Statins should be offered to most patients with diabetes, reported The Times . Research has shown that “a wide range of patients benefited from the cholesterol-lowering drugs irrespective of their type of diabetes”, the newspaper added.
The Times reports are based on a study in almost 90,000 people that has provided further evidence for the use of what is now almost standard treatment for people with diabetes. There are few absolute exceptions mentioned by the authors but these include very young people and pregnant women. These findings further support what is increasingly seen as best practice.
Where did the story come from?
The research was coordinated by the Cholesterol Treatment Trialists Collaboration, based in Oxford and Australia using data from several international studies. Some of the individual studies contributing data were supported by grants from pharmaceutical companies but this new analysis was supported by grants from non-industry UK and Australian sources. It was published in the (peer-reviewed) medical journal: The Lancet .
What kind of scientific study was this?
This was a meta-analysis of pooled data from 14 major randomised trials of statin therapy. The studies included 18,686 patients (mostly with type 2 diabetes), and 71,370 without diabetes. The researchers selected the trials that aimed to modify lipid levels in the blood only and which aimed to treat more than 1000 people for at least two years. The researchers used the definitions of diabetes used when the patients enrolled in the trials to divide them into groups. No data was collected on any new cases of diabetes occurring after the studies had started.
The researchers were interested in the number of deaths (and the cause of death) as well as the number of heart attacks and strokes that occurred per step reduction (1mmol/L) in “bad” LDL-cholesterol. In the statistical analysis, the researchers adjusted the risks reported by individual studies to add weight to those where there had been a greater fall in cholesterol as a result of treatment. In this way, they allowed for the fact that there were differences in all the patients’ baseline cholesterol levels between trials and that some trials – those with a higher average cholesterol at the start – could be expected to achieve larger reductions in LDL-cholesterol
What were the results of the study?
After five years, for every 1,000 people treated with statins, 42 fewer patients with diabetes had major vascular problems, such as heart attacks or stroke. Over an average of 4.3 years, deaths from heart disease fell by nine per cent for every step reduction in “bad” cholesterol. The effect in people with diabetes was similar to the 13 per cent fall for people without diabetes. The reduction in deaths amongst people with diabetes was not significant.
The risk of major cardiovascular events such as heart attacks and stroke fell by about a fifth for every step reduction in bad cholesterol in everyone with or without diabetes. This proportional reduction was about the same if people were separated by age, sex, blood pressure, weight, smoking status, cholesterol level, or the presence of cardiovascular disease at baseline
What interpretations did the researchers draw from these results?
The researchers said that "most people with diabetes should now be considered for statin therapy unless their short term risk is low (e.g. children) or statin therapy has been shown to be unsuitable for them (e.g. pregnant women)".
They also said that it no longer matters whether heart disease, such as angina, has developed or what the level of cholesterol is in the blood, the researchers suggest that having diabetes is reason enough to start taking this medication.
What does the NHS Knowledge Service make of this study?
This is large study collecting data from all the major trials of statin therapy. Only one of these studies was designed to enrol only people with diabetes, however, there were small proportions of people who had diabetes enrolled into each of the other trials. Pooling the data from all the people with diabetes in these trials allows this valuable source of information to be analysed. In this type of study large numbers of people are required to show a convincing effect of the treatment. In the case of statins in diabetic patients, the authors have demonstrated this. But there are limitations to the study.
- One hundred and forty-seven people with type 1 diabetes treated with statins in these trials suffered a cardiovascular event. Although this represents the same proportional improvement seen in people with type 2 diabetes, the result was non-significant. Looking at the pattern of results, this is likely to be because numbers were small.
- Several other studies have reported results since this analysis was designed and these could not be included in the main analysis. The researchers analysed these results separately and found that they did not materially effect their conclusions.
- When the researchers analysed the mortality amongst people with diabetes treated with statins they used a strict definition of statistical confidence, this means that even this large study may have included insufficient people with diabetes to detect a difference in the numbers of deaths from different causes.
Although the statins appear to be reducing the number of cardiovascular events in diabetic patients, it is important to recognise the added improvements that can be gained by lifestyle changes, such as giving up smoking, healthy diet, and regular exercise.
Sir Muir Gray adds...
The biggest impact from reducing risk factors for vascular disease result from dealing with whole groups within the population, for example, people with diabetes, rather than trying to identify individuals. Some evidence suggests that the best strategy for prevention is for everyone over 50 to have a little bit of statin, combined with small doses of other risk reduction drugs.
Analysis by Bazian
Edited by NHS Website
Links to the headlines
BBC News, 11 January 2008
Daily Mail, 11 January 2008
The Daily Telegraph, 11 January 2008
The Guardian, 11 January 2008
The Times, 11 January 2008
Links to the science
Cochrane Database Syst Rev 2002, Issue 3
Cochrane Database Syst Rev 2000, Issue 2
Cochrane Database Syst Rev 2002, Issue 3
Cochrane Database Syst Rev 2007, Issue 4
Lancet 2008; 371:117–125