Blood Glucose Management in ICU Patients
Initiation Threshold and Target Range
Initiate intravenous insulin therapy when blood glucose persistently exceeds 180 mg/dL (measured on two occasions), and maintain glucose between 140–180 mg/dL for the vast majority of critically ill patients. 1, 2
- This 140–180 mg/dL target is based on the landmark NICE-SUGAR trial, which demonstrated that intensive glucose control (80–110 mg/dL) increased mortality by 27.5% versus 25% in the moderate control group, with a 10–15-fold increase in severe hypoglycemia. 1
- The American Diabetes Association, Society of Critical Care Medicine, and American College of Endocrinology all converge on this same recommendation. 1, 2
- Hyperglycemia is defined as blood glucose >140 mg/dL, which should prompt intervention such as minimizing dextrose infusions, but formal insulin therapy begins at the 180 mg/dL threshold. 1
More Stringent Targets for Selected Patients
Consider a tighter target of 110–140 mg/dL only for highly selected patients (cardiac surgery, acute ischemic cardiac/neurological events) and only if achievable without significant hypoglycemia. 1, 2
- The original Van den Berghe surgical ICU study showed 40% mortality reduction with 80–110 mg/dL targets, but this has not been reproducible in broader ICU populations. 1
- Never target glucose below 110 mg/dL due to increased hypoglycemia risk without additional clinical benefit. 2
- For most ICU patients, the risk-benefit ratio strongly favors the 140–180 mg/dL range. 1
Insulin Delivery Method
Use continuous intravenous insulin infusion (CII) as the exclusive method for glycemic control in critically ill ICU patients. 2, 3
- IV insulin has a half-life of <15 minutes, allowing rapid dose adjustments in response to unpredictable changes in clinical status, nutrition, or hemodynamics. 2, 3
- CII typically achieves target glucose levels within 4–8 hours. 2
- Avoid subcutaneous insulin of any type during the acute ICU phase, particularly during hypotension or shock, as absorption is unreliable and has not been formally studied in critically ill patients. 2, 3
Protocol Implementation
Implement validated written or computerized insulin infusion protocols that allow predefined adjustments based on glycemic fluctuations. 1, 4
- Computer-based algorithms have demonstrated lower rates of hypoglycemia and reduced glycemic variability compared to manual protocols. 2
- Protocols should target a low rate of hypoglycemia (blood glucose ≤70 mg/dL) as a primary safety endpoint. 1
- Computerized insulin infusion programs can achieve mean glucose of 147–171 mg/dL with hypoglycemia rates of 3–8% for blood glucose <60 mg/dL. 5
Blood Glucose Monitoring Frequency
Monitor blood glucose every 30 minutes to 2 hours during IV insulin infusion until stable, then every 4 hours thereafter. 3, 4
- More frequent monitoring (every 1–2 hours) is required during the initial titration phase until glucose stabilizes. 3
- Use arterial blood rather than capillary blood for point-of-care testing if arterial catheters are present, as capillary measurements may be inaccurate in critically ill patients. 3
- When point-of-care results do not correlate with clinical picture, confirm with conventional laboratory testing. 4
Adjustments for Nutrition
Coordinate insulin infusion adjustments with nutritional status changes, including NPO status, enteral/parenteral nutrition initiation or interruption. 2, 3
- A simple intervention for mild hyperglycemia is to minimize or avoid dextrose infusions when patients receive other nutritional support. 1
- Never continue insulin infusions without adjusting for interrupted nutrition or NPO status for procedures. 3
- Safe glucose targeting requires integration of an adequate nutrition protocol with insulin titration. 6
Adjustments for Corticosteroids
Anticipate increased insulin requirements during corticosteroid therapy and monitor for rebound hypoglycemia with sudden dose reductions. 4
- Corticosteroids are a major iatrogenic cause of stress hyperglycemia in ICU patients. 7
- Sudden reduction of corticosteroid dose is a common hypoglycemia trigger that requires proactive insulin dose reduction. 4
Adjustments for Renal Impairment
Use higher glucose targets (potentially 160–200 mg/dL) for patients with end-stage renal disease (ESRD) due to markedly elevated hypoglycemia risk. 5
- ESRD patients have 41% incidence of blood glucose <79 mg/dL compared to 17.8% in non-ESRD patients, even with standard protocols. 5
- Reduced renal gluconeogenesis and impaired insulin clearance both contribute to hypoglycemia risk in renal failure. 2
- Mean glucose in ESRD patients can be maintained at 147 mg/dL, similar to non-ESRD patients, but hypoglycemia rates are substantially higher. 5
Adjustments for Hepatic Impairment
Exercise heightened caution with insulin dosing in hepatic impairment due to reduced hepatic gluconeogenesis and altered insulin metabolism. 2
- Malnutrition and hepatic failure are independent risk factors for severe hypoglycemia during insulin therapy. 2
- Consider more conservative insulin titration and more frequent monitoring in patients with significant hepatic dysfunction. 2
Transition to Subcutaneous Insulin
Transition from IV to subcutaneous insulin only when the patient is hemodynamically stable, acidosis (if present) is resolved, glucose measurements are stable for 4–6 hours, and a stable nutrition plan is established. 3, 4
- Start subcutaneous basal-bolus insulin regimen 1–2 hours before discontinuing IV insulin infusion to prevent rebound hyperglycemia. 4
- Use basal-bolus-correction insulin regimen (not sliding scale alone) for patients with good nutritional intake. 4
- For patients with poor oral intake or NPO status, use basal insulin plus correction insulin only. 4
Treatment of Hypoglycemia
Define hypoglycemia by severity: Level 1 (54–70 mg/dL), Level 2 (<54 mg/dL), and Level 3 (altered mental/physical status requiring assistance). 1
- Even a single episode of mild hypoglycemia (blood glucose <80 mg/dL) is associated with increased risk of death and longer ICU length of stay. 6
- Severe hypoglycemia (<40 mg/dL) is associated with cognitive impairment and increased mortality. 2
- Elderly patients fail to perceive neuroglycopenic and autonomic hypoglycemic symptoms, delaying recognition and treatment. 2
Special Populations
Neurosurgical Patients
Maintain strict glucose control in neurosurgical patients, as hyperglycemia may worsen neurological outcomes in patients with brain pathology. 3
- Continue IV insulin infusion throughout the perioperative period for patients with persistently elevated glucose >180 mg/dL. 3
- Rapid glucose control is warranted for ischemic events, though intensive lowering has not shown additional benefit. 2
Elderly Frail Patients
Target 140–180 mg/dL for elderly frail ICU patients with heightened emphasis on preventing hypoglycemia in this vulnerable population. 2
- Elderly frail patients have significantly elevated hypoglycemia risk due to renal failure, malnutrition, malignancies, dementia, and frailty itself. 2
- Do not pursue stringent targets (<110–140 mg/dL) in elderly frail patients unless highly select cases. 2
Common Pitfalls to Avoid
Never use sliding scale insulin as the sole regimen in the ICU, as this approach increases glycemic variability, hospital complications, and is strongly discouraged by all major guidelines. 2, 3, 4
Avoid overly aggressive glucose targets (<110 mg/dL), which increase hypoglycemia risk without improving outcomes and may increase mortality. 1, 2, 4
Do not rely solely on point-of-care glucose meters in critically ill patients, as the FDA has stated these may be inaccurate in this population. 3
Never continue insulin infusions at the same rate when nutrition is interrupted, as this is a major cause of iatrogenic hypoglycemia. 3
Quality Improvement Measures
Consult specialized diabetes or glucose management teams when available, as appropriately trained specialists reduce length of stay, improve glycemic control, and improve outcomes. 4
- Implementation of structured computerized order sets that embed glycemic management guidance improves overall glucose control. 4
- Electronic insulin order templates raise mean glucose toward target without increasing hypoglycemia. 4
- The Joint Commission accreditation program for hospital diabetes care provides valuable resources for standardizing inpatient management. 1, 4