Proposal of a New Approach to Study and Categorize Stress Hyperglycemia in Acute Myocardial Infarction

BackgroundStress hyperglycemia (SH) is a valid prognosticator of in-hospital complications and mortality in the intensive care unit, and is universally available, simple, and cost-effective. Even small refinements of SH can improve the risk stratification of patients with one of the most important diseases today—acute myocardial infarction (AMI).ObjectiveThe aim of the review was to analyze whether SH nomenclature and methodology have been consistent in the medical literature in order to identify possible methodological faults and to suggest possible solutions.DiscussionSH nomenclature and glycemic targets have been relatively uniform in recent years, but there has been a pronounced variability in the methodology. Recent meta-analysis showed that AMI patients with new hyperglycemia had a 3.6-fold increased risk of mortality during hospitalization in comparison to those who were normoglycemic. Four SH methodological mistakes were identified. First, using one cutoff value for SH instead of two different values (one for patients with diabetes mellitus [DM] and one for patients without DM). Second, analyzing, for example, either tertiles or quintiles without dividing AMI patients into subgroups according to their DM status. Third, studying only two subgroups (with SH and without SH), without determining the presence of DM, when DM is not analyzed. Fourth, failure to measure glycated hemoglobin.ConclusionsThe same admission blood glucose (BG) is a marker of different mortality risks in diabetic compared to nondiabetic AMI patients. For example, when admission BG is 108–126 mg/dL (6–7 mmol/L), then the risk of in-hospital mortality is higher in DM patients; however, with an admission BG of 162–180 mg/dL (9–10 mmol/L), the risk is lower in diabetic patients. We can improve the clinical utility of the admission BG in AMI if we analyze four groups of patients (those with and without previously diagnosed DM, and above and below the admission glycemia cutoff values for in-hospital mortality). Those cutoffs should be calculated separately for diabetic and nondiabetic AMI patients.