Glucose Management

BSL is often erratic in the ICU patient. In general, target BSL should be 8-12mmol/L. This is:

Other suggested targets include 8-11mmol/L or 6-10mmol/L.

• Relatively hyperglycaemic
• Safe
  Attempts at tight glucose control typically result in frequent hypoglycaemic episodes.
  • Hypoglycaemia is dangerous and may result in:
    • Permanent neurological injury
    • Cardiac arrest
  • Hyperglycaemia
    Is relatively benign in the short term.

Pathophysiology

Hyperglycaemia:

• Commonly occurs in critically ill patients
  • Secondary to ↓ hepatic response (resistance) to insulin
• Associated with ↑ mortality
• Associated with
  • Impaired immune function
  • Impaired wound healing

Hypoglycaemia:

• Associated with ↑ morbidity and mortality
• Occurs with intensive insulin therapy

Management

Key Studies

BSL control:

• NICE-SUGAR (2009)
  • ~6000 Australasian/American patients with a >3 day ICU admission and no PO intake over that time, without DKA, HHS, or insulin-affecting disease
  • Unblinded, multicentre (38) RCT with minimisation algorithm to minimise differences between regions and surgical/non-surgical patients
  • 6100 patients provides 90% power for 3.8% ARR ↓, assuming baseline 30% mortality
  • Tight vs. conventional BSL control
    • Tight control
      • Aimed BSL 3.5-6mmol/L
      • Insulin ceased if BSL <4.5mmol/L
    • Conventional control
      • Aimed BSL <10mmol/L
      • Insulin ceased if BSL <8mmol/L
  • Tight control required ↑ insulin (97% vs. 69%)
  • Tight control had ↓ BSL (6.4 vs. 8mmol/L)
  • Significant ↑ 90 day mortality with tight control (27.5% vs. 24.9%)
  • No difference in 28 day mortality, ICU length of stay, or need for organ support
    Unclear why the mortality ↑ is not replicated at 90 days if true, given the short-term nature of the treatment.

---

References

1. Intensive versus Conventional Glucose Control in Critically Ill Patients. N Engl J Med. 2009;360(13):1283-1297. doi:10.1056/NEJMoa0810625