Inpatient management of glycemia in people with diabetes has been inadequately studied. Previous randomized trials of intensive insulin therapy in the hospital setting resulted in excessive hypoglycemia. Current ADA guidelines (glucose 140-180 mg/dL) are by consensus with the upper bound defined by observational data and the lower bound by safety concerns. None of the previous studies of intensive glucose management used CGM technology. Whether near normal glucose levels can be achieved without increasing hypoglycemia among hospitalized patients with diabetes with the advent of CGM technology is not known.

There are clear associations between hyperglycemia and poor outcomes in patients with diabetes hospitalized with infection, including COVID-19. The COVID-19 pandemic has increased the urgency to definitively answer the question of whether glucose lowering below 140-180 mg/dL can be achieved without increasing hypoglycemia.

If this proposed study demonstrates intensive management of glucose to a target of 90 to 130 mg/dL without hypoglycemia is achievable in the inpatient setting with CGM technology, a larger study could then be performed to evaluate whether there is clinical benefit including a reduction in morbidity and mortality.

The primary study hypothesis is that glucose management with CGM can achieve a mean glucose of 90-130 mg/dL without increasing hypoglycemia compared with standard care with a glucose target of 140-180 mg/dL. Individuals with diabetes who are hospitalized (non-ICU) for an eligible condition will be randomly assigned to receive standard therapy (glucose target 140-180 mg/dL per ADA guidelines) or intensive therapy (glucose target 90-130 mg/dL and CGM used for monitoring). Real-time CGM will be used in the Intensive Target Group and masked CGM will be used in the Standard Target Group.

The co-primary outcomes, assessed via a hierarchical approach, include a treatment group comparison of mean glucose (superiority) followed by a non-inferiority comparison of time <54 mg/dL measured with CGM.

Protocol Overview:

• The glucose management team (GMT) at each site will identify potentially eligible patients.

• After informed consent and confirmation of eligibility, each participant will be randomly assigned to the Standard Target or Intensive Target group.

• Glucose management in the Standard Target Group will follow the hospital's usual practice using insulin for glucose management, with a target glucose concentration of 140-180 mg/dL.

  - A masked CGM sensor will be worn.

• Glucose management in the Intensive Target Group will include close monitoring by the site GMT to maximize the percentage of glucose values in the range of 90 to 130 mg/dL.

  • Glucose monitoring will include real-time CGM. The hospital's GMT will be the ones monitoring the CGM data. The information from CGM will be used for future management changes to insulin delivery and for safety monitoring, particularly for hypoglycemia. In essence, CGM will be used adjunctively in the study.
  • CGM validation will occur within 2-6 hours following sensor insertion and must be within 20% of a blood glucose result of >100 mg/dL for two consecutive blood glucose measurements. If a blood glucose result is <100 mg/dL, the corresponding CGM reading must be within 20 mg/dL of the blood glucose value. CGM validation will occur at least twice daily.

• Insulin delivery will be based on the site's usual protocol. The hospital nursing staff will follow their institutional SOPs with respect to frequency of BGM testing and use of BG test results for administering meal insulin and corrections. The nursing staff will not be utilizing CGM in this regard.

• Study participation will conclude at the time of hospital discharge, 10 days after randomization, at time of transfer to ICU or death.

Quality Assurance Plan:

Designated personnel from the Coordinating Center will be responsible for maintaining quality assurance (QA) and quality control (QC) systems to ensure that the clinical portion of the trial is conducted and data are generated, documented and reported in compliance with the protocol, Good Clinical Practice (GCP) and the applicable regulatory requirements, as well as to ensure that the rights and wellbeing of trial participants are protected and that the reported trial data are accurate, complete, and verifiable. Eligibility, informed consent, data entry completion and adverse events will be prioritized for monitoring.

Source Data Verification:

Study data will be obtained from the participant, the participant's EHR and from CGM. EHR data will be either transcribed from the EHR onto eCRFs on the study website or will be electronically transmitted to the Coordinating Center.

Standard Operating Procedures:

1. Patient Recruitment and Enrollment:

   Recruitment will involve identification of hospitalized patients at each clinical center who meet the study eligibility criteria. There will be no specific efforts to promote recruitment other than making hospital staff aware of the study.

   Enrollment will proceed with the goal of at least 120 participants entering the randomized trial but no greater than 150. Screening for eligibility will be performed from medical records prior to informed consent being signed. Thus, once informed consent is signed, randomization will proceed quickly. Participants who have signed consent may be permitted to continue into the trial, if eligible, even if the randomization goal has been reached.

   Study participants will be recruited from 3-6 clinical centers in the United States. All eligible participants will be included without regard to sex, race, or ethnicity. There is no restriction on the number of participants to be enrolled by each site toward the overall recruitment goal.

2. Data Collection and Testing:

   • After informed consent has been signed, all potential participants who are females of child-bearing potential will have a serum or urine pregnancy test if one has not been done since hospital admission.
   • Data from the EHR will be recorded in the study database. Medical history information will include diabetes history, other medical conditions, medications, height and weight, HbA1c (during hospitalization or most recent prior to admission), and other relevant laboratory values. The medical condition that prompted hospital admission will be recorded.

3. Data Recorded for Study (in addition to data collected at enrollment):

   At the end of hospitalization or 10 days from randomization, data to be recorded will include:

   • Vitals (blood pressure, heart rate, temperature, and O2 saturation if oxygen is used)

   • BGM

   • Insulin received each day: type of insulin, route (IV or SQ), # of units

   • All medical conditions that developed during the hospitalization

   • Date of transfer to ICU

   • Medications received

   • All coded discharge diagnoses including alive or dead

   • Labs (if measured/available) such as:

     • At admission and prior to randomization (if available)

       • Hematocrit and hemoglobin
       • Electrolytes (sodium, potassium, bicarbonate)
       • Liver function (ALT)
       • COVID-19 testing

     • All measurements (at admission and while active in the study)

       • Glucose
       • HbA1c
       • Creatinine/eGFR

4. Reporting for Adverse Events:

SAEs possibly related to a study device or study participation and UADEs must be reported to the Coordinating Center within 24 hours of the site becoming aware of the event. This can occur via phone or email, or by completion of the online serious adverse event form and device issue form if applicable. If the form is not initially completed, it should be competed as soon as possible after there is sufficient information to evaluate the event. All other reportable ADEs and other reportable AEs should be submitted by completion on the on line form within 7 days of the site becoming aware of the event. The Coordinating Center will notify all participating investigators of any adverse event that is serious, related, and unexpected. Notification will be made within 10 working days after the Coordinating Center becomes aware of the event. Each principal investigator is responsible for reporting serious study-related adverse events and abiding by any other reporting requirements specific to his/her Institutional Review Board or Ethics Committee. Where the JCHR IRB is the overseeing IRB, sites must report all serious, related adverse events within seven calendar days. Upon receipt of a qualifying event, the Sponsor will investigate the event to determine if a UADE is confirmed, and if indicated, report the results of the investigation to all overseeing IRBs, and the FDA within 10 working days of the Sponsor becoming aware of the UADE per 21CFR 812.46(b). The Medical Monitor must determine if the UADE presents an unreasonable risk to participants. If so, the Medical Monitor must ensure that all investigations, or parts of investigations presenting that risk, are terminated as soon as possible but no later than 5 working days after the Medical Monitor makes this determination and no later than 15 working days after first receipt notice of the UADE.

Sample Size:

Sample size has been computed for the treatment group comparison of mean glucose:

A total sample size was computed to be N=110 for the following assumptions: two randomized arms, 90% power, a 25 mg/dL difference in mean glucose between treatment groups, SD of 40 mg/dL, and 2-sided type 1 error of 0.05.

The total sample size has been increased to N=150 to account for potential dropouts and incomplete CGM glucose data due to early hospital discharge.

For percent of time < 54 mg/dL, statistical power will be >99% for demonstrating non-inferiority: two randomized arms, a total sample size of N=110, a standard deviation of 1%, no difference, and alpha =0.025.

Statistical Analysis Plan:

The trial has co-primary outcomes measured using CGM for up to 10 days following randomization; mean glucose tested between groups for superiority and % time <54 mg/dL tested to demonstrate non-inferiority of intensive treatment compared with standard treatment. The intervention will be considered effective only if both endpoints are met.

The null/alternative hypotheses are:

1. Mean Glucose

   1. Null Hypothesis: There is no difference in mean glucose between Standard and Intensive groups.
   2. Alternative Hypothesis: The mean glucose is different in the Standard and Intensive groups.

2. Percent Time <54 mg/dL

   1. Null Hypothesis: There is an absolute mean difference of at least 1% in the percent time <54 mg/dL between the Intensive and Standard group.
   2. Alternative Hypothesis: There is an absolute mean difference of less than 1% in the percent time <54 mg/dL between the Intensive and Standard groups.

Inclusion Criteria:

1. Age >= 18 years old

2. Type 2 diabetes (per investigator assessment); or if not previously diagnosed as having diabetes, HbA1c>=7.0% (laboratory-measured at or since hospital admission or within prior 3-months).

   • Type 1 diabetes, atypical forms of diabetes (including pancreatectomy and pancreatitis) and stress hyperglycemia alone are not eligible.

3. At least 1 blood glucose measurement >180 mg/dL since admission

4. Insulin already initiated since admission or planned to be initiated

5. Non-critical hospitalization with expected length of stay on non-ICU floor >3 days at time of randomization

Exclusion Criteria:

1. Inability to provide written consent

2. Admission to ICU

   • Patients transferred from ICU with an expected length of stay >3 days on a non-ICU floor are eligible

3. Treatment with systemic immunosuppressive agents such as high dose (>7.5 mg/day Prednisone equivalent) steroids or biologics that are not regimented and have been started within the last 3 months at time of enrollment or planned treatment prior to randomization.

4. Suspected or confirmed acute myocardial infarction or stroke as reason for hospital admission or since admission

5. Considered unlikely to survive hospitalization per investigator's judgment

6. Diabetic ketoacidosis (DKA) or hyperosmolar hyperglycemic state (HHS) in the 6 months prior to hospital admission, at hospital admission or prior to randomization during the current hospital admission

7. One or more severe hypoglycemic events within the 6 months prior to hospital admission or prior to randomization during the current admission

8. For females, pregnant or breastfeeding

   • A negative serum or urine pregnancy test will be required for all females of child-bearing potential.

9. CGM other than study CGM being used during hospitalization or planned to be used

10. Blood glucose >400 mg/dL at time of potential enrollment (most recent blood glucose measurement in hospital)

11. Insulin pump being used to deliver insulin during hospitalization or planned to be used

12. Use of IV insulin at time of potential enrollment

13. Hypoxia (O2 saturation <90) present at time of potential enrollment

14. Anasarca present at time of potential enrollment

15. Use of hydroxyurea or high dose acetaminophen use of >4g daily

16. eGFR <20 mL/min or dialysis being received or planned

17. ALT >3X normal or current diagnosis of cirrhosis

18. Cystic fibrosis

19. Expected need for surgery requiring general anesthesia during hospitalization

    • Post-surgical enrollment is permitted

20. Known allergy to medical grade adhesives or a skin condition that may impact CGM performance per investigator discretion

Public data set will include the de-identified participants demographic and CGM data.

1 year post completion

Requests are entered on Jaeb Center for Health Research public website which collects name and contact information for individual requesting data along with reason for request, including planned analyses.