins to men who already have heart disease they are protected against cardiovascular disease. They also have a reduced overall mortality rate (more on this later).
However, if you give statins to men who do not have heart disease, while you do reduce the rate of cardiovascular disease, there is no benefit on overall mortality. None at all. And you don’t need to take my word for it.
The University of British Columbia, which is part of the worldwide Cochrane collaboration – a not-for-profit group that analyses health-care interventions around the world – decided to look at the use of statins in primary prevention, that is, in people with no known pre-existing heart disease. (Secondary prevention is attempting to reduce death from CVD in people who already have heart disease.)
The researchers at the University of Columbia asked the question ‘Do statins have a role in primary prevention?’ And they brought together the data from the major statin studies done around the world. The answer:
• If cardiovascular serious adverse events are viewed in isolation, 71 primary prevention patients with cardiovascular risk factors have to be treated with a statin for 3 to 5 years to prevent one myocardial infarction or stroke.
• This cardiovascular benefit is not reflected in two measures of overall health impact, total mortality and total serious adverse events. Therefore, statins have not been shown to provide an overall health benefit in primary prevention trials.
Dr Graham Jackson, in the UK, also looked at all of the statin trials done up to the year 2000. His conclusion, published in the British Journal of Clinical Pharmacology, was that:
Long term use of statins for primary prevention of heart disease produced a 1% greater risk of death over ten years vs placebo when the result of all the big controlled trials reported before 2000 were combined.
And that, ladies and gentlemen, is the bottom line. Statins do not reduce mortality in men who do not already have diagnosed heart disease, which represents considerably more than 90 per cent of the male population.
As a slight aside, after the publication of these data, someone asked the University of British Columbia researchers the question: ‘What is the evidence of benefit for primary prevention in women, in heart disease?’ Their reply:
There were 10,990 women in the primary prevention trials (28% of the total). Only coronary events were reported for women, but when these were pooled they were not reduced by statin therapy… Thus the coronary benefit in primary prevention trials appears to be limited to men.
In short, in primary prevention, statins not only have zero effect on overall mortality, they also have zero effect on reducing heart disease in women. So you get absolutely no benefits at all. I suppose this may all seem almost unbelievable, given the ludicrous levels of hype surrounding statins, but it’s true.
However, studies such as the one done by the University of British Columbia have been brushed aside by the statin juggernaut, with no discernible effect on anyone, or anything. If anything, the hype has merely accelerated. For example, a couple of years ago a major trial (ASCOT-LLA), was stopped early – an unusual step. The reason for this is because of the ‘massive’ reduction in cardiovascular deaths in those given statins, compared to those poor souls condemned to taking a placebo. The difference was so great even though the trial was blinded that it was considered unethical to continue.*
As the cardiovascular benefits in patients taking Lipitor were highly significant, the independent Steering Committee stopped the cholesterol-lowering arm of the study in October 2002, nearly two years earlier than planned.
Pfizer trumpeted this fact from the rooftops at the time, implying that atorvastatin was just so damned wonderful that the benefits had emerged in double-quick time. This made it unethical not to give the drug. All very fine and noble and public spirited, no doubt.
What was not shouted from the rooftops was the total mortality data. Indeed, in analyses of the trial it wasn’t mentioned at all. Here is one example. It’s a bit long, and gigantically dull, but I just want to make the point that benefit after benefit is being highlighted here, yet there is nothing about total mortality:
Follow-up was planned for an average of 5 years. The ASCOT-LLA was stopped after 3.3 years owing to the superiority of atorvastatin 10 mg over placebo in reducing the primary end point of nonfatal myocardial infarction (MI) and fatal CHD… Cholesterol lowering with atorvastatin was associated with a highly significant reduction in the primary end point of nonfatal MI and fatal CHD (36%, P = 0.0005). The observed benefit was consistent across the secondary end points and the 18 prespecified subgroups. The ASCOT-LLA findings have influenced lipid-lowering guidelines and support the concept that treatment strategies to reduce cardiovascular disease should be based on the assessment of all cardiovascular risk factors, rather than on numerical thresholds of individual risk factors, to determine treatment strategies.’
Am J Med, December 2005; 118 Suppl 12A
In the case of ASCOT-LLA, however, the total mortality data were published. So I took the graph of overall mortality, scanned it, and I have reproduced it for you in (Fig. 27). The dotted line represents deaths in the placebo group, the solid line represents deaths in the atorvastatin group.
Fig. 27 Total mortality: atorvastatin vs placebo
I also did something else to this graph. I added in my own line at 3.3 years. I did this because the trial actually ended at 3.3 years. Yet, as you may notice, the data lines dribble on for another three months. The study was stopped, yet the data analysis went on?
Why did this happen? Well, perhaps it is relevant to point out that that from 3.3 years to 3.5 years the total mortality lines divide sharply in favour of atorvastatin (the only time period when this happened). At 3.3 years, there was hardly any discernible difference between drug and placebo, yet by 3.5 years there were 185 deaths in the atorvastatin group and 212 in the placebo group.
This difference still did not manage to limp its way to statistical significance, but it was considerably better than the difference three months earlier, which was approximately zero. By the way, I haven’t got the exact figures for 3.3 years. I have asked for them, but silence was the stern reply. I have also asked the authors how a trial can end after 3.3 years, yet data collection continues for another 3 months – ‘unblinded’ and uncontrolled. Silence.
However, at this stage I think it’s worth pulling two or three facts together again:
| Fact one: | Statins do not reduce overall mortality in women. |
| Fact two: | Statins do not reduce overall mortality in men without heart disease. |
| Fact three: | Statins do not, therefore, reduce overall morality in > 95% of the adult population. |
Something that I have not really brought up, but perhaps I should, is the following question. If statins reduce death from cardiovascular diseases, yet there is no impact on overall mortality, this means that people taking statins must die at a higher rate from other causes. So what are these other causes?
Well, it’s a good question. But I am afraid that I have no clear-cut answers. In most statin trials there is usually a great echoing silence on the point. In the Heart Protection Study, as you may recall, the researchers didn’t even bother to publish the overall mortality data in women, let alone list exactly what people died of. This type of thing makes it rather difficult to get a handle on what people who are taking statins die of instead of heart disease. The data just ain’t there.
This area is made even more opaque by the reaction of the mainstream researchers to overall mortality data. A recent statin study called ‘Treat to New Targets (TNT)’, failed – once again – to show any benefit on total mortality. According to head researcher of the TNT study, Dr John LaRosa, ‘We need to make the assumption that mortality has been proven, that LDL lowering does in fact lower total mortality rates.’ We need to the make the assumption? Why does a researcher need to make assumptions – especially when all the evidence seems to me to point the other way?
Dr Roger Blumenthal, from Johns Hopkins University Medical Center, in Baltimore, Maryland, said that the TNT mortality finding was ‘unfortunate’ and ‘a bit surprising’, but that the increase in non-cardiovascular mortality was ‘likely due to chance’. Blumenthal went on to claim that, ‘The totality of evidence does not suggest that lowering LDL cholesterol to very low levels is associated with non-cardiovascular mortality.’ If there were fewer cardiovascular deaths in the TNT trial, yet no difference in overall mortality, I would have thought that there is only one conclusion that can be drawn: there were more non-cardiovascular deaths.
At this point, hopefully, you are beginning to realise that statins may not be quite as super-wonderful as you may have thought. You may also be wondering how it is that, despite their almost complete lack of any real benefit – i.e. actually saving lives – statins have been hyped to the very skies. There is a very simple reason for this. It’s called money.
Rosuvastatin (Crestor) was launched a couple of years ago, or so. In the first year of its launch, $1 billion was spent on sales and marketing. To quote Dr Evil from Austin Powers: ‘One… beellion… dollars.’ And this, remember, was only the budget for one of the six statins on the market. One… beellion… dollars, it must be said, buys a hell of a lot of publicity.
At the same time, you could pay journalists to attend international meetings, and provide them with PR company-generated press packs highlighting the wonders of your drug. Positive findings can then be hyped relentlessly, and the health editors of newspapers wined and dined. Ghost authors can then be found to write up the findings of trials, ensuring that the correct marketing spin is applied to the data. Opinion leaders lend their imprimatur to the papers at the end of this process – along with an eye-watering invoice, naturally.
Nowadays, the entire world of clinical trials is controlled to a quite extraordinary degree. To quote Dr Marcia Angell, who used to edit the New England Journal of Medicine (which is one of the top five most influential medical journals in the world – possibly even number one):
It used to be that drug companies simply gave grants to academic medical centres for the use of their clinical researchers to do a study and that was it. It was at arm’s length. The researcher did a study and he or she published the results, whatever those results would be. Now, it’s very, very different. The drug companies increasingly design the studies. They keep the data. They don’t even let the researchers see the data. They analyse the data, they decide whether they’re going to even publish the data at the end of it. They sign contracts with researchers and with academic medical centres saying that they don’t get to publish their work unless they get permission from the drug company. So, you can see that the distortion starts even before publication. It starts in determining what’s going to be published and what isn’t going to be published. This is no longer arm’s length. It’s treating the researchers and the academic medical centres as though they were hired guns or technicians or something. They just do the work. And the drug company will decide what the data show, what the conclusions are and whether it will even be published.
In short, the medical profession is increasingly working closely with pharmaceutical companies.
If you have no role in major pharmaceutical-sponsored clinical trials then you do not speak at major meetings, you do not publish ‘prestige’ papers in high-impact journals, you do not bring in money to your university department. You have little to offer at major international conferences. You live in Backwatersville, man.
A Dr John Kastelein from Amsterdam was utterly outraged that anyone should have objected to the last set of NCEP guidelines because of potential conflicts. He felt that the whole conflict-of-interest issue was being over-hyped. In his words:
I don’t believe a word about the conflict of interest because there is no single opinion leader in the world who has not done any work for a pharmaceutical company in terms of research or trials.
He is supported in his stance by Harvard Medical School Associate Professor Daniel Simon. In his view it’s a mistake to tune out the views of those with potential conflicts of interest because the pharmaceutical industry is driving medical advances. Most unconflicted researchers ‘are not truly expert’. He says.
In an attempt to get some handle on potential conflicts of interest, several of the most prestigious medical journals banded together to demand that those involved in clinical trials, or writing editorials, reveal their connections to the pharmaceutical industry. This is a process known as ‘disclosure’.
To my mind, there are some major problems with disclosure. The first is that if you do not ‘disclose’, absolutely nothing happens – at all. This may not be a victimless crime, but it most certainly is a ‘punishmentless’ crime. If you are rumbled, you can just do a Bill Clinton: ‘But what is disclosure, exactly. Oh, I misunderstood… sorry. You mean being paid vast sums of money by a pharmaceutical company should be disclosed? Gosh. Who would have thought? Golly, I’ll try to remember in the future… promise.’
Entering the debate, the NCEP have said:
The members of experts panels charged with developing guidelines are selected for their scientific and medical expertise, their stature and track record in the field, and their integrity. Individuals who are most expert in a subject area are the ones most suitable to serve on a guideline panel for assessing the science and developing clinical recommendations. They are also often the very people whose advice is sought by industry. Most guideline panels therefore include experts who interact with industry …
I especially liked the bit about integrity, though. How did they measure this? Did they buy a new Acme integrityometer, only £29.99 from Argos? Or did they get all the panel members to eat beans and go up and down in a lift, checking that whoever broke wind owned up each time? Panel members selected for their ‘stature and track record in the field, and their integrity’, is otherwise known as Eminence Based Medicine (EBM).
Eric Topol, just to give one more example, maintains that he has severed his ties with the industry in 2004. Here’s a little something on the matter from his website www.theheart.org:
Topol is unique in that he extricated all ties with industry in 2004. ‘I do not believe that my historical relationship with companies with financial interests in this area is influencing patient care today. I have never ordered a commercial test for aspirin or Plavix resistance for any patient and never advocated the use of such tests for clinical care,’ he writes, adding that it was not mentioned that he published the only article in a peer-reviewed journal warning physicians of the unanticipated potential conflicts of interest in relationships with the investment industry. ‘I have taken a very hard stance on the troubles of the academia-industry megacomplex, have repeatedly challenged industry when there was any question of potential public-health harm, and have tried to set an example of dissociation from industry while still performing important research to advance heart-disease prevention and therapy. It is ironic that an article that purports to unveil bias among physicians besmirches me.’
Again, I will have to leave it up to you to sit on the internet for an afternoon and read the facts that my publisher is reluctant to publish.
If you do so virtually you will find links between all prominent cardiologists and the pharmaceutical industry.
I know that all opinion leaders would be shocked and outraged if you were to suggest to them that they were in any way influenced by the money that they earn from the industry. They consider themselves paragons of virtue. On this issue, however, I defer to the great Robbie Burns:
O wad some Power the giftie gie us,
To see oursels as ithers see us!
It wad frae mony a blunder free us.
THE DAMAGE THAT STATINS CAN DO
Moving on from their complete lack of any benefits – for the vast majority of the population – the next thing to mention is the serious damage that statins are already doing to the NHS budget. Currently, they are the most expensive single item of drug expenditure. While the figures keep on changing, in the very near future, if it has not already happened, statins will cost the NHS over £1 billion per year.
But this is only the costs of the drugs themselves. There are many additional costs:
• Yearly cholesterol tests
• Six-month review by a GP, or nurse
• Payment to GPs for getting blood cholesterol levels down through the Quality Outcome Framework system
Just to tease out one figure in more detail. If we consider that ten million people, at the very least, are supposed to be on statins and they are given a six-month check-up, this amounts to twenty million consultations per year. An average consultation, adding in blood tests, doctor’s time, payment for QoF payment etc., is at least £50, absolute minimum. So this is an extra £1 billion, on top of the £1 billion spent on the statins.
Two billion pounds a year is a not inconsiderable sum of money. What else could you do with the money? Well, you could employ around 70,000 extra nurses a year to start with, which isn’t bad going. Or build two brand, spanking new university-sized hospitals, fully equipped, each year. Take your pick.
But you know. A billion here, two billion there – it’s only money, after all. What I am more interested in looking at here is the potential physical harm that statins can do – apart from possibly causing horribly deformed babies, of course.
At this point, I should state that I do not think that statins are hugely dangerous. In most of the trials statins, have done nothing at all to improve overall mortality, but they don’t seem to have increased the death rate. So putting someone on a statin is unlikely to actually kill them. Having said this, of course, statins may kill. Cerivastatin, the drug withdrawn by Bayer, was implicated in the deaths of at least 100 people before it got withdrawn. Data from the FDA show that simvastatin was established as a direct cause of death in 416 people between 1997 and 2004. All statins have been linked directly to people dying. How many exactly? Who knows? How could you know?
Vioxx, the arthritis drug, was estimated to have been linked to more than 100,000 deaths in the USA in two years. And no one noticed! The fatal effects of Vioxx were only picked up coincidentally as part of a major trial to see if this drug could protect against bowel cancer. The impact on mortality was noted, and highlighted, by a rather heroic employee of the Food and Drugs Administration (FDA), Dr David Graham. For his efforts in protecting the safety of the public he was smeared in the press and ruthlessly attacked. Luckily, a certain Senator Charles Grassley got involved in the case. He wrote a letter to Lester Crawford, the acting commissioner at the FDA at the time. I reprint some sections of it here, because it is an absolute cracker. A symphony in restrained rage:
As Chairman of the Committee on Finance, I have made it clear to you that I expect that Dr David Graham’s right as a federal employee will be fully respected by the Food and Drug Administration. Last Wednesday, November 24, 2004, I requested that the Office of Inspector General (OIG), Department of Health and Human Services conduct a complete and thorough investigation into the facts, events, persons, policies, regulations and laws relating to allegations that a number of management level employees at the FDA may have acted ‘to discredit an outspoken agency safety office who was challenging the FDA’s drug safety policies.’ I referred to the attached article from the Washington Post entitled, ‘Attempt to Discredit Whistle-Blower Alleged.’…
I’d like to reiterate what I have repeatedly stated in writing and have verbally communicated to your agency, namely that this Committee takes its responsibility to protect witnesses and particularly government witnesses very seriously, and that holds particularly true for Dr Graham.’…
I understand that retaliatory action against dissident employees can come under many guises. Therefore, I also request that you address allegations that administrative action may be taken against Dr Graham, including that he may be terminated or transferred against his wishes to a job other than conducting scientific research. Please advise me whether there is any truth to these allegations and, if so, explain what actions are being taken to transfer Dr Graham from his present position and duties at FDA… On at least 6 separate occasions – 3 by letter and 3 in meetings with FDA staff – I have requested that FDA employees be advised that they may come to Congress and speak freely without fear of reprisal. Do you believe that FDA employees are free to speak to members of Congress without advising FDA’s Office of Legislation? If so, when are you going to act on this request?
Ouch!
This sorry saga highlights that fact that a drug could potentially kill hundreds of thousands of people without anyone actually noticing. You may think that this must be impossible. There have to be agencies out there monitoring this sort of thing on a day-to-day basis, combing through the statistics with a fine-tooth comb? Not so – and anyway, how could they? Drug safety is supposed to be fully established in the clinical trial process. Once a drug is out there in the community it could wipe out thousands, unnoticed – witness Vioxx.
In addition to the difficulty of expecting an individual doctor to spot patterns of increased mortality, the majority of people taking statins are usually taking other drugs at the same time. So how can anyone know which drug did what? And what’s more, very few doctors report adverse events at all.
In short, although it seems unlikely that statins are killing thousands of people, it’s as well to remain vigilant. After all, it could take a long time for serious effects to emerge – much longer than any of the statin trials have lasted. Five years may seem sufficient to pick up on serious effects, but if you ran a five-year clinical study on the effects of smoking, you would see no impact on lung cancer – at all. Statins, potentially, are going to be taken for 30, 40, 50 years. Are they safe over this time period? Only time will tell.
Moving on to more immediate effects. The primary way that statins kill people is through a side effect known as rhabdomyolysis, which is breakdown of skeletal muscle. Basically, your muscles dissolve away, the waste products from this process destroy the kidneys and you can then die from kidney failure. This, I hasten to add, is not common. According to one report, in the USA over two and a half years there were only 871 reports of rhabdomyolysis with statins, 38 of which were fatal. How many more cases went unreported is unknown, although several studies have calculated that adverse event reporting underestimates the true number of events by about 95 per cent. Even so, this is not an epidemic by any means.
The major problem with statins though, is not that they kill a few hundred people here and there, it is that they create a huge burden of insidious side effects, most of which go unnoticed, or are dismissed. You feel tired? Well, you are getting older, after all. Muscle pains? Hell, we all get them. Even when you suffer a complete belter of a side effect, most doctors refuse to believe this could possibly have anything to do with the statin you are taking.
Let me introduce you to a doctor in the USA called Duane Graveline. He is a family doctor, but he also trained as an astronaut with NASA, and works closely with airline pilots to assess their fitness to fly. Some years ago, he was found to have a raised cholesterol level and was put on a statin. He had no problem with this, as he fully believed in the cholesterol hypothesis and the benefits of statins. However, he then suffered a highly disturbing episode of memory loss, so he stopped taking the statin. He had no further problems for the next year, so his doctor persuaded him start a statin again, and he did so. Shortly after this, he suffered a much worse episode of memory loss, during which he regressed into his teenage years, unable to recall training as a doctor at all. After regaining his memory he was very shaken by the whole episode and binned the statins for good.
The doctors treating him made the diagnosis of transient global amnesia, cause unknown. They totally refused to accept the possibility that the statin could be the cause, and neither would anyone else. Feeling like a lonely voice in the wilderness, Dr Graveline then published a letter on a website called People’s Pharmacy asking if anyone else taking statins had suffered the same thing. He was immediately inundated by hundreds of cases from distraught patients and relatives. They described a full array of cognitive side effects from amnesia and severe memory loss to confusion and disorientation – all associated with statins, mostly with atorvastatin (Lipitor). The response of the mainstream medical community, however, could be paraphrased thus: ‘You don’t know what you’re talking about. Statins are safe and have very few side effects.’
Here is one letter that was written to Dr Graveline and is reproduced in his book Lipitor, Thief of Memory:
About six weeks ago, my doctor doubled my Lipitor from 20 milligrams to 40 milligrams. For about the past four weeks I have experienced progressive memory loss. I couldn’t remember my brother’s phone number. I couldn’t find my baby’s plate of food after preparing it. I couldn’t remember recent trips. I couldn’t remember to attend a meeting. I couldn’t remember a restaurant I ate in and numerous other similar episodes. This is totally out of character for me. I have called my doctor and am awaiting his return call. For your information I am 39 years old and have been on Lipitor about four years.
Will this be memory loss be ignored by the doctor? Probably. Will this be filed as an adverse event? Almost certainly not. This effect will be considered trivial. However, I think that the ‘mental’ problems associated with statins are far from trivial. As early as the 1960s it was recognised that the people taking cholesterol-lowering drugs tended to die more frequently from violent deaths: accidents, suicide, shootings and the like. This was universally dismissed as a coincidental finding (no matter how many times it cropped up), mainly because no one was able to see how a low cholesterol level could possibly be linked to violent deaths.
I read one post-hoc analysis of a cholesterol-lowering trial in which the authors were so determined to prove that the low cholesterol levels could have nothing to do with dying in a car crash that they pushed the analytical boundaries into another dimension. Their argument was that several of those who died while on statins were actually pedestrians, not drivers. So the statin couldn’t be to blame for the crash. Ha! Just try picking the logic out of that statement.
Anyway, thirty years ago, even twenty years ago, even five years ago, no one knew that cholesterol had anything to do with brain function. This despite that fact that the brain contains over 25 per cent of the total amount of cholesterol in the body, and over 2 per cent of the total weight of the brain is cholesterol (presumably it was thought to be hanging about in the brain by accident?). However, it has more recently been discovered that if you want the brain to function, this requires cholesterol.
A group of researchers, led by Dr Frank Pfrieger, was looking into the function of glial cells in the brain. It was known that these ‘support’ cells had a critical role in the function of synapses (the connections between neurons). Glial cells, it was also known, released a substance that allowed synapses to form, and function. Without this substance your brain would be almost entirely useless. And what was this fantastic, miracle substance?
Isolated neurons in the laboratory survived and grew, but showed only a few of the electrical signals generated by synapses. But when exposed to substances secreted by glial cells they produced strong signs of synaptic activity. The identity of the glial ingredient which triggered synapse formation has remained a mystery until now. But research published in the journal Science suggests that cholesterol is the magic ingredient.
Yes, the magic ingredient was good old ‘deadly’ cholesterol. Without cholesterol, the chemical scourge of mankind, your brain cannot form synapses, and you can’t think properly, or remember anything. Or remember anything.
Maybe it was a tad premature to write off cognitive side effects as a mere coincidence? Especially when it is clear how taking a statin might, just might, cause memory loss, even global amnesia. In fact, it is hard to see how it would not. It might also be possible to see how you would be more likely to die in a car crash – either as a driver or pedestrian – if your brain isn’t functioning properly. ‘Now, do I look left, or right?’
But what about the link between cholesterol lowering, violence, and suicide? Well, in addition to cholesterol’s critical function in synapse formation, it has now been found that a low cholesterol level leads to reduced serotonin levels in the brain. A low serotonin level is one of the key brain abnormalities involved in depression. This is why the most commonly used anti-depressants are designed to boost serotonin levels. They are known as Selective Serotonin Re-Uptake Inhibitors (SSRIs). Prozac is the most famous of this group of drugs.
Low serotonin has also been linked to violence and aggression. And this is far from a theoretical finding. A group of researchers led by Jian Zhang looked at the association between a low cholesterol level and a history of school suspension. They concluded that:
Among non-African-American children, low total cholesterol is associated with school suspension or expulsion and that low total cholesterol may be a risk factor for aggression or a risk marker for other biologic variables that predispose to aggression.
… the results of the current study are consistent with the majority of previous studies examining the associations between low serum cholesterol and various forms of aggression in adults. With few exceptions significant associations have been observed from cross-sectional studies, cohort samples, general population studies, psychiatric patients and criminals, and controlled dietary studies conducted in nonhuman primates. In particular, low total cholesterol has been associated with the onset of conduct disorder during childhood among male criminals.
Added to this, the Royal College of Psychiatry published a paper looking at the role of cholesterol in depression and self harm. It was entitled ‘Low cholesterol may indicate risk of suicide’. Here I take a few sections from the press release:
Lower cholesterol levels were related to higher levels of self-reported impulsivity. The finding of a lower average cholesterol in the DSH (Depression and Self Harm) group confirms other published studies.
The authors hypothesise that the increased death rates in populations with low cholesterol may be the result of increased suicide and accident rates associated with increased tendencies to impulsivity.
It may be that low cholesterol in some way influences the function of the central nervous system, or acts as a marker for factors governing a predisposition to death by trauma and suicide.
It is thought that cholesterol may influence serotonin, a neurotransmitter in the brain, low levels of which are associated not only with depression and suicide, but also with aggression and impulsivity. The latter are often involved before accidents, acts of violence to self and others and attempted or completed suicide.
And so, gentle reader, our scientific knowledge has now advanced to the point where it can no longer be written off as a ‘coincidence’ that people on cholesterol-lowering drugs are more likely to die violent deaths of one sort or another. A clear causal chain now exists, with every link in place.
As a little postscript to this section I would like to quote Frank Pfrieger again: ‘A defective cholesterol metabolism in the brain may impair its development and function.’ So it may also not be a coincidence that you can get serious neurological abnormalities in babies whose mothers were taking statins while pregnant.
How widespread are all of these problems? Who knows? If they exist they will be, almost without exception, underreported. A bit of memory loss here, a lapse into depression there – well, everyone is depressed nowadays, aren’t they? Feelings of anger and aggression, a bit of road rage, some muscle pains. Which patient is going to report such symptoms to their GP? And even if they do, how seriously is the GP going to take it? Remember that Dr Graveline suffered a full-blown episode of global amnesia and he was still hounded to re-start his statins.
What are the other problems with statins? For the sake of brevity, I shall run through them at relatively high speed.
Polyneuropathy
Polyneuropathy, also known as peripheral neuropathy, is characterised by the following:
• Facial weakness
• Difficulty in walking
• Difficulty using the arms, hands, or feet
• Sensation changes (usually of the arms and hands or legs and feet), such as pain, burning, tingling, numbness or decreased sensation
• Difficulty swallowing
• Speech impairment
• Loss of muscle function or feeling in the muscles
• Hoarseness or changing of voice
• Fatigue
Researchers who studied half a million people in Denmark found that those who took statins were significantly more likely to develop polyneuropathy. Just how significantly was highlighted by a study published by the American Academy of Neurology. The researchers found that patients treated with statins for two or more years had a 26.4-fold increase in the risk of definite idiopathic (caused by a drug) polyneuropathy. That is a 2,640 per cent increase in risk.
In general, polyneuropathy is irreversible.
Muscle damage
Although rhabdomyolysis itself is rare, muscle pains and muscle weakness are relatively common with statin use. It is very difficult to get a handle on how common this is. The mainstream view is that about one per cent of people taking statins will suffer muscle pain, or weakness.
However, I have seen much higher figures. Dr James K Liao, director of vascular medicine research at Brigham & Women’s Hospital in Boston and a big supporter of statins, believes muscle pains are much more common, occurring in 15 per cent to 20 per cent of his patients.
In this world, as with much else, seek and ye shall find. A research group in Austria decided to analyse 111 people with FH who had been put on statins, and complained of no side effects whatsoever. But the researchers wanted to know if there were signs of muscle injury anyway. They used a test that is not widely available, but is the gold standard for detecting ‘oxidative’ damage. (They measured 8-epiprostaglandin PGF2alpha, if you really want to know.) To their surprise, they found that 11 of the subjects had significant biochemical signs of oxidative damage to their muscles.
These findings indicate that in the absence of other clinically observable adverse effects, in some of the patients, for an as yet unknown reason, statin therapy may be associated with increased oxidation injury.
Is this clinically important? Who knows? Maybe not for everyone. But this same group of Austrian researchers discovered that statin-related muscle problems are much more likely to occur in those who do a lot of exercise. They looked at a group of professional athletes with FH and found that only 20 per cent of them could tolerate using statins without suffering serious muscle pain and weakness.
Apart from athletes, I am certain that the burden of muscle problems is generally underestimated because such problems tend to creep up slowly. My father in-law takes statins – don’t worry, he wouldn’t dream of listening to me. (Although maybe he will after reading this damned book…) After his statin dose was increased last year he was unable to walk more than a hundred yards without having to sit down and rest. He was persuaded to reduce the dose, and he can now walk for over a mile, easily. Of course neither he nor his doctor believe that his muscle weakness and pain was in any way related to the increased statin dose. After all, he is approaching eighty, he has had a heart attack, he was already slowing down. So what do you expect? We are all remarkably good at dismissing the symptoms of others.
Personally, I believe that a very high percentage of people on statins will suffer some symptoms of muscle ache; most of them they won’t report the symptoms, and the doctor won’t ask. Doctors are not generally very keen on discovering problems with the drugs that they prescribe.
Liver damage
This is reasonably common, although in the main not serious. Mostly it takes the form of raised liver-enzyme levels in the bloodstream. These tend to go away when you stop taking the statin. There have been cases of liver failure while on statins, but these are relatively rare.
This is probably the biggest long-term worry as far as I am concerned. To quote a study from the Journal of the American Medical Association published in 1996:
All members of the two most popular classes of lipid-lowering drugs (the fibrates and the statins) cause cancer in rodents, in some cases at levels of animal exposure close to those prescribed to humans…
Extrapolation of this evidence of carcinogenesis from rodents to humans is an uncertain process. Longer-term clinical trials and careful postmarketing surveillance during the next several decades are needed to determine whether cholesterol-lowering drugs cause cancer in humans. In the meantime, the results of experiments in animals and humans suggest that lipid-lowering drug treatment, especially with the fibrates and statins, should be avoided except in patients at high short-term risk of coronary heart disease.
The pharmaceutical companies have attempted to stomp on this fear by setting up studies that appear to show that statins, rather than causing cancer, actually prevent it. Attack, as they say, is the best form of defence. You may have seen some of the headlines, such as:
New research shows that the popular cholesterol-lowering drugs called statins may slash a person’s chance of developing breast, prostate, and lung tumors in half.
I plucked the above headline from a major medical website called WebMD. So what’s actually wrong with this ‘new research’? Let’s put it this way:
• People with low cholesterol levels are at a much greater risk of dying of cancer.
• People with low cholesterol levels don’t get put on statins (yet).
• People with high cholesterol levels are less likely to die of cancer.
• People with high cholesterol levels do get put on statins.
Therefore, if you find a group of people taking statins, and match them against a group of people not taking statins, hey presto! It is very likely you will see less cancer in those taking statins, because they were protected against cancer in the first place.
Luckily, it doesn’t actually take long to stomp such headlines into the ground. In the major statin trials, the people taking the statins and those taking the placebo are allocated at random, so on average both groups should have identical cholesterol levels. And guess what? When a meta-analysis of the statin trials was done, and published in JAMA, looking for any effect on cancer:
In our current meta-analysis, statins did not reduce the incidence of cancer or cancer death, No reductions were noted for cancers of the breast, colon, gastrointestinal tract, prostate, respiratory tract, or skin (melanoma) when statins were used… the patients in our meta-analysis were primarily treated with simvastatin and pravastatin. As such, we evaluated pravastatin alone and simvastatin alone on cancer incidence and death and found no impact.
In short, the headlines about statins preventing cancer are complete and utter guff, as usual. They reflect marketing spin, pure and simple. But they do their job, the message gets out there: statins don’t cause cancer, they actually protect against cancer. Goodness me, is there nothing these drugs can’t do?
So, while there is no evidence that statins protect against cancer, is there any evidence that they may increase the risk of cancer? There is some, but it is fragmented. A massive (and never-remarked-upon) statin trial in Japan, the J-LIT study, found a small proportion of patients who were ‘hyper-responders’ to simvastatin, i.e. their LDL levels fell dramatically when given the drug. In this group there was a significant increase in deaths from cancer.
In addition to this, in the PROSPER study – one of the few statin trials specifically carried out on older people (for whom cancer risk is increased) – there were 24 more cases of cancer in the statin arm than the placebo arm. This, by the way, more than cancelled out all cardiovascular benefits. In the CARE study there were 12 cases of breast cancer in the statin arm against one in the placebo arm.
None of this is definitive proof that statins cause cancer, by any manner of means. But to my mind, it is something that needs to be carefully monitored. After all, we already have a few disturbing facts at our disposal:
• Statins cause cancer in animals.
• There is some evidence that cancer deaths are increased in the statin trials, especially in those who are ‘hyper-responders’ to statins.
• A low cholesterol level is associated with a high risk of death from cancer.
• It can take many, many years for cancer-causing agents to reveal themselves.
I think that this is more than enough evidence to warrant a high degree of suspicion. So is anyone studying this in any meaningful way? You must be joking. It would be like asking British American Tobacco to fund a study into nicotine addiction.
Heart failure
There are actually a myriad of other statin-related side effects. But you would expect to see this in any drug that was being taken by millions of people. A couple that seem to turn up more regularly than others are acute pancreatitis and severe dizziness. However, there is one potentially very worrying side effect that I would like to draw your attention to, heart failure.
In the late 1980s, when Merck were first readying themselves to launch their two statins – lovastatin (Mevacor) followed by simvastatin (Zocor) – they applied for a patent in which their statins would be combined with a substance known as coenzyme Q10. A substance that, for the sake of brevity, I will call Q10 from now on.
Q10 is found in all cells, everywhere in the body, which is why it is sometimes called ubiquinone (because is it ubiquitous). It is found in particularly high concentrations in high-energy cells such as muscles, and especially cardiac muscle cells, where it plays a key role in the production of ATP. ATP, as you may remember from biology, is to a cell, what fuel is to a car. Conversion of ATP to ADP releases the energy that cells need to work. When ATP runs out, the cell dies. Which means that a reduction in ATP production could be a rather serious matter. Especially in heart muscle, the muscle that can never rest. Indeed, both animal and human studies have shown that reduced Q10 levels can lead to left- and right-sided ventricular dysfunction (heart failure to you and me).
So where is all this going, you may ask. Well, at this point I shall draw another strand into the discussion. You see, Q10 and cholesterol share a biosynthetic pathway. They both originate from Acetyl CoA, and if you block HMG CoA reductase you not only reduce cholesterol synthesis, you also reduce Q10 synthesis. The two pathways do not split until after this point.
If low Q10 levels can lead to heart failure, and statins block the production of Q10, then statins could cause heart failure. Thus, adding Q10 to a statin makes perfect sense, does it not? In the end, though, Merck never acted on their patents (patent 4929437, issued 29 May 1990 and US Patent 4933165, issued 12 June 1990, both entitled ‘Coenzyme Q10 with HMG-CoA Reductase Inhibitors’). Why not? Well, this is the sort of ‘commercially sensitive’ information that Merck have not published. I note, however, that:
• Combination pills are more expensive to make, and it takes longer to get approval from the authorities to launch them.
• Adding Q10 to a statin might be an admission that statins are not totally innocent, cuddly and safe – could they be if they required a built-in ‘antidote’?
• Other statins coming to the market were not going to add Q10 to their statin.
• The risk of causing heart failure in humans by lowering Q10 levels had not been definitively proven.
Having said this, Merck may have been worried enough by something they had seen to fill in two separate patents for a combination pill, a most unusual move. Their research remains unpublished, so I can only speculate on what it showed.
As a little aside, it may be pertinent to this discussion to point out that Merck also make (or should that be made) Vioxx. While they claimed that the increased risk of heart attacks came as a complete surprise to them when this finding first ‘appeared’ in late 2004, during the course of three lawsuits against Merck, it was alleged they deliberately withheld data on the cardiavascular dangers of the drug. You can read about this on the Forbes website: http://www.forbes.com/2005/12/08/merck-vioxx-study-1208markets14.html
Moving on. In addition to the possibility that statins could cause heart failure by blocking Q10 synthesis, a study published in the 3 December 2003 edition of the Journal of the American College of Cardiology reported a strong relationship between lower cholesterol levels and increased mortality in patients with heart failure. The study was conducted in the Royal Brompton Chronic Heart Failure Clinic in London. The researchers measured cholesterol levels in over 400 patients with varying degrees of heart failure and followed their outcomes. They found that people with cholesterol levels below 5.0mmol/l have a two- to three-fold increased risk of dying compared to those whose cholesterol levels were higher. In fact, it has been noted in many studies that heart-failure patients seem to have ‘reverse epidemiology’. By this, I mean that having a high cholesterol level and being obese and having a high blood pressure are associated with prolonged survival.
Bringing all of the strands together, it seems that there are two interrelated mechanisms by which statins could cause heart failure. Firstly, by blocking Q10, thus lowering the heart-muscle energy production; secondly, by lowering cholesterol levels. The ‘Coenzyme Q10 Association’ (yes, there is one) were worried enough about this to have written to the FDA expressing their concern. I believe that they are awaiting a reply – I would advise them not to hold their breath.
There are also cardiologists out there who believe that the current sharp rise in heart failure across the western world is a direct result of the use of statins. Theoretically, at least, they have a strong point. However, I don’t believe that the data yet exists to prove the case one way or the other. Once again, though, there is enough evidence out there to warrant close attention and monitoring.
A final thought:
• Several studies have shown that a low cholesterol level is a major risk factor for people with heart failure.
• Statins significantly reduce the risk of death once you have heart failure.
So, in this case, statins must be working through non cholesterol-lowering effects.
And finally – the damage that statins can do
Leaving aside physical side effects and the eye-watering cost, the ever-increasing use of statins will have, I believe, a major negative impact on society as a whole.
To give just one example of the potential scale of the impact, a couple of years ago a group of researchers in Norway looked at the latest guidelines on CHD prevention published by the European Society of Cardiology. They wanted to see what the full implementation of these guidelines would mean for the Norwegian population – a population which, as the authors point out – has an average life expectancy of 79 – making it one of the longest lived in the history of mankind.
The European guidelines defined a blood pressure of 140/90, and a cholesterol level of 5.0mmo/l, as thresholds for intervention. (Intervention means, in 99 per cent of cases, drug treatment.) Using these two levels, the Norwegian researchers then drew a graph showing what percentage of the population would be in the ‘at risk’ category for high cholesterol and/or a raised blood pressure (see below):
Fig. 28 Drug treatment levels per age group
As you can see, by the age of 25, 50 per cent of the population is ‘at risk’ from their high cholesterol levels. By the age of 50 this has gone up to 90 per cent. And if you in add raised blood pressure, there is hardly anyone left who does not require ‘intervention’.
What’s implied here is that the majority of Norwegians are so unhealthy that they need drug treatment. And, although no one has done it yet, if you applied these latest European guidelines to the UK, you would find an even higher percentage requiring life-long drug treatment.
In my view, this is about as far from the concept of health as it is possible to get. We are in grave danger of converting the vast majority of adults in the country from ‘healthy’, to ‘diseased’, and worried to boot. And the only solution for your ‘disease’? Take a statin for the rest of your life, and never stop.
Can it really be true that 90 per cent of the population need life-long medication? This is bonkers. It is Brave New World, it is a combination of all dystopian nightmares of the future come to life. Health, it would seem, is no longer an absence of disease, but an absence of taking the correct medication.
First, do no harm? I don’t think so.
Statins don’t work by lowering LDL levels
At this point, grudgingly, I will admit that statins do reduce the risk of dying of heart disease in certain populations. Statins definitely reduce overall mortality in men with existing heart disease. So, if you are a man with known heart disease, it may be a good idea to take a statin.(There, I said it – and it did hurt, thanks for asking.) However, before looking at how they actually do this, I think you ought to know in a bit more detail what the figures are.
Here I include a chart copied from Uffe Ravnskov, who wrote the book The Cholesterol Myths, and with whom I communicate regularly. He looked at overall mortality rates from the six biggest statin trials at the time and brought together four different figures related to overall mortality.
• The relative risk reduction
• The absolute risk reduction
• The chances of being alive at the end of the trial on placebo
• The chances of being alive at the end of the trial on a statin
| Relative risk | Absolute risk | Chance of surviving on placebo | Chance of surviving on statin | |
| 4S | −29% | −3.3% | 88.5% | 91.8% |
| WOSCOPS | −21% | −0.9% | 95.9% | 96.8% |
| CARE | −8% | −0.78% | 90.57% | 91.35% |
| AFCAPS/TexCAPS | +3.9% | +0.09% | 97.7% | 97.6% |
| LIPID | −21% | −3.1% | 85.9% | 89.0% |
| EXCEL | +150% | +0.3% | 99.8% | 99.5% |
These figures represent a mixture of primary and secondary prevention. The reason for using them is to give you an idea of the general scale of risk reductions found in the statin trials. Let’s concentrate on three of them in a little more detail: 4S, WOSCOPS and EXCEL.
The 4S trial remains the most positive of all the statin trials. In this study, there was a 3.3 per cent absolute risk reduction in total mortality over five years, which equates to a 0.66 per cent reduction in overall mortality each year. This was purely a secondary prevention trial.
WOCSOPS was a mixed primary/secondary prevention study. However, average cholesterol levels in this trial were 7.0mmol/l, which puts everyone in the study into a ‘high risk’ category. In WOSCOPS there was a 0.9 per cent reduction in overall mortality over five years – which represents a 0.18 per cent reduction in overall mortality per year.
Just to present the other side of the story, in the EXCEL trial, which was entirely a primary prevention study, there was a 0.3 per cent increase in overall mortality over five years. Once again, perhaps these figures are not quite as spectacular as you would imagine, given the hype that surrounds statins.
Before moving on from the statistics, I would like to tackle the use of the term ‘saving lives’. You may remember the press release that accompanied the HPS study:
If now, as a result, an extra 10 million high-risk people were to go onto statin treatment, this would save about 50,000 lives a year – that’s a thousand each week.
Leaving aside the point that this 50,000 figure actually equates to one life ‘saved’ for every 200 people taking the statin – 10 million is an awful lot of people to use as your denominator – the concept of saving lives, suggesting, as it does, that each of the 50,000 whose lives have been saved will go on to live out a full and healthy life, is not best chosen.
In reality, taking a statin can only delay death, not prevent it. By how much? Well, if one in two hundred more people are alive after one year of taking statins, this means that if you wait another two-hundredths of a year (plus another little bit) the statin group will have caught up on the ‘placebo’ group in total number of deaths. This represents an increased life expectancy of slightly under two days.
So, rather than stating that fifty thousand lives would be saved every year by taking statins, it would be considerably more accurate to state that if ten million people (at very high risk of heart disease) took a statin for a year they would all live – on average – two days longer. And if all ten million took a statin for two hundred years, they would all live – on average – an extra year.
If we assume that most people would take a statin for thirty years, maximum, this would lead to an average increase in lifespan of approximately two months. Which doesn’t sound quite as dramatic as saving fifty thousand lives a year, or a thousand a week – or however else you choose to hype up your figures. But there you go, it happens to be considerably more accurate.
Also remember that this benefit would only be seen in men with pre-existing heart disease. Women and men without pre-existing heart disease would live not a day longer. They would just have the dubious pleasure of thirty years of paying for drugs, worry and side effects.
Perhaps you think that I am also manipulating statistics in a way to make my point. Maybe. But an increase in average survival time is how all results are presented in cancer trials. This is one reason why, I think, cancer trials tend to look rather unimpressive when stacked up against the highly dramatic ‘life-saving’ cardiovascular studies*.
Anyway, moving back to the main point of this section. In secondary prevention studies (in men) it seems that statins do lower cardiovascular mortality, and by enough to wipe out any increase in other causes of death. But do they do it by lowering LDL levels, or do they do it through other mechanisms?
This is not, actually, an easy question to answer with absolute certainty. Maybe by now you feel it doesn’t much matter any more – given that you have just been made aware of the minute benefits that statins offer, even in the highest risk groups. However, I think it is important to look at this issue for two reasons:
• If I can prove that statins do not work by lowering LDL levels, then the entire ‘cholesterol hypothesis’ totally disintegrates.
• Currently, the combined might of the pharmaceutical industry plus opinion leaders are pressing hard for ever-greater LDL lowering and they are deliberately blurring the distinction between primary and secondary prevention. I think that this must be resisted, as it will lead to more and more people being put on very high doses of very potent statins, which would be a complete disaster.
But where to start? Where indeed.
When the first statin trials came out, it’s true that they did seem to provide definitive proof that statins worked by lowering LDL levels. After all, they did exactly what they were supposed to: lower LDL and protect against CHD.
But it didn’t take long before whole series of anomalies emerged which suggested that statins might be working through other mechanisms rather than by LDL lowering. The non-lipid actions are now bundled under the heading ‘pleiotropic’ effects (see Glossary for full explanation of this term).
And so a debate has emerged. A rather one-sided debate, given the scale of the respective budgets supporting either faction. But it is a debate nonetheless. Do statins protect by lowering LDL, or in other ways?
In the ‘statins don’t work by lowering LDL levels’ corner are the following facts. This is only a small selection:
• Statins protect against dying of heart failure, despite the fact that a high LDL level is associated with increased survival in this condition.
• The beneficial effects of statins have been seen within weeks, days, even hours of taking a statin. And this finding does not remotely fit with the hypothesis that raised LDL creates gradual plaque build-up over years and years9.
• Statins protect against strokes, but a raised cholesterol level is not a risk factor for stroke.
• Statins provide the same degree of protection no matter if the LDL level is high, average, or low (HPS study).
• Some studies, such as CARE, showed that the greatest degrees of LDL lowering were associated with a rise in deaths from heart disease.
While some of us may say, ‘Who cares? If they work, they work,’ if you have spent a lifetime, and built your glittering reputation, ‘proving’ that raised LDL causes heart disease, you are not going to allow facts like these to remain standing.
Equally, if you are trying to sell a statin-based drug entirely on the awesome power of its LDL-lowering properties, you are not going to take kindly to an alternative explanation jamming up the works. And so, a series of studies were set up designed to prove, for good and all, that the more you lower LDL, the greater the protection against heart disease.
The major studies were PROVE-IT, TNT, A to Z, REVERSAL, ASTEROID and IDEAL. No, there will not be a test at the end of this chapter to remember which was which, or what was what. Or which proved what, why or how. After all, even the acronyms themselves are impossible to remember. And a bit bonkers too – witness ASTEROID (A Study To Evaluate the effect of Rosuvastatin On Intravascular Ultrasound-Derived Coronary Atheroma Burden). Wheee!
Let’s look firstly at the IDEAL study – the biggest. After it ended, Scott Grundy – who, it must be said, is a statinoholic – wrote an article entitled ‘The clinical implications of IDEAL: in the context of Recent Intensive Statin Therapy Trials’ for Medscape. I quote:
We now have very clear evidence that patients with established CAD [heart disease] will benefit from intensive LDL-Cholesterol [LDL-C]- lowering therapy.
In short, case closed: statins work by lowering LDL levels, now will you all please shut up? This, I have to say, I find an extremely interesting interpretation, given the results themselves.
The IDEAL study – it should be pointed out – was very big. Twice the size of 4S, with 8,888 patients, no less. Patients were either put on 20mg or 40mg of simvastatin (the doses used in 4S), or 80mg of atorvastatin (Lipitor). LDL was lowered by 33 per cent in the simvastatin group and 42 per cent in the atorvastatin group. The result: ‘No statistically significant differences were seen in all-cause mortality, cardiovascular mortality, or non-cardiovascular mortality.’ (In fact, cardiovascular mortality was very slightly higher in the atorvastatin group.) Despite this, according to the study’s author, we now have very clear evidence that patients with established heart disease will benefit from more intensive LDL/cholesterol lowering… This is an interpretation that flatly contradicts the results of the study itself. Although, to be perfectly frank, I see a trend of analysis of trials that bears no relation to my interpretation of these trials.
Maybe he misunderstood the key findings that ‘No statistically significant differences were seen in all-cause mortality, cardiovascular mortality, or non-cardiovascular mortality’? Maybe this wasn’t quite clear cut enough for him? Perhaps, he just missed the word ‘No’ at the start of the sentence? Easy to do that, I find.
And what of ASTEROID? You may remember that this study was hyped to the very skies, appearing as a major news item on the BBC, no less, under the headlines ‘Drug can reverse heart disease’ – with pretty pictures of disappearing plaques, just to make the point.
Not everyone was that impressed. Dr Graham Jackson, a UK cardiologist, wrote an editorial on the matter in the International Journal of Clinical Practice. I quote a selection of his views on ASTEROID:
As a marketing exercise it was brilliant. As an educational exercise it exploited sensationalism. And as a scientific exercise it was another own goal for the pharmaceutical industry (AstraZeneca provided a press release which went largely uncriticised and could be seen as part of a direct-to-the-public advertising campaign)… So what was so stunning about ASTEROID? Nothing really…
The study was headed by Dr Steven Nissen. Of the ASTEROID trial itself, Dr Nissen stated that he had never seen regression of atherosclerotic plaques in his entire career before this. This sounds similar to his reaction to the ApoA-1 Milano trial two years previously: ‘We really didn’t think it was going to work,’ Nissen told WebMD. ‘Nobody was more shocked than I was when the statisticians handed me the data… It is unprecedented. Nobody has even seen this kind of plaque regression. It really is an epiphany.’ In relation to another intensive LDL-lowering trial that he was in charge of, the REVERSAL trial:
‘The results were striking’, Dr Nissen said, ‘demonstrating a complete halting of coronary disease progression in the atorvastatin-treated patients and continued progression of disease in the pravastatintreated group.’
Dr Steven Nissen, it should be mentioned, was also the lead investigator in the most positive study on intensive LDL lowering that measured clinical end-points, rather than measuring plaque size with an ultrasound probe. It was called Treating to New Targets (TNT).
They say that lightning never strikes twice in the same place. Yet there have only been four major studies on intensive LDL lowering that have been positive, and Steve Nissen has been in charge of three of them. In fact, if you removed the Steven Nissen-controlled trials, the evidence on intensive LDL lowering would be almost entirely negative.
The only major positive study not run by Steven Nissen was PROVE-IT. Ironically, this study was set up by Bristol-Myers Squibb to prove that intensive LDL lowering had no added benefit to moderate LDL lowering. It seemed the ultimate own goal, as it appeared to end up proving the opposite. But did it?
Firstly, the bare bones of the trial itself. In PROVE-IT, the investigators took 4,162 patients who had been in hospital following an MI, or unstable angina (almost, but not quite an MI). They then split the group in half and gave one half pravastatin (made by Bristol-Myers Squibb), and the other half a much higher dose of atorvastatin (made by Pfizer). As expected, LDL level, or ‘bad cholesterol’ level, was reduced to a greater extent in the atorvastatin group.
• LDL in the treated pravastatin group: average 95 mg/dl (range 79–113)
• LDL in the treated atorvastatin group: average 62 mg/dl (range 50–79)
In short, in the atorvastatin group there was a 32 per cent greater reduction in LDL levels, and there was also a 16 per cent greater reduction in – well – almost everything you can think of: all-cause mortality, MI, unstable angina, hospital readmission, interventional procedures – you name it. It was all quite wonderful. In absolute terms, a somewhat less wonderful 0.5 per cent reduction in overall mortality per year.
My interpretation? Actually, I don’t have one. You see, the researchers gave two different groups of patients two different doses of two different drugs. They then decided that all benefits seen were due to greater LDL lowering. How can they draw this conclusion? The answer is that they cannot. It’s impossible to do so, unless you know that the two drugs are absolutely identical in all of their actions, other than the impact they have on LDL lowering.
If, for example, atorvastatin has non-LDL lowering benefits on CHD that pravastatin does not, this would be the reason for the difference. But how can anyone know. This has never been tested.
Ironically, the possibility that atorvastatin is better at preventing CHD than pravastatin, no matter what the LDL level, is supported by Dr Steven Nissen. In the REVERSAL study, which also used low dose pravastatin and high dose atorvastatin, he found the following:
Surprisingly, despite attaining a low LDL level on pravastatin, these patients showed highly significant progression for percent atheroma volume and percent obstructive volume… At any LDL level, progression was less on atorvastatin than on pravastatin. When I started this study, I believed that any reduction in progression would just be due to lower LDL levels, but now I’m not so sure. This analysis suggests that it may be more than just LDL it seems to be the drug as well… Yes, this is a post-hoc analysis and should be considered hypothesis generating, but I would say it is a robust finding.
Goodness me, of all people.
This finding actually highlights the hopeless weakness at the heart of all the intensive LDL-lowering trials. They have almost all used different doses of different drugs. This is not a scientific technique that I would recommend if you ever want to actually prove anything, ever.
Even if they had used different doses of the same drug, you would not be able to say that it is the LDL lowering that created any benefit. It could have been another dose-dependent ‘direct’ drug effect that you haven’t even measured, or don’t even know exists.
Having said this, there has been one, and only one, major study done in which different doses of the same statin were used – the A to Z trial, using simvastatin. Guess what? Despite major differences in the LDL levels attained, there were no benefits seen from taking the higher dose of simvastatin on cardiovascular – or overall – mortality. This was despite the fact that the ‘low-dose’ group weren’t actually given a statin for the first four months of the study. Bonkers.
Frankly, the intensive LDL-lowering studies have not actually proved anything at all – except, perhaps, that simvastatin and atorvastatin are superior in reducing the risk of CHD than pravastatin for – as yet – unknown reasons. And anyone who argues differently needs to be given a copy of How to do Scientific Studies – for Five-Year-Olds.
Rule one: If you have more than one uncontrolled variable in your study you can’t prove anything.
Rule two: If you think you have proved something in a study with more than one uncontrolled variable, Rule one shall apply.
Summary
At this point I shall attempt to draw all the strands on the use of statins together. First the positive data:
• If you are a man with pre-existing heart disease, statins reduce your risk of dying of anything by a maximum of 0.66 per cent per year. (This figure is based on the most positive data from the most positive study – 4S. Study run by Merck, primary data analysis carried out by Merck employee.)
• If you are a man without pre-existing heart disease, statins can reduce your risk of dying of cardiovascular disease – by a small amount.
• If you are a woman at very high risk of heart disease, statins reduce the risk of dying of cardiovascular disease (that is, strokes and heart disease).
• If you are a woman, no matter what your level of risk, statins will not increase your life expectancy by one day. Deaths from cardiovascular disease reduced; deaths from other causes increased.
• If you are a man without heart disease, statins will not increase your life expectancy by one day.
Then the negative data:
• Statins, cholesterol tests and GP appointments and screening are costing the NHS alone billions of pounds a year.
• Statins cause muscle pains and muscle weakness in up to 20 per cent of people who take them.
• Statins cause rhabdomyolysis, which can be fatal.
• One type of statin, simvastatin, over a period of six years, caused 416 deaths in the USA alone.
• Statins cause polyneuropathy.
• Statins cause memory loss, depression, confusion, irritability and dizziness.
• Stains cause major birth defects.
Finally, a couple of worrying, though unproved, possibilities:
• Statins may increase cancer risk.
• Statins may cause heart failure.
It is also, as yet, not remotely proven that statins protect against heart disease by lowering LDL levels. The current hyping of the intensive LDL-lowering trials has been driven purely by the pharmaceutical industry. They claim to have proved beyond doubt that the more the LDL is lowered, the greater the protection against heart disease, and they have tried to use this ‘fact’ to press for ever-greater cholesterol lowering in the entire population.
*Clinical trials are usually ‘blinded’. That is, neither the doctor nor the patient knows if they are getting the drug. However, some researchers have access to certain blinded data, to check up on whether or not too many deaths are occurring in the different patient groups for the trial to continue.
*Actually, it is an increase in ‘median’ survival that is used. However, while the difference between average and median can be technically important, in most cases the two things are virtually the same. The reason why ‘median’ increase in survival is not used in cardiology is that it would take a hell of a long time for 50 per cent of the people to die. Not unfortunately usually a problem in cancer trials.