David Klonoff: 0:15

Hello, welcome to Diabetes Technology Report. I'm Dr David Klonoff. I'm an endocrinologist at Mills Peninsula Medical Center in San Mateo. Today we have a very special guest and my co-moderator, Dr David Kerr, will introduce him.

David Kerr: 0:34

Welcome everyone. I'm David Kerr. I'm speaking to you again from Santa Barbara, california, and our very special guest today is Guillermo Umpierrez, who probably doesn't need any introduction to most people who are involved in diabetes care. Hello, guillermo, welcome. We like to begin these podcasts with just finding a little bit about how you ended up being a superstar in diabetes technology, and particularly in hospital diabetes care. What kind of led you to this place?

Guillermo Umpierrez: 1:07

Well, david and David, I'm an endocrinologist and done diabetes clinical care and research for the past 40 years. So my first job description at Grady Hospital at Emory University was take care of diabetes at Grady and we have a large number of admissions, many of them with multiple complications. So I've always been in the hospital dealing with patient care and of course it was very natural that if I was going to do research it was going to be in this area of inpatient glycemic control. So I started doing full research. Well, a significant percentage of my time doing research in the late 1980s or no 1990s, sorry, and always been in diabetes care.

Guillermo Umpierrez: 2:01

So with severe hyperglycemia complications, amputations and, of course, in the last few years, related to technology.

David Kerr: 2:12

And what are you working on at the moment? What's the frontier of technology in diabetes research that's keeping you awake at night?

Guillermo Umpierrez: 2:21

awake at night. Well, right now we are completing two studies with technology. One is can you improve glycemic control with the use of continuous glucose monitoring? Can you target a glucose between 90 to 130 instead of 140 to 180? That has been the traditional target and using CGM? So we did it with real-time CGM and patients were treated in six different centers in the United States and with insulin therapy. Most of them received basal bolus and one had a real time. The other has a blinded CGM. And the answer is there was a trend to improve glycemic control but we did not see significant difference.

Guillermo Umpierrez: 3:17

And it's very, very hard in the hospital, especially for those patients who have significant hyperglycemia with elevated hemoglobin A1C, basically greater than 9%. It's almost impossible to type control them with so few insulin. So they are glucotoxic and the standard here is to start insulin in 0.5, 0.3, 0.5 units per kilo. That's maybe not enough. So what we learned in that study it was just presented a couple of months ago at the ADA is that if you have a patient with a hemoglobin A1c less than eight, it is much easier to have good glucose control If the hemoglobin A1c is greater than 9, we did not achieve good glucose control with the average pro glucose about 170, 175, even if you use or do not use a continuous glucose monitoring. So that was a big lesson for us. We knew that those patients, especially those who come to greater heart growth with hemoglobin 1C of 9, 10, 11, those are very tough people to control.

David Klonoff: 4:30

Guillermo, how do you balance the risks of delivering insulin that can lead to hypoglycemia versus the benefits of keeping people at a target glucose level?

Guillermo Umpierrez: 4:44

Yeah, and that's the big challenge. That's right, because hypoglycemia less than seven has been reported in different studies. We're between 8% to 28%, and there are a significant number of retrospective data that hypoglycemia is associated with poor outcome. We also reported that about 45% of patients with glucose less than 70 do not have symptoms. So it's a challenge.

Guillermo Umpierrez: 5:13

You're clinical how do you discover, how do you recognize hypoglycemia and what is the impact? So the only way that you will be able to recognize these asymptomatic hypoglycemia is with glucose monitoring or continuous glucose monitoring. And how do you do it? Well, we, just in the last few years, we have relaxed our target. So you remember those papers from Belgium in the ICU, 80 to 110. Now we know that's not the case. They don't help at all. In the non-ICU we used to say 70 to 130 as the goal for inpatients in the general world. Now we say somewhere between 100 to 180. And then if you want to go in the lower end, well, maybe use agents not associated with hypoglycemia. If not, just keep the blood glucose between 100 to 180. And that's the best way to prevent hypoglycemia. If you talk at 100 to 180, the hypoglycemia rate is less than 10%.

David Klonoff: 6:18

Guillermo, I was at your presentation at the ADA when you discussed this data, and I remember some of the doctors were having trouble with patients not getting fed either the correct foods or getting fed at the correct time. That can be as big a problem as, let's say, not finding the ideal dose of insulin. What do you think?

Guillermo Umpierrez: 6:37

Yeah, and if you don't eat or the insulin is given before meals and you don't eat, there's a risk of hypoglycemia. But more importantly, it's the challenge that we have with our nursing staff and food delivery, the trade delivery. Now we want the insulin to be given before meals, but many times it's during or after meals and in many hospitals, at least in Atlanta, we have short staff with nursing and there is a lot of nursing turnover that is very hard to educate them in a consistent basis.

Guillermo Umpierrez: 7:11

So you are right, David, that's the main problem that we have in the inpatient the timing between insulin and tray delivery and food intake.

David Kerr: 7:22

Guillermo, I was going to ask you about something you just said. So the HbA1c when you get into hospital, say for an elective procedure, seems to determine what's going to happen. So is your advice for people with diabetes who are maybe going to have an orthopedic procedure or whatever, that they really should optimize their glucose control before they even get to hospital, because it's so important?

Guillermo Umpierrez: 7:49

Yes, and the recent Endocrine Society guidelines published about a year, year and a half ago. One of the big questions was what is the ideal? Hemoglobin in one seed and ideally it's less than 8%. Those patients are going to be able to very well control in the hospital less hyper and hypoglycemia, better glucose control, less complications. Now that's the ideal for elective procedures. We don't want to hold surgery or cancel surgery because you're hemoglobin 1, ch9, knowing that maybe you need to be more aggressive or use ways to implement better glucose monitoring to try to achieve better glucose control.

David Kerr: 8:33

What about the other end of the hospital experience, the discharge end? Do you think we're doing a good job there at the moment? People who say have diabetes when they come into hospital, but there's also people where diabetes is diagnosed when they are admitted to hospital. Are you happy with what we're doing for the discharge back into the community?

Guillermo Umpierrez: 8:59

Yeah, so you just highlighted extremely important questions, right? We know that poor glucose control is associated with increased rate of readmission to the hospital and it has more complications. We know that if you use just point-of-care testing, finger sticks, somewhere around 20% 25% of them within three months have low blood glucose. So David Klonoff he has a smile on his face because he's the principal investigator of a randomized control study that we just finished, investigator of a randomized control study that we just finished and that study shows that the use of continuous glucose monitoring recognize much more hyper and hypoglycemic events compared to point-of-care testing. So you can ask the patient do finger stick four times a day? But unfortunately they don't do it.

Guillermo Umpierrez: 9:55

So we send the patient home with a blind CGM and with a Libre tool and those patients who work on the CGM have better. There was a trend, there was a pilot study just a hundred patients that have a significant trend on better glucose control lower time above range greater than 180 and 180,. Lower time below range less than 70, and much better what is called timing range. In that study we defined timing range 70 to 180. So I think that that study really opened up the field that you should be monitoring those patients that are sent home with insulin with a better monitoring tool, and that's the current recommendation. That's right. So patients with type 1, patients with hyperglycemic, even patients just on type 2 and insulin should go home with a continuous glucose monitoring.

David Klonoff: 11:00

Well, Guillermo, the study that you're talking about, that we did together. It showed improvement in outcomes, but they were not statistically significant and the number of subjects we looked at was probably not enough where we could have proven it. Do you think that there'll be future studies with larger numbers of patients to show statistical significance?

Guillermo Umpierrez: 11:22

Yes, yes, absolutely. And we in the large population base that we publish now in Diabetes Technology and Therapeutics, a few months earlier this year, a couple months ago, in the implementation of CGM in Andalusia, the south of Spain, in a couple hundred thousand patients, you reduce the number of hypoglycemia, hypoglycemia admissions and in patients with type 1 diabetes, implementation in a population, large population, will reduce the rate of diabetic ketoacidosis in type 1 by half. So I think that we have been doing finger stick for the past 50 years. I think that we have been doing finger stick for the past 50 years. I believe that we are now in the time that we should consider better glucose monitoring In the hospital. We recognize hypoglycemia and hyperglycemia After discharge. Now we know that you can improve glycemic control In an observational study. You reduce the number of readmissions and you reduce the number of diabetic ketoacidosis. So a randomized control trial I'm waiting for you to get the second paper, second study, run and we will recruit patients with you, guillermo.

David Klonoff: 12:40

another condition that you've done more work on, I think, than anybody else, is stress hyperglycemia. Could you explain that and how it differs from diabetes, if someone comes into the hospital with a high blood sugar?

Guillermo Umpierrez: 12:54

Yeah, our first study, I think it was 22 years ago. We noticed that about 28% of patients in the hospital come with high blood glucose and of them, somewhere around 10% to 15% have no previous history of diabetes. And now we have looked into this and we know that somewhere around 20% of patients no previous history of diabetes, having even a glucose less than 126 and hemoglobin 1C less than 6.5, develop hyperglycemia in the hospital, defined as a glucose greater than 140 and greater than 180. And both are associated with increased rate of complications, increased rate of surgical infections, admissions and even a trend in mortality. So stress hyperglycemia defined as newly diagnosed hyperglycemia in those patients, even with hemoglobin O and C less than 6.5, is common and is associated with increased rate of complication, maybe even the same or worse than those patients with prehistory diabetes.

David Kerr: 14:08

Guillermo, I've got to take advantage of you being on the podcast and ask you the red-hot, controversial question about GLP-1 receptor agonists and elective surgery. Do you stop them? If you do, when do you stop them? Should you stop them? Where are you in this red hot state?

Guillermo Umpierrez: 14:29

So you're following this recommendation of the anesthesiology group and I think that's a little. It's unfunded by good scientific data. Most of these were case reports, retrospective data.

Guillermo Umpierrez: 14:45

We do know that GLP-1, one of the mechanisms that it works is because it's delayed gastric emptying. But there is a tachyphylaxis. So we have symptoms that appear for the first few weeks but then patients do not have symptoms. So if somebody who do not have any symptoms of gastrointestinal nausea, vomiting, who have been on stable doses of GLP-1, do you really have to stop the medicine? So by observation data doesn't really help us much. Dr Klanov Group published a paper in, I think, jama not too long ago with a large number of patients and it shows that there is no increased rate of pulmonary complications. We, on the other hand, have reviewed the literature and published data suggesting that you have to individualize care. So if you have a patient coming to me, let's say an orthopedic surgeon says hemoglobin 1 CO9, tune him up. Surgeon says hemoglobin 1-CO9, tune him up.

Guillermo Umpierrez: 15:48

I'm not going to use GLP-1 if the patient is going to have surgery within the next one or two weeks, because I know that is when they have nausea, vomiting or they have gastric emptying syndrome.

Guillermo Umpierrez: 15:55

But if you have a patient that has been on stable doses with diabetes and without diabetes, maybe it's safe to continue to use GLP-1, but alert the anesthesiologist that the patient is in this medicine, they can follow the full stomach protocol.

Guillermo Umpierrez: 16:12

I mean, anesthesiologists have been doing anesthesia with everybody for many years, so they know how to deal with that. Now if the patient has symptoms, then you have to say, well, it's not urgent, maybe you stop the medicine. But if the patient has history of diabetes, you have to be concerned that the glucose is going to go out. So I think that we need to individualize, we need more data, but I don't think that this one way of doing things, stopping this Yelp to one, makes sense to me, for everybody, and which is the data so far that is coming out suggests that the rate of pulmonary complications is not as common. But of course all depends on when do you use it, how do you use it. The other thing, the final thing about your question, david, is if you have a group of people who are using this long-acting GLP-1,.

Guillermo Umpierrez: 17:11

Where are those with diabetes and those? The others are people who are just doing for weight loss. So in the people with diabetes, many of them already have gastroparesis, so those patients may be at a higher risk. But in patients who are obese, if they have been stable, I mean, I don't see much of concern that we have related to pulmonary complications or aspiration. But the real expert in this field is Dr Klanov. So, david, any comments?

David Klonoff: 17:44

Yes, a group of us, including Dr Kerr, reviewed a database of 130 million people. We looked at nine common surgeries and six potential complications of surgery that could be due to GLP-1, and we found no increased risk for any of these six complications. Guillermo, I want to ask you something else now. You recently led a very important international project to rewrite the treatment goals for hyperglycemic crises, and I know that you and the team presented this data at the ADA meeting last month. What does that project mean to you and what effect do you think it will have?

Guillermo Umpierrez: 18:39

So that was an update. We were invited to write an update of the ADA guidelines for the treatment of diabetic ketoacidosis and I was honored to serve as the chair. So my first request to the ADA was this has to be a global consensus. It just can't be ADA. So we invited the Diabetes Technology Society, we invited AIDS, we invited the European Association of Studies of Diabetes, uk Diabetes, and we have an international group of people working together to review the literature, because the last guidelines were written in 2009. And there was over 7,000 articles that have been published in the area. So we updated the consensus, providing new data on epidemiology, highlighting the increased number of patients with hyperglycemic admissions. We reviewed the diagnostic criterias and ways to treat people with hyperglycemic emergencies and, finally, we updated on complications and how to go to prevent readmission and how to go to prevent complications of treatment. But it was a global consensus, so with major organizations that are involved in diabetes care in Europe and, of course, in the United States.

David Klonoff: 20:10

Well, it had been 15 years and it was definitely time. Guillermo, I would like to thank you on behalf of both Davids for speaking with us today. You are definitely the expert in diabetes treatments and we're very fortunate that you joined us For the audience. We look forward to you joining us for future Diabetes Technology Reports. We're available on Spotify, the Apple Store and the Diabetes Technology Society website. So, until our next Diabetes Technology Report, we'll catch each other later. Bye-bye, thank you very much.

Guillermo Umpierrez: 20:50

Bye-bye, thank you.