Blood glucose and survival following myocardial infarction

Transient hyperglycemia (raised blood glucose concentration) is a common feature of serious acute illness, including myocardial infarction. For many years, this so-called "stress hyperglycemia" was considered of little clinical significance unless it was particularly severe (blood glucose > 12 mmol/L) or associated with diabetes. 

That is no longer the case because there is accumulating evidence that blood glucose concentration at the time of admission is an important predictor of outcome following myocardial infarction. The largest trial to examine the relationship between admission blood glucose and outcome for non-diabetics who have suffered a myocardial infarction was recently published. 

The study population comprised 1604 consecutive non-diabetic patients who, following a diagnosis of myocardial infarction, were admitted to intensive care units across France during the month of November 2000. All had blood sampled for glucose estimation at the time of admission. 

In addition to blood glucose, a wealth of other baseline data for each study patient was recorded, including age, gender, BMI, blood pressure, heart rate, presence or absence of hyperlipidemia, smoking habit, detail of current MI (STEMI or non-STEMI) and evidence of heart failure (Killip Class). The main outcome measures were in-hospital mortality and 1-year mortality.

Of the 1604 study patients, 796 had an admission blood glucose of 6.9 mmol/L or less; 275 had a blood glucose in the range of 6.9-7.9 mmol/L; 264 had a blood glucose in the range of 7.9-9.3 mmol/L and for the remaining 269 patients, blood glucose was greater than 9.3 mmol/L. 

Only 3.7 % of the 796 patient whose admission blood glucose was less than 6.9 mmol/L died before leaving hospital, compared with 6.5 % of those whose blood glucose was between 6.9 and 7.9 mmol/L, 12.5 % of those whose blood glucose was between 7.9 and 9.3 mmol/L. 

The highest  mortality rate (15.2 %) occurred in those with the highest admission blood glucose (> 9.2 mmol/L). A similar pattern was observed when 1-year mortality rates were calculated: 92.5 % of those with the lowest admission blood glucose were still alive at 1 year compared with just 75 % of those with the highest blood glucose. 

Further statistical analysis demonstrated that the relationship between admission blood glucose and mortality remained in all patient subsets and was independent of age, severity of infarct, clinical condition and mode of treatment. 

This study has provided the best evidence to date of the adverse effect of raised blood glucose in the context of myocardial infarction and contributes to a growing body of evidence that suggests normalization of blood glucose should be a therapeutic objective for all heart-attack victims, not just those with diabetes.