Here is a concise but thorough overview of glycaemic control in the ICU, based on authoritative textbook sources.
Glycaemic Control in the ICU
Why Hyperglycaemia Occurs in Critical Illness
Critically ill patients commonly develop stress hyperglycaemia — a transient rise in blood glucose driven by:
- Elevated counter-regulatory hormones (catecholamines, cortisol, glucagon)
- Cytokines and inflammatory mediators causing insulin resistance
- Impaired GLUT4-mediated glucose uptake in liver and skeletal muscle
- Exogenous glucose from parenteral nutrition and dextrose-containing IV fluids
- Drugs such as glucocorticoids and vasopressors
Blood glucose levels in the ICU correlate with severity of illness — the sicker the patient, the higher the glucose.
Why It Matters
Hyperglycaemia in the ICU impairs multiple defensive mechanisms:
- Immune function: reduced neutrophil chemotaxis, phagocytosis, and intracellular bacterial killing
- Endothelial function: decreased nitric oxide production, impaired vascular reactivity
- Wound healing: delayed repair and increased surgical site infection risk
- Neurological and cardiac outcomes: worsens outcome after stroke and myocardial infarction
Additionally, glucose variability (not just the absolute level) is an independent predictor of mortality. Three measures are used: standard deviation (SD), mean amplitude of glycaemic excursion (MAGE), and the glycaemic lability index (GLI) — the last being the best discriminator for mortality.
Key Clinical Trials
| Trial | Finding |
|---|
| Van den Berghe (Leuven Surgical, 2001) | Tight control (80–110 mg/dL) reduced ICU mortality vs conventional (180–200 mg/dL); also reduced infections, need for dialysis, transfusions, and critical illness polyneuropathy |
| Leuven Medical Trial (2009) | No significant difference in hospital mortality with tight control |
| NICE-SUGAR (2009) | Landmark multicenter RCT (n = 6,104): intensive control (target 81–108 mg/dL) led to higher 90-day mortality and more severe hypoglycaemia (6.8% vs 0.5%) vs conventional control (144–180 mg/dL) |
The NICE-SUGAR trial fundamentally shifted practice away from "tight" glycaemic control. The excess mortality was largely attributed to hypoglycaemic episodes, particularly in patients with distributive shock.
Current Recommendations
All major guidelines now recommend moderate glycaemic targets:
| Organisation | Target Glucose |
|---|
| Surviving Sepsis Campaign (2021) | Start insulin at ≥180 mg/dL; maintain <180 mg/dL |
| American Diabetes Association | 140–180 mg/dL (perioperative/ICU) |
| American College of Physicians | 140–180 mg/dL |
| Society of Thoracic Surgeons (cardiac surgery) | 150–180 mg/dL |
| Institute for Healthcare Improvement | <180 mg/dL |
Avoid hypoglycaemia below 70 mg/dL — this carries its own mortality risk.
Treatment
- Insulin is the cornerstone of therapy in the ICU
- Intravenous insulin infusion is preferred over subcutaneous routes in critically ill patients (variable absorption due to haemodynamic instability, oedema, and poor tissue perfusion)
- Sliding-scale insulin alone is inferior to scheduled insulin regimens
- Oral hypoglycaemics are generally avoided in the ICU (slow onset, insufficient potency, GI absorption issues)
- Nutritional strategies can also help: low-carbohydrate enteral formulas, hypocaloric feeding
Practical Summary
Control glucose, but not too tightly. Target 140–180 mg/dL in most ICU patients. Start insulin when glucose ≥180 mg/dL. Prevent hypoglycaemia (<70 mg/dL) just as vigilantly. Monitor for glucose variability and minimise swings.
Sources: Barash Clinical Anesthesia 9e; Current Surgical Therapy 14e; Goodman & Gilman's Pharmacological Basis of Therapeutics