Loading summary
Transcript58 lines
- [00:03]
A
Welcome to this special three part series of Diabetes Core Update where we will discuss the cutting edge topic of automated insulin delivery for people with type 2 diabetes. Yes, I said for people with type 2 diabetes and I know that might be surprising for some, not for everyone, but who's listening? And we're going to explain why we're talking about it as we go along. I'm your host, Dr. Neal Skolnick, professor of Family and Community Medicine at the Sidney Kimmel Medical College of Thomas Jefferson University. And this special series of Diabetes Core Update is sponsored by Insulet. Joining us for today's episode, we are so fortunate to have a good friend of mine, Davida Kruger. Davita has been a certified nurse practitioner in diabetes for more than 40 years at Henry Ford Health in Detroit, Michigan. Her roles there include both clinical practice and research. She is past chair of the American Diabetes Association Research foundation and has served on the American Diabetes Association's Research Policy Committee. She is also a past President, Healthcare and Education of the American Diabetes Association. She served as editor of diabetes spectrum from 2005-8. She was editor in Chief. I'll be done soon, don't worry. Of clinical diabetes from 2011-16. She was just awarded the ADA's Rockmill Levine Award for Distinguished Service and has been an author and over 100 publications, including relevant to today's discussion. Two different trials published this year, one in JAMA Network Open and the other in the New England Journal of Medicine, looking at automated insulin delivery and type 2 diabetes. Davida, it's a good thing you haven't done more than you have or we'd have no time left to talk about our topic.
- [02:06]
B
We barely have time for what we need. I'm so glad.
- [02:10]
A
Thanks so much for joining us.
- [02:11]
B
It is my pleasure and my honor. I'm really excited to be here today.
- [02:16]
A
Davita Traditionally we've thought about automated insulin delivery aid for people with type 11 diabetes. Everyone with type 1 diabetes needs insulin. Those regimens are often complex and often quite unforgiving. Why should we now be thinking about aid for people with type 2?
- [02:35]
B
Well, first of all, I think I challenge almost anything everything you just said. So I think the way I look at it, first of all, everybody should be on a sensor. I always think of CGM as a right, not a privilege. And so as we meet all these patients, they should be on a continuous glucose monitor. Then 7.9 million people in the United States are on insulin. And we're doing an imperfect job because if I give you an injection of insulin, you know, the horse is out of the barn. I don't know when it's going to onset, peak duration, absorption. I'm not sure you're going to remember to take your lunchtime or your dinner injections. And so when I start thinking about insulin, I think of the insulin pump as another vehicle. You have syringes, you have, you know, different devices. And an insulin pump is one other way to deliver insulin. And I think we've had a fear of insulin, that they are very complicated. And what we learned doing these two trials is they're not as complicated as we thought and that patients are very able to do this. And we took people from where they were. Some of them had more education than others. Some of them were carb counting, some of them weren't. We just said, hey, you want to do this study? And we were blown away at how well patients did.
- [03:58]
A
Is that right? You know, it's interesting, I, I sometimes wonder. We say patients feel it'll be too complicated, but I think it's often physicians and clinicians that think things will be too complicated. So we're hesitant to then roll that out to our patients because we know that listening to our podcast, we have a lot of primary care clinicians and endocrinologists. I know the endocrinologists are very familiar with automated insulin delivery, but many primary care clinicians aren't.
- [04:32]
We're all familiar with long acting basal insulin, short acting mealtime insulin. But can you explain what automated insulin delivery is and where it might fit in, kind of how it works to accommodate for all those.
- [04:48]
B
Sure. So this is what I spend my whole life doing. If you're not wearing a bump, you're not seeing me. So automated insulin delivery, you have to have a continuous glucose monitor, which is your first step, and the sensor and the pump talk to each other. And so depending on. And each company has their own algorithm, which in of itself is a whole other thing, but easy enough to learn. But depending on which the pump is, you set it up with basal background insulin just like you would in an injection, and then you set up the algorithm for insulin to carb ratio correction factors, max basal insulin, and a number of other things that the pump needs to have to get started. And so then the sensor talks to the pump and based on what the blood sugar is, it may increase the insulin or decrease the insulin. And some of the companies call that an extra bolus. Some of them call it just giving extra insulin. There's a whole lot of different terms for it. I don't think we have to get caught up in that. The most important thing about aid, and the most exciting thing about aid, is that the pump talks to the sensor, which we know are accurate. And based on what the blood sugar is and the numbers we've given it, it will increase the insulin, decrease the insulin, and if 30 to 60 minutes, again depending on the pump, before you're starting to go low, the pump lowers the insulin and will shut it off so that you don't get hypoglycemic. So think about that. Think about what the patient is most concerned about, hypoglycemia, what the healthcare provider is most concerned about, hypoglycemia. And by marrying the sensor and the pump, we can take care of so much of that and making sure the patient gets injections.
- [06:45]
A
So it's sensing the direction that things are going in. Not always where it is right at this moment, but where it's heading and then it's adjusting for that. Is that right?
- [06:56]
B
Well, yes, but one of the newer pumps that just came to the market, called the Twist, its algorithm actually looks at where you are now, where you're supposed to be, what your goal is and how to get you there in the next number of hours. And it adjusts for that too. So it's really kind of geeky.
- [07:18]
A
Yeah, no, that, that's interesting. And I just want to go back to basics for a minute because you used the term, it has a basal insulin rate, but in the pump itself, there's not actually long acting basal insulin.
- [07:32]
B
Correct.
- [07:33]
A
So what's in the pump?
- [07:35]
B
It's all rapid acting insulin. So the patient only has one kind of insulin. They fill a cartridge of some kind, depending on the brand of the pump, that lasts for up to three days. Some of them are cartridges that just attach to the body. No tubing. Some of them have tubing. And again, we learned that patients can do both. And after two or three days, depending on how much insulin they use, they change the whole setup. The patient is responsible for changing the whole setup. And they can do it. It's amazing.
- [08:10]
A
Again, I'm chuckling because you have so much experience with.
- [08:19]
Patients being able to do this, and I have so much experience with clinicians being afraid that patients can't do this. And I think that's something is important to acknowledge. And that's why in the second episode, which you're also going to be joining us for, we're really going to get down to basics and talk about how to do this, how to teach patients. But before we get into that, which is Our next episode, you were involved in two of the most important trials of aid for people with type 2 diabetes, both of them published this year. Can you share with us what those trials demonstrated? So we have a sense of that kind of bird's eye view picture. How well does this actually work?
- [09:06]
B
Okay, so the, the first, the Insulet trial, which was published in jama, was. There was, there was no one off the pump. Everybody got pump. So it wasn't a randomized control. Everybody got palm. And I mean, I can go into details and nitty gritty of the A1Cs and you probably know it off the top of your head, but here's what I think is really important. If your A1C was 6.5, you stayed at 6.5. If your A1C was 9 or 10, you came down by 2%.
- [09:42]
And so the higher the A1C, the better the patient did. So here's the thing, everyone got better A1Cs for the most part. And you could argue that why would you give it to somebody who was at 6.5? And I would tell you it's because it's a better lifestyle. Those people who have 6.5 struggle to get there with all the things they have to do in diabetes. So that's another whole conversation because I would give it to you whether you're 6.5 or 13. But here's the thing. Some of the patients were on over 100 units a day.
- [10:14]
And 31% of the individuals who entered the Insulet study were on over 100 units a day at the end of the 13 weeks. That's a short study. Only 10% were on greater than 100 units.
- [10:29]
A
Now, now why would that be?
- [10:31]
B
So I think there's a couple things. When you start a pump, you take all the insulin down by 25% of what they were taking. The absorption's better. You don't have a long acting insulin, you just have rapid acting insulin. The body does better with acting insulin, or patient wasn't taking what was prescribed and now it's attached to them. And so all the background insulin or basal insulin, whatever we want to call it, but we're going to call it basil because that's what it is. It's not long acting, it's the basal insulin that's rapid and it's attached to you. You want to bolus, you just bolus your mealtime bolus. But we also looked at the data that individuals who bolus once a day, twice a day, three times a day and four times a day. And there was a smidgy difference in the A1C, like maybe 0.2 difference between one time a day and four times a day, which also tells you that when blood sugars are high, the pump does correct. So. And for years and years and years, we've said you can't have a pump unless you carb count. And that's a misnomer. You do not have to carb count with any of these insulin pumps. We'd like you to take a mealtime bolus. But we also have proven that you don't have to take a mealtime bolus. We saw that in Insulet study. We saw that in tandem study. The Medtronic people presented their data, and that was not the case either with a Medtronic pump. So I'm also doing a sequels pump and type 2 diabetes. And we're not mandating carb counting. We're asking for boluses.
- [12:06]
A
Interesting. Now, I realize I'm moving a little bit back and forth between what's going on and some practical things, but as you've mentioned mealtime insulin a few times just now, I want to clarify and understand. So normally mealtime insulin, let's say you're out at a restaurant and you eat a little more than usual and you want to take mealtime insulin. There's a challenge to that. You have to lift one shirt up, you have to take out a syringe or go to the, you know, go to the bathroom to excuse yourself. Just so everyone understands, how is mealtime insulin bolus done? When you, you've got a pump on.
- [12:48]
B
On your cell phone now, you can use the device attached to you.
- [12:54]
A
Right.
- [12:55]
B
But most of people use their cell phone. So there's an apnea.
- [12:59]
A
They look like everyone else I see.
- [13:01]
B
At dinner at a table. Exactly. So that's the other magnificent thing you can on some of the pumps, not all the pumps. You can bolus off the pump or off your phone. But most of my patients opt to bolus off the phone and there's no reason why they can't.
- [13:20]
A
Yeah, it just makes it so much easier. And we'll come back to that issue that you alluded to, how challenging it is for many people and how this just makes life easier. I want to go back now to some of the data you mentioned that from the.
- [13:39]
From the JAMA trial. What did the New England Journal trial show?
- [13:44]
B
Almost similar stuff. They had about the same lowering of the A1C. Carb counting was not important. A high insulin doses came down. It was like, there we Go. And it reminds you that, that the people that need insulin pumps will do well. And, and here's the other thing from my perspective, Neal, is that it's not my diabetes. It's very important that I understand every pump on the market, which is a full time job, because you just need to figure out what the algorithm is. Only because I have to go to a different website and I need to understand it's not hard to do. But I have to know every pump because I want the patient to pick what's best for their lifestyle.
- [14:31]
A
I will say you, as you know, truly an expert in the area.
- [14:38]
Need to know every pump. I'm going to suggest for most of us clinicians, just like we might not know every medicine in a certain class of medicines we get most used to using, we need to know medicines in a particular class and we get used to using one so that we know where the edges are and are very comfortable with it. That might be the same here, that for many of us who don't see the volume of patients that you do, we choose one to become comfortable with. Now, I just want to be clear on two other things. Degree of hypoglycemia, you alluded to what the study showed and also time and range, which increasingly is another metric which.
- [15:23]
B
Absolutely. Both studies we're talking about achieved greater than 70% time and range. And when you had the option to change the target, like in the, like in the omnipod, you can't change the target in the tandem, but you can change the targets. And the highest targets 150 down to 110, there was no more hypoglycemia by changing the target down to 110, but you had a better increase in time and range. So if a patient sat at 150, and we don't do that, we usually just use 110 because all the other pumps on the market are 110. And so if I took that patient from 150 down to 110 as a target for the pump to work towards. So the pump has to have a goal and a target kind of thing that it works towards. So if I did 150 and then decided to change that, it increased the time and range by 10% without hypoglycemia.
- [16:22]
A
That's interesting. And when you mention those targets, it strikes me that that's not all that different than when I often start basal insulin for someone just a once daily injection. I will often have a more lenient initial target because we're all afraid of what will happen and then incrementally go to a more rigorous level of control. So it sounds like the same can be done with the automated system.
- [16:51]
B
Absolutely. But I would tell you, with the automated system, because the pump is so smart and does protect the patient, we don't have to be as afraid to get the patient a treatment goal. So it's remarkable. We only had 13 weeks in both of these studies and the amazing stuff that we were able to accomplish, what the patients were able to do, and it's all on the patients, because all I do is say, yes, we'll change that, we'll put that up, we'll put that down, and thank you for coming to your visit and doing this. They do all the heavy lifting and so I have to say they did remarkable without a lot of extra support. And there's always extra support in research, but there's also real life, real world data that followed the studies that actually showed the same thing as we did in clinical research.
- [17:51]
A
That is fascinating. Are there any downsides?
- [17:55]
B
You know, you're asking the wrong person because I live and breathe pumps, but I think the downside is getting the healthcare provider comfortable with pumps. Because the reality is, as we move into the world of type 2 diabetes, there's no way Endo can manage it all. So whether, whether it's picking one pump or picking two, I don't just do it. We have to move it into primary care. And for some people, they don't like anything attached to their body and some people won't wear a sensor, but I think that's happening less and less and less than I've ever seen because you really need a sensor to be integrated. But you know, I can't think if you're going to be on insulin, this is the vehicle of choice to deliver that insulin. That protects you from hyperglycemia and protects you from low blood sugars.
- [18:47]
A
So davita, even starting someone on once a day insulin, not just people who are on multi dose insulin, is that right?
- [18:54]
B
Yes, we did that in the studies, both studies, and we're doing that in the sequel study as well, because we miss where they need extra insulin, quite honestly. And so the pump will adjust. And if they had a bigger meal than usual, you can't see that they were hyperglycemic unless you're looking at the sensor. And if you're just doing basal insulin in primary care, many of those people are not on sensors. And so you're kind of dragging along, waiting to add. And so I think that, yes, we did that and the patients did very well. The other thing, and here's the thing that blew my mind and you know, it takes a lot to blow my mind, is that there was a group of patients that were on a stable dose of GLP1 and SGLT2 inhibitor. By FDA regulations, we had to include a certain number. And so many people think, oh, I did my GLP1, they're at maximum dose. I can sit back, I'm done as the clinician. No, the A1C lowering, the magnitude of the A1C lowering was the same in the patients on stable GLP1 as it was on people that weren't on GLP1, which proved to me that I was wrong, that my patients on GLP1 most likely also at some point in their career of diabetes will need insulin.
- [20:22]
A
Well, it's interesting. I mean, in many ways GLP1s push off the time until you need insulin. Yes, right. Because they really are incredibly powerful agents. And when I know studied, and we've talked about this on other podcasts, when studied head to head against basal insulin actually have greater A1C reduction. If I remember correctly, one of the trials, tirzepatide versus one of the basal insulins, basal insulin got about 1.2% decrease. Tirzepatide was found 2.2. But that doesn't mean that you're not going to eventually have that pancreas where it just isn't having that much there to squeeze out anymore.
- [21:09]
B
Yeah, yeah. And it doesn't have to do with the duration of your diabetes. It has to do with the individual. And so that would be what I would say is, yes, GLP1 is our go to SGLT2. Inhibitors are important, but watch and monitor that patient. And if you're thinking that patient needs insulin and you should be, at some point, think about an insulin pump, even if it's only basal.
- [21:32]
A
You know, it's interesting, Davita, because that issue of therapeutic inertia is something that we've talked about before and is very real, that you have Someone and their A1C is creeping up over time and it can take a long time until basal insulin is started. And I'm going to say there's an end. If I remember correctly, and this is years ago, the Treat to target trial of.
- [21:59]
Basal insulin of largine had people on Entry had an A1C in the low to mid eights, but only half of them ended up on basal insulin alone, getting to an A1C below 7.
- [22:15]
Which is why we then go to multi dose insulin. But the point being that we still end up in that place where we need something more.
- [22:24]
B
Well, the other thing Neil and I know you know these numbers is that less than half of the people with type 2 diabetes have an A1C less than 7. And there's a ton hanging out between 8, 9 and 10%. And so that's, that's awful that only 50% of. And think of the numbers of people we have that have type 2 diabetes and type 2, only 50% of them are less than 7. Now, you and I could argue whether 7 is the goal. And I would say to you, it is the goal. If I can put you on a sensor and I can put you on a pump, that will protect you from hypoglycemia, because you know that hypoglycemia is the rate limiting factor. Why am I not going to give you an A1C or help you get an A1C of less than 7%? And as our patients age, they still are vibrant, they're still working, they're still doing all kinds of things. And I get very offended when we talk about what age, what is old. That's a personal problem. But, but I think we need to recognize that the, these people that are not in an A1C of 7%, we're not giving them the right therapies. And because we have A1, we have CGM, because we have GLP1s, because we have SGLT2 inhibitors, we should be able to get these patients to treatment goal. And if not, then they should be considered for insulin and we're not doing it soon enough. Too much inertia.
- [23:47]
A
I think you're absolutely right. And you know, there's always a discussion that goes back and forth between patient and clinician on starting insulin because I do think it's something that patients are afraid of. They've heard different people's experience and often misinterpret that experience. We've all seen this story where a patient will say so and so a relative was started in insulin and then they had a complication, an amputation. And they relate, not that it was someone waited too long to control that A1C but rather that there was a causal relation. Nonetheless, there's a lot of burden when insulin is started and particularly increased burden when we go to multi dose insulin. What has been your experience with the degree of burden that is entailed by automated insulin delivery? The fancy word is diabetes distress, but I like to think of it as personal burden.
- [24:48]
B
Yes. Well, first of all, I'll go back to my comment before, when I said that we had patients in Both studies that A1Cs between 6.5 and 7 and their A1Cs didn't necessarily come down, but their burden came down, their personal distress came down. Because now they didn't have to find injections four or five times a day and know where their insulin was and know that they had two injections a day, four injections a day, two different kinds. And the risk of hypoglycemia because even with a sensor it lets you know that your blood sugars are going down, but the insulin can't be stopped. So I think that yes, you have to fill this device every three days and you have to wear it, but that's better than four injections a day. And I do think I talk to patients all the time and you know, in the best scenario, there's several injections or more that are missed per week that affect the A1C. Now if I have something attached to you, there's a better chance. And if that thing that's attached to you has a basal rate that adjusts the insulin, even if you miss your mealtime insulin, that really does decrease the patient's distress and also family members. So we saw that there were questionnaires in both studies that asked pre and post about the distressed and there was a huge decrease for the majority of the patients.
- [26:20]
A
And that's really a big deal because we always need to address two worlds. One is the hard outcomes, things like A1C and its relationship to long term complications. But at the same time, in parallel, the truth is patients care about their day to day experience and when we can have the two of them line up.
- [26:46]
B
Isn'T that marvelous? It really is. And here's the thing. Even though I'm very, you know, I like to hang on to everything, the reality is I can't. I have to share at this point in my career because there's too many people that will benefit by this. And so many of those people are just not being seen by endocrinologists. So I really think it beholds us all to spend the time to explain this to the primary care world and see who raises their hand that says, I can do this, I want to do this and I want to bring this to my practice and focus on those individuals so that we have a place for all people with diabetes that can get the kind of care that they deserve.
- [27:29]
A
That makes sense. You've convinced me.
- [27:34]
So aid for people with type 2 diabetes is still relatively new. Which devices are approved for use in people with type 2.
- [27:42]
B
So we're very lucky right now. Tandems pump is Approved, the Control IQ plus insulets, OmniPod 5, Medtronic 780G, and we should be done with SQL's Twist by February. And fingers crossed they'll have FDA approval as well. So the data is that strong that there really was not an issue for any of these pumps with magnificent algorithms that really do improve the lives of people diabetes. And I have to say, and again, I have a prejudice because I live and breathe insulin pumps and have my entire career. It really makes your life easier as a clinician and I think that's where we have to take people to understand that not only is it a better way for people diabetes, but as a clinician, you have less burden as well. And so I think that's, you know, it's like someone looking at an A1C versus looking at time and range. You know, you look at time and range and your world opens. If you put someone on an insulin pump and you can see what they're doing and what's not happening, it makes your job so much easier. So I hope that's where we can take this. We've got the pumps, we've got the data. I could tell you my patients love it and they have type 2 diabetes. We just need to convince them, the healthcare providers, because that's how we're going to get this done.
- [29:07]
A
That makes so much sense. Davita, we're about out of time.
- [29:14]
Any final thoughts you'd like to share?
- [29:16]
B
You know, I think the big thought is that if you're thinking about insulin, you should be thinking about an insulin pump as the vehicle of choice. It does decrease the burden for the patient and it enhances the clinician's ability to practice at a whole new level. And I think these pumps are simple enough, easy enough for both the patient and the healthcare provider. And I really do hope that in the next phase of what we're all trying to do, we can bring it to primary care.
- [29:45]
A
Davida, it is always such a pleasure talking to you. Thank you so much for joining us and to our listeners, of course. Thank you. You're the most important part of this podcast and this is the first of a three part series on automated insulin delivery for people with type 2 diabetes. In this first part, boy, you heard it. A very convincing overview of some really important benefits with a new technology. And new technology is always hard to adapt to. We all have different levels of comfort with technology. But I think the data that we heard today was really convincing. And I'll say for me, not just the data, but the way that it makes people's lives easier while improving the lab values that we look at is what's particularly convincing. Nonetheless, it still can be intimidating. So in the second episode, we're going to discuss the details of patient selection. And I think davita made a point. Patient selection is not going to be the big issue here, but we're going to go over also practical advice about how to start automated insulin delivery for a patient with type 2 diabetes. I think that is going to be really important. This special series of Diabetes Score Update is sponsored by Insulet. We thank you for listening. The American diabetes association. I'm Dr. Neal Skolnick. Till next time, stay safe and keep learning.
- [31:27]
Sam.