Much Ado About Inpatient Insulin Therapy

A CME symposium supported by an unrestricted educational grant from Eli Lilly and Company

Claresa S Levetan, MD, FACE
Silvio Inzucchi, MD
Guillermo E Umperriez, MD, FACP, FACE

Reported by Joelle Escoffery, PhD

Hyperglycemia in the inpatient in the inpatient setting is the cause of considerable mortality, both for patients with diagnosed diabetes and patients with newly discovered hyperglycemia. The risk of inpatient mortality is far greater among patients with new hyperglycemia compared with either normoglycemic patients or patients with a known diabetes history. Additionally, inpatient hyperglycemia increases the risk of sepsis, dialysis, transfusion, increased duration of hospital stay, increased need for ventilator support, and poorer outcomes following stroke. There are a variety of mechanisms by which hyperglycemia can lead to poorer outcomes, including alterations in immune function, deleterious cardiovascular effects, increased thrombosis, increased inflammation, endothelial cell dysfunction, and neuronal damage. The beneficial effect of inpatient insulin therapy is due in part to correcting hyperglycemia, but there is also evidence to suggest that insulin itself has beneficial effects, including improvements in endothelial cell dysfunction, stimulation of NO, and anti-inflammatory effects. The American Diabetes Association (ADA) and the American Association of Clinical Endocrinologists (AACE) both have established guidelines for the management of inpatient hyperglycemia. Although the targets vary somewhat between the 2 organizations, ICU targets are lower (typically below 110 mg/dL) and less stringent targets are used in noncritical care settings. Although there are many strategies available for administering insulin in the inpatient setting, use of the sliding scale is not recommended, as it is associated with an increased risk for both hyperglycemia and hypoglycemia.

Hyperglycemia has been shown to be associated with negative outcomes in a variety of critical care settings, including the medical ICU, the cardiothoracic ICU, and the CCU. Large clinical trials such as the Portland Diabetes Project, the Diabetes Insulin-Glucose Infusion in Acute Myocardial Infarction (DIGAMI), and the Leuven Study all clearly demonstrated the beneficial effect of tight glucose control in the inpatient setting. The DIGAMI 2 and Leuven 2 studies were less clear in their findings, with the former being stopped early due to poor recruitment, relatively minor between group differences, and a lower overall mortality rate. Although the Leuven 2 study showed a beneficial effect of tight inpatient glycemic control on mortality among patients in the ICU for longer than 3 days, patients who were in the hospital less than 3 days actually showed increased mortality in the intensively-treated group. There are numerous barriers to achieving tight glycemic control in the ICU setting, including fear of hypoglycemia, lack of appreciation of the benefits of tight glucose control, and suboptimal infrastructure. The ideal insulin protocol should be standardized, easily ordered, easily implemented, effective, and safe.

Although there is a sizeable body of research addressing hyperglycemia control in the critical care setting, there are very little data available to guide insulin therapy on the general ward. As is the case in the ICU setting, mortality is considerably higher among patients with new hyperglycemia compared with patients who are normoglycemic or who have a known diabetes history. Surgery among patients with hyperglycemia is associated with a greater infection rate, impaired wound healing, increased length of hospital stay, and higher operative and in-hospital mortality. A target of <180 mg/dL is recommended by the ADA and AACE; in spite of recommendations, two thirds of hospitalized patients are not optimally controlled. There are a different set of barriers associated with meeting glycemic targets in the noncritical care setting, including holding therapeutic regimen, use of the insulin sliding scale, and delays in initiating therapy. Basal-bolus therapy offers greater flexibility than conventional therapy and is an effective strategy in this group. Finally, hospitalized patients can be educated on proper diabetes care, including insulin administration, blood glucose monitoring, treating hypoglycemia, meal planning, and sick day management.