Use of CGM for Diabetes Management Following a Hospitalization.

Overview

Patients with diabetes hospitalized for heart failure or acute myocardial infarction will have a continuous glucose monitor (CGM) placed on the day of discharge which will be downloaded at their outpatient follow up clinic visit 6-14 days later. At the follow-up visit medications may be modified based on downloaded glucose data. This study is designed to determine the frequency that patients with diabetes hospitalized for heart failure or acute myocardial infarction have significant hypoglycemia or hyperglycemia requiring medication adjustment in the 1-2 weeks following hospital discharge. The outcome will be a binary- yes, diabetes medications were adjusted or no, diabetes medications were not adjusted. As a secondary outcome 30-day hospital readmission and mortality will be compared to historical hospital data collected by the inpatient cardiovascular service as part of their quality improvement monitoring. Researchers do not expect to see a significant difference in 30-day readmission or mortality as this study will be under-powered. In addition, our inclusion criteria will specifically target the patients at highest risk for bad diabetes outcomes, thus researchers may see worse outcomes compared to historical controls, but this assessment will help us design future randomized studies.

After the participant has signed the consent form a continuous glucose monitor (CGM) sensor will be placed on the arm of the participant. If consent is signed prior to the day of discharge the sensor placement will be delayed until the day of discharge. The participant will be provided with instruction to avoid bumping the sensor and to avoid dislodging when bathing and dressing. If the sensor should fall off before the follow up appointment, they will be instructed to bring it with them for a partial download. During the already scheduled post-discharge follow up appointment the CGM sensor data will be downloaded by clinic staff. The diabetes medications will be reconciled, and the downloaded data will be reviewed with the patient. Based on the download, a PharmD will have the option of increasing or decreasing insulin doses by a maximum of 10% to reduce hypoglycemia and/or hyperglycemia. The goal will be to adjust medications, if needed, to target blood sugars between 90-250mg/dl greater than 80% of the time. This is a conservative goal aimed at reducing risk for severe hypo or hyperglycemia with good glycemic control rather than to target tight glycemic control. If adjustment of oral agents or a change of greater than 10% is required this will be done in consultation with an attending physician. These adjustments will follow standards of care and the expectation is that this will benefit the patient. All changes will be communicated to the participant's primary care physician (PCP) or endocrinologist. If the patient does not have an endocrinologist and it is felt that significant adjustments will be required over time to improve the participant's glucose control they will be given the option of a referral to the Diabetes Cardiovascular Clinic, General Endocrinology Clinic, or if having excessive hypoglycemia the Diabetes Technology Clinic for consideration for long term use of CGM. If no changes are required, this will also be communicated to the patient's PCP or endocrinologist. The downloaded data from the CGM will be scanned in the media tab in the electronic medical record and any adjustments will be included as part of the medical record. Between 30 and 40 days post-discharge the participant's medical record will be accessed again. At this point it will be determined if the patient was readmitted to the hospital or died in the 30 days post-discharge. In addition, clinic notes will be examined to determine if there is any evidence that the patient developed worsening hypoglycemia or worsening hyperglycemia in response to medication adjustments made during the post-discharge clinic. This data will be collected to assist in understanding the efficacy of the intervention and to ensure that the intervention is not causing unintended harm. This access of the medical record would not be expected to benefit the patient but could provide benefit to future participants or patients.