Abstract
Background
Increased hemoglobin A1c levels indicate a poor cardiovascular prognosis; however, the associations between glycemic control, platelet reactivity assessed by thromboelastography (TEG), and outcomes after coronary stenting in the Chinese population are unclear. Our study aimed to determine whether HbA1c levels are related to platelet reactivity after the administration of a loading dose of clopidogrel and aspirin; therefore, there is an effort to intensify antiplatelet treatment in patients with poor glycemic control after percutaneous coronary intervention.
Methods
HbA1c levels were categorized into 3 groups: <6.5% (n = 1103, 48.12%), 6.5–8.5% (n = 341, 36.9%), and > 8.5% (n = 177, 14.9%). ROC (resistance to clopidogrel, ROC) and ROA (resistance to aspirin) were defined as the ADP (adenosine diphosphate.) inhibition rate of less than 30% and an AA (Arachidonic Acid.) inhibition rate of less than 50% by TEG, respectively. Multivariate Cox regression analysis was performed to evaluate HbA1c levels and the independent impact of the ROC curve on adverse events. Univariate logistic analysis was used to evaluate the associations between HbA1c levels and ROC. A restricted cubic spline was used to explore the nonlinear relationship between HbA1c and MACEs. Receiver operating characteristic analysis was conducted to assess the predictive value of HbA1c for MACEs.
Results
The incidence of ROC was 9.3% (151 of 1621), whereas that of ROC was 16.5% (268 of 1621). The ROC for each of the 3 groups significantly increased with increasing HbA1c levels [4.3% vs 7.1% vs 10.1%, p = 0.006]; however, the ROA did not [16.4% vs 17.7% vs 14.3%, P = 0.694]. Both HbA1c > 8.5 and 6.5 ≤ HbA1c ≤ 8.5 were significantly associated with ROC. However, after adjustment for other covariables, only the HbA1c > 8.5 subgroup remained significantly associated with ROC (3.356 [1.231, 9.234], p = 0.009). Compared with the HbA1c < 6.5 subgroup, the HbA1c˃8.5 subgroup was significantly associated with nonfatal MI (2.297 [1.275, 4.137], P = 0.006), ischemia-driven revascularization (3.845 [2.082, 7.101], p < 0.001) and MACE (3.142 [2.346, 4.206], < 0.001), but not all-cause mortality (2.371 [0.551, 10.190], 0.246), and remained associated even after fully adjusting for confounders at the 36-month follow-up. When considered a continuous variable, HbA1c had a positive correlation with MACEs, as indicated by the RCS. The AUC of HbA1c for the prediction of 36-month MACEs was 0.636 (0.612, 0.624; P < 0.001).
Conclusion
In our two-center large-scale study, HbA1c levels were positively correlated with ROC, but the adverse effect on clinical outcomes was driven by elevated HbA1c, which forecasted nonfatal MI and ischemia-driven revascularization, constituting an argument to intensify glycemic control in subjects with diabetes after intracoronary stent placement.