Previously, I have written about the attempts by Professor Sir Rory Collins to have two articles in the BMJ withdrawn. Both of them cast considerable doubt on the effectiveness of statins, particularly for those at relatively low risk of cardiovascular disease. However, both pieces also misquoted a piece of research finding that statins were associated with adverse effects in 17.4 per cent of those who take them. Both articles rounded this figure up to 18-20 per cent, and also implied that statins caused the adverse effects seen (the study – ‘observational’ or ‘epidemiological’ – type did not allow this inference to be made).

Once alerted to the issue, the BMJ promptly withdrew the comments about adverse effects. But Professor Collins was not happy, and insisted that the BMJ retract both articles in their entirety. However, to my knowledge, neither Professor Collins not anyone else had countered the main thrust of the articles (one was principally about the ineffectiveness of statins, and the other was the lack of association between saturated fat and heart disease).

The editor-in-chief of the BMJ – Dr Fiona Godlee – convened a committee to look into the matter, and to rule whether there were grounds for the one or both of the articles to retracted.

Elsewhere in the BMJ, one of Professor Collins’ previous collaborators – Professor Peter Sever – referred to Professor Collins’ detailed case for retraction. I emailed Professor Collins asking if I could see his detailed arguments. I was particularly interested to read what Professor Collins’ had to say about the claims that statins are, on the whole, quite ineffective. To his credit, Professor Collins replied, but only to tell me I would need to wait until the committee issued their report and all the documentation.

Well, last week, the committee published their report. It found no grounds whatsoever for retraction of either of the articles. You can read the report here, and here is a list of all the relevant documentation.

I read four (SP17-20) letters Professor Collins submitted to Fiona Godlee to make his case. Each letter, I think, has a similar theme: the claims about statins side effects were inaccurate and overblown, and as a result people may be discouraged from taking them or continuing them (exposing them to increased risk of cardiovascular disease). However, these issues had been acknowledged by the BMJ and the authors of the articles, and the statements had already been withdrawn.

However, on the subject of the effectiveness (actually, ineffectiveness) of statins, Professor Collins has nothing much to say. My guess is that Professor Collins would prefer not to engage with this critically important area, because erhaps he knows statins are extremely limited in their effectiveness in the vast majority of people who take them. Better, I suggest, to bang on and on about some other topic that maybe provides a convenient diversion.

I don’t know, but I have a sense that Professor Collins’ crusade to have the papers retracted was as much about censoring the truth about statin effectiveness as writing any wrong to do with adverse effects. And I am delighted to say he did not get away with it.

In fact, I suspect rather that Professor Collins’ ‘shock and awe’ tactics have spectacularly backfired. I think he has come across as bullying and aggressive (and wrong). But there is worse in store for him: The committee ad the editor of the BMJ have drawn our attention to the fact that the data held my Professor Collins and his colleagues at the CTT is for their eyes only. No-one else can have access to it nor is in a position to verify their conclusions about statins. In the interests of transparency and scientific integrity, they are calling for this data to be made public.

My personal opinion is that Professor Collins and some of his colleagues have done considerable damage to the reputation of medical and research fraternities. I’d say his actions will have contributed to the growing mistrust members of the public appear to have for ‘experts’. Being eminently credentialed does not guarantee that someone will always seem keen to uphold the basic tenets of good medicine and good research.

This week saw the publication of a study that aimed to assess the impact of NHS health checks [1]. The researchers compared in outcomes general practices that undertook screening, with those in patients from practices where health checks were not routine. The researchers focused on five conditions: high blood pressure, heart disease, kidney disease atrial fibrillation (a type of heart rhythm disturbance) and diabetes.

This 3-year study found that the rates of these conditions was no higher in screened patients than in those who received normal care. In other words, health checks did not pick up an additional health issues that would not have been picked up in the normal course of events. By the way, in this study the total number of screens performed was 16,669. Think of all the time, effort and resources that went into these screening endeavours, only for them to prove to be a complete dud.

More important than this finding, I think, is the impact of health checks not on the apparent prevalence of a condition, but on actual health. Do health checks translate into a reduced risk of disease or death?

The impact of health checks on health outcomes was assessed by researchers from the so-called Cochrane Collaboration (specialising in performing ‘meta-analyses’ of health interventions) [2]. The review pooled together the results of 14 studies where the health outcomes of people who underwent a health check were compared with those of people who didn’t.

Here are the results:

Risk of death from cancer – no benefit

Risk of death from cardiovascular disease – no benefit

Overall risk of death – no benefit

The authors concluded that:

General health checks did not reduce morbidity [illness] or mortality, neither overall nor for cardiovascular or cancer causes, although they increased the number of new diagnoses. Important harmful outcomes were often not studied or reported.

Yet, despite this, the NHS Choices website continues to promote health checks and makes quite grand claims about the ‘expected’ benefits in terms of lives saved and cases of disease prevented. The NHS Choices does not provide the evidence that backs up these claims (which, remember, run counter to the evidence presented here).

Should you get an invite for a health check, it’s perfectly legitimate to attend. It’s also perfectly legitimate, though, to ask your doctor for the evidence that supports this initiative. It’s high time we doctors were made more accountable for the recommendations we make and services we offer, including those sanctioned by our Government.

References:

1. Caley M, et al. The impact of NHS Health Checks on the prevalence of disease in general practices: a controlled study. BJGP 2014;64:625 e516-e521

2. Krogsbøll LT, et al. General health checks in adults for reducing morbidity and mortality from disease: Cochrane systematic review and meta-analysis. BMJ 2012;345:e7191.3

There are three main classes of agents that are known to boost HDL levels: the B-vitamin niacin (vitamin B3), fibrates, and inhibitors of what is known as ‘cholesteryl ester transfer protein’ (CETP-inhibitors). Recently, a review of the impact of these agents on health was published in the British Medical Journal [1].

The review assessed data from studies which had compared the impact of these agents on their own, or when added onto treatment with statins.

When added to statins, none of the agents brought any benefits in terms of endpoints such as non-fatal heart attack, fatal heart attack or overall risk of death (overall mortality).

When not taken in conjunction with statins, however, niacin and fibrates were found to reduce the risk of non-fatal heart attack (but there were no mortality benefits). CETP inhibitors did no good in any setting, and one CETP (torcetrapib) actually increased the risk of death (it has been withdrawn from sale).

An accompanying editorial [2] from the head of cardiology at Sydney University suggests that we should not give up on HDL modification just yet. The editorial suggests that there is some evidence that HDL may be genuinely protective for cardiovascular disease through mechanisms that include the removal of cholesterol from cells involved in the early stages of atherosclerosis (foam cells), immune modulation, and amelioration of diabetes.

However, even if HDL is genuinely protective for cardiovascular disease, that does not mean that something that raises HDL levels will automatically be beneficial for health: if arsenic and cyanide were found to boost HDL levels, that would not be an argument for taking arsenic and cyanide each day.

While the author of the editorial may not want to give up on HDL as a potential marker for cardiovascular disease and a target for therapy, he does concede that: “…it probably is time to abandon our assumption that fibrates, niacin, or CETP inhibitors will improve clinical outcomes in contemporary populations taking statins simply because they have favourable effects on surrogate lipoprotein biomarkers.”

This is a key point: we cannot judge health effects of pharmaceutical agents on their effect on so-called surrogate markers such a cholesterol levels. The only important thing is their effect on overall health. What a shame, then, that we still have regulatory agencies like the Food and Drug Administration in the US and the National Institute for Health and Care Excellence in the UK recommending the use of drugs based solely on their impact on surrogate markers (the drug ezetimibe is a case in point).

It is lamentable that those whose job it is to provide reliable guidance on medical treatments are so stupid and/or corrupt as to recommend doctors prescribe treatments which have no proven benefits and may in fact do considerable harm.

References:

1. Keene D, et al. Effect on cardiovascular risk of high density lipoprotein targeted drug treatments niacin, fibrates, and CETP inhibitors: meta-analysis of randomised controlled trials including 117 411 patients. 2014;349:g4379

2. Kritharides L, et al. Not so “good” cholesterol. BMJ 2014;349:g4664 (Published 18 July 2014)

I was interested to see a recently published study which links shift work with an increased risk of type 2 diabetes, and this association was particularly strong in men [1]. This study cannot be used to conclude that shift work somehow causes diabetes (only that they are associated). The same is true of studies which have linked short-sleep with an increased risk of diabetes [2].

However, there is some evidence that short sleep has the potential to impair blood sugar control (glycaemic control).

In one study, nine healthy adults were tested to see what effect sleep deprivation has on insulin resistance [3]. On one night, individuals were allowed to sleep for up to 8.5 hours (23.00 – 7.30 hrs). On another night, sleep was only permitted for four hours (01.00 – 05.00 hrs). The actual average sleep times were 7 hours 34 minutes and 3 hours 46 minutes respectively.

The results showed that in the sleep-deprived state there was evidence of insulin resistance. ‘Endogenous’ sugar production (internal production of sugar, say, from the liver) was higher, and clearance of glucose (say, into muscle cells) was lower in the sleep-deprived state.

Now, less than four hours sleep is not much sleep. But then again, increased insulin resistance was seen in individuals after just one night of sleep deprivation. It’s possible that less extreme sleep deprivation or sleep disruption (such as shift work) over longer periods of time also poses hazards regarding diabetes risk.

In one study, the impact of ‘shift work’ on disease markers, including those for diabetes [4]. 26 healthy individuals were allowed to sleep for up to 10 hours a night for 10 nights. Then, they were restricted to no more than 5 hours sleep for 8 nights. In this setting, some individuals took their sleep at night. Others, however, simulated ‘shift work’ by taking their sleep in the day on four of the eight days.

Short sleep led to a reduction in insulin sensitivity (as has been noted in previous studies), as well as an increase in markers of inflammation. Also, in men, reductions in insulin sensitivity and increases in inflammation were about twice a bad when they were sleeping in the ‘shift’ setting, compared to those sleeping normally (at night).

The evidence as it stands supports the idea that both duration and pattern of sleep can have a bearing on risk of type 2 diabetes, and other chronic health issues too.

References:

1. Gan Y, et al. Shift work and diabetes mellitus: a meta-analysis of observational studies Occup Environ Med doi:10.1136/oemed-2014-102150

2. Chaput JP, et al. Sleep duration as a risk factor for the development of type 2 diabetes or impaired glucose tolerance: Analyses of the Quebec Family Study. Sleep Med. 2009;10(8):919-24

3. Donga E, et al. A single night of partial sleep deprivation induces insulin resistance in multiple metabolic pathways in healthy subjects. J Clin Endocrinol Metab. 2010;95(6):2963-8

4. Leproult R, et al.^, Circadian misalignment augments markers of insulin resistance and inflammation, independently of sleep loss. Diabetes. 2014;63(6):1860-9

One way to gauge the extent of ‘heart disease’ is to measure the thickness of the walls of coronary artery – known as the ‘coronary intima-media thickness’ (CIMT), assessed via ultrasound performed via a catheter inserted into the vessel via the groin or arm. In a recent study, individuals who had had surgical repair for coarctation of the aorta were treated with a high-dose statin to see what effect it had on CIMT as well as health outcomes [1].

There were 155 individuals in the study, about half of which were treated with 80 mg of atorvastatin (Lipitor) each day, while the other half got no treatment. The trial lasted for 3 years.

The atorvastatin led to a significant reduction in both total cholesterol levels as well as supposedly harmful ‘LDL’ cholesterol. However, those taking atorvastatin saw no reduction in their CIMT measurements compared to those who took nothing.

The researchers also assessed the impact of high-dose atorvastatin on more important measures such as risk of heart attack and stroke and death from these things. There were no benefits to be had here, either.

In this study with this group of patients, long-term reduction in cholesterol did not translate into any benefits at all. This challenges the concept that cholesterol reduction is fundamentally healthy. However, we have much more evidence that challenges the ‘lower cholesterol is better’ paradigm, including the fact that several drugs that successfully reduce cholesterol have been found to have no benefits whatsoever and some of which actually harm health. Dietary modification of cholesterol has been found to be ineffective too.

Should we be too surprised then, that the ‘benefits’ of cholesterol reduction do not always turned out ‘as expected’

References:

1. Luijendijk P^, et al. Beneficial effect of high dose statins on the vascular wall in patients with repaired aortic coarctation. Int J Cardiol. 2014 Jun 28 [Epub ahead of print]

However, blood sugar control is a spectrum: between what are thought to be ‘healthy’ blood sugar levels and type 2 diabetes exist states which are not extreme enough to be called diabetes, but are associated with enhanced risk of developing this condition.

For example, if the glucose concentration 2 hours after the glucose load is 7.8-11.1 mmol/L (140-200 mg/dL), then a diagnosis of ‘impaired glucose tolerance’ is made. The term ‘Impaired fasting glucose’ is also used to define individuals with fasting blood glucose levels of 6.1-6.9 mmol/L (110-125 mg/dL) (according to the World Health Organization).

Another relatively newly-coined term is ‘pre-diabetes’. This diagnosis is based on a measurement known as the HbA1c which gives an indication of the overall blood sugar (glyaemic) control over the preceding 3 months or so. One of the supposed advangates of this test is that it does not require fasting nor an oral glucose load. According to the American Diabetes Association, a HbA1c of 5.7%-6.4% indicates ‘pre-diabetes’, though some bodies have recommended a slightly higher threshold for this condition. The basic thinking here is that by identifying pre-diabetes, individuals may be helped with interventions including drug therapy.

While this all may seem eminently sensible on the surface, there are a number of problems with this approach. These are explored in quite some detail in a paper published this week in the British Medical Journal co-authored by Professor John Yudkin from University College London and Professor Victor Montori from the Mayo Clinic in the US [1].

The article explains that the relationship between glycaemic control and health risks are a continuum, and where we begin to diagnose pre-diabetes is utterly arbitrary. If we apply the criterion recommended by the ADA, then apparently half the Chinese population (that’s almost half a billion people) have pre-diabetes. The authors question whether we even have the resources to treat vast swathes of the population according to these arbitrary criteria.

Of course a stronger case might be made for those arguing for population treatment if this had been tested. But as the authors point out, there simply is no evidence that earlier intervention based on the ADA’s criteria leads to either improved health or a reduced risk of death.

What we do know if that certain interventions appear to be able to delay the onset of diabetes by about 2-4 years. The real effects in terms of reducing healthcare costs or disease burden, though, are unknown.

And even if genuine benefits exist, let’s not forget that applying a diagnosis of ‘pre-diabetes’ to someone is not without risk. We have risks associated with medication, say, but also there may be issues with self-image, health and life insurance (particularly in the US).

One of the first-line drugs used to treat pre-diabetes is metformin. The authors point out that data suggest that using this drug may reduce the risk of developing type 2 diabetes by about 31 per cent. However, as the authors point out, what this means is that treating pre-diabetes in this way means individuals now have a “100% chance of using metformin with the goal of reducing by 31 per cent their risk of developing a condition that might require them to use metformin.”

This approach with metformin or some other drug may have some sections of the drug industry rubbing their hands with glee, I suppose, but it’s certainly true that overall benefit is not assured, and there is even considerable potential for harm.

The authors are sceptical regarding any supposed benefits, and also warn that we risk deflecting from the real issues with diabetes and pre-diabetes relating, for instance, to diet and exercise. The article ends with three points that doctors should discuss with patients. These are:

• A diagnosis of pre-diabetes does not mean that you will develop diabetes. In fact, of 100 people like you, fewer than 50 are likely to develop diabetes in the next 10 years

• There are ways of reducing your risk of developing diabetes that involve changing your diet and being active. These can result from efforts you make as well as changes in your environment (food supply, workplace conditions, education, and other social determinants of health)

• There are drugs to delay diabetes, but these are the same drugs you will need if you do develop diabetes, and the value of starting them before you have developed diabetes is unknown

The footnotes to the article state that: “The authors are clinical academics with a shared interest in patient centred diabetes care and shared decision making.” And this comes loud and clear in their article, I think. Their focus really does seem to be on taking an objective look at the data and giving patients the very best information on which a truly informed decision can be made.

The article also states that it is: “…part of a series on overdiagnosis looking at the risks and harms to patients of expanding definitions of disease and increasing use of new diagnostic technologies. The underlying message here is that more medicine is not necessarily better (and may in fact be a bad thing). We need more articles like this, and we need more high profile medical journals willing to publish them.

References:

1. Yudkin JS, et al. The epidemic of pre-diabetes: the medicine and the politics. BMJ 2014;349:g4485

Another quite-common approach to reducing risk of heart attack or stroke is to take a statin. Actually, statins don’t work particularly well for this, and another potential problem is that these drugs deplete levels of the substance coenzyme Q10 (CoQ10) in the body. This is important because CoQ10 performs a vital function in the production of energy in the mitochondria. Low CoQ10 levels could, therefore, stifle energy production and may, theoretically at least, negate some of the benefits one might expect from physical exercise.

Last week, the journal Sports Medicine published a review of the potential impact statins have on effects of regular physical exercise [1]. The authors refer to more than one study in which statins appeared to have no ill-effect on fitness levels (as adjudged by maximum VO2). However, as the authors point out, these studies have generally been done in healthy individuals. This, and other factors such as the short duration of the studies and relatively small sample sizes may mean they do not necessarily tell the whole story.

They refer specifically to one study that provides good evidence, I think, that statins can put a serious brake on the fitness benefits exercise usually offers [2]. In this research, 37 adults underwent 12 weeks of aerobic training. Half of the group just participated in the exercise programme, while the other half took 40 mg of the statin simvastatin (Zocor) each day. Fitness was assessed prior to the study starting, and at the end too.

In those not taking statins, fitness increased by an average of 10 per cent of the 12-week study. However, in those taking simvastatin, fitness increased by only 1.5 per cent.

As part of this study, mitochondrial activity was assessed by measuring activity of the enzyme citrate synthase (a key enzyme involved in energy production in the mitochondria). Citrate synthase activity increased by 13 per cent in those who adopted exercise, but actually fell by 4.5 per cent in those taking the drug.

In the recent review [1], the authors also quote animal research in which statins have been found to reduce endurance and mitochondrial activity.

The evidence suggests, therefore, that statins have the potential to negate at least some of the benefits of exercise. But there is another major problem, too: those who may be fatigued from taking statins or perhaps are suffering from muscle pain (myalgia) as a result of these drugs may be less inclined to take exercise in the first place. The authors of this review cite evidence suggesting that as many as 25 per cent of those taking statins experience muscle symptoms on taking exercise, and that about 40 per cent of people with myalgia avoid even moderate exertion.

Some may say these problems, when they happen, are a small price to pay for the ‘obvious benefits’ of statins. However, it is probably worth bearing in mind that in most people who take statins (those with no history or heart attack or stroke), the research suggests statins have no net health benefit and don’t reduce the risk of death either [3].

References:

1. Murlasits Z, et al. The Effects of Statin Medications on Aerobic Exercise Capacity and Training Adaptations. Sports Medicine epub 11 July 2014

2. Mikus CR, et al. Simvastatin impairs exercise training adaptations. Journal of the American College of Cardiology. Epub 10 April 2013

3. Do statins have a role in primary prevention? An update. Therapeutics Letter Issue 77 Mar-Apr 2010 Therapeutics Initiative University of British Columbia

This week saw the publication of a study in which women of average age 71 were treated with magnesium or placebo for 12 weeks, to see what effect this had on certain tests of physical functioning. The magnesium came in the form of magnesium oxide at a dose of 900 mg per day, providing 300 mg of actual (elemental) magnesium each day.

The three prime tests of functionality in this study were a test of balance in the standing position, speed of walking over a 4-metre distance, and the time it took for women to rise out of a chair five times with their arms folded across their chests.

In the women supplemented with magnesium, both the latter two tests improved significantly compared to the women taking placebo. These results suggest that magnesium supplementation may help to improve physical functionality in older people, and may also help to prevent decline in functionality as we age.

One of the interesting things about this study is that it utilised magnesium in the form of magnesium oxide. This particular form of magnesium is known not to be very well absorbed (bioavailable), and it might be the better results would have been achieved with more bioavailable forms of magnesium such as magnesium citrate or magnesium malate.

Another interesting thing about the study was that it excluded women with relatively low levels of vitamin D. This is interesting because vitamin D is another nutrient that is critically important to muscular function. It might be, that the full benefits of magnesium supplementation are more likely to be seen in individuals not suffering from vitamin D deficiency.

Overall, though, I think this is an interesting study and one that suggests magnesium supplementation should at least be considered for individuals keen to preserve their functionality as they age.

References

1. Veronese N, et al. Effect of oral magnesium supplementation on physical performance in healthy elderly women involved in a weekly exercise program A random mice controlled trial American Journal of Clinical Nutrition Epub 9 July 2014