Login | Register
Crit-iq-ue
Home > resources > whats news

What's News

1 April 2009

Intensive Insulin Therapy  & the NICE-SUGAR trial

Summary
The NICE-SUGAR study 1 , published in the NEJM (26th Mar 2009) is a landmark paper in the critical care literature. The finding of increased 90-day mortality in critically ill patients receiving intensive insulin therapy (IIT) is a practice changing result for many ICU’s that adopted IIT following Van den Berghe’s initial paper 2 .
Debate has centred around the interpretation and generalisability of Van den Berghe’s initial paper, the conflicting results in subsequent studies 3 4, and concerns regarding the high incidence of hypoglycaemia when maintaining blood glucose levels (BGL) between 4.4 and 6.1 mmol/l.
The developments in this field since 2001 have been of great interest to the critical care community, and is a must know subject for critical care trainees.

Major Papers
Intensive Insulin Therapy in Critically Ill Patients. Van den Berghe G, et al. NEJM 2001.
In 2001 the benefits of preventing hyperglycaemia in intensive care patients became the focus of international attention following the report of  reduction in mortality and morbidity in surgical ICU patients randomised to IIT (4.4-6.0 mmol/L) compared to conventional glucose control (10.0-11.1 mmol/L) 2 .

The 1548 patient trial, stopped at interim analysis due to decreased mortality in the treatment arm, reported a decrease in mortality (8% vs 4.6%), and morbidity including new renal failure (8.2% vs 4.8%), ventilation, critical illness polyneuropathy (51.9% vs 28.7%), and sepsis (7.8%v s 4.2%). The improvement in outcomes were attributable to effects on patients with an ICU LOS of greater than 5 days. Insulin use, both the percentage of patients receiving and actual dose, was increased with IIT, as were the episodes of hypoglycaemia
Analysis of the cost of implementation of IIT showed a reduction in average total cost of healthcare resource utilisation of 2638 euros per patient with IIT. This represented an increase in average cost of insulin delivery from 72 euro to 144 euros with IIT, compared to a decrease in average cost of resources used for ventilation, vasopressor support, dialysis, antibiotics, and LOS from 10,569 euros to 7931 euros for IIT
Criticism of this paper was based on the single centre design, higher than expected baseline mortality, aggressive parenteral feeding practice, high incidence of hypoglycaemia, and implausibility of the treatment effect. Nevertheless, glucose control became an important issue in critical care medicine.

Intensive Insulin Therapy in the Medical ICU. Van den Berghe G, et al. NEJM 2006.
In 2006, Van den Berghe et al repeated their original study study in a medical ICU patients 4 . This trial again compared IIT to conventional therapy, but in 1200 critically ill patients medical ICU patients who were assumed to require at least 3 days ICU care. A pre-hoc subgroup analysis was planned for patients who did spend a third day in ICU. 
Overall IIT was associated with no difference in mortality, and a reduction in newly acquired kidney injury, earlier weaning from ventilation, and decreased ICU and hospital LOS. In the predefined group with an ICU LOS of 3 days or greater, IIT was associated with reduced ICU and hospital mortality, earlier weaning from mechanical ventilation, decreased ICU and hospital LOS, reduction in newly acquired kidney injury, reduction in hyperbilirubinaemia, hyperinflammation, and cumulative TISS-28 scores. The authors have recently presented further analysis of outcomes showing a significant reduction in critical illness polyneuropathy in the IIT group (39% vs 50%, p=0.017).
With a primary mortality outcome not confirming the results from the previous trial, and less convincing secondary outcomes, this trial did not effectively resolve the issue of validity and generalisability of IIT.
 Intensive Insulin Therapy and Pentastarch Resuscitation in Severe Sepsis. The SepNet Investigators. NEJM 2008.

In 2008 the international review board stopped the  German SepNet groups IIT in severe sepsis trial after 537 patients were enrolled because of a high incidence of hypoglycaemia in the IIT group (17.6% vs 4.5%). At the time of stopping there was no difference in mortality or morbidity between the IIT and conventional treatment arms. The contribution of this trial to the current literature is limited, as the incidence of hypoglycaemia was similar to the Leuven trials, and failure to detect benefit in the IIT group occurred with inadequate recruitment. 3

The NICE-SUGAR Study Investigators. Intensive versus conventional glucose control in critically ill patients. NEJM 2009.
The NICE-SUGAR study, a collaboration between the ANZICS Clinical Trials Group and the Canadian Critical Care Trials Group, reported on mortality in over 6000 critically ill hyperglycemic patients randomly assigned to either IIT (4.4-6.0 mmol/L) or conventional therapy (<10 mmol/L). The primary outcome was 90 day mortality, with multiple secondary outcomes relating to mortality and organ failures.
The original analysis was designed to enroll 4000 patients, however this was increased to 6100 patients following Van den Berghes Medical ICU paper reporting a reduced mortality benefit. The study was powered at 90% to detect a 3.8% absolute mortality difference, assuming a base-line mortality of 30%.

 

Patients expected to require treatment for greater than 3 days were enrolled from 42 adult medical/surgical ICU’s, with 6104 patients randomised between December 2004 and November 2008. The study groups were similar at baseline, and as expected the IIT arm received more insulin, more often, and had lower blood glucose levels, suggesting the intervention was successfully delivered.
The management of patients during the study reveals a number of points of interest. Firstly the calories delivered were less than the Van den Berghe trials, reflecting the differences in feeding practices in the 2 trial populations. Secondly the IIT arm received more corticosteroids than the control arm (34.6% vs 31.7%, p 0.02), a finding that could represent an unknown effect of insulin therapy.
The study’s key conclusion was that relative mortality at 90 days was 10% higher in the IIT group (829/3010) than the control group (751/3012), with this difference primarily reflecting more cardiovascular deaths in the tight control group. This represents a number need to harm of 38. The IIT arm had an expected increase in severe hypoglycaemia, with no sequelae reported.
Conclusion
The finding of increased mortality in the treatment arm of the NICE-SUGAR trial is unexpected, and differ from the previous Van den Berghe trials. NICE-SUGAR was a larger, multicenter trial, with a longer-term outcome of 90-day mortality, and although the mechanism of increased mortality due to cardiovascular death is not understood, it cannot be ignored. As such the use of intensive insulin therapy to control BSL in the 4.4-6.0 mmol/L range cannot be recommended.

What others have said
The Endocrine Society
The Endocrine Society commends the NICE-SUGAR investigators for producing an important and provocative addition to the medical literature and draws the following conclusions and recommendations from their data. First, near-normalization of blood sugar does not clearly improve outcomes in all critically ill hyperglycemic ICU patients, and there is even a suggestion that such an approach may worsen outcomes. Second, looser control of hyperglycemia, i.e., target blood glucose of 144-180 mg/dl, is a reasonable, and perhaps preferable, option in this particular group of very sick patients. Third, it is essential to assess clinically meaningful outcomes, such as mortality, as well as surrogate or intermediate endpoints, such as blood sugar level, in studies of diabetes treatment as the NICE-SUGAR study has done; improvement of blood sugar control may not always translate to better clinical results.
Finally, the rush to deploy difficult and resource-intensive protocols in ICU's may be premature until there is a better understanding of the reasons that the NICE-SUGAR results differ so markedly from those of an earlier study by Van den Berghe et al. , which showed that tight control of blood sugar in critically ill hyperglycemic patients seemed to improve outcomes. There are certainly differences in study design and target patient populations between these two studies, and the situation is further clouded by van den Berghe's failure to replicate the results of her original study, which was done in surgical patients, in a subsequent study of critically ill medical patients.
Close analysis of these study differences, including examination of the results in various sub-groups, may give rise to important questions that need to be answered by further studies. For example, are outcomes after tight glucose control the same in those with pre-existing diabetes as in those without it? Does the number or severity of hypoglycemic episodes, or some other aspect of tight glucose control, impair the future counter-regulatory and autonomic nervous system responses to hypoglycemia in this patient population? What is the temporal relationship of hypoglycemia to death? Would the use of continuous glucose monitoring improve outcomes?
Aggressive outpatient management of hyperglycemia has been the hallmark of diabetes care for almost two decades since the DCCT4 and UKPDS5 studies showed reduced microvascular complications with such an approach. Potential benefits of tight glucose control in reducing macrovascular complications and death have been more difficult to demonstrate, with recent studies (ACCORD, ADVANCE, and VADT) 5-8 showing that such aggressive management of hyperglycemia can be associated with substantial risks and/or few benefits.
The Endocrine Society believes that we have entered an era of more nuanced and patient-appropriate recommendations as a result of these recent large, well-done outpatient and inpatient studies. We believe physicians should individually tailor their approach to glycemic control in their ICU patients, perhaps targeting glucose values between 144-180 mg/dl, until we better understand the reasons for these somewhat counterintuitive findings.


References

  1. Finfer S, Chittock DR, Su SY, et al. Intensive versus conventional glucose control in critically ill patients. The New England journal of medicine 2009;360(13):1283-97.
  2. Van den Berghe G, Wouters P, Weekers F, et al. Intensive insulin therapy in critically ill patients. The New England journal of medicine 2001;345(19):1359-67.
  3. Brunkhorst FM, Engel C, Bloos F, et al. Intensive Insulin Therapy and Pentastarch Resuscitation in Severe Sepsis. The New England journal of medicine 2008;358:125-39
  4. Van den Berghe G, Wilmer A, Hermans G, et al. Intensive insulin therapy in the medical ICU. The New England journal of medicine 2006;354(5):449-61.
  5. Diabetes Control and Complications Trial Research Group. The effect of intensive treatment of diabetes on the development and progression of long-term complications in insulin-dependent diabetes mellitus. N Engl J Med 1993;329:977-986.
  6. UK Prospective Diabetes Study (UKPDS) Group. Intensive blood-glucose control with Sulphonylureas or insulin compared with conventional treatment and risk of complications in patients with type 2 diabetes (UKPDS 33). Lancet 21998;352:837-853.
  7. The Action to Control Cardiovascular Risk in Diabetes Study Group. Effects of intensive glucose lowering in type 2 diabetes. N Engl J Med 2008;358:2545-2559.
  8. The ADVANCE Collaborative Group. Intensive blood glucose control and vascular outcomes in patients with type 2 diabetes. N Engl J Med 2008;358:2560-2572.
  9. Duckworth, W, Abraira, C, Moritz, T, et al. Glucose control and vascular complications in veterans with type 2 diabetes. N Engl J Med 2009;360:129-139.

by Neil Orford