The metabolic syndrome

Posted 11 Dec 2015

Dr Ed Warren, FRCGP, GP, Sheffield and GP trainer,

Dr Ed Warren, FRCGP, GP, Sheffield and GP trainer,
Barnsley VTS

Recent reports in the national press suggested that people with high blood pressure were ‘60% more likely to develop diabetes’. While the association between the two conditions is well known, what is the truth behind this claim? Practice Nurse investigated

THE CASE

Approaching 60 years old, Gordon is a nervous accountant with a rather gloomy view of existence. He is meticulous in his bodily habits, has never smoked, walks the dog daily, and obsessively consumes his five portions of fruit and vegetables a day (and follows whatever other food fad is fashionable). You see him quite regularly in your minor illness clinics, usually just needing reassurance, and always claiming that he did not want to come, but had been sent by his wife.

Predictably, he was one of the first patients on the phone to book when the NHS Health Checks started. Everything, including the blood tests, had been normal, but his first blood pressure reading was 160/84mmHg: on repeat checking this settled to 138/78mmHg. He seemed relatively content.

Today, however, he phones you in high alarm. He has just opened his copy of The Times to find a report from a reputable research institution declaring that high blood pressure could be an indication of future diabetes. Because he had a high reading at his health check, he wonders how frequently he should be tested for diabetes.

THE NHS HEALTH CHECK

Announced by the Prime Minister in 2008 and rolled out in 2009, the NHS Health Check claims that it will benefit people between 40 and 74 years old, by ‘Helping you prevent heart disease, stroke, diabetes, kidney disease and dementia’.¹These are indeed bold claims. Logically, prevention is better than cure, but the good sense that is a characteristic of general practice nurses should warn that anything like this that is being promoted as a panacea must prompt reservations and caveats by the bucketful. Responsibility for the Checks was passed over to local councils by the Health and Social Care Act 2012, but then those councils invariably commissioned local general practices to actually do the work. So you have probably had a steady drizzle of attenders.

The Checks are budgeted to cost £6bn over the next 10 years,² so it would be nice to think that diverting resources of this magnitude away from other things at a time when NHS finances are firmly ‘in the red’ actually does some good. Unfortunately the evidence is not compelling. The Checks were implemented in the absence of a robust body of evidence that they would do any good,^2,3 and even the Daily Mail (not usually given to criticism of anything in healthcare except NHS workers) has come down firmly against the Checks.⁴What they can certainly do, as in Gordon’s case, is generate anxiety.

In a rash moment you promise Gordon that you will look into things and try to find out about his risk of diabetes.

WHAT IS THE METABOLIC SYNDROME?

Raised blood pressure is common. About 30% of people in the UK over age 45 have a blood pressure that is above the currently recognised threshold of 140/90mmHg.⁵This rises with age so that by age 75, 70% of people are above the target. These are the people who carry a diagnosis of hypertension: there are more who have had a few high blood pressure readings, but have not been identified as potentially benefiting from treatment. So raised blood pressure is common, and isolated high readings are even commoner especially if you are anxious about something.

Diabetes is rather less common, but is getting more common. In the UK there are an estimated 2 million people diagnosed with diabetes, and another 750,000 cases yet to be diagnosed.⁶With such high prevalences it is not surprising that there is an overlap between those with raised blood pressure and those with diabetes.

In addition, it has been known for some time that diabetes and vascular disease commonly exist together. This is why the routine follow-up of diabetic patients includes checks of blood pressure – the treatment blood pressure target for people with diabetes is lower than for non-diabetics,⁶ and also cholesterol and kidney function. Correspondingly the routine follow-up of hypertension would normally include checking for diabetes. If you have diabetes, you have a 75% chance of dying of heart disease, and a 15% chance of dying of a stroke, so that only 10% die of something other than vascular disease.

The metabolic syndrome (also sometimes called ‘syndrome X’) is a bit of a slippery fish. A syndrome is a set of medical signs and symptoms that frequently occur together, and which may be associated with a specific disease. There are many different definitions of the metabolic syndrome. The one given in Box 1 is a joint effort from the International Diabetes Federation and the American Heart Association published in 2009.⁷ It used to be thought that the common factor for the signs and symptoms of the metabolic syndrome was insulin resistance (the reduced ability in some people to react normally to insulin), but this simplicity is now disputed.⁸

On this definition, 30% of Europeans over 50 years are affected, and numbers are increasing. There appears to be a genetic vulnerability to developing the syndrome, but this is aggravated by significant lifestyle factors such as lack of exercise, obesity and poor diet. Looked at like this, metabolic syndrome looks like a disease deserving treatment. However, as a diagnosis this is a no better way of assessing cardiovascular risk than the numerous other CHD risk tools, and is some ways performs worse.⁷Treatment is with diet, exercise, weight loss, blood pressure and cholesterol control: this is the familiar CHD risk management routine, and there is nothing special that can be applied to those with a diagnosis of metabolic syndrome.

So is a diagnosis of metabolic syndrome a useful concept when thinking of the link between raised blood pressure and diabetes, or just an excuse for people who are fat and lazy? That is not quite fair as some people do appear to carry a genetic risk of acquiring the metabolic syndrome. In addition, much of the work of primary care is with people who would not need care if they were able to follow healthier lifestyle choices, and in that respect metabolic syndrome is no different.

WHAT CAN YOU SAY TO GORDON?

Suspecting that Gordon will not be the last of the ‘worried well’ who frightened by the publicity that this research has generated, you might logically decide that your best way of offering Gordon and others a knowledgeable response is to look at the actual research. At first you look at the press release,⁹ which announces a ‘major new study’, and then at the short report in the BMJ¹⁰ with a rather less grandiose heading. But you also find that the Journal of the American College of Cardiology has published the research online for you to look at.¹¹

You are aware that all research can be criticised if you look deeply enough, and that the perfect research paper has never been written. You are also aware that current ideas of what constitutes good evidence are relatively recent, and so that your daughters and certainly your grand-daughters may well one day be laughing at you for your naivety. However, this paper looks promising. It is published in a reputable journal by a respected research group based at the oldest and arguably the best university in England (Oxford). The analysis is based on over 4 million patients all registered in general practices, using data derived from their practice computers, i.e. the sorts of patients that you will be seeing every day. So far so good – it looks like this evidence could be useful to you in your work.

But as you read on, the questions begin to increase. This is research done by a UK team on UK patients, so why was it published in America? The US has some of the most prestigious medical journals in the world, but then so does the UK. You then find that it is in fact two pieces of work: one a new study; and the second a meta-analysis of previous work.

The new work is based on ‘one or more’ blood pressure readings over a 23 year period. You know that blood pressure rises with age,⁵ and that a single blood pressure reading is next to useless when trying to identify a patient’s true BP. However, looking at single BP readings does correspond to what practice nurses actually do in their clinics, so may have relevance. In the research the initial recorded BPs are ‘banded’ into clumps of 10mmHg for both systolic and diastolic pressure starting at 65mmHg: that’s OK, but it does not correspond with the NICE guidance on BP stratification which starts at 90mmHgi.e. NICE thresholds end with a ‘0’ and not a ‘5’. The authors recognise that single BP readings are unreliable, so they work a statistical fiddle (albeit one that has a track record in previous research) to derive an estimate of actual blood pressure. It is this estimate that is used to check for diabetes risk.

Experience has taught you to beware of any research that presents its results as ‘relative risk’, as this paper does. The use of relative risk is a way of fooling a reader into believing that a risk is greater than it actually is. Doubling a risk of 1% still only gives you a risk of 2%. But this is a curious lapse as diabetes is so common that even a small rise in relative risk represents a lot of patients. Accordingly you wonder about editorial quality control.

Next you turn to the meta-analysis part of the paper. Thirty-nine pieces of research were looked at, but most of these were not from the UK and so may not be applicable to your practice population. Even though individual papers showed variable relationships between blood pressure and the incidence of diabetes, the pooled results do show an effect. However, the meta-analysis was done on published work only: it is known that most research that gets started does not get finished, and of that which is finished, not all gets published. Also research is less likely to be offered for publication if the results do not show what the researchers wanted.¹²So the meta-analysis is OK as far as it goes, but would you bet your pension on it?

You are forced to conclude, as does the press release, that this work probably shows an association between single blood pressure measurements and future diabetes, but it does not prove that one causes the other. More research is needed. Furthermore, the paper does not suggest that there is anything you can do about it. In the past few years there has been a lot of interest in preventing diabetes because the dramatic increase in diagnoses appears to be aggravated by unhealthy lifestyles. Diabetes UK currently recommends ways of delaying the onset of or preventing diabetes. The suggested lifestyle changes are: healthy eating with weight control and increasing exercise.¹³Sound familiar? These are the sorts of things that we are already recommending to patients, whether they are at future risk or already have diabetes.

IMPLICATIONS FOR GORDON

But what of poor Gordon, who is now trembling in the corner. He has not read the original research paper, but is mightily impressed that you have, and is now relying on you to offer him sensible advice. Gordon has had one dubious BP reading, which settled on repeating. He has an impressively healthy lifestyle, so for him diabetes prevention is just doing what he is already doing. If he is not placated, you may decide to do an HbA1c to reassure him. In terms of follow-up, he will be invited for another Free NHS Health Check in 5 years (unless a future government by then has realised that they are probably ineffective), which is about right. However, he always has the option of returning earlier if he gets any actual symptoms.

The metabolic syndrome

title

title

title

title

title

title