ACP Guideline Discourages Intensive Insulin Therapy

Janis C. Kelly

February 14, 2011

February 14, 2011 — The American College of Physicians (ACP) has issued a new clinical guideline for glycemic control in hospitalized patients that discourages the use of intensive insulin therapy (IIT). The guideline is based on a systematic review, which concluded: "No consistent evidence demonstrates that IIT targeted to strict glycemic control compared with less strict glycemic control improves health outcomes in hospitalized patients. Furthermore, IIT is associated with increased risk for severe hypoglycemia."

The clinical practice guideline, published online February 14 in the Annals of Internal Medicine, recommends against "using intensive insulin therapy to strictly control blood glucose in non-surgical intensive care unit (SICU)/medical intensive care unit (MICU) patients with or without diabetes mellitus."

The guideline also recommends against using IIT rather than conventional glucose control to normalize blood glucose in patients in the SICU/MICU, whether they have diabetes or not. Finally, the guideline recommends a target blood glucose level of 7.8 to 11.1 mmol/L (140 - 200 mg/dL) if insulin therapy is used in patients in the SICU/MICU.

Data supporting the new guideline were drawn from a systematic review by Devan Kansagara, MD, and colleagues. Their meta-analysis of 21 trials in ICU, perioperative care, myocardial infarction, and stroke or brain injury settings found that IIT did not improve short-term mortality, long-term mortality, infection rates, length of stay, or the need for renal replacement therapy. Furthermore, IIT was associated with a 6-fold increase in risk for severe hypoglycemia in all hospital settings.

The ACP Clinical Guidelines Committee, chaired by Paul Shekelle, MD, PhD, used the data to shape the 3 recommendations for the use in IIT in hospitalized patients with or without diabetes.

Current Results Consistent With NICE-SUGAR Trial

The ACP paper marks another brake on the formerly fast rush to adopt tight glycemic control for critically ill adults. This followed a 2001 study by van den Berghe et al, who reported that targeting normoglycemia in ventilated patients in the SICU reduced the risk for in-hospital death by one third. However, this result was not confirmed by the subsequent Normoglycemia in Intensive Care Evaluation — Survival Using Glucose Algorithm Regulation (NICE-SUGAR) randomized comparison of intensive vs conventional glucose control.

Simon Finfer, FRCP, FJFICM, one of the NICE-SUGAR investigators, told Medscape Medical News that the new ACP guideline reinforces both the American Diabetes Association (ADA) guidelines for diabetes care in hospital (glucose target of 140 - 180 mg/dL, 7.8 - 10 mmol/L) and the similar Institute for Healthcare Improvement guidelines.

Dr. Finfer, who is senior staff specialist in intensive care at the Royal North Shore Hospital of Sydney, Australia, also said: "There are no subgroups for whom intensive glucose control is proven beneficial."

ADA Weighs in on the ACP Guidelines

Sue Kirkman, MD, senior vice president, Medical Affairs and Community Information, ADA, told Medscape Medical News that the ACP reviewers did a very thorough and systematic review of the data and that the new guideline is basically consistent with the ADA recommendations.

Dr. Kirkman said, "The upper limit of the range is higher than we'd recommend. It's not clear why they picked 200 mg/dL rather than 180 mg/dL, and 140 to 180 mg/dL was the target for the control group in the NICE-SUGAR study." Dr. Kirkman explained that 180 mg/dL is also the point at which the kidneys start to spill glucose. Theoretically, patients could get dehydrated above that level, but this would mostly be a risk at much more elevated blood glucose readings, she explained.

"In actuality, there isn't much difference between the 2 recommendations," Dr. Kirkman said.

The ACP guideline is expected to affect not only diabetes specialists but also hospitalists, critical care specialists, and primary care providers.

Dr. Kirkman said, "I think some may still want to aim for the lower range of the ADA recommendations (110 - 140 mg/dL) for some patients, such as post–coronary artery bypass graft patients in the ICU. The ACP statement and other guidelines, such as ADA's, acknowledge that severe hyperglycemia (>200 mg/dL) is certainly linked to adverse outcomes for hospitalized patients, so one would hope that the pendulum won't swing too far towards not being concerned with hyperglycemia."

Both Dr. Finfer and Dr. Kirkman noted that the ACP guideline applies only to patients in the ICU.

Dr. Kirkman said, "By far the biggest unanswered question is what the goals should be for non-ICU patients, which is most patients in the hospital. Also, the links of hypoglycemia to adverse outcomes need to be better understood. Is this causative, or is it a marker for a sicker patient? Better systems to prevent hypoglycemia are needed."

Unanswered Questions

Dr. Finfer pointed out that there is still a need to define quality standards for tight glycemic control, to find affordable methods for frequent and accurate measurement of blood glucose in the ICU, and to conduct multicenter efficacy studies to determine whether tight control can reduce mortality under optimal conditions.

Support for development of the guidelines came from the ACP operating budget. Dr. Shekelle has received grant funding from the Agency for Healthcare Research and Quality, as well as royalties from UpToDate. Another author of the guidelines is an employee of ACP and of Pfizer. Support for the review by Dr. Kansagara and colleagues came from the US Department of Veterans Affairs Health Services Research and Development Service and the ACP. Dr. Kirkman and Dr. Finfer have disclosed no relevant financial relationships.

Ann Intern Med. 2011;154:260-267, 268-282. Abstract Abstract