GPs should periodically screen patients on statins for new-onset diabetes, say researchers who have found a dose-dependent increase in the risk of developing diabetes while taking statins.
An analysis of five clinical trials included 32,752 non-diabetic patients taking statins over 4.9 years, and found that 8.2% went on to develop diabetes and 20.4% experienced a cardiovascular event.
Patients on intensive statin therapy - simvastatin 80mg or atorvastatin 80mg - were 12% more likely to develop new-onset diabetes than those on moderate dose therapy - either placebo, simvastatin 20 0r 40mg, atrovastatin 10mg or pravastatin 40mg.
However, high-dose patients were 16% less likely to experience cardiovascular events than those on moderate doses.
There were 149 more cases of incident diabetes in the 1,500 participants assigned to intensive statin treatment than the 1,300 receiving moderate therapy, and 416 fewer patients with cardiovascular events in the group that recevied intensive-dose therapy.
Previous studies have shown an increased diabetes risk from statins, but this is the first to establish a dose-dependent relationship.
Lead author Dr David Preiss, clinical research fellow at the University of Glasgow's cardiovascular research centre, said: ‘We haven't informed patients of these side-effects before. We should, so that they get good information before they start treatment.
‘It's probably a good idea for doctors to be screening for diabetes every one or two years for patients taking statins.'
He said the findings suggested that the risks may outweigh the benefits for some groups of patients. ‘Two large statins trials show no benefit for patients with heart failure, and we know that diabetes and heart failure are a pretty poor combination. For that group there's a question about treating with statins at all.'
The meta-analysis population were already at high risk of cardiovascular events, and included patients with acute coronary syndromes. ‘But if you look at different levels of cardiovascular risk, or those who have a high risk of diabetes, the risk-benefit ratio from statins may also start to change,' said Dr Preiss.
Dr Malcolm Kendrick, a GP in Manchester, disagreed with the findings. ‘There are no studies longer than five years, so any assumption of prolonged benefit from statins is an extrapolation of unknown data,' he said.
He criticised the practice of combining all adverse cardiovascular events and comparing these with only one possible adverse effect of the medication.
‘Trials which use combined end points are meaningless - you have to look at overall mortality and serious cardiovascular events.'
JAMA. 2011;305(24):2556-2564
