Severe hypoglycemia and serious adverse events were more common in the intensive therapy group.
Intensive insulin therapy — designed to normalize blood glucose levels — is advocated increasingly for hyperglycemic patients in intensive care units. Nevertheless, convincing evidence of benefit comes mainly from only one trial conducted primarily in critically ill patients who had just undergone cardiac surgery (Journal Watch Nov 16 2001). In this new study, German researchers randomized about 500 ICU patients with severe sepsis to intensive insulin therapy (initiated when glucose levels exceeded 110 mg/dL; target, 80–110 mg/dL) or to conventional insulin therapy (initiated when glucose levels exceeded 200 mg/dL; target, 180–200 mg/dL). To compare colloid and crystalloid therapy, patients were also randomized to receive hydroxyethyl starch (HES) or Ringer’s lactate for fluid resuscitation.
The trial was halted early for safety reasons. At 1 month, morbidity and mortality were virtually identical in the intensive and conventional insulin groups, but severe hypoglycemia and serious adverse events were significantly more common in the intensive insulin group. In the fluid resuscitation portion of the trial, morbidity and mortality were similar in the HES and Ringer’s lactate groups, but renal failure was more common with HES.
Comment: In patients with severe sepsis, intensive insulin therapy was not beneficial and in fact was responsible for adverse events. These findings, coupled with the ambiguous results from a recent trial of intensive insulin therapy in medical ICU patients generally (Journal Watch Feb 10 2006), indicate that proponents of intensive insulin therapy to achieve euglycemia in the medical ICU are still on shaky ground.
Published in Journal Watch General Medicine January 24, 2008
Brunkhorst FM et al. Intensive insulin therapy and pentastarch resuscitation in severe sepsis. N Engl J Med 2008 Jan 10; 358:125.