Dysglycaemia in the critically ill and the interaction of chronic and acute glycaemia with mortality

PURPOSE

Hyperglycaemia is common in the critically ill. The objectives of this study were to determine the prevalence of critical illness-associated hyperglycaemia (CIAH) and recognised and unrecognised diabetes in the critically ill as well as to evaluate the impact of premorbid glycaemia on the association between acute hyperglycaemia and mortality.

METHODS

In 1,000 consecutively admitted patients we prospectively measured glycated haemoglobin (HbA1c) on admission, and blood glucose concentrations during the 48 h after admission, to the intensive care unit. Patients with blood glucose ≥7.0 mmol/l when fasting or ≥11.1 mmol/l during feeding were deemed hyperglycaemic. Patients with acute hyperglycaemia and HbA1c <6.5% (48 mmol/mol) were categorised as 'CIAH', those with known diabetes as 'recognised diabetes', and those with HbA1c ≥6.5% but no previous diagnosis of diabetes as 'unrecognised diabetes'. The remainder were classified as 'normoglycaemic'. Hospital mortality, HbA1c and acute peak glycaemia were assessed using a logistic regression model.

RESULTS

Of 1,000 patients, 498 (49.8%) had CIAH, 220 (22%) had recognised diabetes, 55 (5.5%) had unrecognised diabetes and 227 (22.7%) were normoglycaemic. The risk of death increased by approximately 20% for each increase in acute glycaemia of 1 mmol/l in patients with CIAH and those with diabetes and HbA1c levels <7% (53 mmol/mol), but not in patients with diabetes and HbA1c ≥7%. This association was lost when adjusted for severity of illness.

CONCLUSIONS

Critical illness-associated hyperglycaemia is the most frequent cause of hyperglycaemia in the critically ill. Peak glucose concentrations during critical illness are associated with increased mortality in patients with adequate premorbid glycaemic control, but not in patients with premorbid hyperglycaemia. Optimal glucose thresholds in the critically ill may, therefore, be affected by premorbid glycaemia.