Interpretation of Continuous Glucose Monitoring Data Q&A

Sherri Horvat, BSN, RN, CDE

1. How accurate is continuous glucose monitoring (CGM) compared to blood glucose testing?

Each device is going to be different, but what you want to look at is the MARD, which is the Mean Absolute Relative Difference. This uses a lab analyzer to determine how different the reading is from a lab glucose value. Based on fingersticks being done accurately, CGM is getting more and more accurate with MARD in some models being in the single digit values. When blood glucose was first being measured by fingersticks and treatment decisions were being made, they were very similar to what current CGM models offer. We see this gap narrowing over time with the improved accuracy of CGM devices. In addition, accuracy is very dependent on appropriate and accurate calibrations.

The Dexcom G5 Mobile System is the first CGM System to receive FDA approval to make CGM-based treatment decisions ("non-adjunctive use"). What does this new indication mean? It means the Dexcom G5 Mobile can now be used to make diabetes treatment decisions without confirming the reading with a blood glucose meter.* Previously, all CGM devices were indicated for use only as a complement (adjunct), not a replacement, for blood glucose meters.

*Fingersticks required for calibration, or if symptoms or expectations do not match readings, or when taking medications containing acetaminophen.”

2. Please comment on recent paper that was published in Diabetes Care regarding people with diabetes reluctance to adopt diabetes technology.

Unfortunately, there are always perceived barriers when new technology is introduced to individuals, and until the benefits outweigh the perceived disadvantages, people will hold back. As clinicians we need to help them see what the actual benefits are. I have found that when explaining these newer technologies to individuals, letting them know they do not have to try this forever can help. I ask them to give it a six months before making up their minds as it takes that long to get to where it is an innate behavior as to what to do and how to adjust to new technology such as insulin pump therapy. With the variety of options available, the newer technologies are becoming not only more and more accurate, but also more and more discrete. With wireless bolusing available, the ability to manage their diabetes without others being aware is more of a reality. In addition, the infusion sets and sensors are also smaller and less bulky, allowing them to be more easily concealed depending on where they choose to place the sites. I believe this paper confirms what we as clinicians have long believed to be the perceptions of persons in regards to technology and can address each of these head on with the clear benefits of improved glycemia not only in the here and now, but moving into the future.

Meeting the Challenge of Inpatient Glycemic Management in the Non-Critical Care Setting Q&A

Jane Seley, DNP, MPH, GNP, BC-ADM, CDE, CDTC, FAAN, FAADE

1. How do you address hospitalists who don't want to order a different regimen for discharge because "this is the PCPs job" so they send the patient home on their past medication without changing things.

There is no magic answer to this great question. I try to appeal to the hospitalists by pointing out that sending the patient home on an individualized diabetes self-care regimen based on their current status post an acute illness is key to preventing readmissions. I tell them about a recent study we completed (and others have found at other hospitals) to prevent readmissions that found that diabetes education was the most successful strategy to prevent readmission.

2. What strategies have you used to have blood glucose monitoring (BGM) times within that 30 minute window before meals?

Having enough point of care blood glucose meters is essential so that the RN doesn’t have to wait. Our written policy is currently within 60 minutes of meal tray delivery, but I teach to aim for 30 minutes whenever possible.

3. Do you recommend that the staff nurse give the diabetes self-management education (DSME) inpatient?

The staff RN knows the patient/family best and is in the best position to teach survival skills while performing them. Every time an RN checks their patient’s blood glucose (BG) or administers an insulin dose presents an opportunity to involve the patient/family in self-care. The diabetes educator should be called in when the patient is having difficulty mastering the survival skills and/or needs accommodations.

4. Do you have recommendations for giving prandial insulin for breakfast after the patient has had a hypoglycemic event? I find here that nurses often hold the prandial insulin without an order.

The proper order of tasks would be to treat the hypoglycemia until the patient’s BG returns to a safe level. We use >80 mg/dl (4.4 mmol/L) without symptoms or >100 mg/dL (5.6 mmol/L) with symptoms), then give rapid acting insulin to cover the amount of carbs about to be eaten. RNs need to be educated about the difference between correction and prandial insulin, and the importance of treating low BG first instead of the meal serving as the treatment for hypoglycemia.

5. Would it be beneficial once hospitalized to do a fructosamine level instead?

Serum fructosamine levels or glycated protein are indicative of glycemic control over the course of 2-3 weeks as opposed to the HbA1c which is 2-3 months. Values can vary based on conditions that affect serum albumin production such as nephrotic syndrome, thyroid disease and cirrhosis therefore HbA1c’s are preferred.

6. What do you tell staff to do when there is meal dose insulin ordered but blood sugar test is not done pre-meal?

This decision needs to be made on an individual basis. If the patient started eating before the BG is done, than the RN could get an order to give the insulin dose for a BG of 100-150 mg/dL (5.6-8.3 mmol/L).

7. Would you still give the ½ dose of prandial insulin an hour after a meal, if the nurse had not given it with the meal?

This needs to be decided on an individual basis, considering how much carbohydrate the patient has eaten, fat content in meal (which will prolong the breakdown of the meal) and the current BG. The higher the carbohydrates eaten and higher the BG, the safer I would feel giving ½ the dose. Our policy says up to 1 hour, not longer.

8. Can you provide evidence based practice for carbohydrate counting in the hospital?

Ryan, D., Swift, C. The mealtime challenge: Nutrition and glycemic control in the hospital. Diabetes Spectrum; 27:163-168, 2014.

9. How do you deal with patients being discharged on certain insulin or meters and then when they arrive at the pharmacy this is not covered, so patients cannot afford their supplies? Our case management does not verify insulins and diabetes supplies are covered by insurance. It is quite time consuming and almost impossible to get a definitive answer from Medicare or the patient’s insurance.

We are encouraging patients to use the pharmacy on premises to reconcile the prescriptions and get new prescriptions to match formulary as needed. We are also writing prescriptions for BGM with generic names.

10. Do you have any policy for patients self-administering and adjusting their own insulin doses in the hospital? We have some physicians who are writing orders to let patients use own insulin from home and adjust their own doses.

We have a written policy, Patient Taking Own Medication, that delineates how to do this and endocrine has to make sure the patient is competent.

11. We would also be interested in a program on the use of U500 insulin in the hospital setting.

I have had success switching a lot of the U500 pts to U100 during hospitalization and many have done well. Do a literature search, there have been a few papers published recently on U500.

12. Do you have any ideas for improving the management of patients who have recurrent admissions for diabetic ketoacidosis (DKA) or recurrent flare ups of gastroparesis?

I try to see patients with recurrent DKA and try to figure out what went wrong and discuss prevention in detail. We have a few experts in gastroparesis that we refer those patients to for consults.

13. What difference does it make testing BG 30 minutes or 60 minutes before the meal?

The longer you wait to give the insulin and the meal, the more the BG could potentially change and the insulin dose may not be relevant.

14. Any thoughts on telemedicine for transitional assistance?

There are some programs available that use text messaging support for transitional care. WellDoc is one of them - see BlueStar https://www.welldoc.com/product/bluestar.

15. What is the best regimen for steroid use in hospital on non-critically ill patient?

There are so many opinions on which insulins to use and how, and the guidelines are vague. The Society of Hospital Medicine Glycemic Control Workbook has some examples: SHM Glycemic Control Task Force, The Glycemic Control Implementation Guide: Improving Glycemic Control, Preventing Hypoglycemia, and Optimizing Care of the Inpatient with Hyperglycemia and Diabetes, and Society of Hospital Medicine website, Glycemic Control Quality Improvement implementation toolkit http://www.hospitalmedicine.org.

16. Do you use nocturnal BG monitoring in your facility? Have you seen a reduction in hypoglycemia?

We do bedtime BG as a standard, and there is an option to order 3 am BG if needed, but it is rare.