Diabetes is associated with a considerable risk of coronary events and with a resultant poor outcome. The presence of diabetes is equivalent to the presence of pre-existing coronary artery disease (CAD) as a risk factor for future myocardial infarction (MI). Patients who have both diabetes and CAD are at the highest risk of coronary events. A 7-year follow-up study of Finnish men and women in late 1950s revealed an overall incidence of MI of 12.9%. Patients with no pre-existing CAD had an incidence of MI of 3.5%, whereas similar patients with diabetes had an incidence of MI of 20.2%, comparable to patients without diabetes but with pre-existing CAD who had an incidence of 18.8%. Patients with both CAD and diabetes had a 45% chance of having an MI during the observation period. The presence of diabetes was also associated with a poorer prognosis post-MI (2). Reviews of large cohorts of patients presenting with acute MI reveal that patients with diabetes have a higher 30-day (3) and 1-year (4) mortality. The prognosis is worse with the severity and duration of their diabetes (3,4). Patients with diabetes who survive an acute MI with left ventricular function compromise (ejection fraction less than 40%) have more cardiovascular events and deaths on long-term follow-up (3.5 years) than patients who do not have diabetes (5). The increased cardiovascular risk in patients with diabetes is well established.

Pre-existing diabetes is associated with worse prognosis after an MI, and acute hyperglycemia in the setting of an acute MI compounds this. Over 700 patients with diabetes admitted with non-Q-wave MI over a 10-year period were followed for 2 years. This comprised 16% of all patients admitted with a non-Q-wave MI. Patients in the highest blood glucose (BG) quartile (BG over 275 mg/dL, 15.3 mmol) were more likely to be women, had longer duration of diabetes (more than 10 years), had congestive heart failure on admission, were on treatment for hyperlipidemia, had a non-Q-wave MI, and had an elevated creatinine on admission (6). These patients had a higher hospital mortality, 30-day mortality, and 2-year mortality, with a hazard ratio of 2.46, 2.41, and 1.69, respectively, compared with the lowest quartile BG on admission (less than 153 mg/dL, 8.5 mmol) (6). However, hypoglycemia (BG less than 55, 3 mmol) was also associated with a higher 2-year mortality. The prognosis after an MI is worse with more severe hyperglycemia, prolonged duration of diabetes, and renal compromise.

Hyperglycemia is a complicating factor in many patients who present with ACS but are not known to have diabetes. Hyperglycemia in patients with established diabetes who present with ACS is anticipated and requires treatment. Hyperglycemia newly noted in the setting of ACS is also significant. In 1,664 consecutive patients presenting with acute MI hyperglycemia, defined as BG over 200 mg/dL (11.1 mmol), was noted in 11.1% of patients without diabetes and 62.5% patients with diabetes. Although diabetes was an independent predictor of in-hospital mortality, newly noted hyperglycemia was an even stronger predictor. In the whole population, predictors of mortality
were the presence of peripheral vascular disease, use of insulin on admission, age, previous heart failure or MI, being female, and having renal compromise. Use of aspirin, beta-blockade, statin, and angiotensin-converting enzyme inhibition were protective. All patients presenting with an acute MI were included, regardless of whether they received percutaneous coronary intervention (7). Acute hyperglycemia in patients with and without diabetes was noted to be associated with higher risk of the “no-reflow” phenomenon, as measured by intracoronary myocardial contrast echocardiography after reperfusion. In that study blood glucose (BG) of 160 mg/dL (8.9 mmol) was a dividing point between patients with and without reflow. Lack of reflow was not related to glycosylated hemoglobin or the incidence of pre-existing diabetes mellitus. Patients with acute hyperglycemia also had significantly greater infarct size measured as a rise in creatine kinase and a decline in wall motion score, as determined by echocardiography (8). Similarly, acute hyperglycemia (over 180 mg/dL or 10 mmol) was associated with lower left ventricular ejection fraction (LVEF) on presentation and on discharge in patients with first anterior wall MI. The improvement in LVEF from admission to discharge was smaller in patients with hyperglycemia. However, a pre-existing diagnosis of diabetes mellitus did not predict acute or discharge LVEF. The 30-day mortality correlated with acute hyperglycemia, rising at levels over 144 mg/dL (8 mmol) at presentation, with the plasma glucose being an independent predictor of mortality (9). The same group also found that acute hyperglycemia abolished the benefits of ischemic preconditioning in patients presenting with the first anterior wall MI (10).

Meta-analysis of 15 studies revealed increased risk of death and pump failure following MI in patients with “stress hyperglycemia,” but no diagnosis of diabetes prior to presentation. The risk of death was 3.9-fold higher in patients with BG 110 to 144 mg/dL (6.1 to 8 mmol) than in patients presenting with BG levels in the normal range, or under 110 mg/dL (6.1 mmol). Further increases of BG were also associated with increased risk of congestive heart failure and cardiogenic shock. The prognostic implications of hyperglycemia in patients with pre-existing diabetes were not as significant as in patients with hyperglycemia and not known to have diabetes (11).

Acute hyperglycemia can have implications on short- and long-term prognosis. An analysis of nearly 1,000 patients presenting with acute MI revealed a correlation between admission BG and postdischarge mortality (up to 50 months) but not in hospital death. Mortality increased by 4% for each 18 mg/dL (1 mmol) increase in BG for patients without known diabetes and 5% for patients with established diabetes (12). Glucose levels at presentation are more predictive of adverse cardiac outcome than a diagnosis of pre-existing diabetes. Hyperglycemia at presentation was noted to be a more important prognosticator than the presence of microvascular disease.

It is estimated that about 30% of people with diabetes do not know that they have this disease. Hence, patients presenting with hyperglycemia may have diabetes that has yet to be diagnosed. Analysis of 146 patients presenting with acute MI and hyperglycemia revealed that 18.5% of patients had known diabetes, 9.2% of patients not previously diagnosed with diabetes were noted to have an elevated glycosylated hemoglobin on admission suggesting pre-existing hyperglycemia, and the rest had no evidence of pre-existing hyperglycemia. In this cohort acute hyperglycemia, along with reduced renal function, was an independent predictor of 5- and 28-day mortality. Admission hyperglycemia was associated with higher mortality, whereas the glycosylated hemoglobin was not (13).

The risk of developing diabetes is increased after an MI. In patients without diabetes presenting with acute MI and BG less than 200 mg/dL (11.1 mmol), mild hyperglycemia at the time of hospitalization was predictive of persistent glucose intolerance on follow-up. Only 34% of these patients had normal glucose tolerance on follow-up, 25% had diabetes, and the remainder had impaired glucose tolerance. Glucose tolerance at the time of discharge was similar to glucose tolerance at 3-month follow-up (14). On 34-month follow-up, cardiovascular morbidity and mortality were associated with abnormal glucose tolerance, all eight deaths occurred in patients with abnormal glucose tolerance (15). In Finland, where a national health registry is used to track outcomes, 16% of men and 20% of women presenting with their first MI had diabetes. Those who did not have diabetes at the time of their MI had an increased risk of developing diabetes subsequently, a 2.3-fold increase for men and 4.3-fold increase for women (16). Abnormal glucose tolerance after MI predicted glucose intolerance on follow-up. Newly noted glucose intolerance was associated with high risk of future cardiovascular events.

The beating heart has a large energy requirement: a 300-gram heart synthesizes and utilizes about 5 kg of adenosine triphosphate (ATP) each day. At any time the heart contains only about 750 mg of ATP, thus the turnover of ATP is rapid, and the heart requires a constant production of ATP. Normally FFAs are the substrate for 60% to 70% of myocardial oxygen consumption; however, in the diabetic heart this is up to 90%. Because there is no anaerobic pathway for the consumption of FFA, there is increased dependence on glucose oxidation in the setting of ischemia, leading to increased lactate production (17). Insulin deficiency, hyperglycemia, and high FFA levels result in impaired glucose uptake, glycolysis, impaired pyruvate oxidation, impaired lactate uptake, and greater dependency on fatty acid as a source of Krebs cycle substrates. Reducing dependence on fatty acid oxidation and increasing pyruvate oxidation would benefit the diabetic heart during and following ischemia (18). Improved glucose uptake is protective in the setting of myocardial stress. Transgenic mice that overexpressed the GLUT-1 glucose transporter had better survival in the setting of left ventricular stress and hypertrophy than wild-type mice (19). Insulin may act as an antiapoptotic mediator by activating protein kinase B, which has dual effects of increasing glucose uptake via increased translocation of the glucose transporter
GLUT-4 to the plasma membrane and inhibition of the activation of proapoptotic peptide (20). Any compromise in the delivery and uptake of substrates for energy production, increase in reliance on oxidation, or slowing in the washout of toxic by-products will result in myocardial injury.

Acute hyperglycemia results in impaired vascular function. Peripheral vascular function has been studied under various conditions of experimental hyperglycemia. Flow-mediated vasodilation in the brachial artery was inhibited during an oral glucose tolerance test in human subjects with normal glucose tolerance, impaired glucose tolerance, or diabetes. The degree of flow-mediated vasodilation was inversely proportional to plasma glucose level (21). Hyperglycemia is associated with increased production of free radicals (superoxide anion), which results in the inactivation of nitric oxide (NO), an endothelium-derived vasodilator (22). Beckman and his colleagues noted impaired endothelium-dependent vasodilation in human volunteers in the presence of acute hyperglycemia. They studied methacholine chloride-stimulated forearm blood flow under euglycemic and hyperglycemic conditions and in the presence and absence of a protein kinase C-β inhibitor. Protein kinase C-β inhibits endothelium-dependent vasodilation by decreasing endothelium-derived NO synthesis. Because the inhibition of vasodilation in the presence of hyperglycemia was corrected with inhibition of protein kinase C-β, these authors concluded that acute hyperglycemia is associated with activation of protein kinase C-β and impaired NO synthesis (23). A possible mediator of increased oxidative stress may be the enzyme heme oxygenase, which catalyzes the oxidation of heme into biliverdin, carbon monoxide, and free iron. The presence of acute hyperglycemia is associated with lower levels of this cell-protective enzyme (24). While hyperglycemia is noted to reduce vascular relaxation in some sites, such as the forearm (25), this was not observed in the coronary microcirculation (26). This results in greater oxidative stress, decreased NO, and decreased endothelium-dependent vasodilatation.

Acute and chronic hyperglycemia is associated with higher inflammatory tone, a process that may accelerate plaque rupture and stimulate coagulation pathways (27,28). Inflammation is a precursor to ACS, is sustained long after the event, and is associated with poorer outcome (29). Patients with hyperglycemia in the setting of an acute MI had higher inflammatory markers and more evidence of cytotoxic T-cell activation than patients with BG levels under 126 mg/dL (7 mmol) (30). Hyperglycemia results in increased platelet adhesion and increased adhesiveness of the endothelium through the activation of endothelial adhesion molecules such as VCAM-1 (31). Hyperglycemic patients who received insulin had a smaller rise in C-reactive protein levels over 24 hours after MI (32). Treatment with glucose and insulin in patients presenting with acute MI and receiving reperfusion therapy resulted in an attenuated rise in inflammatory markers and plasminogen activator inhibitor-1 (33). Because acute insulin administration reduces inflammation in MI and reperfusion, insulin administration may be therapeutic in this setting.

Hyperglycemia predisposes the myocardium to rhythm abnormalities. Increased ST-segment elevation was noted in patients presenting with hyperglycemia and normal glycosylated hemoglobin levels (34). In experimental models hyperglycemia is associated with prolongation of the QT interval in vivo and in vitro (35,36).