Acute Glycemic Control Reduces Mortality from Acute MI in Diabetics

Abstract & Commentary

By Jonathan Abrams, MD, Professor of Medicine, Division of Cardiology, University of New Mexico, Albuquerque. Dr. Abrams serves on the speaker’s bureau for Merck, Pfizer, and Parke-Davis.

Synopsis: Acute hyperglycemia, not long-term glycemic control, are not predictive of survival after MI.

Source: Cao JJ, et al. Relation of Chronic and Acute Glycemic Control on Mortality in Acute Myocardial Infarction with Diabetes Mellitus. Am J Cardiol. 2005;96:183-186.

This report echoes a number of recent studies implicating hyperglycemia as an adverse risk factor for individuals admitted with an acute MI. The GIK hypothesis of Sodi-Pallares of many years ago still echoes. Several recent studies have demonstrated a substantial risk elevation in diabetics, including UKPDS 35 and the DIGAMI glucose infusion trials. The current report is a retrospective analysis of 827 diabetics with acute MI admitted to the CCU of Henry Ford Hospital of Detroit during 1997-2001. Admission glucose levels were available for the majority, and a substantial number of individuals (349) had 2 or more measurements of glyco-hemoglobin (HbA1c) within the 2 years prior to admission. These individuals were analyzed separately; the average number HbA1c measurements in this cohort was 3.5 ± 2.2 in the 2 years prior to acute MI. In-hospital mortality was 9.3%. A multivariate logistical regression analysis to identify predictors of in-hospital death found increased admission glucose levels, older age, and elevated cardiac markers as independently associated with increased mortality.

The mean glucose in the 752 subjects who survived hospitalization was 234 ± 117mg/dL, whereas in the 75 individuals who died, the mean glucose level was 316 ± 143, P = < 0.0001. Although HbA1c levels were higher in the individuals who died, HbA1c quantities were not predictive of mortality. Mean levels were “significantly greater in patients with AMI and elevated admission glucose values. However, the mortality rate did not differ significantly among quartiles of HbA1c, with an odds ratio of 0.64, 1.01, and 1.08, respectively, comparing the highest quartiles to the lowest quartile of HbA1c” (NS). The lack of predictive power of HbA1c was also true for the first 6 months after admission. The adjusted odds ratio of AMI mortality in relation to admission glucose was consistent throughout the cohort, with quartiles of risk ranging from 1.03, 2.60, and 5.24, comparing the greater quartiles to the lowest; P = 0.001. Cao and colleagues conclude that “acute hyperglycemia, not long-term glycemic control, affects in-hospital AMI mortality. Acute hyperglycemia is a potent predictor of death after AMI.”

Many metabolic and pathophysiologic factors are known to be associated with hyperglycemia, and these are briefly discussed by Cao et al. Increased catecholamines, glucocorticoids and, most importantly, free-fatty acids are well known to be adverse markers and probably players in the adverse outcomes of diabetics with poor glycemic control. Cao et al suggest that acute hyperglycemia itself “is also likely toxic to myocardium,” and that “a result of insulin insufficiency further augments the toxicity of free-fatty acids.” These data suggest that even in non-diabetics, hyperglycemia is related to adverse outcomes after AMI. It isn’t clear why HbA1c levels, which assess long-term glycemic control, are not predictive of survival after MI, though many of these individuals with high levels also had high levels of glycemia. “More importantly, . . .the findings should encourage clinical efforts targeted at ‘tight’ acute glycemic control in all diabetic patients after AMI.”


Even though these data are not prospective, they do not surprise. Most cardiologists today recognize that there is an adverse relationship between diabetes, as well as hyperglycemia and adverse outcomes in CAD and particularly AMI. Over the past few years, cardiologists and endocrinologists have begun a conversation regarding the many adverse outcomes relating to diabetes (and the metabolic syndrome); with the accumulation of much outcome data, there has been an increasing trend towards tighter glucose control for acute CAD events and even bringing the cardiologist into the choices of therapy for hyperglycemia. The American College of Cardiology is, at present, introducing a diabetes module for all of its members. Many papers, editorials, etc. stress the 2-4 fold increase in mortality and morbidity in diabetics who have CAD. The worst outcomes are with acute infarction, but adverse events are also greater in diabetics who undergo coronary re-vascularization, even in stable CAD. This report from one center from Detroit, while not a detailed prospective analysis of abnormal glucose metabolism or clinical presentation of the patients, speaks for itself, and strongly supports that hyperglycemia per se is a substantial adverse phenomenon. Efforts to aggressively treat elevated glucose levels, starting from the first day in the hospital, are still woefully inadequate; that cardiology focus is more towards opening up the infarct artery and supporting left ventricular function than optimizing glycemia; endocrinologists are not called in for routine elevated blood sugars, particularly in diabetics.

Commonly (in my experience at least) in acute coronary syndromes, hyperglycemia in non-diabetics, or yet to be identified diabetics, is often ignored and not followed up, including the failure to institute vigorous insulin therapy for lowering blood sugar. We should and will be hearing much more about this subject in the years to come.