Diabetes and pre-infarct angina. Time to rethink comorbidities in the reperfusion-injury phenomenon?

Abstract

Abstract Background Pre-infarct angina (PIA) has been shown to reduce reperfusion injury and infarct size in patients with ST-elevation myocardial infarction (STEMI) and currently represents the most efficient form of myocardial conditioning yet discovered. The role of diabetes on ischemic preconditioning remains controversial – while some pre-clinical studies suggest that diabetes blunts ischemic conditioning, clinical studies are lacking. Methods We retrospectively evaluated consecutive patients with STEMI admitted in our hospital from January 2008 to August 2018 who underwent primary angioplasty (PCI). PIA was defined as chest, arm or jaw pain during the preceding 48h before STEMI diagnosis. Peak creatine kinase and peak Troponin T levels were used as a surrogate of infarct size. Ischemic time (IT) was defined as the time between the onset of symptoms to the restoration of flow after either guidewire passage, thrombus aspiration or first balloon inflation. Results Of the 1143 included patients, 74% (n=845) were male and mean age was 62.6±13.1 years. A quarter of STEMI-patients had diabetes (25%, n=285). Almost a third of the patients (32%, n=359) had a history of angina in the preceding 48h before STEMI (PIA). The proportion of PIA was similar between diabetic and non-diabetic patients. In patients with diabetes, PIA was associated with lower creatine kinase (CK) (1144 [500–2212] vs 1715 [908–3309] U/L, p=0.0029) and Troponin T (TnT 3.30 [1.90–6.58] vs 4.88 [2.50–9.58] ng/mL, p=0.0022) despite similar IT as compared to those without PIA (328 [200–554] vs. 258 [180–530] minutes, p=0.1365). In non-diabetic patients, PIA was not significantly associated with infarct size (TnT 3.74 [2.23–7.11] vs 4.56 [2.44–7.77] ng/mL, p=0.1945; CK 1549 [910 - 2909] vs 1793 [996 - 3078] U/L, p=0.0653) even after adjustment for the increased ischemic time (240 [150–550] vs. 210 [140–405] minutes, p=0.0128) (β=−0.12, p=0.085 for CK and β=−0.11, p=0.183 for TnT). A significant interaction was observed between the existence of PIA and diabetes on peak TnT (p=0.026 for interaction) and CK (p=0.047 for interaction), which was independent of the culprit vessel and IT. During a median follow-up period of 18.0 [12.1–25.5] months, 268 (24.0%) MACE events have occurred (165 deaths, 27 strokes, 46 myocardial infarctions and 26 target vessel revascularization). PIA was associated with a significant reduction in the incidence of MACE (HR 0.66 (95% CI: 0.48–0.89)) driven by a reduction on mortality (HR 0.44 (95% CI: 0.28–0.70)). Diabetes was associated with an increased incidence of MACE (HR 1.42 (95% CI: 1.07–1.89)). No interaction was found between diabetes and PIA on their effect on MACCE events. Conclusion PIA is a strong predictor of favourable outcomes in the setting of STEMI. The effect of PIA on myocardial protection in patients with STEMI undergoing primary PCI seems to be modulated by the presence of diabetes. Distribution of Peak CK and Peak TnT Funding Acknowledgement Type of funding source: None