Author + information
- Richard L. Kennedy, MD, PhD⁎ (, )
- Venkat Vangaveti, MSc, PhD and
- Usman H. Malabu, MBBS
- ↵⁎School of Medicine, Deakin University, Waurn Ponds Campus, Geelong, Victoria 3220, Australia
Waters et al. (1) discuss further evidence relating to increased incidence of new diabetes with statins. Superficially, the message from these and other studies is reassuring: statins have transformed the management of atherosclerotic disease; intensive therapy shows greater benefit than “standard” therapy; the rate of incident diabetes with statins is relatively small; benefits of statins are still apparent in patients with diabetes. It is further reassuring from this study (using data from the TNT [Treating to New Targets] and IDEAL [Incremental Decrease in Endpoints through Aggressive Lipid-lowering] studies) that those who develop diabetes are those who appear to be at risk.
The authors imply, but do not state, that all 4 risk factors for diabetes considered were equally important, and that there was no difference in incident diabetes between those with no or 1 risk factor. It is surprising that age was not considered as a factor in their analysis. Recent analyses of data from major lipid trials have documented the incidence of diabetes based on fasting glucose, but postprandial data are not available. The rate of incident diabetes may thus be underestimated. The effect of statins on diabetes incidence may also be underestimated because many of the patients in lipid trials have already been exposed to statins prior to the trial. Also, the comparators in the TNT and IDEAL studies (10 mg atorvastatin or 20 mg simvastatin) may themselves increase incident diabetes, thus partially masking the influence of intensive therapy.
It is unlikely that statins cause diabetes, but rather exacerbate a pre-existing dysglycemic state. Their impact almost certainly goes beyond the reported annual 2% increase in new diabetes. In the TNT and IDEAL studies, around 17% of patients had diabetes at baseline (previously diagnosed or increased fasting plasma glucose [FPG]). Deterioration in glycemic control in diabetic patients using statins is well documented (2) but has not been as extensively studied as the increase in incident diabetes. The increase in HbA1c with statins is similar in magnitude to the decrease seen with the newer classes of oral hypoglycemic agents. The mechanism is not known. Impaired β-cell function may well be involved (2), but impaired insulin sensitivity has also been documented (3). If impaired insulin secretion is the predominant mechanism, then increased FPG may be the earliest diagnostic criterion for diabetes to be satisfied. If the problem is impaired insulin sensitivity, then increased postprandial glucose may appear first. Postprandial insulin and lipid excursions are well known to contribute to progression of macrovascular disease.
Development of diabetes or worsening of its control has the potential to impair quality of life through increased need for treatment, increased clinic visits, side effects of treatment, and development of complications. These considerations are not minor but we should certainly not deny patients the considerable benefits of statins where they are indicated. Statins can be lifesaving, but it is noteworthy that intensive statin therapy was not associated with decreased overall or cardiovascular mortality in either the TNT or IDEAL studies. Careful consideration of the risks and benefits are needed in each case, and we need greater understanding of the differing effects of class members on glucose homeostasis.
- American College of Cardiology Foundation