FAQ's Hospital Glucose Control


Non-ICU Setting

  1. What percentage of patients in the non-ICU setting have diabetes?
    1. Approximately 30% of the patients admitted to the non-ICU setting have diabetes (95% are type 2). This means U.S. hospitals admit around 8 million patients with diabetes into the non-ICU setting on an annual basis. There are approximately 150,000 patients with diabetes in the non-ICU setting on a daily basis who would benefit from improved glucose control.
  2. What is the current standard of care for glucose control in the non-ICU setting?
    1. Multiple daily injections of subcutaneous insulin (100 units/mL) by nurses is the current standard of care in the non-ICU setting. In a typical scenario, the nurse measures the patient’s blood glucose level before every meal and at nighttime (e.g., 9 PM) and based on these results and the patient’s insulin orders, administers a dose of short and/or long acting insulin.
  3. Is the current standard of care effective at controlling glucose levels in the non-ICU setting?
    1. No. Multiple daily injections using either a sliding scale with short acting insulin or a basal-bolus technique with long/short acting insulin is not effective at maintaining glucose levels in the desired range of 70-180 mg/dL while avoiding hypoglycemia (<70 mg/dL). Studies using the current standard of care basal-bolus insulin regimen continue to report hypoglycemia (<70 mg/dL) rates > 1% and percent time in the desired glucose range 70-180 mg/dL of <70%.
  4. What level of hyperglycemia is associated with worse outcomes in the non-ICU setting?
    1. Glucose levels > 180 mg/dL in patients with diabetes, and > 140 mg/dL in patients without diabetes are associated with increased complication and mortality rates, and prolonged hospital length of stay.
  5. How prevalent is hyperglycemia in the non-ICU setting?
    1. Very prevalent. A U.S. study of 2.8 million non-ICU patients that looked at 37 million point of care glucose measurements noted that 32% of all patient days had glucose measurements > 180 mg/dL (Swanson, Endocrine Practice, 2011).
  6. Does hyperglycemia only occur in patients with diabetes in the non-ICU setting?
    1. No. In a study of 1,886 adult admissions to a single hospital, it was noted that 38% of the patients developed hyperglycemia (>=200 mg/dL), and out of these patients one-third had no history of diabetes (Umpierrez, J Clinical Endocrinology & Metabolism, 2002).
  1. Does hyperglycemia adversely affect the outcome of patients without diabetes?
    1. Yes. In fact, when patients without diabetes develop hyperglycemia, their outcomes are worse than when patients with diabetes develop hyperglycemia. In a prospective study of 2,558 patients, it was shown that patients without diabetes but with hyperglycemia (>=200 mg/dL) versus patients with diabetes and hyperglycemia had higher hospital acquired infection rates (27.5% vs 9.1%) and higher mortality rates (15.5% vs 2.8%) (Barmanray, J Clinical Endocrinology & Metabolism, 2024).
  1. Do attempts to more aggressively control glucose levels of non-ICU patients with diabetes decrease their hospital length of stay?
    1. Yes. In a study of 603 patients with diabetes, when patients with hyperglycemia (>180 mg/dL) were managed by a specialist diabetes team (nurses, pharmacist, nutritionist, endocrinologist) versus the usual care, they were noted to have a lower prevalence of glucose levels > 180 mg/dL (36.5% vs 51.6%) and shorter hospital lengths of stay (12.1 days vs 15.3 days) (Swee, Int J of Clinical Practice, 2017).
    2. Yes. In a systematic review on the effects of introducing a diabetes inpatient specialist nurse (DISN) on the hospital length of stay of patients with diabetes, a median reduction of 0.5 – 3 days in the length of stay was noted due to the improved glucose control achieved through use of the DISN (Akiboye, Diabetic Medicine, 2021).
    3. Yes. In a retrospective study of 11,477 patients with diabetes, it was shown that consultation by a diabetology team within 48 hours of admission led to a 1.56 day reduction in the length of stay (Sheahan, Endocrinology Diabetes & Metabolism, 2020).
  1. Is hypoglycemia prevalent in non-ICU patients with diabetes?
    1. Yes. In a retrospective study of 2,582 patients with diabetes, it was shown that 7.7% of the patients developed hypoglycemia (<= 50 mg/dL), and that hypoglycemia was associated with a 2.5 day increase in the length of stay (Turchin, Diabetes Care, 2009).
  2. Does hypoglycemia in non-ICU patients with diabetes adversely affect their outcomes.
    1. Yes. In a retrospective study of 921,306 patients with diabetes admitted to a non-ICU setting, it was noted that 5% of the patients had a secondary diagnosis of hypoglycemia, and that the presence of hypoglycemia increased the odds ratio for death by 24% (Zapatero, Endocrine Practice, 2014).
  3. Will the introduction of continuous glucose monitors (CGM) to the hospital setting, by itself, improve the glucose control of patients with diabetes in the non-ICU setting?
    1. No. In a randomized study of 110 patients with type 2 diabetes, the use of CGM (Dexcom G6/G7) glucose values by a hospital glycemic management team to control their patients (N=60) to a glucose range of 90-130 mg/dL did not result in improved glucose control versus standard care in patients (N=50) where the goal was to control to a glucose range of 140-180 mg/dL, by the same glycemic management team. The mean glucose level of CGM guided care versus standard care was not statistically different (170 mg/dL vs 175 mg/dl, p=0.25), despite the fact the CGM guided group was being controlled to a lower glucose control range (Hirsch, Diabetes Care, 2025).
  4. Why don’t hospitals use the closed loop glucose control systems (artificial pancreas) already approved by the FDA?
    1. The FDA has only approved these systems (e.g., Tandem, Medtronic, Omnipod) for use in the outpatient setting. As of April 2025, the FDA has not approved the use of any artificial pancreas systems for use in the non-ICU setting.


ICU Setting

  1. What percentage of patients in the ICU setting have diabetes?
    1. Approximately 30% of 6 million ICU patients admitted annually in the U.S. have diabetes (95% are type 2). Thus, on an annual basis, there are approximately 1.8 million patients with diabetes, and 4.2 million patients without diabetes admitted to U.S. ICU’s.
  2. Is glucose control important for ICU patients without diabetes?
    1. Yes. It turns out it is actually more important to prevent hyperglycemia in patients without diabetes as they have a higher mortality rate for a given time weighted average glucose level versus patients with diabetes (Fong, BMC Anesthesiology, 2022).
  3. What percentage of patients in the ICU setting develop hyperglycemia?
    1. In a large prospective study of 6,104 ICU patients assessing glucose control, 83% of the patients required insulin treatment to control their glucose levels (Finfer, New England Journal of Medicine, 2009).
  4. What is the current standard of care for glucose control in the ICU setting?
    1. Intravenous (IV) insulin (1 unit/mL) is used to control glucose levels in the ICU setting. The nurses adjust the IV rate of insulin every 1-4 hours depending on the patients measured glucose level and the instructions from the locally developed IV insulin protocol or a web based clinical decision support system (e.g., Glucommander). Using this method, nurses can spend up to 2 hours/patient/day in glucose control efforts.
  5. Is the current standard of care effective at controlling glucose levels in the ICU setting?
    1. No. Adjusting the rate of IV insulin every 1-4 hours is not effective at maintaining glucose levels in the desired range of 100-180 mg/dL while avoiding hypoglycemia (<70 mg/dL). ICU studies using the current standard of care continue to report hypoglycemia (<70 mg/dL) rates > 1% and percent time in the desired glucose range 100-180 mg/dL <70%.
  6. What level of hyperglycemia is associated with worse outcomes in the ICU setting?
    1. Time weight average glucose levels > 130 mg/dl in patients without diabetes, and > 160 mg/dL in patients with diabetes are associated with increased mortality rates in ICU patients (Fong, BMC Anesthesiology, 2022).
  7. Does hyperglycemia only occur in patients with diabetes in the ICU setting?
  1. No. Clinically significant hyperglycemia occurs in patients with and without diabetes in the ICU setting.
  1. Does hyperglycemia adversely affect the outcome of patients without diabetes in the ICU setting?
    1. Yes. In a study of 259,040 ICU admissions, it was shown mean glucose levels were predictive of mortality rates in patients without diabetes. Patients without diabetes who had mean glucose levels in the range of 70-110 mg/dL had an overall mortality rate of 7.6%, whereas patients with mean glucose levels of 146-199 mg/dL had an overall mortality rate of 19.8% (Falciglia, Critical Care Medicine, 2009).
  1. Do attempts to more aggressively control glucose levels of ICU patients decrease their hospital length of stay?
  1. Yes. In a prospective study of 447 ICU patients who had undergone pancreatic surgery, those randomized to a post-operative glucose control range of 80-110 mg/dL had a hospital length of stay of 18.2 days, whereas those randomized to a glucose control range of 140-180 mg/dL had a hospital length of stay of 23 days (Okabayashi, Diabetes Care, 2014).
  2. Yes. In a trauma study of 2,026 ICU patients, changing the glucose control range from 80-200 mg/dl to 80-110 mg/dL, statistically significantly decreased the hospital length of stay from 12.3 to 11 days and the mortality rate from 21.5% to 14.7% (Eriksson, Journal of Emergencies, Trauma & Shock, 2011).
  1. Is hypoglycemia prevalent in ICU patients and does it affect patient outcomes?
    1. Yes. In a retrospective study of 4,496 ICU patients, 22.4% of the patients experienced hypoglycemia (< 81 mg/dL). The mortality rate of the patients who experienced hypoglycemia was statistically significantly higher than the group of patients who did not (36.6% vs 19.7%) (Egi, Mayo Clinical Proceedings, 2010).
    2. Yes. In a retrospective study of 16,669 ICU patients, 12.7% of the patients experienced hypoglycemia (<= 70 mg/dL). The mortality rate of the patients who experienced hypoglycemia was statistically significantly higher than the group of patients who did not (48.8 % vs 15.9%) (Yeh, Scientific Reports, 2025).
    3. Yes. In a retrospective study of 4,986 ICU patients, 7.7% of the patients experienced hypoglycemia (<= 70 mg/dL). The mortality rate of the patients who experienced hypoglycemia was statistically significantly higher than the group of patients who did not (22% vs 10%) (Wernly, Medical Principles and Practice, 2019).
  2. Will the introduction of continuous glucose monitors (CGM) to the ICU setting, by itself, improve the glucose control of patients?
    1. Possibly. In a randomized study comparing glucose control using a continuous glucose monitor (Medtronic Guardian) versus a point of care glucose meter (Bayer, Contour Plus) with both groups (N=48 each) controlled to a glucose control range of 70-180 mg/dL, the CGM group versus the point of care group had statistically significant lowering of time weighted average glucose values (182 mg/dL vs 200 mg/dL) and an increase in time in range 70-180 mg/dL (64.5% vs 46.4%) (Chu, Journal of Critical Care, 2024).
  3. Why don’t hospitals use the closed loop glucose control systems (artificial pancreas) already approved by the FDA in the ICU?
    1. The FDA has only approved these systems (e.g., Tandem, Medtronic, Omnipod) for use in the outpatient setting. As of April 2025, the FDA has not approved the use of any artificial pancreas systems for use in the ICU setting.