A more liberal approach to glucose control in patients with diabetes in the intensive care unit does not appear to increase rates of hospital acquired complications, Victorian research shows.

With up to 30% of patients in ICU having diabetes, current recommendations specify a target of 6-10 mmol/L for blood glucose, aimed at reducing potential complications such as infections from hyperglycaemia. However this tight control approach may risk hypoglycaemic episodes, and therefore a new study involving 400 critically ill patients with diabetes at Melbourne’s Austin Hospital compared outcomes before and after a trial of a liberal glucose target of 10-14 mmol/L versus a conventional target of 6-10 mmol/L.

The study found a similar proportion of patients in each group had at least one hospital-acquired complication during their stay (82.5% v 88.5%).

There were no significant differences in rates of all infectious complications combined (18.5% v 16%) or rates of sepsis, pneumonia, urinary tract infections or wound infections between the groups.

Similarly there were no differences in cardiovascular complications combined (7.5% v 6%) or rates of myocardial infarction, pulmonary oedema, heart failure, pulmonary embolism or DVT with a liberal versus conventional strategy.

Acute kidney injury rates were similar in both groups (10.5% v 11%) as were neurological complications (16.5% v 14.5%), predominately delirium.

Hospital mortality was similar in both groups (20% v 16%) as was ICU and hospital length of stays (2.7 v 2.1 days and 13 v 11 days respectively).

Patients managed with a more liberal approach to glycaemic control had fewer relative hypoglycaemic episodes than other patients (32.5% v 48.9%).

The researchers said the findings imply that hyperglycaemia may be a physiological response to acute illness, rather than being the source of further complications in this patient population.

“Although our observations should be considered exploratory, they support the safety of cautious ongoing exploration of more liberal glycaemic targets in ICU patients with diabetes mellitus.”