ACP issues guideline on glucose management for hospitalized patients

ACP's director of clinical policy discusses a new guideline for intensive insulin therapy.

It's been almost a whole decade now that blood glucose management has been a cause of turmoil and debate in hospital medicine. In November 2001, Van den Berghe and colleagues published a study in the New England Journal of Medicine finding that intensive insulin therapy with glucose targets of 80 to 100 mg/dL reduced morbidity and mortality in surgical intensive care unit (ICU) patients. Many hospitals and hospitalists changed their management practices accordingly.

Then, in the late 2000s, other studies, including the NICE-SUGAR trial, began to provide evidence that intensive insulin with a normoglycemic target harmed patients. Physicians, hospitals and guidelines began shifting away from intensive control in light of the emerging evidence against it, but questions and uncertainties remained.

In an effort to answer some of those questions, ACP experts recently gathered the available evidence and developed a guideline regarding intensive insulin therapy for general and ICU inpatients, published in the Feb. 15 Annals of Internal Medicine. Lead author Amir Qaseem, PhD, FACP, director of clinical policy for the College, recently spoke to ACP Hospitalist about the guideline.

Q: Why were these guidelines needed?

A: Hyperglycemia is a very common finding among medical and surgical patients with or without diabetes. It's associated with increased morbidity, mortality and cost, and also poor immune response, increased cardiovascular events, thrombosis, etc. Most clinicians make an effort to prevent and control this hyperglycemia in the inpatient setting.

But what is the optimal target for the hospitalized patient in terms of glycemic control? We wanted to address this issue. We address the management of hyperglycemia and evaluate the benefits and harms associated with the use of intensive insulin therapy to achieve tight glycemic control in hospitalized patients with or without diabetes.

Q: What are you seeing in current practice regarding inpatient glucose management?

A: The earlier trials that came out and strongly suggested that intensive insulin therapy leads to better outcomes—it's something that has been widely adopted in hospitals and in the ICU. But now there has been new evidence suggesting otherwise. This has caused some confusion on what are the optimal targets. Is targeting normal glycemia a good strategy or not a good strategy? Does it lead to better outcomes in patients or not? In this paper, we define intensive insulin therapy as blood glucose between 80 and 110 mg/dL (4.4-6.1 mmol/L).

Q: Two of the new guideline's three recommendations tell physicians what not to do, rather than specifying what glycemic levels they should target. Why is this?

A: You're right, some of the places we are saying what not to do because that's also important. You can end up harming patients in doing certain things.

In the first recommendation, ACP recommends not using intensive insulin therapy to strictly control blood glucose in non-surgical ICU/medical ICU [SICU/MICU] patients with or without diabetes, which is a strong recommendation based on moderate-quality evidence. What we found is that there is no reduction in mortality with the target blood glucose levels between 80 and 180 mg/dL (4.4-10 mmol/L) compared to higher levels. The [lower targets] did not have any impact in patients who had myocardial infarction, stroke, brain injury or folks who are under perioperative care.

On the other hand, harms are likely to increase when target blood sugar levels are too low. The reason we did not specify the target level is because in all these trials, the range varied. However, we found that avoiding targets less than 140 mg/dL (7.8 mmol/L) definitely should be a priority, because that was leading to more harm. It leads to hypoglycemia and increased length of stay and even increased mortality, although the consequences of increased hypoglycemia were not really clear. Again, it goes back to the [evidence being] so unclear. For non-SICU/MICU patients, we do not really specify a target.

Q: How about the second and third recommendations?

A: For patients who are in the SICU and MICU, we recommend not using intensive insulin therapy to normalize blood glucose in patients with or without diabetes. This is a strong recommendation, based on high-quality evidence, because what we found is having targets between 80 and 110 mg/dL (4.4-6.1 mmol/L), you're doing more damage than good. It's associated with increase in mortality.

But what target level do you use in these patients? We recommend that between 140 and 200 mg/dL (7.8-10 mmol/L) is a reasonable target. That's a weak recommendation, based on moderate-quality evidence. This is based on the levels utilized in the trials, so that's all we can base our recommendation on. We do not have a precise and narrow range for blood glucose level. What we found from the trials is if you keep the blood glucose levels between 140 and 200 mg/dL (7.8-10 mmol/L), it's associated with similar mortality outcomes as if you had it between 80 and 110 mg/dL (4.4-6.1 mmol/L).

Q: Is additional research needed to come to more definitive conclusions?

A: Yes, there's always more research that can be done. We probably can have more trials on these insulin protocols in non-ICU settings. It's still not really clear what is the impact of hypoglycemia or what specific target levels can be used in ICU settings and non-ICU settings. More of these studies that can narrow down the range of the target blood glucose level will always be helpful.

Q: The evidence and guidance on inpatient glucose management have varied widely in recent years. Has the pendulum on this issue returned to a middle point where it's likely to remain for some time?

A: This is all dependent on the evidence. In the past, the evidence showed that tight glycemic control leads to better outcomes. But now we have enough evidence that shows that tight glycemic control is actually doing more harm than benefit.

I always say our guidelines are based on current evidence. If some other high quality trials came out that show something completely different, that may change something. But I do think that tight glycemic control is definitely harmful. There is high quality evidence that goes in that direction. But for now, this guideline is based on best available current evidence.

Q: Any other advice you would offer hospitalists on this subject?

A: Intensive insulin therapy is associated with a high risk for severe hypoglycemia, especially when the blood glucose level target is less than 6.7 mmol/L (120 mg/dL). Given that the harms outweigh the benefits, we should not routinely implement strict targets for blood glucose control in hospitalized patients.