Daniel R. Brown, M.D., Ph.D.

ontemporary anesthetic practice is associated with low perioperative morbidity and mortality. While research in areas such as genomics promise to improve patient care, investigating the impact of intraoperative care on functional recovery and long-term outcomes has shown that anesthesiologists have much more to contribute to the care of the perioperative patient. Studies have shown, for example, that administration of perioperative beta-adrenergic antagonists have beneficial effects long after the patient leaves the operating room. Recent studies in critically ill patients suggest that other aspects of perioperative care managed primarily by anesthesiologists, including ventilator management, antimicrobial administration and transfusion practice, are likely to affect patient outcomes as well.

Glycemic control in critically ill patients is actively being debated among intensivists and endocrinologists. Available data support an association between hyperglycemia and increased morbidity and mortality in many different surgical and medical populations. The link between hyperglycemia and poor outcomes led to studies designed to determine if management of hyperglycemia would improve patient care. The DIGAMI trial enrolled diabetic patients admitted with acute myocardial infarction and randomly assigned patients to intensive insulin therapy with intravenous insulin or to routine antidiabetic therapy.¹ Intensive insulin therapy was shown to be associated with significantly reduced long-term mortality. This observation encouraged investigation in other patient populations.

A recent prospective, randomized trial in critically ill adult patients was stopped early due to a significant reduction in intensive care unit (ICU) mortality in those patients managed with intensive insulin therapy.² In this trial, patients were randomized to intensive insulin therapy (goal glucose 80-110 mg/dl) or conventional therapy (treat for glucose > 215 mg/dl, maintain 180-200 mg/dl). In addition to decreased ICU mortality, in-hospital mortality and bloodstream infections also were reduced with intensive insulin therapy. This study is notable in that the vast majority of patients were postsurgical with more than 60 percent admitted following cardiac surgery.

Other investigators have suggested improved outcomes in postcardiac surgery patients with improved glycemic control. A study of more than 3,500 diabetic patients undergoing coronary artery bypass grafting reported an absolute and risk-adjusted decrease in mortality of 57 percent and 50 percent, respectively, and a significant decrease in major infectious complications associated with improved postoperative glycemic control.³

While no data exist to guide therapy for intraoperative glycemic management, postoperative data and pathophysiologic reasoning would suggest that improved glycemic control may improve patient outcomes. Until recently no guidelines or position statements regarding inpatient or perioperative glycemic control had been proposed. In December 2003, the American Association of Clinical Endocrinologists <www.aace.com> convened a consensus development conference on inpatient metabolic control. The purpose of the conference was to bring together international thought leaders in glucose management to develop medical guidelines for inpatient glycemic control, including the perioperative period. Implementing such guidelines during the perioperative period would involve anesthesiologists directly in a variety of settings, and while ASA was invited to provide commentary, there was no opportunity to provide input for development of the position statement.

The position statement considered intravenous insulin to be indicated for glycemic management in critically ill patients during the perioperative period and labor and delivery and following high-dose glucocorticoid therapy.⁴ The following guidelines were proposed as upper limits for glycemic targets: ICU patients <110 mg/dl; non-ICU patients <110 mg/dl preprandial, <180 mg/dl maximum; labor and delivery patients <100 mg/dl. The consensus panel members considered 110 mg/dl as the upper limit during the perioperative period.

These new guidelines, if accepted and implemented as suggested, would have profound impact on current anesthesiology practice. The guidelines are intended for all patients, not just those with a diagnosis of diabetes. Furthermore, with a blood glucose >110 mg/dl as an indication for insulin therapy, a large percentage of patients would receive insulin therapy during the perioperative period. With obesity increasing in the United States, it would be expected that the population at risk for hyperglycemia would only increase in the coming years.

The apparent link between hyperglycemia and outcomes raises other questions as well, including:

• What is the impact on preoperative testing? Is the preoperative medical examination an appropriate time to screen for impaired glycemic control by measuring HbA1c? Does the preoperative metabolic state as measured by serum glucose and glycosylated hemoglobin (HbA1c) predict perioperative hyperglycemia?

• In those patients found to be hyperglycemic, either preoperatively or during the perioperative period, what further testing and care should be performed? Who should provide this care, and how will coordination of these activities occur?

• Is glycemic control indicated in all patients for all procedures? What is the optimal blood glucose level with respect to the risk/benefit of hypoglycemia?

• Determination and management of hyperglycemia requires increased resources. While there may be overall cost savings considering length of stay, prevention of complications and improved quality of life, who will pay for the increased testing and therapy?

• How will institutions review and implement guidelines for inpatient glycemic control? What role will anesthesiologists play in their implementation?

At the least, anesthesiologists need to be aware of the emerging data linking poor outcomes with inpatient hyperglycemia. Practices are changing in ICUs, and, in all likelihood, anesthetic practice will need to change as well. While many questions remain unanswered, anesthesiologists will likely be asked by surgeons, intensivists and perhaps regulating agencies to achieve glycemic control during the perioperative period. The scope of this problem is large, but the potential impact on patient care is even greater.


References:

1. Malmberg K, Norhammar A, Wedel H, Ryden L. Glycometabolic state at admission: Important risk marker of mortality in conventionally treated patients with diabetes mellitus and acute myocardial infarction: Long-term results from the diabetes and insulin-glucose infusion in acute myocardial infarction (DIGAMI) study. Circulation. 1999; 99(20):2626-2632.

2. Van den Berghe G, Wouters P, Weelers F, et al. Intensive insulin therapy in critically ill patients. N Engl J Med. 2001; 345:1359-1367.

3. Furnary AP, Gao G, Grunkemeier GL, et al. Continuous insulin infusion reduces mortality in patients with diabetes undergoing coronary artery bypass grafting. J Thorac Cardiovasc Surg. 2003; 125(5):1007-1021.

4. American College of Endocrinology Position Statement on Inpatient Diabetes and Metabolic Control. Endocrine Practice. 2004; 10(1):77-82. <www.aace.com/clin/guidelines/InpatientDiabetesPositionStatement.pdf>.

Daniel R. Brown, M.D., Ph.D., is Chair, Division of Critical Care Medicine, Department of Anesthesiology, Mayo Clinic, Rochester, Minnesota.