Author + information
- David D. Waters, MD⁎ (, )
- S. Matthijs Boekholdt, MD, PhD and
- Terje R. Pedersen, MD, PhD
- ↵⁎Division of Cardiology, San Francisco General Hospital, 1001 Potrero Avenue, San Francisco, California 94110
We thank Dr. Kennedy and colleagues and Dr. Koh for their interest in our paper (1). In response to Dr. Kennedy, we previously compared the strengths of the 4 major risk factors for diabetes across 3 large statin trials (2). Fasting blood sugar is the strongest predictor, which is not a surprise because it is used for the diagnosis of diabetes. Age was considered, but was not a predictor of diabetes in any of these 3 trials (2).
We agree with Dr. Kennedy that the incidence of diabetes may be underestimated in these trials. Unfortunately, the risk of diabetes with statins was not recognized when these trials were designed and executed. Thus, endpoint ascertainment is highly reliable for cardiovascular events, but less so for diabetes. Meta-analyses clearly show a stepwise increase in incident diabetes from placebo to low-dose to high-dose statin therapy, with the increments being 9% and 16%, respectively (3,4).
Data from clinical trials indicate that the deterioration in glycemic control related to statins in patients with diabetes may not be as large as that suggested by Dr. Kennedy and colleagues. In CARDS (Collaborative Atorvastatin Diabetes Study), where 2,838 patients with type 2 diabetes were randomized to atorvastatin 10 mg/day or placebo and followed for a median of 3.9 years (5), the adjusted mean difference in HbA1C between the treatment groups at the end of the study was only 0.105% (p = 0.03). In the ASPEN (Atorvastatin Study for Prevention of Coronary Heart Disease Endpoints in Non-insulin-dependent Diabetes) study, where 2,410 patients with type 2 diabetes were randomized to atorvastatin 10 mg/day or placebo and followed for 4 years, the changes in HbA1C were identical, 0.2%, in the 2 treatment groups (6). A random sample of 1,087 participants with diabetes in the Heart Protection Study had HbA1C measurements at baseline and after an average of 4.6 years of follow-up. The increase in HbA1C was slightly but not statistically significantly higher in the simvastatin 40 mg group compared to placebo (7).
Neither the TNT (Treating to New Targets) study nor the IDEAL (Incremental Decrease in End points through Aggressive Lipid lowering) study was powered to show a significant decrease in overall or cardiovascular mortality. Indeed, in the meta-analysis of more versus less intensive statin therapy (which included the TNT and IDEAL studies), the overall incidence of coronary heart disease (CHD) death was only 0.7% (8). Nevertheless, the combined endpoint of CHD death or nonfatal MI was reduced (relative risk: 0.71; 95% confidence interval: 0.58 to 0.87 per 1 mmol/l reduction in low-density lipoprotein cholesterol), as were other important endpoints such as ischemic stroke and coronary revascularization. The benefits of more intensive statin therapy are clearly established.
We disagree with Dr. Koh's conclusion that “higher dose statin therapy does not have any benefit compared with lower dose statin therapy,” both for the reason described in the preceding paragraph, and because the impact of new onset diabetes is relatively minor compared to the cardiovascular events included in our analysis: CHD death, myocardial infarction, resuscitated cardiac arrest, and fatal or nonfatal stroke. Comparing one hazard ratio to another is misleading because the types of events are not equivalent. As noted in our paper, nearly 20% of patients with an event had more than one of them, and other cardiovascular events that are prevented by statin treatment such as coronary revascularization, new onset angina, and transient ischemic attack were not included in our analysis.
Dr. Koh speculates that the long-term adverse effects of new onset diabetes with statins might generate a relative increase in deaths, and cites data from the JUPITER (Justification for the Use of statins in Prevention: an Intervention Trial Using Rosuvastatin) trial as support. The median follow-up of JUPITER trial patients was only 1.9 years, and despite a significant increase in incident diabetes in the rosuvastatin treatment group, all major cardiovascular events were reduced by nearly half. Even among the patients who did develop diabetes during the trial, rosuvastatin was associated with a risk reduction similar to that seen in the overall group (9). We agree with the conclusion of that report, that “the cardiovascular and mortality benefits of statin therapy exceed the diabetes hazard, including in participants at high risk of developing diabetes.”
- American College of Cardiology Foundation
- Waters D.D.,
- Ho J.E.,
- Boekholdt S.M.,
- et al.
- Waters D.D.,
- Ho J.E.,
- DeMicco D.A.,
- et al.
- Newman C.B.,
- Szarek M.,
- Colhoun H.M.,
- et al.
- Knopp R.H.,
- d'Emden M.,
- Smilde J.G.,
- Pocock S.J.
- Cholesterol Treatment Trialists' (CTT) Collaborators