Diabetes Care in the Hospital Setting

2016 ADA Guideline Recommendations
Types of Diabetes Mellitus
Main Diabetes Page
Diabetes type I
Diabetes type II
Gestational Diabetes Mellitus
2016 ADA Standard of Medical Care Guideline Recommendations
Strategies for Improving Care
Classification and Diagnosis of Diabetes
Foundations of Care and Comprehensive Medical Evaluation Diabetes Self-Management, Education, and Support Nutritional Therapy
Prevention or Delay of Type II Diabetes
Glycemic Targets
Obesity Management for Treatment of Type II Diabetes
Approaches to Glycemic Treatment
Cardiovascular Disease and Risk Management Hypertension and Blood Pressure Control Lipid Management Antiplatelet Agents Coronary Heart Disease
Microvascular Complications and Foot Care Diabetic Kidney Disease Diabetic Retinopathy Diabetic Neuropathy Diabetic Footcare
Older Adults with Diabetes
Children and Adolescents with Diabetes Management of Cardiovascular Risk Factors in Children and Adolescents with Diabetes Microvascular Complications in Children and Adolescents with Diabetes
Management of Diabetes in Pregnancy
Diabetes Care in the Hospital Setting

Editor-In-Chief: C. Michael Gibson, M.S., M.D. [1] Associate Editor(s)-in-Chief: Anahita Deylamsalehi, M.D.[2]; Shivani Chaparala M.B.B.S [3]; Tarek Nafee, M.D. [4]

• There are some available clinical practice guidelines and systematic reviews that suggest basal-bolus insulin may lower blood sugar more efficient, nevertheless it is more likely to cause hypoglycemia with no change in clinical outcomes. .
  • Key studies include:
    • Basal-bolus versus a basal plus correction insulin
    • Basal-bolus versus sliding scale insulin
    • Basal-bolus versus basal Plus sliding scale versus sliding scale alone.
    • Insulin analogues versus human insulin.
    • Basal-bolus insulin versus sliding scale insulin using the Glucommander eGlycemic Management System.
• Formerly guidelines advocated to stop all oral antidiabetic agents in the hospital settings of diabetes management, nevertheless some clinical trials support the effectiveness of oral antidiabetic drugs, solitary or in combination to insulin therapy, in some hospitalized individuals. Countries such as England, Israel and India practice the usage of oral antidiabetic agents such as, metformin and sulfonylureas for inpatient diabetic patients.
  • Use metformin with caution in hospitalized patients especially if renal or hepatic failure, sepsis and shock is present due to high chance of lactic acidosis. To prevent such an adverse effect, patients with GFR in range of 30–45 mL/min per 1·73 m² should recieve lower doses of metformin. Furthermore metformin must be discontinued in patients with GFR lower than 30 ml/min per 1·73 m².
  • Metformin should be discontinued prior to iodinated contrast imaging if eGFR <60 mL/min per 1·73 m², history of hepatic disease, acute heart failure, alcoholism or received intra-arterial contrast medium.
  • Compared to other oral antidiabetic agents, metformin showed higher rate of gastrointestinal adverse effects in the hospitalized patients.
  • Frequent hypoglycemic events have been reported in sulfonylureas usage in the hospital setting, hence experts don't recommend it for treatment of diabetic inpatients. Facrors such as old age, renal insufficiency and concurrent insulin use have been related to higher risk of hypoglycemia. Although UK recommendations proposed it's effectiveness in mangement of hyperglycemia due to glucocorticoid use.
  • Adverse effect such as heart failure (due to water retention and late onset of action turned thiazolidinediones into a ineffective treatment in hospital settings.
• A landmark trial done in 2009 demonstrated an increased mortality risk with intensive insulin treatment, specifically in critically ill patients. Chance of hypoglycemia rises with intensive insulin therapy and can further complicate the situation.
• Continuous insulin infusion is recommended for critically ill diabetic patients (such as ICU patients), which should be replaced by subcutaneous insulin when patient is stable.
  • The following factors should be considered when transition from continuous insulin infusion to subcutaneous insulin is planned:
    • Patients should have a steady glucose concentration, at least for last 4–6 hours
    • Normal anion gap must be achieved (if patient have been presented with DKA, acidosis should have been resolved before transition).
    • Patients should be hemodynamically stable with out vasopressors
    • Patients should be on a stable diet
    • Patients should be on a steady intravenous infusion rate
• Based on a systematic review, universal usage of premixed insulin regimens is not trusted in hospitalized diabetic patients.
• The following table is a summary of treatment in non-critically ill hospitalized patients with diabetes:

Patient status	Mild hyperglycaemia	Moderate hyperglycaemia	Severe hyperglycaemia
Definition	Blood glucose < 200 Patients who are taking less than 2 anti-diabetic drugs (such as oral anti-diabetic drug or GLP-1 receptor agonists)	201 < Blood glucose <300 Patients who are taking multiple anti-diabetic drug (such as oral anti-diabetic drug or GLP-1 receptor agonists) Patients who are taking less than 0·6 U/kg insulin per day	Blood glucose > 301 Patients who are taking multiple anti-diabetic drug (such as oral anti-diabetic drug or GLP-1 receptor agonists) Patients who are taking more than 0·6 U/kg insulin per day
Approach	Low dose basal insulin OR oral anti-diabetic drug†, if there are no contraindications. Further blood glucose correction can be applied by rapid-acting insulin (before meals or every 6 hours)	Basal insulin OR oral anti-diabetic drug†, if there are no contraindications. Initial insulin dose: 0·2–0·3 U/kg per day (start from 0·15 U/kg per day (if using basal insulin alone) or 0·3 U/kg per day (if using basal–bolus) for patients with high risk of hypoglycemia). Further blood glucose correction can be applied by rapid-acting insulin (before meals or every 6 hours)	Basal–bolus insulin regimen Initial insulin dose: Reduce patient's home insulin regimen by 20% OR 0·3 U/kg per day (half basal and half bolus) If patient has poor intake, hold the prandial insulin.

_{†One of the options which has been studied in randomized controlled trials is dipeptidyl peptidase-4 inhibitor. Although metformin use is common, use it with caution due to high risk of lactic acidosis, especially in high risk patients (such as sepsis, hypoxia, renal insufficiency, shock and hepatic failure)}

• The following table is a summary of treatment in critically ill hospitalized patients with diabetes:

Patient status	Patients with surgical or other medical conditions‡	Mild to moderate DKA	Severe DKA or HHS
Approach	Continuous insulin infusion §	Continuous insulin infusion OR subcutaneous insulin (consider DKA protocol)	Continuous insulin infusion

_{‡Continuous insulin infusion specially could be beneficial in hypoglycemia due to steroid use or in solid organ transplant patients. §Prompt treatment is recommended in patients with myocardial infarction or ischemic stroke due to possible further harm due to hyperglycemia. ALthough intensive treatment is not recommended due to higher chance of hypoglycemia.}

• Based on a systematic review published in 2021, numerous subcutanoeus insulin regimen have been studied on non-critically ill diabetic patients. The following table is a summary of the afformentioned studies:

Umpierrez et al (2007)	Glargine–glulisine (basal-bolus) Vs sliding-scale insulin Study has been done on 130 patients with diabetes type 2 who were on oral antidiabetic agents or low dose insulin (less than 0·4 U/kg per day), with glucose concentrations of 140–400 mg/dL. Reported average glucose concentration in basal-bolus group has been 166 mg/dL (SD: 1.8) Vs average glucose concentration of 193 mg/dL in sliding-scale insulin group (P value < 0·001). Non of the patients developed hypoglycemia (glucose concentrations less than 40 mg/dl).
Umpierrez et al (2009)	Human insulin (NPH and regular) † Vs detemir–aspart (basal-bolus) Study has been done on 130 patients with diabetes type 2 who were on various type of treatments, with glucose concentrations of 140–400 mg/dL. Reported average glucose concentration showed no difference. 4·5% of patients in detemir–aspart group developed hypoglycemia (glucose concentrations less than 40 mg/dl Vs 1·6% of patients in human insulin.
Umpierrez et al (2011)	Glargine–glulisine (basal-bolus) Vs sliding-scale insulin Study has been done on 211 surgical patients with diabetes type 2 who were on oral antidiabetic agents or low dose insulin (less than 0·4 U/kg per day), with glucose concentrations of 140–400 mg/dL. Reported average glucose concentration showed better glycemic control in basal-bolus group (P value = 0·003 ). Patients in basal-bolus group showed lower rate of general complications₪, nevertheless rates of hypoglycemia were higher (4% in basal-bolus group Vs 0% in sliding-scale insulin group).
Schroeder et al (2012)	NPH and regular‡ Vs sliding-scale insulin Study has been done on 141 orthopaedic surgery patients with diabetes type 2 or history of frequent hyperglycemia (>180 mg/dL). Reported average glucose concentration in NPH and regular group has been 161·2 mg/dL Vs average glucose concentration of 175·8 mg/dL in sliding-scale insulin group (P value < 0·0005). Two episodes of severe hypoglycemia have been reported in NPH and regular group.
Umpierrez et al (2013)	Basal-plus (glargine–glulisine) Vs basal-bolus (glargine–glulisine) Vs sliding-scale insulin Study has been done on 375 patients with diabetes type 2 who were on oral antidiabetic agents or low dose insulin (less than 0·4 U/kg per day), with glucose concentrations of 140–400 mg/dL. Reported average glucose concentrations were akin in both basal-plus and basal-bolus groups. Treatment failure was higher in sliding-scale insulin group (19%), compared to basal-bolus (0%) or basal-plus (2%) groups.
Mader et al (2014)	Glargine–aspart (basal-bolus) Vs standard treatment (such as oral antidiabetic agents), insulin or combination of both Study has been done on 74 patients with diabetes type 2 who were on various type of treatments, with glucose concentrations higher than 140 mg/dL. Reported average glucose concentration in basal-bolus group (33%) was higher compared to standard treatment group (23%) (P value 0·001) ¶
Bueno et al (2015)	Glargine–glulisine (basal-bolus) Vs NPH and regular₳ Study has been done on 134 patients with diabetes type 2 who were not a surgical case. Reported average glucose concentrations were akin in both NPH and regular and basal-bolus groups. Hypoglycemic events were reported as 35% in basal-bolus group, compared to 38% in NPH and regular group.
Bellido et al (2015)	Glargine–glulisine (basal-bolus) Vs premixed NPH and regular (70/30)¥ Study has been done on 72 patients with diabetes type 2 who were either medical or surgical cases, who were on oral antidiabetic agents, insulin or a combination of both.§ Reported average glucose concentration showed no difference. Objectionably high rate of hypoglycemia have been reported in premixed NPH and regular group.
Vellanki et al (2015)	Glargine–aspart (basal-bolus) without bedtime addition Vs Glargine–aspart (basal-bolus) with bedtime addition Study has been done on 206 patients with diabetes type 2 who were medical or surgical cases with glucose concentrations of 140–400 mg/dL. Patients were on oral antidiabetic agents, insulin or a combination of both. Reported average glucose concentration showed no difference. Non of the patients developed hypoglycemia (glucose concentrations less than 40 mg/dl).
Gracia-Ramos et al (2016)	Glargine–lispro (basal-plus) Vs premixed insulin analog (lispro 25/75) ¤ Study has been done on 54 patients with diabetes type 2 who were on oral antidiabetic agents or low dose insulin (less than 0·4 U/kg per day), with glucose concentrations of 140–400 mg/dL. Reported average glucose concentrations showed no difference. Rate of hypoglycemia was simillar in both groups (16%).
Pasquel et al (2020)	Basal-bolus (glargine U300–glulisine) Vs basal-bolus (glargine U100–glulisine) Study has been done on 176 patients with diabetes type 2 who were on various type of treatments, with glucose concentrations of 140–400 mg/dL. Reported average glucose concentration showed no difference. Significant lower rate of hypoglycemic events in glargine U300 group. (P value = 0·023)

_{†Two-thirds before breakfast and one-third before dinner
₪Patients in basal-bolus group showed lower rate of general complications such as postoperative wound infection, bacteremia, pneumonia, renal insufficiency and respiratory failure.}
^{‡Three times a day}
_{¶Related article didn't provide the absolute numbers
₳NPH insulin has been used twise a day. Regular insulin has been used before meals
¥60% before breakfast and 40% before dinner
§Study terminated after interim analysis.
¤Two-thirds with breakfast and one-third with dinner}

• There are numerous data rejecting use of subcutaneous sliding scale insulin in management of inpatient diabetes. However it usage could be accepted in management of stress hyperglycaemia in nondiabetics.
• If home insulin regimen of a patient has been higher than ≥0·6 U/kg each day, 20% reduction in general insulin dose should be considered at the time of admission. This effort is recommended to lower the chance of hypoglycemia in poor oral intake.
• Basal-bolus approach has been reported as an effective treatment for hyperglycemia, nevertheless it's usage in patients with mild hyperglycemia (<200 mg/dl) has been warned due to high risk of iatrogenic hypoglycaemia (with rate of 12–30% in some studies).
• Another approach of insulin therapy such as basal-plus has been recommended for patients with mild hyperglycemia, surgical patients and low oral intake. The mentioned approach is consisted of a basal dose of insulin (0·1–0·25 U/kg/day) and corrective insulin doses if hyperglycemia developed (Monitor glucose concentration every 6 hours in NPO patients, and before every meal in others). This approach has been associated with lower risk of hypoglycemia and is recommended in patients with low oral intake or surgical cases.
• High rate of iatrogenic hypoglycaemia in premixed insulin therapy turned this approach into an untrsuted hyperglycemia treatment in hospitalized diabetic patients. However using premixed insulin therapy in patients taking enteral nutrition have been recommended (there are not enough supporting data).

2016 ADA Standards of Medical Care in Diabetes Guidelines

"1. Consider performing an A1C on all patients with diabetes or hyperglycemia admitted to the hospital if not performed in the previous 3 months. (Level of Evidence: C)"

"2. Insulin therapy should be initiated for treatment of persistent hyperglycemia starting at a threshold ≥180 mg/dL (10.0 mmol/L). Once insulin therapy is started, a target glucose range of 140–180 mg/dL (7.8–10.0 mmol/L) is recommended for the majority of critically ill patients(Level of Evidence: A)and noncritically ill patients (Level of Evidence: C)"

"3. More stringent goals, such as 110–140 mg/dL (6.1–7.8 mmol/L) may be ap- propriate for selected critically ill patients, as long as this can be achieved without significant hypoglycemia (Level of Evidence: C)"

"4. Intravenous insulin infusions should be administered using validated written or computerized protocols that allow for predefined adjustments in the insulin infusion rate based on glycemic fluctuations and insulin dose. (Level of Evidence: E)"

"5. A basal plus bolus correction insulin regimen is the preferred treatment for noncritically ill patients with poor oral intake or those who are taking nothing by mouth. An insulin regimen with basal, nutritional, and correction components is the preferred treatment for patients with good nutritional intake. (Level of Evidence: A)"

"6. The sole use of sliding scale insulin in the inpatient hospital setting is strongly discouraged (Level of Evidence: A)"

"7. A hypoglycemia management protocol should be adopted and implemented by each hospital or hospital system. A plan for preventing and treating hypoglycemia should be established for each patient. Episodes of hypoglycemia in the hospital should be documented in the medical record and tracked. (Level of Evidence: E)"

"8. The treatment regimen should be reviewed and changed if necessary to prevent further hypoglycemia when a blood glucose value is <70 mg/dL (3.9 mmol/L). (Level of Evidence: C)"

"9. There should be a structured discharge plan tailored to the individual patient. (Level of Evidence: B)"