Outcome studies have shown evidence of an association between hyperglycemia in the perioperative period and adverse events. These include a longer length of hospital stay as well as higher incidences of surgical site infection, pneumonia, infections of intravascular devices, sepsis, acute renal failure, blood product transfusion, neurocognitive dysfunction after cardiac surgery, and mortality. The effects of hyperglycemia are known to be mediated through a number of mechanisms: impaired neutrophil and monocyte function, endothelial cell dysfunction, production of interleukin-6 (IL-6) and tumor necrosis factor-alpha (TNF-α) by mononuclear cells, induction of platelet activity, and elevated levels of fibrinogen and von Willebrand factor. The tasks for the anesthesia provider are (a) to treat vigorously both preoperative and perioperative hyperglycemia in patients who are known to be diabetic—it has been suggested that diabetic patients having elective surgery should have glycemic control optimized to yield a HgbAIC of less than 7%—and (b) to be vigilant in watching for possible instances of hyperglycemia in patients without a previous diagnosis of diabetes (diagnosed or undiagnosed metabolic syndrome, perioperative steroid use, etc.)

To do this, the anesthesiologist must be aware that the definition of hyperglycemia is changing. It varies depending on the study, but in general the window of normoglycemia is narrowing—several outcome studies have defined hyperglycemic patients as those having a blood glucose of 150 mg/dL or higher. Also, in 2004, the American Association of Clinical Endocrinologists issued a position statement recommending intravenous insulin therapy in critically ill and non-critically ill patients during the perioperative period and for labor and delivery patients. Members of the consensus panel agreed on 110 mg/dL as the acceptable upper limit of blood glucose during the perioperative period, mainly for critically ill patients, and 100 mg/dL for labor and delivery patients.

The DIGAMI trial was one of the first trials devised to examine whether managing hyperglycemia would improve patient outcomes. Diabetic patients with an acute myocardial infarction were enrolled and randomized to routine care of hyperglycemia versus intravenous insulin for at least 24 hours
followed by four daily subcutaneous injections continuing for 3 months after discharge. Patients were followed for a mean of 3.4 years. Mortality at that time was 11% lower in the intensive insulin therapy group (33% vs. 44%). The greatest benefit was seen in those without prior insulin use and without a prior cardiac history. The results of this study led to the investigation of intensive insulin therapy in other populations. One important study (by Van den Berghe) of mechanically ventilated patients admitted to a surgical intensive care unit (ICU) was stopped early because of the significant reduction in mortality in patients managed with intensive insulin therapy in the immediate postoperative period.