Author + information
- André J. Scheen, MD, PhD⁎ ( and )
- Delphine Legrand, MD
- ↵⁎CHU Liège, Division of Diabetes, Nutrition, and Metabolic Disorders, Department of Medicine, CHU Sart Tilman, University of Liège, Liège 4000, Belgium
We read with interest the study by Angiolillo et al. (1) demonstrating that patients with type 2 diabetes have increased platelet aggregation on dual aspirin–clopidogrel therapy and that patients with insulin-treated diabetes mellitus (ITDM) have greater adenosine diphosphate (ADP)-induced platelet aggregation compared with patients with non–insulin-treated diabetes mellitus (NITDM) (1). This is an important observation as 42% of 15,603 randomized patients (in the CHARISMA [Clopidogrel for High Atherothrombotic Risk and Ischemic Stabilization, Management, and Avoidance] trial) had diabetes (17% treated with insulin), with no significant better cardiovascular protection of the clopidogrel plus aspirin combination versus aspirin alone (2). However, Angiolillo et al.’s (1) finding and interpretation might be considered by cardiologists as a counterproductive effect of insulin therapy and represent an erroneous argument for not switching to insulin those numerous type 2 diabetic patients with poor glucose control while on oral treatment.
Type 2 diabetes is a progressive disease. Even if it is true that patients with ITDM are at a more advanced stage of their metabolic disorder, the need to switch to insulin reflects profound insulin secretory defect rather than more severe insulin resistance (3,4), as erroneously stated by Angiolillo et al. (1). That the higher proportion of women among ITDM subjects may be considered as an argument supporting the insulin-resistance hypothesis should also be challenged, as greater insulin resistance in women than in men is not a classical finding if appropriately measured (5).
Besides glucose-lowering therapy, the most important clinically relevant difference between the 2 groups was the 1% difference in hemoglobin A1c(HbA1c) level (7.9% in ITDM patients vs. 6.9% in those with NITDM, p < 0.001), that is, the same difference as that reported in the intensive group versus the conventional group in the United Kingdom Prospective Diabetes Study (6). Angiolillo et al. (1) suggested that this 1% difference could not explain the difference in platelet reactivity as HbA1clevels were not correlated with any of the platelet-function assays performed. This finding is in contrast to other observations showing a significant influence of glucose levels on platelet reactivity and effect of antiplatelet agents (7). According to Angiolillo et al. (1), the study was conducted in a tightly controlled diabetic population, which led to a limited variability in HbA1clevels; however, the reported 6% coefficient of variation of HbA1clevels looks astonishingly low with regard to the mean ± SD data of the 2 subgroups (7.9 ± 1.5% vs. 6.9 ± 1.0%).
A key message from the study by Angiolillo et al. (1) is that aggressive and/or tailored antithrombotic regimens for high-risk patients such as diabetic patients may be warranted. However, emphasizing in the “therapeutic implications” section that “treatment with insulin is typically considered a surrogate of increased atherothrombotic risk” may be misleading. Although this remains a controversial issue, numerous data do not support this statement (8). As diabetologists, the key objective is to obtain adequate metabolic control (HbA1c<7% and ideally <6.0%), in combination with aggressive management of all other cardiovascular risk factors, including effective antiplatelet therapy (9). In numerous patients, insulin therapy is a necessary and often irreplaceable partner to tackle hyperglycemia and reach HbA1ctargets. Please do not throw the baby out with the bathwater!
- American College of Cardiology Foundation
- Angiolillo D.J.,
- Bernardo E.,
- Ramirez C.,
- et al.
- Scheen A.J.,
- Lefèbvre P.J.
- American Diabetes Association