October 23, 2017

expert reaction to study looking at statins and type 2 diabetes risk

A new study, published in BMJ Open Diabetes Research & Care, examines the increased risk of developing type 2 diabetes from long-term statin use.

Prof. Jeremy Pearson, Associate Medical Director, British Heart Foundation, said:

“The increased risk of diabetes in those taking a statin is already known.  The risk reported in this publication is slightly higher than that previously reported in the largest analysis of trial data so far published (see Collins et al, Lancet November 2016), which is likely to be more accurate.  That analysis indicated that statin treatment equally effectively reduces cardiovascular events in people with or without diabetes (by around 1000 cases per 10,000 people in 5 years) but increases the onset of diabetes (by around 50-100 cases per 10,000 people in 5 years).

“Statins are an important and proven treatment for lowering cholesterol and reducing the risk of a potentially fatal heart attack or stroke.  This study indicates that statins can increase the onset of diabetes in some people.  However, it does not mean that people should stop taking their statins as there is no doubt they save lives.  Instead the findings reinforce the need to regularly monitor the blood sugar levels of patients taking statins.  As the authors of the current study conclude, the lower absolute increase in diabetes onset needs to be balanced against the substantial benefit of statin treatment.

“People who are prescribed statins should not worry, and should continue to take their statins as normal.  They can speak to their GP or call the BHF’s heart helpline on 0300 330 3311 if they have any concerns.”

Dr Tim Chico, Reader in Cardiovascular Medicine / consultant cardiologist, University of Sheffield, said:

“This study further confirms that there is a small increase in risk of diabetes with statin treatment.  It is disappointing that the authors only report ‘relative risk’ which is very prone to being misreported or misunderstood.  So, although statins may be associated with an increase in the relative risk of diabetes by 33-37% in this study, that does not mean that 33-37% of people given statins developed diabetes.  Instead, it means that statins may have multiplied the risk of diabetes in the non-statin treated group by 1.33-1.37.

“These finding are consistent with other studies, such as the Jupiter study that found that statins increased the risk of diabetes by about 25% relative risk.  Putting that into context, it meant that 0.16% of people on placebo developed diabetes, while 0.2% of people on statins did, but despite this statins reduce the risk of heart attack and stroke.  The fact that the allocation to statins was not randomised in this new study does make it hard to be certain that the higher risk of diabetes in statin treated patients was truly due to the statin.

“Both cardiovascular disease such as heart attack and stroke, and type 2 diabetes (the most common form, which statins may increase the risk of) are caused by a combination of lifestyle and genetic factors, and no single drug or behaviour will completely protect against them.  Type 2 diabetes is largely caused by being overweight, and having a poor diet and a sedentary lifestyle, and more attention must be directed to addressing these issues in society.  Statins play a major role in reducing heart disease in high risk patients, but all drugs have possible side effects and their risks and benefits need to be properly discussed between the doctor and the patient.”

Prof. Stephen O’Rahilly, Director of the MRC Metabolic Diseases Unit, University of Cambridge and Society for Endocrinology member, said:

“It has long been known that while statins are very effective at reducing heart attack and stroke risk, there is evidence that they can also increase the risk of Type 2 diabetes to some degree.  This new study suggests that the risk of developing type 2 diabetes is somewhat higher than previously estimated – however the study was not specifically designed to address this issue, and it relies on modelling to take into account confounding factors underlying the reasons for prescribing statins.  Therefore these results cannot be viewed as definitive.”

Prof. Alun Hughes, Professor of Cardiovascular Physiology and Pharmacology, UCL Institute of Cardiovascular Science, said:

“This paper adds slightly to previous evidence from randomised clinical trials that statin use is associated with a modest increase in the risk of new onset diabetes.  Allocation of statin was not randomised or blinded in this study and so it’s important to bear in mind that the findings could be biased.

“Furthermore, even if it is assumed that any bias is small, it is difficult to assess the importance of these observations since the authors do not give the results in terms of absolute risk increases.  Presenting results as hazard ratios or relative risks always makes effects look large – ‘a 30 per cent heightened risk of developing type 2 diabetes’ sounds like a big effect but based on figure 2 it looks like statins may be associated with an increase in the annual rate of new diabetes from about 6 per 100 people to approximately 8 per 100 people over a 10 year period.  I should stress this is a ‘back-of-the envelope’ calculation since the authors don’t give the required data (but it’s probably not wildly wrong).  An association of this size isn’t trivial but it isn’t large enough to outweigh the well-established benefits of statins in preventing cardiovascular disease, and shouldn’t dissuade people from taking statins.”

Prof. Colin Baigent, Director of the MRC Population Health Research Unit, Clinical Trial Service Unit & Epidemiological Studies Unit, University of Oxford, said:

“Although it is known that statins do increase the risk of diabetes, the best available evidence about the size of this excess risk comes from randomised trials, which have suggested a much smaller risk than was found in this study.

“The main weakness of the new research was that the doctor’s decision to start a statin might have been linked to a patient’s future risk of diabetes.  This complicates calculations, and means that the increased diabetes risk of about one third is really quite unreliable.  More importantly, though, in people at increased risk of diabetes, the evidence shows clearly that statins reduce their risk of heart attacks and strokes substantially irrespective of whether they develop diabetes whilst taking a statin, so any adverse health effects of diabetes caused by statins are massively outweighed by their benefits.”

Prof. Naveed Sattar, Professor of Metabolic Medicine, University of Glasgow, said:

“This is not new news.  We already know from randomised trials comparing statin versus placebo (which are far superior to observational studies) that statins slightly increase the risk of diabetes by around 12% compared to placebo.  The present study is not a randomised trial of statins so the usual issues of confounding and treatment biases cannot be overcome and the headline figure of 30% increased diabetes risk is therefore far less precise and could be an over-estimate.

“Even so, any modest diabetes risk is not a concern for the vast majority of patients at elevated cardiovascular risk where the heart benefits they gain from statins will substantially outweigh any risk of developing diabetes in the future.  Furthermore, in clinical practice doctors should now even more strongly emphasise patients make positive lifestyle changes at the same time as starting a statin, so that any risk of diabetes could be attenuated or even fully overcome.  Any lifestyle changes also benefit patients’ hearts so it’s a win-win.  There is also another positive in that doctors are more regularly checking for risk of diabetes in those about to start statins and, in this way, we are picking up more people at elevated diabetes risk who can gain considerably from more targeted lifestyle interventions to mitigate such risks.”

* ‘Statin use and risk of developing diabetes: results from the Diabetes Prevention Program’ by Jill P Crandall et al. published in BMJ Open Diabetes Research & Care on Monday 23 October 2017.

Declared interests

Prof. Jeremy Pearson: “No declarations of interest.”

Dr Tim Chico: “No conflicts.”

Prof. Stephen O’Rahilly: “No conflicts of interest.”

Prof. Colin Baigent: “I jointly coordinate, with Professor Rory Collins, the Cholesterol Treatment Trialists’ Collaboration meta-analysis of all large-scale randomized trials of statin-based treatments.  CTSU has received funding from the pharmaceutical industry to conduct randomized trials, including trials of statin therapy, but this work is conducted independently of the source of funding and the trials are sponsored by the University of Oxford.”

Prof. Naveed Sattar: “I have consulted for Amgen and also previously written / published on this topic on several occasions.”

None others received.