So I'd like now to pull all the things we've heard together and talk about best practice and a roadmap for people with diabetes across the full spectrum of diabetes using see GM technology. And we've come a long way in diabetes from landmark clinical trials like D. C. C. T. And U. K. Pds onto developments in insulin pump technology and then the development of continuous glucose monitoring somewhat challenging and imprecise initially. But improvements and standardized CGM parameters that have become available only in the last few years That are driving changes in practice and the fact that we can use these systems for much much longer. Currently up to 14 days the freestyle libre has Been available for a few years now and in 2018 had optional alarms. And now we have liberated three which is bringing yet more advances into practice. Let me now share with you some new data that you're probably not aware of. One of these is a study presented a few weeks ago at the A. T. T. E. D. Conference in europe. And this was a meta analysis of 74 studies after literature search at over 30,000 participants with type one and type two diabetes mainly type one using the freestyle liberal system. And across all age groups, reductions in a onesie at three months were similar for adults with type one. All type two diabetes about a .5% reduction in a one c. That continued cool 24 to 7.5 months in type one diabetes as well as in uh in type one diabetes both type one and type two diabetes. And then those with the highest starting A one C. At greater reductions in A one C. As we've seen before. And this occurred in both in Type one and type two. What is important is that these changes in a. one c. would then sustained for 24 months in people with type one diabetes and for 12 months in people with type two diabetes, that doesn't mean it won't be sustained longer. It's just that there's not enough data in Type two for knowing whether it will sustain longer than that. So Evans had all concluded that meta regression analysis of real world evidence confirms that using freestyle libre is associated with not just in significant reductions in a one c. But that these reductions can be sustained over a long period of time. Up to 24 months and now I want to share with you a study that is being presented at this meeting here in new Orleans at these scientific sessions. Looking at people who are on basil insulin. So, and they particularly focused on ambulatory glucose profile to drive the treatment decisions. So very often with people on basal insulin, we only look at the fasting glucose and make changes based on fasting glucose. Now with this technology we could look at the profile across the day and that's very easily done by reviewing the ambulatory glucose profile which will allow you to make changes in their daytime treatment as well as based on their fasting glucose. So it helps evaluate healthcare providers understanding of the A. G. P. And developing a rationale for making changes. So in this study 136 people with an A. One C. Of 7 to 10% ah were enrolled in 10 sites and they were they used freestyle, liberate pro sensors and had at least seven days of sensor data to generate an A. G. P. All these uh physicians and health care providers received training on a. G. P. Interpretation. So here's what happened 94%, that's 99 out of 105 of the patients were recommended a therapy change. Uh This is just based on a few days of C. G. M. That basil insulin was increased About 40% of the time They initiated Bolus insulin in 18%. In a few people possibly because they were getting hypoglycemic. They decreased the dose of basal insulin in a few and I would sort of speculate these were people with profound postprandial glucose excursions who may not have wanted to go on to parental insulin GLP one receptor agonist was initiated In another 11% existing medications were increased and then about 11% and SGL- T2 innovator was initiated. A few people had a decrease in existing medications. It may be due to hypoglycemia or other observations on the a. g. p. and about 9% of people had uh and change in their diet and lifestyle appropriate recommendations for people based on the possibly showing postprandial glucose variability. So just a few days of C. G. M monitoring and looking at the A. G. P. Allwards a change in therapy for about 95% of people. So let's look at three patients from this study ah patient one oh average glucose of 1 62 and A. G. M. I. Of 7.2 and years the pattern. And if you look at the a. one c. 6.9, if you would just doing a one c, you'd be very happy with this patient on basil insulin with fasting glucose. That wasn't too bad. Uh And uh and a one C. Of 6.9%. However, when you look at the A. G. P. You see that the patient is getting some hypoglycemia during the night and very high peaks during the day, particularly after meals. If you look at patient, two Same enemy same average glucose slightly higher a one c. And here you see that the overall curve is flat, but the patient is getting some peaks, no hypoglycemia like the first patient, but some peaks after some meals. And that is sufficient to be driving up the A one C by 1.3%. And you could have a discussion with this patient about maybe choosing Brandel insulin or GLP one, maybe an SGL T two inhibitor to decrease the variability or an appropriate dietary intervention And then finally years patient three again, same average glucose and same GMI, but the a one c. 7.6%. This patient has a lot of post dinner hypoglycemia which could be addressed by either maybe a pre dinner in insulin injection or a shifting of the diet. No hypoglycemia during the daytime. So you don't need any intervention for lunchtime And a high glucose after breakfast, which may mean a high carbohydrate load at breakfast time. So, a discussion about all this with the patient would be important and occasionally some hypoglycemia drift down during the night, which occurs in many people, but on some occasions is dropping below 77%. So uh A one C alone would not give you this information. So much of this has been reviewed in a review article recently published to understand the clinical implications of differences between the G. M. I. And the duplicated hemoglobin. Some of it is understandable that GM is based on the current glucose profile and the current blood glucose over the previous for a few days. Whereas the blockaded hemoglobin reflects what has happened over the A 2-3 Month period, maybe blockaded albumin would give you a shorter term metric to make a comparison to, but Glad created a hemoglobin is also based on the fact that people get like eight proteins at different rates. So there's some high gladiators and some local educators and that causes some confusion and this high H. G. I. Has been taught to be genetically mediated and an important factor in hypoglycemia and mortality. So what this allows us to do and this was also presented at A. T. T. D. Is to try and develop a personalized glucose A one C. Relationship for the management of individuals with with diabetes. Uh if you look at the C. G. M. Data and the A. One C data, Here's the average a one c. And the years, the average glucose. You see quite a scatter. It varies a lot and it varies sometimes based on uh ethnicity. It's you see higher A one season african americans that's actually very well known. You see uh the lower ones in younger patients and perhaps higher in older patients. So race and gender, race and age matter but not gender. And this relationship between glucose and A one C. Is very well known to be discrepancy and actually very important and in in predicting outcomes. So calculating a personal average glucose A Gr will help develop individualization of the target. So you don't have an inappropriate the target for somebody who's probably got a high H. G. I. And therefore has a falsely elevated A one C in relation to the to the average blood glucose and vice versa. Mhm. So what we want is the whole picture everything is important. Sometimes the A one seat helps in the long term uh and we have to recognize it's different from what's going on in the previous couple of weeks, which we can see on uh the A. G. P. The other metrics matters. Getting time and range as you heard is very important. And you can see that sometimes people will go to a one C. Might have a very poor target range with both high and low blood glucose and that's not good for the for the patient I mentioned the accuracy. I want to emphasize again that we've come a long way with accuracy. And compared to the gold standard, the Yellow Springs, most sensors today have very good accuracy Uh for concentrations below 70 it's accurate and about 90% of people uh within about 15-20 mg from the true blood glucose and uh at least Within a plus of -40 and 99%. If you go into people who are above 70, the metric there would be want to be within 15-20%. It's close to 90% again there and if you take a little wider range uh it's it's almost 100%. So we have the ability to have sensor accuracy for all results within plus or -20 in about 93% of individuals. Again, compared to this yC reference range at different levels, you have the mean absolute relative difference way lower than most other metrics that we are in most other techniques for estimating glitchy mia that we have. So in conclusion, continuous glucose monitoring has become a foundational core strategy and technology for optimizing care of people with diabetes across a wide spectrum, both for Type one and type two diabetes. Thank you very much for your attention, and we welcome questions.
Related Presenters