Over the last 25 years, more than twice as many patients have been discharged from U.S. hospitals with a diagnosis of diabetes mellitus (DM). In 2006, the number reached an estimated 5.2 million. The increase stems from many factors, including the overall rise in obesity, which parallels the increase in type 2 diabetes. Typically, about 25% of hospital patients have a diagnosis of DM or hyperglycemia during their hospital stay.
Historically, managing hyperglycemia in the hospital has been seen as secondary to managing the admitting diagnosis. But a growing body of literature supports targeted glucose control, because hyperglycemia in hospital patients can prolong lengths of stay, increase the infection risk, and raise mortality. This article, which addresses glucose management in hospital patients who aren’t critically ill, is based largely on guidelines from the American Diabetes Association (ADA) and the American Association of Clinical Endocrinologists (AACE).
Classifying DM in hospital patients
Generally, hyperglycemia in hospital patients is classified as known DM, newly diagnosed DM, or stress hyperglycemia.
- Known DM applies to patients with preexisting type 1, type 2, or gestational diabetes.
- Newly diagnosed DM refers to patients newly diagnosed during their hospital stay who meet ADA diagnostic criteria. A hemoglobin A1c (HbA1c) level of 6.5% or higher indicates DM and reflects an average blood glucose (BG) level of 140 mg/dL. The HbA1c test indicates the average BG level over the preceding 2 to 3 months; an elevated HbA1c level indicates the patient’s BG level was high before admission. Hospital patients with HbA1c levels of 6.5% or higher usually are classified as newly diagnosed, even though their DM probably preceded their hospital admission.
- Stress hyperglycemia in the hospital is illness-related and usually transient, with BG levels normalizing once the patient’s other illness resolves.
Caregivers need to know which type of DM a patient has, as this can greatly influence treatment. For example, if your patient has type 1 DM, avoid withholding insulin doses; this patient doesn’t produce insulin, and withholding insulin can cause severe hyperglycemia, diabetic ketoacidosis, or both. If you’re concerned about an insulin dose potentially causing hypoglycemia, call the prescribing provider to discuss adjusting the insulin dosage.
In noncritically ill patients, the AACE and ADA consensus statement on inpatient glycemic control recommends a preprandial BG target below 140 mg/dL and a random BG target below 180 mg/dL. BG guidelines should be evaluated using clinical judgment based on the patient’s medical condition and current outpatient glycemic control. For instance, for patients accustomed to more intensive glycemic targets in an outpatient setting, a more aggressive treatment regimen may be used to achieve a preprandial level closer to 100 mg/dL, as long as it can be achieved safely with minimal hypoglycemia.
In older patients and those with impaired mental status, hypoglycemia should be avoided. They may especially benefit from current BG targets of 140 to 180 mg/dL, which aren’t particularly stringent.
In the hospital setting, insulin therapy is preferable to oral antidiabetic agents for hyperglycemia treatment, even in patients who’d been using oral agents before admission. Insulin achieves faster glycemic control than oral agents, which can take days or weeks to reach a therapeutic effect. Also, insulin can be adjusted based on the patient’s nutritional status (such as nothing-by-mouth [NPO] status), and it causes fewer adverse effects and contraindications than most oral agents. (See Oral antidiabetic agents and noninsulin injectables by clicking the PDF icon above.)
In noncritically ill hospital patients, subcutaneous insulin most commonly is used for glycemic management. The most effective subcutaneous regimens attempt to mimic normal physiologic insulin production. Physiologic insulin production includes insulin release from the pancreas in two different ways:
- A small amount of insulin is released continuously to suppress hepatic glucose production (called basal insulin).
- Insulin also is released in response to eating, to control the postprandial glucose rise (called bolus insulin).
Basal insulin can be long-acting (insulin glargine or insulin detemir) or intermediate-acting (insulin isophane suspension [NPH]). Or it can be given at a continual basal rate via a continuous subcutaneous insulin infusion pump. Basal insulins are given to mimic physiologic continuous insulin release. Long-acting basal insulins have a duration of about 24 hours and usually lack a significant peak effect, helping to smoothly control hepatic glucose output.
Intermediate-acting insulin (NPH) has a variable peak action of 4 to 12 hours and a duration of 18 to 24 hours. Because of its strong peak effect and duration shorter than 24 hours, NPH insulin doesn’t mimic physiologic basal insulin as closely as long-acting insulins. The hypoglycemia that can occur with NPH may happen when BG levels aren’t typically monitored; thus, hypoglycemia may go unnoticed unless the patient recognizes signs and symptoms and alerts hospital staff. For example, the morning NPH dose peaks a few hours after lunch, which may coincide with the peak of short-acting insulin given at lunch, increasing the hypoglycemia risk. The bedtime NPH dose peaks in the early morning (about 4 a.m.).
Bolus insulins, either rapid-acting (insulin lispro, insulin aspart, or insulin glulisine) or short-acting (regular insulin), are used to control both postprandial glucose rises and hyperglycemia. Rapid-acting insulins generally are preferred because
of their fast onset, earlier peak to better cover postprandial glucose increases, and shorter duration, which avoids overlapping insulin doses when given at every meal. Optimal rapid-acting insulin scales have an order for an insulin dose that covers food eaten, plus a correction dose that covers hyperglycemia. However, rapid-acting insulin scales commonly are ordered as a “correction” for hyperglycemia.
Rapid-acting insulins more closely mimic normal physiologic prandial insulin release than regular insulin, which has a slower onset, later peak, and longer duration (about 6 hours or more). The longer duration of regular insulin means that if it’s given every 4 hours, it overlaps, causing insulin “stacking” and increasing the hypoglycemia risk. Also, the insulin dosage in a typical regular “sliding-scale” insulin regimen is based only on the BG level and is the same whether the patient eats a meal or is on NPO status. (See Timing is everything by clicking the PDF icon above.)
Hospital subcutaneous insulin protocols or order sets that mimic normal physiologic insulin production to best control BG use long-acting basal insulin with fast-acting insulin for coverage of meals and hyperglycemia. Key issues to consider with subcutaneous insulin protocols are minimizing complexity, ensuring adequate staff training, including standardized hypoglycemia treatment, and making guidelines available for glycemic goals and insulin dosing for initiation and adjustment. Nurses are critical to administration of subcutaneous insulin injections and proper protocol implementation.
During a patient’s hospital stay, insulin dosages usually must be titrated daily to achieve glycemic goals. Many factors can influence insulin administration and BG levels, including timing of meal-tray delivery, appetite changes, bedtime snacking, unanticipated NPO status, changes in enteral or parenteral nutrition rates, steroid dosage changes, delays caused by procedures and treatments, and overtreatment of hypoglycemic episodes. Be sure to document factors affecting BG levels, so the care team can make informed decisions for insulin prescribing and customize patient education. (See Continuous subcutaneous insulin infusion by clicking the PDf icon above).
Hypoglycemia episodes and protocols
Hypoglycemia is defined as a BG level below 70 mg/dL; severe hypoglycemia refers to a BG level below 40 mg/dL. If a patient shows a sudden change in mood or mental status, checking the BG level is an appropriate nursing action. In addition to mental changes, hypoglycemia may cause tachycardia, sweating, agitation, elevated blood pressure, myocardial ischemia, angina, cardiac arrhythmias, transiently prolonged QT intervals, seizures, and sudden death.
Many factors can increase a hospital patient’s hypoglycemia risk, including sudden reduction of corticosteroid dosages, altered ability
to report signs and symptoms, reduced oral intake, emesis, new NPO status, inappropriate timing of short- or rapid-acting insulin
relative to meals, reduced I.V. dextrose administration rate, and unexpected interruption of enteral or parenteral nutrition.
Early recognition and treatment of hypoglycemia can prevent adverse outcomes. A hypoglycemia protocol initiated by nursing staff
is crucial to timely hypoglycemia treatment. The protocol should include orders for appropriate treatment. For instance, in patients who can tolerate oral intake, 15 g of rapidly digested carbohydrates (such as 4 oz of fruit juice or regular soda, one tube of glucose paste, or four glucose tablets) are recommended. In patients who are NPO or can’t take oral treatment, dextrose 50% I.V. or glucagon given I.M. or subcutaneously are appropriate treatments.
Patients who’ve had a hypoglycemia episode should be monitored closely for 24 hours afterward because they’re at increased risk for another episode. Notify the oncoming nurse of hypoglycemia episodes that occurred during your shift.
The hypoglycemia protocol should include provider notification to discuss the possible cause of the hypoglycemic event and make insulin adjustments if needed, to avoid additional hypoglycemia. Tracking hypoglycemia episodes and analyzing causes are important performance improvement activities.
Conditions that can trigger hyperglycemia
Glucocorticoid therapy and enteral or parenteral nutrition can lead to hyperglycemia.
Glucocorticoids increase hepatic glucose production and inhibit
glucose uptake into muscle cells, causing hyperglycemia. Approximately 64% of patients with or without DM may develop hyperglycemia when receiving high-dose corticosteroids. Patients with no previous history of DM who receive these drugs should be monitored for BG increases. In patients with preexisting DM, monitor BG levels closely, as a rise is likely.
Insulin therapy is the standard treatment for glucocorticoid-induced hyperglycemia. In patients requiring high glucocorticoid dosages, such as those who’ve received organ transplants, an insulin infusion protocol (IIP) may be the best choice. The IIP should be nurse-driven, and nurses should be well educated about the IIP. Also, the IIP should include frequent glucose monitoring, with insulin dosage adjustments based on the glycemic rate of change; it should also include hypoglycemia treatment guidelines. An adequate nurse-to-patient ratio is important, too, because of the increased testing, monitoring, and interventions required. As glucocorticoid dosages are tapered, the patient may be transitioned to subcutaneous insulin therapy.
Studies on best practices for treating glucocorticoid-induced hyperglycemia are limited. But because glucocorticoids’ major effect is to inhibit glucose uptake in muscle and adipose tissue, causing a greater postprandial BG rise, using a special correction scale with rapid-acting insulin may help control the glucose level.
Some evidence suggests NPH insulin may be useful in patients receiving glucocorticoids. Clore and Thurby-Hay proposed a simple equation for use of intermediate-acting NPH insulin, in which the NPH dosage is based on the prednisone dosage and the patient’s weight. Overall, keep in mind that insulin dosages should be adjusted as glucocorticoid dosages are decreased, to avoid hypoglycemia. When caring for patients receiving glucocorticoids, consult hospital providers trained and experienced in glucose management.
Enteral nutrition (EN) formulas generally are high in carbohydrates, contributing 45% to 90% of total daily calories. Various types of insulin regimens can be used to control BG levels when a patient’s receiving EN. Basal and bolus subcutaneous insulin can be used together.
Be aware that because of its long duration, long-acting insulin may increase the hypoglycemia risk if EN is disrupted, so be sure to use it thoughtfully. One approach is to start basal insulin at a lower dosage and use bolus insulin for correction dosages. Then the basal insulin dosage can be increased slowly to help achieve euglycemia.
For cyclic EN, treatment may involve an intermediate-acting insulin at the start of EN administration, with correctional rapid- or short-
acting insulin as needed. Bolus tube feedings can be treated as meals and covered with rapid- or short-acting insulin at bolus time.
Hyperglycemia from parenteral nutrition (PN) may stem from multiple factors, including patient age, clinical stability, and dextrose infusion rate. Various treatment regimens are available, including subcutaneous insulin, insulin added to the PN solution, or both. No controlled trials have established which strategies are best. Adding incremental insulin to total PN is an option, but this approach may take days to determine the correct insulin dosage. Subcutaneous insulin protocols may include basal insulin, bolus insulin, or both.
Preventing hypoglycemia in patients receiving enteral or parenteral nutrition
To prevent hypoglycemia in patients receiving insulin along with EN or PN, obtain a PRN order to start I.V. dextrose solution if EN suddenly is interrupted (as when a feeding tube is pulled out or EN is withheld due to high residuals). Commonly, the dextrose solution is started at the same rate at which the nutritional formula was infusing.
Preparation for discharge to an outpatient setting should begin at the time of the patient’s hospital admission. Obtaining an HbA1c value (if one isn’t available from the last 3 months) helps evaluate the patient’s preadmission glucose control and aids discharge planning for the medication regimen. For example, staff can anticipate that a patient on oral antidiabetics who has an elevated HbA1c level may need to receive insulin on discharge and will need insulin education and time to practice self-injections.
Early assessment of the patient’s knowledge of diabetes self-management allows you to tailor education to the patient’s needs during the hospital stay. Before providing education, assess for and address any barriers to learning, such as reductions in cognitive ability, literacy level, motor skills, or visual acuity. For example, a patient with poor visual acuity may need to use an insulin pen at home, rather than an insulin bottle and syringe.
diagnosis, BG goals and self-monitoring at home, hyperglycemia and hypoglycemia management, consistent eating patterns, medication administration, sick-day management, syringe use and disposal, and identification of care providers for follow-up.
Besides providing verbal instructions, offer easy-to-understand written instructions to help the patient retain the material and clarify medication administration. For additional self-management education, provide a referral to an outpatient diabetes education program. Instruct the patient to keep follow-up appointments with the primary care provider or an endocrinologist, preferably within 1 month of discharge.
Patient education and advocacy
Although hospital patients with hyperglycemia are a unique population, they certainly aren’t rare. The Centers for Disease Control and Prevention predicts that DM cases may triple within the next 40 years, so DM is sure to be an increasing part of hospital nursing care. Understanding glycemic goals and best practices for insulin regimens helps nurses advocate for patients with DM to get the best possible care, and can improve hospital outcomes. And because self-management is a large part of diabetes care at home, nurses also play a key role in providing education and opportunities to practice self-care.
Visit www.AmericanNurseToday.com for a complete list of references.
Stacey A. Seggelke is a clinical nurse specialist in the adult diabetes program at the University of Colorado Denver School of Medicine. Bridget Everhart is an inpatient diabetes educator at the University of Colorado Hospital in Aurora.