Myocardial injury in diabetic patients with multivessel coronary artery disease after revascularization interventions

Myocardial revascularization interventions are routinely indicated during the course of the treatment of patients with chronic coronary artery disease (CAD). The most important benefit observed with these interventions is the improvement in angina and exercise capacity in patients with limiting symptoms [1, 2].

However, during these interventions, myocardial damage may occur, and this may also influence outcomes during follow-up [3]. Periprocedural myocardial injury (PMI) has been extensively studied but is still not completely understood. The diagnosis is still challenging, and the appropriate treatment is yet to be defined. Especially with the emergence of highly sensitive markers of myocardial necrosis, the diagnosis of myocardial injury and infarction after coronary interventions has become even more debatable [4, 5].

Many known and unknown factors may influence myocardial injury following interventions. Although this is still controversial, diabetes mellitus may change myocardial functions and potentially make the myocardium more susceptible to ischemic injuries [6].

Experimental studies that addressed this question have shown contradictory results about the ischemic myocardial sensitivity of diabetic animal models. While some studies have demonstrated higher cardiac damage after ischemic insults [7, 8], others have shown no differences [9] or even a protective role of diabetes [10‐12].

Additionally, clinical studies in the scenario of acute coronary syndromes have shown indirect evidence of a higher susceptibility of diabetic patients to such things as greater risk for the development of heart failure [13] and higher morbidity and mortality rates [14]. However, uncertainties still remain about whether these are caused by a higher sensitivity to myocardial ischemia due to diabetes mellitus, or if these are related to other reasons such as more diffuse and aggressive coronary disease or due to the microvascular complications potentially caused by diabetes. Moreover, direct evidence from human studies about myocardial responses to ischemic insults in diabetic populations is still inconclusive and scarce.

In this context, the medicine, angioplasty or surgery study V is a prospective trial that assessed myocardial injury after myocardial revascularization interventions in patients with chronic multivessel CAD who underwent percutaneous coronary intervention (PCI), on-pump, or off-pump coronary artery bypass surgery (CABG) [15]. Cardiac biomarkers, troponin, and CK-MB were systematically collected before and after procedures to assess myocardial injury by the perspective of high-sensitive biomarkers released after interventions. Patients also underwent cardiac magnetic resonances (CMR) with late gadolinium enhancement (LGE) before and after procedures to assess new areas of myocardial infarction by the perspective of a high-sensitive imaging study. In this post hoc analysis, myocardial injury was analyzed in patients with type 2 diabetes mellitus and compared with patients without diabetes.

The medicine, angioplasty, or surgery study V (MASS-V) is a prospective nonrandomized trial that aimed at studying PMI after PCI, on-pump, or off-pump CABG [15]. Patients with angiographically documented multivessel coronary stenosis of more than 70% by visual assessment and myocardial ischemia due to angina symptoms or positive stress tests were included. Only patients with stable symptoms and preserved systolic ventricular function were included. They were excluded if they have experienced a recent acute coronary syndrome or other thromboembolic phenomena in the last 3 months, systemic inflammatory disease, or kidney dysfunction (creatinine above 2.0 mg/mL).

Between May 2012 and March 2014, 326 patients with multivessel CAD were analyzed for study enrollment. From these, 219 were included in this prospective, single-center, nonrandomized trial and were assigned to PCI, on-pump or off-pump CABG. Seventeen patients were excluded after the procedures (11 due to claustrophobia during CMR, four due to cerebrovascular accidents, and two due to septicemia). Thus, 202 patients were analyzed in this study. Of these, 66 were assigned to PCI, 67 to off-pump, and 69 to on-pump CABG. Ninety patients (44.5%) had diabetes mellitus at study entry. The baseline characteristics of diabetic and nondiabetic patients are presented in Table 1. Both groups were statistically similar regarding main demographic characteristics, despite glycemic levels. Table 2 shows the SYNTAX scores of patients with and without diabetes, stratified by treatment groups.

The analysis of the data from this study shows that patients with diabetes have similar myocardial sensitivity to that of nondiabetic patients after revascularization interventions. The evidence of similar release of both cardiac biomarkers (troponin and CK-MB) after three distinct revascularization strategies, PCI, on-pump, or off-pump CABG, was confirmed by the findings of a high-sensitive and specific imaging technique to detect cardiac ischemic damage, the cardiac magnetic resonance, by the analysis of necrosis as well as myocardial edema.

Although the release of markers of myocardial necrosis after interventions are multifactorial and do not necessarily represent ischemic insults, the findings of myocardial necrosis and edema at CMR are mostly related to ischemic events during cardiac procedures and can be considered as strong evidence to confirm these findings in our sample. In fact, the strength of the present study lies in the assessment of multiple time points of both cardiac biomarkers collected in a standardized fashion, with a minimum loss of data, and by bringing the relevant information of CMR results, for both myocardial necrosis and edema. Thus, the analysis of the release of these biomarkers and the findings of cardiac resonance shows similar myocardial sensitivity between diabetic and nondiabetic patients in this sample.

An interesting study from Verdoia et al. evaluated the effect of diabetes in PMI and myonecrosis after PCI treatment [18]. Using the release of CK-MB and troponin to define PMI and myonecrosis, the authors found that diabetes is not associated with increased risk of cardiac damage after PCI therapy. Thus, their findings agree with the results from the present study. However, adding to the information from that study, the present one also assessed the release of biomarkers after on-pump and off-pump bypass surgery and the evaluation of cardiac magnetic resonance.

Other studies have assessed the release of cardiac biomarkers in the setting of acute coronary syndromes and also evaluated mortality rates in diabetic patients [13, 19‐22]. Interestingly, although the majority of them have found higher mortality rates in the diabetic population, the release of cardiac biomarkers was similar to that in the nondiabetic population, the same finding seen in the present study after coronary interventions. Thus, these studies show that myocardial sensitivity is probably not influenced by diabetes status, and other factors might be associated with the higher morbidity and mortality rates observed in the diabetic population. Potential reasons for this observation are speculative and still under debate, but might include greater extent of coronary artery disease [23], systolic dysfunction [24], diastolic dysfunction [8], and a theoretically pro-thrombotic state that might be associated with diabetes [25]. Although some argue that myocardial intrinsic properties such as ischemic preconditioning might be altered by diabetes [26], our group has recently shown similar findings between diabetic and nondiabetic patients who underwent sequential exercise tests, as part of a preconditioning protocol [27].

In agreement with the findings of this study, Alegria et al. [28] have also evaluated mortality rates at 6 months and infarct size by myocardial scintigraphy in patients who suffered an acute myocardial infarction treated by thrombolytic therapy. They have shown higher mortality rates associated with diabetes that persisted after adjustment for confounding factors. However, this higher mortality could not be totally explained by very modest differences in myocardial infarct size and reduced ejection fraction in diabetic patients. Moreover, the authors could not completely rule out that these very small differences could not still be due to the remaining influence of confounders.

Furthermore, a recent study by Grodzinsky et al. [29] has also assessed angina symptoms over 1 year after PCI treatment. In this study, angina prevalence and severity were similar between patients with and without diabetes. These findings also confirm the findings of the present study: ischemia assessed by the presence of angina and its effects in the myocardium of patients after coronary interventions occur independently of diabetes status.

Interestingly, the analysis of correlation between glycemic and A1c levels and peak troponin and CK-MB did not show any correlation among these variables, although graphically it seems that the higher levels of troponin and CK-MB were found in the lower levels of glycemia. These findings might probably be because the higher peak levels of cardiac markers occurred within the most frequent observations of glycemic levels. In addition, the correlation coefficients do not support any further conclusions. These results are also confirmed by the analysis of cardiac damage assessed by magnetic resonance that did not show any association of PMI with glycemic and A1c level groups at baseline.

The results from this study deserve some consideration. First, it is possible that the small sample size could limit the study power to detect differences between the two populations. On the other hand, the multiple measurements of cardiac biomarkers in addition to the findings of cardiac resonance make this possibility less reasonable. Second, the well-controlled glycemic status, noted by the well-controlled levels of fasting glycemia and A1c in the diabetic population might have influenced study findings. Different findings could be observed in a diabetic population with poor controlled glycemic levels.

In this study, type 2 diabetes mellitus did not add risk to the development of myocardial injury after percutaneous coronary intervention or bypass surgery in patients with multivessel coronary artery disease.