Major findings
By investigating the association between categories of abnormal glucose metabolism based on various definition and MACE in patients with intermediate lesions, the current study found that prediabetes based on IEC HbA1c-based definition (6.0 ≤ HbA1c < 6.5%) was associated with significant increased MACE risk compared with NGT, which was consistent with the best cut-off point identified based on the Youden’s index. Newly diagnosed diabetes was associated with increased MACE risk compared with normal glycemia based on all the currently widely used definitions. The current study supported the use of IEC HbA1c-based definition to identify high-risk patients of MACE, who may benefit from early lifestyle interventions.
Reasons For Selecting Patients With Intermediate Lesions
The current study enrolled patients with angiographically confirmed coronary intermediate lesions to represent patients with stable coronary heart disease for the following two reasons: On the one hand, patients with coronary intermediate lesions had similar degree of coronary stenoses (DS% of 50–70%), and thus the effect of lesion stenosis severity on prognosis may be reduced. On the other hand, patients with coronary intermediate lesions various significantly in short-term prognosis. In patients without functional significant lesions and deferred from revascularization therapy, MACE occurred in approximately 4% of the population in one-year follow-up [10], suggesting that further investigation of prognostic factors will assist in risk stratification and outcome improvement.
Explanations For The Superiority Of Hba1c Over Fpg
Our findings showed that prediabetes defined based on HbA1c, but not fasting plasma glucose, identified a group of patients at high-risk of MACE. Explanations for the superiority of HbA1c over fasting glucose to identify patients at risk for MACE are proposed as follows: Glycated hemoglobin values reflect the three-month average endogenous exposure to glucose, including postprandial spikes, and show low intraindividual variability, particularly in people without diabetes[11]. In addition, HbA1c is a useful marker for other glycated molecules, such as advanced glycation end-products, which are likely drivers of vascular inflammation and subsequent plaque progression and rupture, leading to major adverse events in patients[12]. These features support the role of HbA1c as a novel biomarker in risk stratification.
Comparison With Previous Studies
Growing number of studies explored the association between prediabetes defined based on HbA1c value and MACE in the setting of CAD[1]. However, most studies enrolled patients with acute coronary syndrome [13, 14] or those who received revascularization[14–16], and the current study add new data in this field by examining this association in patients with stable CAD and not undergoing revascularization. Our study found that prediabetes defined according to IEC HbA1c-based definition was associated increased risk of MACE. Our findings are in consistent with previous studies showing that prediabetes was associated with poorer prognosis in patients who underwent PCI and treated with contemporary DES[16, 17]. In contrast, some previous studies reported no significant association between HbA1c-defined prediabetes and prognosis[15, 18]. The contradictory results may be explained by the difference in prediabetes definition and endpoint. In the above two studies, prediabetes was defined by HbA1c-ADA definition, which is HbA1c of 5.7%-6.4%. Similarly, when prediabetes was defined based on HbA1c-ADA definition in our study, approximately half of the study population were classified as prediabetes, and only 15% of the study population were classified as normal glucose metabolism. This may explain the non-significant association between prediabetes and outcome.
Possible Underlying Mechanisms
Several plausible biological mechanisms have been proposed to explain a possible direct relationship between chronically elevated blood glucose levels and CHD (24). Glucose can react with many different proteins, creating advanced glycation end products, which contribute to long-term complications in diabetes as well as to endothelial dysfunction, plaque formation and progression. In addition to the direct effect of elevated glucose on atherosclerosis, HbA1c may also contribute to plaque progression by diabetic dyslipidemia, hypertension, and inflammation, which can accelerate vascular injury and cardiovascular disease risk[19]. Diabetes has been proposed to accelerate atherosclerosis via oxidative stress, and increased inflammation[20].
Clinical Significance
Our study suggested that prediabetes based on the IEC HbA1c-based definition predicts MACE risk in patients with stable CAD. As discussed above, HbA1c reflects the average endogenous exposure to glucose and have low variability compared with fasting glucose. In addition, it less time-consuming compared with OGTT, the gold standard for diabetes diagnosis[21]. These characteristics may contribute to the superiority of glycated hemoglobin over other diagnostic methods for long-term risk stratification. Of note, newly diagnosed diabetes across all the four definitions was associated with a significant increased MACE risk compared with normal glycemic metabolism. Since patients with prediabetes have a significant higher risk of progression to diabetes, efforts including dietary and exercise intervention should be made in all patients with CAD and abnormal glucose metabolism[22]. Regarding pharmacologic interventions, no pharmacologic agent has been approved currently by the U.S. Food and Drug administration specifically for diabetes prevention or the treatment for prediabetes[23]. However, recent cardiovascular outcomes trials indicated cardiovascular benefits of novel glucose-lowering drugs, which included sodium-glucose cotransporter-2 inhibitors and glucagon-like peptide-1 receptor agonists. Therefore, these drugs may be recommended in patients with prediabetes to prevent or delay the onset of diabetes, which requires validation in future studies[24].
Limitations
Our study has several limitations: Firstly, OGTT tests were not performed in the majority of patients, and the relationship between post-load glucose value, and impaired glucose tolerance, another form of prediabetes defined by the 2h-postload glucose level was not evaluated. Secondly, only baseline HbA1c was collected, while the association between variations in HbA1c during follow-up was not assessed. Finally, the current study was a single center retrospective study with moderate size, and unmeasured confounders could not be excluded. Our findings need further validation in large-scale prospective cohort in future studies.