Secondary Logo

Journal Logo

A balancing act: Achieving glycemic control in hospitalized patients

Kubacka, Beata MSN, RN, CDN, RD

doi: 10.1097/01.NURSE.0000438706.66979.52
Feature: CE Connection


EVERY YEAR, all over the world, the number of people with diabetes increases. According to the CDC, diabetes affects 25.8 million people in the United States. Of these, an estimated 7 million are undiagnosed.1,2 (See Diabetes takes a heavy toll.)

At any given time, 12% to 25% of hospitalized patients have diabetes or some degree of hyperglycemia.3 Patients who experience hyperglycemia during a hospitalization tend to have a longer length of stay, averaging 1 to 3 additional days.3 Poor inpatient glycemic control is associated with increased morbidity, mortality, hospital admissions, length of stay, and medical costs.4

Because most hospitalized patients with diabetes aren't admitted for treatment of diabetes as a primary diagnosis, diabetes isn't the focus of the hospitalization.4 As a comorbidity, diabetes has a great impact on the course of the hospitalization and patient outcomes, so appropriate management is a critical aspect of inpatient care.

This article presents the latest evidence-based guidelines about caring for hospitalized adult patients with diabetes. By tapping into the following targeted questions and answers, nurses can make sure patients obtain appropriate monitoring and treatment, avoid hypo- and hyperglycemia, and receive timely follow-up care.

Back to Top | Article Outline

What are blood glucose targets for inpatients?

The American Diabetes Association (ADA) and the American Association of Clinical Endocrinologists (AACE) have issued a consensus statement on inpatient glycemic control. In all patients, the goal is to avoid hypoglycemia, defined as a blood glucose level of less than 70 mg/dL.5 (See Targeting inpatients.) Blood glucose goals for inpatients are as follows:

  • for critically ill patients, 140 to 180 mg/dL
  • for noncritically ill patients, less than 140 mg/dL fasting and less than 180 mg/dL for random blood glucose.

Higher levels may be acceptable for terminally ill patients or those with severe comorbidities.6 To improve patient outcomes and glycemic management, hospital staff must properly develop and use insulin protocols.

Back to Top | Article Outline

What do nurses need to know about blood glucose monitoring?

All patients with diabetes should have a prescription for blood glucose monitoring with results available to members of the healthcare team. Glucose monitoring should also be initiated in any patient receiving therapy associated with an increased risk for hyperglycemia, such as enteral or parenteral nutrition or therapy with glucocorticoids, octreotide, or immunosuppressants.6

At a minimum, patients receiving oral nutrition require premeal and bedtime blood glucose monitoring. For patients not receiving any kind of nutrition (oral or enteral), blood glucose monitoring is recommended every 4 to 6 hours. Patients receiving I.V. insulin infusions require more frequent blood glucose monitoring, every 30 minutes to 2 hours.6

Keep these points in mind:

  • Never share lancing devices among patients because of the risk of transmitting blood-borne diseases.6
  • Significant discrepancies among capillary, venous, and arterial plasma samples have been observed in patients with low or high hemoglobin concentrations or hypoperfusion, and in the presence of interfering substances, particularly maltose, contained in immunoglobulins. In addition, the FDA allows for point-of-care meters to have a +/- 20% error rate. Any glucose result that doesn't correlate with the patient's clinical status should be confirmed through a venous blood glucose level.6
Back to Top | Article Outline

Why is hyperglycemia a problem for hospitalized patients?

Hyperglycemia is defined as any blood glucose level above 140 mg/dL.6 Substantial evidence links hyperglycemia during hospitalization to poor patient outcomes.6 Although the exact prevalence of hyperglycemia in the hospital setting isn't known, the evidence suggests that most patients experience at least one blood glucose value above 250 mg/dL.3

Adverse outcomes associated with inpatient hyperglycemia include the following:

  • increased rates of healthcare-associated infection and sepsis
  • longer periods when mechanical ventilation is required
  • increased risk of mortality after myocardial infarction or cardiac surgery
  • greater risk of acute kidney injury and poor wound healing
  • increased length of stay.4

Patients with new-onset hyperglycemia have a higher risk of mortality compared to patients with a history of diabetes and those with normoglycemia.1 Surgical patients with at least one blood glucose level greater than 220 mg/dL on postoperative day one have a risk of postoperative healthcare-associated infection that's 2.7 times higher than that of patients whose blood glucose level is maintained at less than 220 mg/dL, regardless of history of diabetes.1 Research shows that patients with blood glucose levels of 200 to 300 mg/dL and more than 300 mg/dL have twice and almost three times the odds of mortality, respectively, than normoglycemic patients.3

Clearly, stress-induced hyperglycemia is associated with a higher risk of complications and mortality. When hyperglycemia has been documented, initiating glycemic control should be considered.

Back to Top | Article Outline

How should hyperglycemia be managed in a critically ill patient?

According to the ADA and AACE consensus statement, critically ill patients should be placed on an insulin infusion to control hyperglycemia, starting with a threshold no higher than 180 mg/dL. The goal of an insulin infusion is to maintain glucose levels between 140 and 180 mg/dL. Targets below 110 mg/dL are no longer recommended.6

Investigators for the Normoglycemia in Intensive Care Evaluation and Survival Using Glucose Algorithm Regulation (NICE-SUGAR) study observed two groups in an ICU setting. In one group, intensive glucose control (81 to 108 mg/dL) was attempted. In the other, the goal was moderate glucose control (144 to 180 mg/dL). The investigators found that patients in the intensive glucose control group had increased mortality and more frequent severe hypoglycemia. The NICE-SUGAR investigators ultimately proposed that no additional benefit was gained from lowering blood glucose below a target range of 140 to 180 mg/dL. This study, along with a study by Van den Berghe et al. that found similar results, is the basis for the most recent recommendations by the ADA and AACE task force on management of inpatient hyperglycemia.1

Most of the published insulin infusion protocols are outdated because their glycemic targets are set too low, according to the latest guidelines. All organizations should evaluate their insulin infusion protocols for compliance with current recommendations. The safest insulin infusion protocols require frequent blood glucose monitoring and take into account the patient's current and previous blood glucose levels and rate of infusion to guide changes in the infusion rates. Protocols should be simple to prescribe and follow, achieve glycemic goals in a reasonable time frame, maintain goals with minimal risk for hypoglycemia, and include instructions for transitioning to subcutaneous insulin. They should also include guidelines for treating hypoglycemia.7,8



The half-life of I.V. insulin is only minutes. In patients with normal renal function, I.V. insulin is cleared within 20 minutes.9 This is why patients need subcutaneous insulin before the infusion is stopped. The recommendations are that regular or rapid-acting insulin be given 1 hour before the infusion is stopped, or intermediate or long-acting insulin be given 2 to 4 hours before the infusion is stopped. Failure to allow for overlap in the I.V. and subcutaneous insulin can result in rapid development of hyperglycemia and, in patients with type 1 diabetes, a risk for diabetic ketoacidosis due to the lack of circulating insulin.9

Insulin infusions are recommended outside the ICU only if enough staff is available to monitor the patient closely and staff is well trained to support the safe implementation of the protocol, particularly the frequent monitoring of blood glucose levels. Potential candidates for an insulin infusion include a patient with type 1 diabetes who's N.P.O., a patient with uncontrolled gestational diabetes, and a patient with persistent hyperglycemia (>300 mg/dL) despite aggressive scheduled subcutaneous insulin regimens.7

Those receiving oral nutrition while receiving an insulin infusion may have increased risk for hyperglycemia and hypoglycemia. Infusion protocols are typically developed for fasting patients and aren't constructed to prevent postprandial blood excursions. As a result, hypoglycemia can occur when the insulin infusion dose is increased in response to meal-related elevations. Supplemental subcutaneous insulin administered separately from the infusion may help control postprandial elevations without increasing the risk of hypoglycemia.7

Back to Top | Article Outline

Why do we give different types of insulin?

In the ICU, I.V. is the preferred route for insulin, but outside the ICU, the subcutaneous route is preferred. Scheduled subcutaneous insulin administration should be modeled after normal pancreatic insulin secretion. In a person without diabetes, a steady amount of insulin is secreted throughout the day; this is called basal insulin. Basal insulin helps control blood glucose between meals and during sleep. Bolus insulin is the extra insulin the pancreas releases in response to postprandial elevations in blood glucose levels specifically related to carbohydrates.8

Insulin administration should include basal, nutritional (or prandial) bolus, and correction doses to account for hyperglycemia above the desired target. The goal is to adjust basal and nutritional bolus so that correction doses can be minimized. Day-to-day decisions about insulin dosing must take into account the patient's clinical and nutritional status and concurrent medications that can affect blood glucose levels. Nurses need to be aware that insulin shouldn't be held if premeal blood glucose is in good control because prandial elevations need insulin coverage.6

Although monotherapy sliding scale insulin (SSI) continues to be the most frequently prescribed insulin regimen, it's no longer recommended because it offers only a reactive and ineffective approach to treating hyperglycemia after it's occurred. SSI fails to meet a patient's basal insulin requirements, and meeting these requirements is an absolute necessity in patients with type 1 diabetes. Many patients with type 2 diabetes may also need to have their basal insulin requirements met to adequately manage their blood glucose levels.3,4,6

(See Getting to know basal and nutritional bolus/correction insulin.)

Back to Top | Article Outline

Why should oral hypoglycemic medications be discontinued upon hospitalization?

Insulin is the cornerstone of inpatient glycemic management. Oral medications should be discontinued upon hospitalization because they're difficult to titrate for tight glycemic management and are contraindicated in many acute conditions.10 In addition, many hospitalized patients can't take oral medications due to illness, inability to eat, or N.P.O. status.



  • Metformin is contraindicated in patients with renal impairment, hypoperfusion, or hypoxemia, and for patients undergoing diagnostic testing requiring intravascular contrast media.
  • Sulfonylureas, such as glyburide, increase the risk of hypoglycemia, especially if nutritional intake is interrupted.
  • Thiazolidinediones, such as pioglitazone, are associated with peripheral edema, and should be used with caution, especially in patients with heart failure.8
Back to Top | Article Outline

How are insulin dosages prescribed to meet a patient's needs?

Clinicians have multiple approaches to insulin dosing, and each patient has a unique response to insulin. For those previously on insulin, preadmission insulin doses can serve as a guide for inpatient treatment. If the patient is insulin-naive, the first step is to calculate a total daily dose (TDD) of insulin. TDD includes all the insulin required in a day, both basal and bolus, and can be initiated based on body weight. Starting at lower doses is recommended for patients who are insulin-sensitive (patients who are new to insulin, thin [that is, body mass index of less than 18.5], older than age 70, or with stress hyperglycemia) and at the higher doses for those who are insulin-resistant (patients who are obese, take corticosteroids, or have uncontrolled diabetes mellitus). Stress hyperglycemia can occur due to illness. This stress response results from the release of counterregulatory hormones, such as cortisol, glucagon, and adrenaline, which raise blood glucose by promoting the breakdown of glycogen in the liver. Patients with renal dysfunction may need a lower insulin dosage due to impaired insulin clearance.

Of the calculated TDD, 50% should be provided as basal insulin and 50% as a bolus to cover nutritional elevations. In addition to the bolus insulin, correction insulin should be prescribed to cover any hyperglycemia.8,9

Daily titration is often needed to reach glycemic goals. Looking at the difference between bedtime and fasting glucose levels can help assess the adequacy of basal insulin: If the basal insulin dose is correct, the bedtime and fasting glucose levels won't differ by much. If the fasting blood glucose is significantly higher than the bedtime value, an increase in basal insulin dose is suggested to cover hepatic glucose production. If the fasting level is significantly lower, a decrease in basal insulin is suggested.8 To maximize absorption when the dose is greater than 50 units, basal insulin should be given as two injections.9

Bolus insulin doses can be evaluated by looking at the prelunch, predinner, and bedtime glucose levels. If the bolus dose is adequate and about 4 hours have passed since the last dose of insulin, the blood glucose goal is less than 140 mg/dL. Two-hour postprandial checks, with a goal of less than 180 mg/dL, can also be used to determine the adequacy of the bolus dose.8

Ideal insulin regimens require at least daily reassessment and adjustment to meet the patient's changing metabolic needs. Things to consider include the amount of insulin required in the last 24 hours, the patient's response, and the patient's nutritional status.4

Back to Top | Article Outline

How is insulin adjusted when patients aren't eating?

Changes in patients' carbohydrate intake can prove challenging because about half of insulin is provided in response to carbohydrate intake. In the hospital, blood glucose levels can be unpredictable due to changes in patients' meal intakes and medical status. Even if they're not eating, patients with type 1 diabetes should always receive basal insulin to prevent ketosis. I.V. dextrose may also be prescribed to support blood glucose levels.

Anticipated length of N.P.O. status needs to be a factor in determining insulin dosing for a patient with type 2 diabetes. When the duration of N.P.O. status is expected to be short, the general rule is to continue basal insulin while removing the bolus component and continuing with correction doses. Using basal insulin in patients with extended N.P.O. status poses a risk for prolonged hypoglycemia and may not be advisable without the addition of I.V. dextrose (such as I.V. dextrose 5% in 0.9% sodium chloride solution or D5W, depending on electrolyte status). In patients who are N.P.O. for extended periods, short-acting insulin may be given every 6 hours to prevent hyperglycemia.9

Back to Top | Article Outline

Can patients keep their insulin pump in the hospital?

An insulin pump delivers continuous subcutaneous insulin. In most cases, rapid-acting insulin is administered throughout the day to deliver both basal and bolus doses. A basal dose is delivered in regular intervals, usually every hour, while a bolus dose is delivered by the patient as needed. Patients deliver bolus doses based on the target blood glucose range and insulin-to-carbohydrate ratios.6,8

Patients with continuous subcutaneous insulin infusion (CSII) pump therapy can continue to self-manage provided they have the mental and physical capacity and the supplies to do so. Because of the many types of insulin pumps on the market, it's not cost-effective for hospitals to stock supplies.6,8

For adult patients, it's not advised that a family member manage an insulin pump in the hospital setting. Hospitals should have policy and procedures to guide inpatient CSII therapy.6,8

Nurses should document basal and bolus doses at least once a day. Any change in status or need for procedures such as diagnostic imaging or surgery may warrant the discontinuation of the pump.6,8 If the insulin pump is discontinued in the hospital, the patient will require a prescription for subcutaneous insulin.

Back to Top | Article Outline

How can hypoglycemia be avoided?

Many practitioners undertreat hyperglycemia to avoid hypoglycemia. Fear of causing hypoglycemia when using basal insulin is generally unfounded; the exception is patients with a prolonged N.P.O. status in the absence of a carbohydrate source.

The most common cause of hypoglycemia is inadequate adjustment of insulin when a patient's oral intake is altered. Intensive daily monitoring of blood glucose levels and oral intake and adjusting the medication regimen as indicated can prevent hypoglycemic events.4 (See What contributes to hypoglycemia?)

Back to Top | Article Outline

Why is hypoglycemia dangerous? What should nurses do if a patient is hypoglycemic?

Hypoglycemia is defined as a blood glucose level less than 70 mg/dL, the threshold for the release of counterregulatory hormones. The prevalence of hypoglycemia was reported to be 5.7% of all point-of-care blood glucose tests in a 2009 survey of 575 hospitals.4,11

At levels of 50 mg/dL, cognitive impairment begins.6 Severe hypoglycemia in hospitalized patients has been defined as less than 40 mg/dL. As with hyperglycemia, hypoglycemia is linked to adverse outcomes, including cardiovascular and cerebrovascular events, patient falls, seizures, coma, and death.8 Cardiovascular events include tachycardia and other dysrhythmias, hypertension, myocardial ischemia, angina, and QT interval prolongation. Even mild hypoglycemia can have severe consequences in patients with underlying cardiac or cerebrovascular disease.

Hypoglycemia can occur in patients whether or not they have diabetes. Early detection and treatment are crucial, and nurses must initiate hypoglycemia protocols to promptly treat hypoglycemia. The recommendation for patients who can tolerate oral intake is 15 g of rapidly digested carbohydrate (4 oz [120 mL] juice or regular soda, 1 tube of glucose gel, or 3 to 4 glucose tablets); if blood glucose is below 50 mg/dL, 30 g of carbohydrate may be given. For patients who are N.P.O. or who can't tolerate oral intake, administer 50 mg (25 g) dextrose 50% I.V. push or glucagon injection.8

Recheck blood glucose 15 minutes after treatment to assess the effectiveness of treatment. If blood glucose remains low, re-treat the patient as indicated until levels have stabilized.9 Don't provide meals until blood glucose has risen to an acceptable level.

Making changes to the antidiabetic regimens shortly after hypoglycemia has occurred can prevent further episodes.3 The insulin regimen should be reassessed if blood glucose falls below 100 mg/dL. Modifications are required when blood glucose values are less than 70 mg/dL unless these events can be explained by other factors, such as missed meals.

Maintaining appropriate care requires effective communication between healthcare providers and nurses.6,11 Tracking and analyzing causes for hypoglycemia are important quality improvement activities.

Back to Top | Article Outline

What's the benefit of evaluating A1C?

Hospitalization is an opportunity for clinicians to evaluate long-term glucose control and initiate new therapy in patients previously diagnosed with diabetes. It's also an opportunity to diagnose diabetes and initiate long-term care in those previously undiagnosed.3

A1C testing is the gold standard for assessing glycemic control over several months. It can help clinicians determine if the patient is experiencing stress-induced hyperglycemia, has previously undiagnosed diabetes, or in the case of a patient already diagnosed with diabetes, is managing diabetes effectively. Hemolysis and recent blood transfusion (within 3 months) are examples of when A1C results are unreliable.9

The A1C reflects average glucose concentration for about 3 months, the average lifespan of a red blood cell. A1C values of 6.5% or greater indicate diabetes in a previously undiagnosed patient. The goal for patients with diabetes is to maintain A1C at 7.0% or less. Elevated A1C indicates poor long-term glycemic control.6

Keep in mind that frequent episodes of hypoglycemia will decrease the mean A1C level. A discussion with the patient about previous glucose monitoring and frequency of hypoglycemia may be warranted if questions about A1C results arise.

An A1C level greater than 7.0% has been associated with increased risk for microvascular and macrovascular complications. Microvascular complications include retinopathy, nephropathy, and neuropathy. Macrovascular complications include coronary artery disease, stroke, peripheral arterial disease, heart failure, and cardiomyopathy.

Back to Top | Article Outline

What follow-up will the patient need?

Transitioning a patient from inpatient to outpatient care may require adjusting the insulin regimen shortly after discharge because patients may have been treated at higher or lower doses than they required before admission. An inadequate adjustment may put the patient at risk for hyper- or hypoglycemia shortly after discharge, which could lead to readmission.10

Patients with hyperglycemia in the hospital without a prior diagnosis of diabetes will require follow-up testing and care after discharge. Follow-up within 1 month with a primary care provider or endocrinologist is recommended.5,6,8

Diabetes discharge education should include these basic “survival skills”:

  • understanding the diabetes diagnosis, including the recognition, treatment, and prevention of hyperglycemia and hypoglycemia
  • self-monitoring of blood glucose levels
  • medications
  • meal planning
  • how to manage diabetes when sick
  • a follow-up plan of care.5,6,8

Diabetes education is essential for self-management. Both face-to-face and printed instructions should be part of the teaching plan.5,6,8 Referral to an outpatient diabetes center for follow-up education provided by certified diabetes educators will help patients meet their self-care needs.

Back to Top | Article Outline

Facing the challenge

Glycemic control becomes even more challenging during times of hospitalization. The bottom line? Patients will feel better, return to health more quickly, and avoid long-term complications when their blood glucose levels are maintained at goal levels.

Back to Top | Article Outline

Diabetes takes a heavy toll

The risk of death among people with diabetes is about twice that for people of similar age who don't have diabetes. Diabetes, the seventh leading cause of death in the United States, is the leading cause of chronic kidney disease, nontraumatic lower limb amputations, and new cases of blindness, as well as a major cause of cardiovascular disease. Patients with diabetes have two times the risk of myocardial infarction and stroke compared to the general population and have worse outcomes, including mortality, after surviving a cardiovascular event. People with diabetes have more frequent and longer hospitalizations.1,2,4,7

Estimated total direct and indirect costs of diabetes in 2007 were $174 billion, which rose to $245 billion in 2012, reflecting a 41% increase in 5 years. The average medical cost is 2.3 times higher than for people without diabetes.1,2

Back to Top | Article Outline

What contributes to hypoglycemia?6,11

Nutritional causes

  • Changes in nutritional intake
  • N.P.O. status
  • Emesis
  • Unexpected interruption of enteral or parenteral nutrition
Back to Top | Article Outline

Drug-related causes

  • Sulfonylureas
  • Excessive insulin
  • Inappropriate timing of short- or rapid-acting insulin in relation to meals
  • Reduced administration of I.V. dextrose
  • Corticosteroid tapering
  • Insufficient glucose with insulin for acute treatment of hyperkalemia
Back to Top | Article Outline

Contributing conditions

  • Heart failure
  • Renal insufficiency (glomerular filtration rate less than 60 mL/min)
  • Liver disease
  • Malignancy
  • Infection
  • Shock
  • Sepsis.
Back to Top | Article Outline


1. Shogbon AO, Levy SB. Intensive glucose control in the management of diabetes mellitus and inpatient hyperglycemia. Am J Health Syst Pharm. 2010;67(10):798–805.
2. CDC. Diabetes Public Health Resource. 2011 National Diabetes Fact Sheet.
3. Moghissi ES. Reexamining the evidence for inpatient glucose control: new recommendations for glycemic targets. Am J Health Syst Pharm. 2010;67(16 suppl 8):S3-S8.
4. Johnston JA, Van Horn ER. The effects of correction insulin and basal insulin on inpatient glycemic control. Medsurg Nurs. 2011;20(4):187–193.
5. Moghissi ES, Korytkowski MT, DiNardo M, et al. American Association of Clinical Endocrinologists and American Diabetes Association consensus statement on inpatient glycemic control. Endocr Pract. 2009;15(4):353–369.
6. American Diabetes Association. Standards of Medical Care in Diabetes—2013. Diabetes Care. 2013;36(suppl 1):S11–S66.
7. Lleva RR, Inzucchi SE. Hospital management of hyperglycemia. Curr Opin Endocrinol Diabetes Obes. 2011;18(2):110–118.
8. Seggelke SA. Hitting the target for inpatient glycemic management. Nurse Pract. 2011;36(5):24–31.
9. Lien LF, Cox ME, Feinglos MN, Corsino L. Glycemic Control in the Hospitalized Patient. New York, NY: Springer Science + Business Media; 2011.
10. Hsia E, Draznin B. Intensive control of diabetes in the hospital: why, how, and what is in the future. J Diabetes Sci Technol. 2011;5(6):1596–1601.
11. Elliott MB, Schafers SJ, McGill JB, Tobin GS. Prediction and prevention of treatment-related inpatient hypoglycemia. J Diabetes Sci Technol. 2012;6(2):302–309.
© 2014 by Wolters Kluwer Health | Lippincott Williams & Wilkins.