NICE-SUGAR: Intensive glucose control in the ICU increased mortality

Data suggest review of current glucose-lowering guidelines.

Intensive glucose control in critically ill adult patients increased the risk for death by 10%, according to the Normoglycemia in Intensive Care Evaluation — Survival Using Glucose Algorithm Regulation, the NICE-SUGAR trial.

The trial compared intensive vs. conventional glucose control delivered by IV insulin infusion in 6,104 adults admitted to the ICU. Patients were randomly assigned, with 3,054 patients to intensive control (81 mg/dL to 108 mg/dL) and 3,012 patients to conventional glucose control (<180 mg/dL).

The NICE-SUGAR investigators reported that intensive glucose control increased the absolute risk of death at 90 days by 2.6 percentage points, and that the difference in mortality was significant even after adjusting for potential confounders. They said that severe hypoglycemia was “significantly more common” in the intensive control group.

During 90 days, 829 patients assigned to intensive glucose control died compared with 751 patients assigned to conventional glucose control (OR=1.14; 95% CI, 1.02-1.28).

“Intensively lowering blood glucose to a target of 81 mg/dL to 108 mg/dL does not benefit critically ill patients and may well increase their risk of dying. There is no benefit to be gained over a target of <180 mg/dL,” Simon Finfer, MBBS, FJFICM, senior staff specialist in intensive care at Royal North Shore Hospital in Sydney, Australia, told Endocrine Today.

Patients intensively treated in the NICE-SUGAR trial were more commonly treated with corticosteroids than those in the coventional treatment arm. The excess deaths in that arm were predominantly from cardiovascular casues.

The two groups did not differ in the median number of days in the ICU or the hospital and days on mechanical ventilation or renal replacement therapy. The treatment effect was similar between operative patients and nonoperative patients, the researchers said.

Glucose control guidelines

“Previous, smaller research studies have produced conflicting results and overall suggested that intensive blood glucose control didn’t affect death rates in critically ill adults. It’s essential that international guidelines reflect this new evidence,” said Dean R. Chittock, MD, a NICE-SUGAR researcher and head of the division of critical care medicine at the University of British Columbia, Vancouver.

“Many professional organizations recommend very tight glucose control for ICU patients — they will now need to take this new evidence into consideration and adjust recommendations accordingly,” Chittock said in a press release.

Silvio E. Inzucchi, MD, and Mark D. Siegel, MD, in an accompanying editorial in The New England Journal of Medicine, said that further exploration of the increased mortality with intensive control is warranted in this population, perhaps using a per-protocol analysis.

“In retrospect,” they said, “it may turn out that we have been overly enthusiastic in our attempts to attain euglycemia during critical care. Similar and well-intentioned exuberance for rigid glucose targets in outpatient care was challenged this past summer by the jarring results from the ACCORD trial.

“Until further evidence becomes available, it would seem reasonable to continue our attempts to optimize the management of blood glucose in our hospitalized patients, especially to avert the extremes of hyperglycemia and hypoglycemia.” – by Katie Kalvaitis

Finfer S. N Engl J Med. 2009;360:1283-1297.
Inzucchi SE. N Engl J Med. 2009;360:1346.

The hypothesis that complete euglycemia in critically ill patients is beneficial requires that methods be available to assure that such treatment can be accomplished with complete safety. The message of the NICE-SUGAR study, therefore, appears to be one illustrating the lack of safety with current approaches to institution of such treatment protocols, paralleling the findings of the ACCORD study, in which an overly aggressive approach to outpatient glycemic control led to similarly unexpected adverse outcomes. The implication, then, is that the development of new therapeutic approaches will be of great importance in allowing us to adequately assess the hypothesis that glycemic control can lead to improvement in outcomes, whether such studies are undertaken in the acute setting of critical illness or in the chronic setting of treatment of diabetes.

Looking further at the article, a point which I have not seen discussed is the peculiarity that the increase in mortality began well after 20 to 30 days, but the duration of intensive glycemic treatment was only four days. I am not sure what mechanism one would invoke to explain a delayed effect of excess hypoglycemia on mortality or indeed of any other adverse consequence of insulin other than the possibility of excessive glycemic variation, and this further leads me to wonder, 'what went wrong here?' We need to remember what Joslin said many decades ago: 'insulin is a treatment for the wise and not for the foolish, whether they be patients or doctors.'

– Zachary T. Bloomgarden, MD

Endocrine Today Editorial Board member