Am Fam Physician. 2006;73(4):591-594
Diabetes has reached epidemic proportions in the United States. This is true whether patients are outpatients or inpatients; diabetes is the fourth most common comorbid condition complicating all hospital discharges. In 1997, diabetes was present in 9.5 percent of all patients discharged from hospitals and in 29 percent of patients undergoing cardiac surgery.1 Diabetes is associated with a two- to four-fold increase in hospitalization rates; its presence increases the length of hospital stays by one to three days, depending on the admission diagnosis. In-hospital hyperglycemia is an important marker of poor clinical outcomes and mortality in patients with or without diabetes.2
Although several organizations have issued guidelines for outpatient management of diabetes, no guidelines have been formulated for inpatient management. The American College of Endocrinology, in collaboration with several other organizations, has developed a consensus statement on inpatient diabetes and metabolic control.3 Because so many groups are involved in the care of hospitalized patients who may experience hyperglycemia, a multidisciplinary group was formed to develop a single set of standards and guidelines; the recommendations were cosponsored by the American Association of Diabetes Educators, the American Diabetes Association, the American Heart Association, the American Society of Anesthesiologists, the Society of Critical Care Medicine, the Society of Hospital Medicine, the Society of Thoracic Surgeons, the Endocrine Society, and the American College of Cardiology. The rationale for developing guidelines to achieve tight glycemic control in the hospital was based on increasing evidence that metabolic regulation of hyperglycemia translates into improved outcomes in patients with diabetes and in those who develop hyperglycemia in the hospital (see accompanying table4–10).
Hyperglycemia |
Complications of MI4 |
Complications of stroke5 |
Complications of vascular and cardiac surgery6 |
Mortality in critically ill patients7 |
Mortality after coronary bypass surgery |
Aggressive insulin therapy |
Improved cardiac surgery outcomes8; mortality risk reductions of 50 percent or greater have been observed in patients treated with insulin infusions for the first three postoperative days. |
Improved ICU outcomes8–10 (e.g., 34 percent reduction in hospital mortality rates, 46 percent decrease in rates of sepsis, 41 percent decrease in rates of acute renal failure, 50 percent reduction in transfusions, 44 percent reduction in critical-illness polyneuropathy)9 |
Improved post-MI outcomes4 (e.g., 28 percent improvement in long-term survival rates in diabetic patients with acute MI) |
Reductions in infection rates, intubation times, length of hospital stays, and cost |
The consensus panel reviewed research with the original investigators to formulate standards for diabetes management in the hospital and to suggest techniques by which these goals and targets may be achieved. The data clearly show that all patients, especially those previously undiagnosed with diabetes, benefit from intensive management of hyperglycemia with insulin. The panel concluded that patients in intensive care units should have a target glycemic level of 110 mg per dL (6.1 mmol per L) and that medical and surgical patients should have a target preprandial level of 110 mg per dL and a maximal level of 180 mg per dL (10.0 mmol per L).
Insulin, whether administered intravenously or subcutaneously, was determined to be the primary means of effective glycemic control in the hospital setting. The use of standardized protocols for continuous intravenous insulin therapy is associated with improved glycemic control and low rates of hypoglycemia; the use of sliding scales is discouraged.
The next great challenge will be implementation of these standards. Hospital systems will have to change to achieve the goals defined above. Hospital- and ward-wide protocols for administration and monitoring of blood glucose levels and insulin infusions will be needed, as will protocols for risk management for hypoglycemia. Furthermore, a broad base of medical and surgical specialists must participate if we are to be successful in reducing diabetic inpatient mortality and morbidity rates; greater integration of care across units and the support of nursing and pharmacy staff will be needed. Clearly, there is a role for the family physician in the implementation and coordination of these guidelines.
The occurrence of significant hyperglycemia in the hospital will require close follow-up after discharge. In patients with previously diagnosed diabetes and an elevated A1C level, the preadmission diabetes care plan requires revision. In patients without previously diagnosed diabetes, the differentiation between hospital-related hyperglycemia and undiagnosed diabetes requires follow-up testing (e.g., fasting blood glucose measurements, two-hour oral glucose tolerance tests) once patients are metabolically stable. Outpatient diabetes self-management training is critical to minimize the risk of future complications.