Management of Diabetes and Hyperglycemia in the Hospital: A Practical Guide to Subcutaneous Insulin Use in the Non-Critically

Monitoring Glucose levels and HbA1cClick to expand

The HbA1c is an important tool to incorporate into the inpatient evaluation. It is now accepted as a diagnostic test for diabetes, and values can supplement patient clinical history in determining the effectiveness of pre-hospitalization treatment regimens. The test can help to identify some of the 25 percent of patients with diabetes mellitus (DM) that had previously been undiagnosed and recognition of DM while inpatient may reduce their risk of readmission.1,17 Since it is relatively easy and inexpensive, it has become an “opt out” test on many order sets, is recommended in some guidelines and testing it on all patients with hyperglycemia without a value documented within the last 60 days is a required component for Joint Commission Certification in Advanced Inpatient Diabetes Care.13

Guidelines recommend that all patients have blood glucose testing upon admission and those with a BG value >140mg/dL have ongoing testing for at least 24-48 h.13 POC glucose meters are currently the method most widely used for monitoring glucose because of the rapid turnaround and need to make timely adjustments to therapy. However, there are a wide number of potential issues (shock, hypoxia, dehydration, extremes in hematocrit, elevated bilirubin and triglycerides) impacting the accuracy.18,19 Additionally, issues with some monitors and lack of Food and Drug Administration (FDA) approval for critical care or hospital use have also been raised.20,21 Outside of these analytical issues, pre-analytical issues with site selection and interface with results for the correct patient being available to caregivers are important influencers on the quality of care being provided. Incorporating meter quality control and the processes for testing and recording values are important components for multidisciplinary committees to consider. Procedures aligned with The Joint Commission (TJC) must be established to assure ongoing quality control of the monitors, process of patient testing and recording the results. Additionally, with such threats to valid POC test results, clinical condition should guide treatment with the use of confirmatory venous samples when in question. Hospitalized patients require frequent testing due to the fluctuations and many variables influencing blood glucose. Testing is generally recommended to be performed before meals and at bedtime in patients who are eating or receiving bolus tube feeds13 although recent studies have called into question the value of the bedtime value.22 In patients who are NPO or receiving continuous enteral or parenteral nutrition, glucose monitoring is performed every four to six hours depending on whether short-acting or rapid-acting analog insulin is used. For patients on intravenous insulin, the ADA and AACE recommend hourly blood glucose monitoring except for patients with stable blood glucose within the target range, for whom monitoring can be performed up to every two hours.

Glycemic TargetsClick to expand

Critically Ill

In 2001, van den Berghe published results demonstrating significant improvements in mortality in surgical ICU patients treated with IV insulin therapy targeted at normalizing blood glucose.23 Following this, numerous specialty organizations published recommendations targeting “very tight” glycemic control and many hospitals implemented infusion protocols to normalize blood glucose values in their critically ill patients.24,25 However, subsequent trials had mixed results and finally NICE SUGAR ultimately demonstrated increased mortality and hypoglycemia with “very tight”/normalized blood glucose versus moderate control.26, 27 Revised guidelines now recommend that insulin therapy be initiated if the blood glucose is >180 mg/dL and then maintained between 140 and 180 mg/dL in critically ill patients. Glucose targets somewhat lower than these may be appropriate in selected patients although strong evidence is lacking. The Society of Critical Care Medicine guidelines continue to recommend a target of 100-150 mg/dL but emphasize avoiding hypoglycemia.28 For these purposes, critical illness generally refers to patients requiring invasive mechanical ventilation, those requiring pressor support and patients with multisystem organ failure.

Special Populations

For post-operative cardiovascular surgery, The Joint Commission’s Surgical Care Improvement Project (SCIP) updates from January 2014 set targets of blood glucose levels in the 18-24 hour window after anesthesia end time of <180mg/dL; but these SCIP glycemic metrics are currently on hold as of early 2015 with new metrics still to be determined at the time of this publication.29

Non-Critically Ill

With very few prospective, RCT data for establishing specific guidelines in non-critically ill patients, the ADA/AACE recommendations in 2009 are based on clinical experience and judgment.13


Hypoglycemia has associated risks demonstrated in various studies, and strategies are necessary to avoid it. Studies have used variable definitions for hypoglycemia but the ADA has defined hypoglycemia as a blood glucose level <70 mg/dL and severe hypoglycemia as <40 mg/dL.30

Methods to Achieve ControlClick to expand

Inpatient hyperglycemia is best managed with insulin only. Insulin works reliably, and doses can be rapidly adjusted depending on changes in glucose levels and food intake. Oral agents should be discontinued during acute illness in most circumstances.13 Most are associated with some risks and are limited in their ability to be rapidly adjusted to achieve goals.31, 32

  1. Subcutaneous insulin: Scheduled subcutaneous administration of insulin is the preferred method for achieving and maintaining glucose control in non-critically ill patients with diabetes or stress hyperglycemia. The recommended components of inpatient subcutaneous insulin regimens are a basal, a nutritional and a correctional element. Prolonged therapy with sliding-scale insulin (correction only without scheduled insulin) as the sole regimen is ineffective in the majority of patients (and dangerous in those with type 1 diabetes). However, initial therapy with correction insulin only may be appropriate in patients with type 2 diabetes who:
  • are well controlled (HbA1c <7% or normal BG values) with only diet or a low dose-oral agent
  • have mild hyperglycemia and
    • are NPO on no nutritional replacement
    • are on new or tapering steroids
    • hypoglycemic risk factors including but not limited to end-stage liver or kidney disease, elderly patients or those with an unknown drug overdose

The general recommendations for dosing subcutaneous insulin are derived from published studies and current ADA/AACE and Endocrine Society guidelines. They generally involve:

  1. Estimating patients’ total daily insulin requirement, or total daily dose (TDD)
  2. Dividing the TDD into approximately 50 percent basal insulin and 50 percent nutritional
  3. Adding a correction scale based on the patient’s estimated insulin sensitivity and then
  4. Adjusting the doses daily

  1. Insulin infusion and transition to subcutaneous: On the basis of the available evidence, insulin infusion should be used to control hyperglycemia in the majority of critically ill patients or post-operative patients in the ICU setting, specifically patients requiring invasive positive pressure ventilation and moderate to high dose vasopressor support.33 In addition to critically ill patients, some patients may have improved control with insulin infusion including but not limited to those with uncontrolled hyperglycemia on glucocorticoids, TPN, hyperglycemic crisis and labor and delivery. Many hospitals have been able to safely administer insulin infusions outside of the critical care setting b y providing staff education and adjusting nurse-to-patient ratios to allow continued frequent blood glucose monitoring and adjustments.34

    Following resolution of their indication for insulin infusion, patients will need to be transitioned to subcutaneous insulin. Patients with type 1 DM, on insulin prior to admission to the hospital and those requiring insulin infusion rates more than two units/hour, will need to be transitioned to basal/bolus insulin regimens. The average infusion rate over the preceding six to eight hours is then multiplied by 24 to determine the predicted requirement for the next day. Given that patients are likely to be continuing to improve and have reduced insulin requirements, that daily estimate is further reduced to 60-80 percent. This value can be used as the total daily requirement if they were receiving significant nutritional support in the ICU (TPN or tube feeds) or the basal amount with nutritional layered over time as oral intake increases.35-40

Transition to the Outpatient SettingClick to expand

Preparation for transition to the outpatient setting is an important goal of inpatient diabetes management and begins with the hospital admission. The HbA1C is an important laboratory test that should be ordered in nondiabetic hyperglycemic patients and diabetic patients who have not had a recent test. An HbA1c value >6.5% is diagnostic of diabetes, 5.7-6.4% is “pre-diabetes” and one over 9 percent has been recommended by some to indicate the need for newly aggressive treatment strategies.41 Modification of the outpatient regimen should be done with careful planning and assessment of patient and facility-specific resources as well as with communication with the outpatient provider.42 There are several published algorithms for diabetes treatment that can be used to guide such decisions but it is most important to factor in individual patient needs.47 All patients with diabetes should receive teaching in diabetes self-management education focused on survival skills.13

Communication to the outpatient provider regarding the education provided, relevant lab values including the HbA1c and medication changes is essential for patient safety and a requirement of The Joint Commission’s Inpatient Diabetes Disease Certification.

Glycemic Control Project Team

The first edition of this Toolkit was supported in part by a non-educational grant from sanofi-aventis US, LLC.
The second edition of this Toolkit is supported in part by an educational grant from Novo Nordisk in 2015.

The Glycemic Control Implementation Toolkit is an online resource for visitors to the Society of Hospital Medicine's (SHM’s) website. All content and links have been reviewed by the Glycemic Control Implementation Toolkit Project Team; however, the Society of Hospital Medicine does not exercise any editorial control over content associated with the external links that have been made available via this website.