Normalization of blood glucose does not necessarily benefit the critically ill

Transitory hyperglycemia (raised blood glucose) is a common feature of critical illness that was once viewed as a benign effect that could be safely left untreated, unless it was particularly severe. 

This traditional attitude to hyperglycemia in the critically ill changed dramatically in 2001 with the publication of a landmark Belgian study that demonstrated that normalization of blood glucose with intensive insulin therapy significantly reduced mortality and morbidity. Strict blood glucose control soon became a standard of critical care recommended by expert bodies and observed in some form by the majority of intensive care physicians. 

Over the intervening years since 2001 subsequent studies have provided conflicting evidence of benefit, and intensive insulin therapy has become an increasingly controversial treatment. The long awaited results of the largest ever randomized controlled trial of intensive insulin therapy, known as the NICE-SUGAR study, were recently published. 

Results suggest, in stark contrast to those of the 2001 Belgium study, that patients gain no benefit from normalization of blood glucose by intensive insulin therapy. For the NICE-SUGAR study 6104 critically ill patients admitted to intensive care units of all types in Australia, New Zealand, Canada and United Sates were randomly assigned within 24 hours of admission to one of two treatment groups. 

The aim for the first group was to maintain blood glucose within normal limits (4.5-6.0 mmol/L) by intensive insulin treatment. Those in the second (conventionally treated) group were only given insulin if blood glucose exceeded 10.0 mmol/L. Primary end point was death from any cause within 90 days of admission. 

Of the 3054 patients assigned to intensive insulin therapy, 829 (27.5 %) died within 90 days, compared with 751 of 3050 (24.9 %) patients in the conventionally treated group. 

There was no difference between the groups for several measures of morbidity (length of stay in intensive care, length of time on mechanical ventilation, etc). NICE-SUGAR study investigators conclude that normalizing blood glucose concentration "does not necessarily benefit patients and may be harmful". In an accompanying editorial the authors discuss the implication of the provocative results of this important study for routine care of the critically ill.