New study sparks debate about benefits of intensive insulin in medical ICU

February 01, 2006

February 1, 2006

Leuven, Belgium - A new study has found that intensive insulin therapy may be of benefit in the medical intensive care unit (ICU). The research builds on previous work by the same team in 2001, which showed that intensive insulin saved lives in the surgical ICU [ 1 ]; as a result, ICUs around the world adopted its use, lead author Dr Greet van den Berghe (Catholic University of Leuven, Belgium) explained to renalwire .

Van den Berghe and her colleagues report their latest findings in the February 2, 2006 issue of the New England Journal of Medicine [ 2 ]. They show that intensive insulin therapy significantly reduced morbidity but not mortality among patients in the medical ICU, and subsequent death and disease were reduced among those who spent more than three days there. Van den Berghe says that although she considers the findings "positive" and her hospital will now adopt use of intensive insulin in the medical ICU, they do require confirmation in larger, multicenter studies.

But in an accompanying editorial [ 3 ], Dr Atul Malhotra (Brigham and Women's Hospital and Harvard Medical School, Boston, MA) says the study "must be considered negative¿¿¿¿[but] the subgroup analyses are interesting."

However, van den Berghe told renalwire that "even if you are very conservative, you can't see this as a negative trial."

Is intensive insulin harmful? Jury still out

Van den Berghe said the new trial was designed using the information they obtained from their surgical ICU study, with the primary end point being mortality for those staying three days or more in the ICU. They randomized 1200 patients to strict normalization of blood glucose levels (80-110 mg/dL [4.4-6.1 mmol/L]) with the use of intensive-insulin infusion or to conventional therapy (insulin administered only when the blood glucose level exceeded 215 mg/dL [12 mmol/L]), with the infusion tapered when the level fell below 180 mg/dL (10 mmol/L). There was a history of diabetes in 16.9% of the patients.

For the 767 patients who stayed in the ICU for three days or more, in-hospital mortality was 43% in the 386 who received intensive insulin therapy compared with 52.5% in those receiving conventional therapy (p=0.009), and morbidity was also reduced. In his editorial, Malhotra acknowledges that "the greatest benefit was seen [in this group]¿¿¿¿a finding similar to the previous [surgical] study."

But, he points out, on the basis of an intention-to-treat analysis, mortality rates did not differ significantly (26.8% in conventional treatment group vs 24.2% in the intensive-insulin group; p=0.30)

Van den Berghe says they knew that not all patients would stay in the ICU for longer than three days, because some would be moved out as they were going to die while others left because they improved. "But it's impossible to predict who will spend more than three days in ICU, and we would have needed 5000 patients or more to have enough power to show a difference in overall mortality," she explains.

But Malhotra told renalwire : "This postrandomization allocation of patients to a long-stayer subgroup is not scientifically valid." To use another figure, he says those who work in the ICU use 28-day mortality as a well-established outcome, and that was 29.9% among those in the intensive-insulin arm and 30% in the conventional-treatment arm (p=0.9). "That again is essentially identical."

I'm not trying to downplay the great contribution van den Berghe has made to this field, but the therapy is unproven and there are some potential concerns.

Among the 433 patients who spent less than three days in the ICU, mortality was greater among those receiving intensive insulin therapy (56 deaths vs 42 deaths). However, this difference was not significant, van den Berghe points out. But Malhotra says this is "immaterial, because the study was not powered to look at any potential toxicity of insulin."

He goes on to discuss whether intensive insulin could actually be detrimental in this group of patients, and he told renalwire that a recent German study, the VISEP trial [ 4 ], found evidence of possible toxicity in medical or surgical severe sepsis patients. That trial was stopped early because there was no difference in mortality between the conventional and intensive-insulin arms and there was frequent hypoglycemia in the intensive-insulin arm (12.1% vs 2.1%; p<0.001). In fact, even in this new study by van den Berghe, hypoglycemia "was an independent predictor of mortality," he adds.

"I'm not trying to downplay the great contribution van den Berghe has made to this field, but the therapy is unproven and there are some potential concerns," he adds.

Less dialysis, less ventilation, and earlier discharge from ICU with intensive insulin

The improvements in morbidity found in van den Berghe's new study resulted from the prevention of acquired kidney injury, thereby reducing the amount of dialysis required; earlier weaning from mechanical ventilation; and earlier discharge from the medical ICU and the hospital among patients who received intensive insulin therapy compared with those who did not, the researchers say.

These findings, in particular, have prompted them to adopt intensive insulin therapy in their hospital, van den Berghe explained. "We are not going to wait for other results, we are doing this. It saves us 3000 per patient," she says, adding that a cost-effectiveness analysis of the study has been submitted for publication to Critical Care Medicine.

We are not going to wait for other results, we are doing this. It saves us ? 3000 per patient.

She adds, "The kidney protection we found was really significant, a very important finding. There are few therapies to prevent renal injury in the ICU."

For cardiologists, she says that while there were few cardiac patients in the current study, there were a lot of cardiac-surgery patients in their previous surgical ICU study, and there has been a more recent observational study in the US in cardiac-surgery patients that found the first three days of intensive insulin therapy "were crucial."

Also, she notes, a recent study on diabetics who had suffered acute myocardial infarction showed a huge mortality benefit of intensive insulin, she notes.

Two large-scale, multicenter studies ongoing

Malhotra says it remains unclear at present which patients should receive intensive insulin therapy as they enter the ICU, but two ongoing, large-scale, multicenter studies are examining the issue of glycemic control in both the surgical and medical ICU?the GLUControl study in Europe in 3500 patients and the NICE-SUGAR trial in Australia, New Zealand, and Canada, which is enrolling 4500 patients, with results expected in 2007.

In the meantime, Malhotra says there are several alternatives open to doctors. "One option would be to withhold intensive insulin therapy until conclusive data are available. Another would be to administer intensive insulin therapy to all critically ill patients on the assumption that more patients will benefit than will be harmed," he suggests. He also proposes a third strategy to be less aggressive in the first three days, giving insulin to achieve target glucose values of less than 150 mg/dL (8.3 mmol/L), but if critical illness persists beyond three days, then more intensive insulin therapy can be considered.

Van den Berghe counters: "My advice to doctors would be 'just do it,' but if you're unsure, participate in one of the ongoing studies, and then you can revise your strategy, if necessary, when the data from these become available."


  1. Van den Berghe G, Wouters P, Weekers F, et al. Intensive insulin therapy in critically ill patients. New Engl J Med 2001; 345: 1359-1367.

  2. Van den Berghe G, Wawilmer A, Hermans G, et al. Intensive insulin therapy in the medical ICU. New Engl J Med 2006; 354: 449-461.

  3. Malhotra A. Intensive insulin in intensive care. New Engl J Med 2006; 354: 516-518.

  4. Brunkhorst FM, Kuhnt E, Engel C, et al. Intensive insulin therapy in patients with severe sepsis and septic shock is associated with an increased rate of hypoglycemia?results from a randomized multicenter study (VISEP). Infection 2005; 33: 19-20.


Comments on Medscape are moderated and should be professional in tone and on topic. You must declare any conflicts of interest related to your comments and responses. Please see our Commenting Guide for further information. We reserve the right to remove posts at our sole discretion.