Intensive Insulin Therapy Not Recommended for Critically Ill Patients
Aggressive blood glucose management does not reduce mortality or incidence of infection
The risks of aggressively lowering blood glucose levels to near normal in critically ill patients with or without diabetes may not outweigh the benefits, according to new guidelines from the American College of Physicians (ACP) published in the American Journal of Medical Quality.
The ACP calls for a target blood glucose level of 140 to 200 mg/dL when insulin therapy is used in the surgical or medical intensive care unit. Targets lower than 140 mg/dL should be avoided as they are associated with harm—namely hypoglycemia—according to the guidelines.
Why the Recommendation Was Made
Hyperglycemia is associated with significant morbidity and mortality in critically ill patients. In the past, several studies suggested benefits of strict glycemic control in critically ill patients and several organizations called for strict glycemic control strategies that were commonly implemented in the intensive care unit (ICU) setting, explained Amir Qaseem, MD, PhD, co-author of the guidelines and Director, Clinical Policy, American College of Physicians.
However, new evidence clearly shows the potential for harm associated with this practice, leading the ACP to examine the current literature on this issue. “We found that bringing glycemic level to normal or near normal in critically ill patients does not reduce mortality or the incidence of infection,” Dr. Qaseem said. “On the other hand, the risk of harms—namely hypoglycemia—is likely to increase with intensive insulin therapy,” he said.
Implications for Inpatient Care
“I think the recommendations will serve mainly to reinforce changes medical centers have already made, rather than push centers to make new changes,” said Gregory Maynard, MD, MSc, SFHM, Senior Vice President, Center for Hospital Innovation and Improvement, Society of Hospital Medicine (SHM). “Intensive insulin therapy requires tight monitoring with hourly glucose testing. Since the efforts to reach euglycemia were associated with increased hypoglycemia, the burden of extra testing and managing those events also increased healthcare costs,” he explained.
“On the other hand, many centers are very successful in reaching more modest glycemic targets with minimal hypoglycemia, and better control may well reduce some complications like surgical site infections and renal insufficiency,” Dr. Maynard continued. “I hope medical centers don’t ‘throw the baby out with the bathwater’ and abandon efforts to improve glycemic control in the hospital, when the real message is just to moderate the goals of therapy. Centers using somewhat lower target ranges should make sure they are tracking hypoglycemia rates by high quality methods, such as the methods SHM offers in their glycemic control data and reporting center, to ensure they are not achieving better glycemic control at the expense of higher hypoglycemia rates,” he said.
“There are parallels in this experience for settings other than the ICU,” Dr. Maynard continued. “Non–critical care glycemic control, where subcutaneous insulin—rather than intravenous insulin—is the most common agent of control, also has glycemic targets in roughly the same range. While uncontrolled hyperglycemia is not desirable, neither is being overly aggressive about glycemic targets, and the targets should be adjusted up further if patient situations demand it. For example, a patient with frequent hypoglycemia or who is on comfort measures only should not have the same glycemic target as patients without that situation,” he said.
Glycemic Control in the Outpatient Setting
“In the outpatient arena, it is even more important to take the patient’s individual goals and other medical issues into account when choosing a target for glycemic control,” Dr. Maynard said. “Hemoglobin A1C is a marker of glycemic control over time. Patients with chronic comorbid conditions and/or older age groups may be at more risk for hypoglycemic events, and less likely to benefit from tight glycemic control. For example, patients with dementia, limited life expectancy, a history of frequent hypoglycemia, or other complicating conditions should probably not have the same glycemic target as a 40-year-old patient with diabetes and limited comorbidity. Therefore, one size (or A1C target) does not fit all, and patients should ask for individualized A1c targets based on their special circumstances,” he said.