Blood Glucose Control in Critical Care

Stress-Induced Hyperglycemia

Stress-induced hyperglycemia is common in critically ill patients. There are no accepted criteria to define this acute hyperglycemia, unlike “chronic diabetes mellitus.” In acute illness, “stress” in response to tissue injury or infection can have profound effects on carbohydrate metabolism. This type of hyperglycemia occurs despite elevation in insulin levels (insulin resistance). It is assumed that several mechanisms contribute to this stress-induced hyperglycemia.

• Decreased glucose uptake and utilization: Insulin-stimulated glucose uptake and utilization is achieved by skeletal muscle for 80% to 85% of all peripheral glucose uptake and by adipose tissues for 5%. In skeletal muscle, exercise is an important stimulating factor for glucose uptake and utilization. However, in critical illness, this exercise-stimulated glucose uptake is decreased, because patients are typically bed-bound. Furthermore, in critically ill patients, glucose transporter-4 (GLUT-4)–dependent insulin-stimulated glucose uptake is inhibited.

• Increased glucose production: The liver is the dominant organ for glucose production from glycogen (gluconeogenesis). In the fasting phase, the liver can produce 2 µg/kg/min of glucose, which represents 85% of whole body gluconeogenesis in healthy subjects. In critically ill patients, this hepatic gluconeogenesis increases because of increased levels of glucagon, cortisol, growth hormone, and cytokines.

• Depressed glycogen production: The production of glycogen from glucose (glycogenesis) is one of the key roles of the liver. In critically ill patients, increases in the level of glucagon, epinephrine, and cytokines inhibit glycogenesis by inactivation of glycogen synthase through increased glycogen synthase-kinase.

• Increased free fatty acids: In critically ill patients, free fatty acid and triglyceride production from adipose tissue increases secondary to increased activity of hormone-sensitive lipase. The increase in the blood level of glucagon and adrenalin enhance such activity, which in turn decreases peripheral glucose uptake.

It is well known that stress-induced hyperglycemia reflects severity of illness and is associated with mortality and morbidity in various patient groups, such as those with acute myocardial ischemia, cerebral infarction and hemorrhage, and multiple trauma and burns. Until recently, it was suggested that stress-induced hyperglycemia may be an adaptive response, promoting glucose uptake into brain and red cells and facilitating wound healing. Even relative hypoglycemia was considered dangerous and was to be avoided. Thus “optimal” blood glucose concentrations were considered to be in the range of 160 to 200 mg/dL (8.8–11.1 mmol/L). Insulin administration was appropriate only when blood glucose exceeded 215 mg/dL (12 mmol/L), because at such levels it may induce osmotic diuresis and fluid shifts that may be clinically undesirable.

Normoglycemia in Critically Ill Patients

In 2001, in a single-center randomized controlled study, van den Berghe et al. found that IIT reduced mortality and morbidity in selected surgical patients (Leuven I). In this trial, 1548 mechanically ventilated surgical patients requiring intensive care were allocated randomly to the IIT group (target glucose: 80–110 mg/dL [4.4–6.1 mmol/L]), starting insulin administration when blood glucose levels exceeded 110 mg/dL (6.1 mmol/L), or to a conventional treatment group (target glucose range: 180–200 mg/dL [10.0–11.1 mmol/L]), starting insulin administration when blood glucose levels exceeded 215 mg/dL (11.9 mmol/L).

In this trial, ventilated postoperative intensive care unit (ICU) patients allocated to IIT had a 43% relative risk reduction for ICU mortality (8.0% vs. 4.6%, p = .04), when compared with patients receiving conventional glucose control. The benefit of IIT occurred particularly in the patients receiving intensive care for more than 5 days (ICU mortality: 20.2% vs. 10.6%, p = .005) and with multiple-organ failure with proven septic focus. IIT also decreased the duration of ventilatory support and ICU stay; reduced the need for blood transfusions; and reduced the incidence of bloodstream infections, critical illness polyneuropathy, and acute renal injury. Logistic regression analysis indicated that the reduction of blood glucose levels, not the administration of insulin, explained the clinical benefit. However, because more than 60% of patients in this trial were postcardiac surgery patients, the benefit of IIT may be altered in other ICUs with a different case mix.

In 2006 the same research group assessed the benefit of IIT in a medical ICU (Leuven II). The protocol of blood glucose management was the same as previously reported in the Leuven I trial. In the intention-to-treat analysis of 1200 patients, IIT did not significantly reduce hospital mortality (40.0% in the conventional treatment group vs. 37.3% in the intensive treatment group, p = .33). However, morbidity was significantly reduced by the prevention of newly acquired renal injury, accelerated weaning from mechanical ventilation, and accelerated discharge from the ICU and the hospital. However, in a post-hoc analysis, among 433 patients who stayed in the ICU for less than 3 days, mortality was greater among those receiving IIT. In contrast, among 767 patients who stayed in the ICU for 3 or more days, mortality was reduced in the IIT group from 52.5% to 43.0% ( p = .009); morbidity also was reduced ( Table 79.1 ).

In 2009 the Normoglycemia in Intensive Care Evaluation-Survival Using Glucose Algorithm Regulation (NICE-SUGAR) study compared IIT (target glucose: 81–108 mg/dL [4.5–6.0 mmol/L]) with conventional glucose control (target glucose: ≤180 mg/dL [≤10.0 mmol/L]) in 6104 ICU patients treated in 42 hospitals across Australia, New Zealand, and Canada. In contrast to the Leuven studies, patients randomized to IIT in the NICE-SUGAR study had a 2.6% excess 90-day mortality compared with patients receiving conventional glucose control (24.9% vs. 27.5%, p = .02).

In addition to the Leuven studies and the NICE-SUGAR study, two single-center and four multicenter RCTs have compared IIT with conventional blood glucose control in a total of almost 6000 patients. Neither of these studies found a significant difference in mortality between the two glucose management strategies in patients with severe sepsis, in patients with septic shock, or in mixed ICU patients (see Table 79.1 ).

Other Key Studies of Glucose Control and Insulin Therapy