Thank You!

Thank you for your interest! Here is the video:

Transcript:

CARBAJAL: Hello, everyone, and welcome to today’s webinar, Enhancing Diabetes Care with Technology: Key Strategies from Emory University and Grady Health System. On behalf of Becker’s Healthcare, thank you so much for joining us today.

So before we begin, I’m going to walk through just a few quick housekeeping instructions here. First, we will begin today’s webinar with a presentation, and we will have time at the end of the hour for a question-and-answer session. You can go ahead and submit any questions that you might have throughout the session by typing them into the Q&A box that you should see on your screen. Today’s session is also being recorded, so it will be available after the event, and you can go ahead and use the same link you used to log into today’s webinar to access the recording once it’s available. If at any time you don’t see your slides moving or if you have any trouble with the audio, try refreshing your web browser, and you can also go ahead and submit technical questions into the Q&A box. We’re here to help with those as well.

With that, I am pleased to welcome today’s speakers. First, Dr. Francisco Pasquel. He is an assistant professor of medicine at Emory University. As a clinician-scientist and trained epidemiologist, Dr. Pasquel conducts translational research focused on hyperglycemic crisis in clinical trials evaluating the role of different glucose-lowering medications for the management of patients with diabetes. With colleagues from Emory and other institutions, he created a website to understand transformation in diabetes care during the pandemic. Dr. Pasquel’s research is partially supported by NIH to study metabolism during stress and to study the interaction of COVID in diabetes.

Next, we have Jordan Messler. He is the executive director of clinical practice with Glytec. He trained in internal medicine at Emory University in Atlanta and subsequently served as an academic hospitalist at Emory University for several years after residency. He is the former medical director for the Morton Plant Hospitalist group in Clearwater, Florida, serving BayCare Health, where he continues to work as a hospitalist. He is the current physician editor for the Society of Hospital Medicine’s blog, The Hospital Leader. In addition, he previously chaired the Society of Hospital Medicine’s Quality and Patient Safety Committee and has been active in several of their national mentoring programs, including Project BOOST and Glycemic Control. Thank you both for being here today, Dr. Pasquel and Dr. Messler. I’ll now go ahead and turn the floor over to you.

JORDAN MESSLER: Great, thank you very much. Thank you all for joining us today. I’m excited to be here. And really, the first half here, I get to tee up the conversation and the excellent case study that you’ll hear from Dr. Pasquel. So in the next 40 to 50 minutes, we’ll talk about a few things, review tactical applications for driving hospital change – in particular, talking about technology to drive change and improve patient safety. We review some strategies used to create buy-in for new technologies and practice changes, and in particular, the importance, as we know, of a team-based culture to need to support change. It just doesn’t happen, and we need to get that buy-in from our team. And then really sprinkled throughout, the notions of diabetes technology in the hospital and where the future of that is headed.

So excited to be here and talking about accelerating change, which really we’ve seen dramatically over the past year because of the pandemic, during the pandemic, on numerous areas. You could probably double this list. We know research has really exploded. Just the amount of general research has gone up 100%. There’s been over 300,000 articles. Some areas recorded about 300,000-plus articles on COVID, about 200,000 of them medical-related, so a lot of research over the past year. Telehealth, we know, has accelerated. It was increasing outpatient and certainly continues to increase, and particularly in the inpatient setting, where I don’t think many of us expected it was going to take root so quickly.

I’ve seen a lot of my colleagues, the physicians, taking different roles, people in specialty areas being relegated to the front lines during the pandemic for the first time. In my career, I’ve talked to physicians that had to have been furloughed or let go and have difficulty finding jobs. Medical education, we know, has been transformed by forcing remote learning. And then face-to-face clinical opportunities with patients have clearly been reduced and trying to understand what that impact will be as well in the future.

Also, our volume has changed with low volumes in the spring and surges throughout the year. We’ve seen a variety of other changes around remote monitoring, just how we educate, conferences, use of social media, and in particular, this last one which we’ll highlight, rapidly adopting new technology. And who’s led that transformation? You might have seen this circulate – the leaders in transformation. We looked at the C-suite. But in this particular instance, it was something that forced all of us to change, which is COVID-19 and the pandemic.

Change has happened around digital technology throughout the world outside of healthcare. This, from McKinsey and Company, showing globally throughout the world, customer interactions that became digital, going from around 20% to 50% over the past year. In particular, I like this quote from Fareed Zakaria of CNN talking about this forced change in digital life – “The pandemic served as a forced mass product testing for digital life, and for the most part, our technological tools passed.”

And thinking this kind of re-framework here – we certainly can recognize this in how we did meetings, how we’re doing this meeting today, for convenience initially, and then recognizing, well, I can’t leave my home for safety. And then in many areas, it became a model of efficiency that, again, our tools passed and are going to stick around. I think you’ll see this theme, and Dr. Pasquel talks about CGM in the hospital that also went through these kind of steps, as well.

So let’s look at some of the technology in the hospital. I’m going to start thinking through a framework. There’s been a variety of ways for change for you and the leadership out there at your system at your hospital – probably have a framework to think this through. I like this one that I saw in Harvard Business Review. It really embodies a variety of ways that we’ve been thinking about changing and how change has occurred. I think this serves as a good model post-pandemic when we want to change and change effectively – strategies to think through.

So, one, forcing us to see the system differently. Two, that bias that we need to have to truly implement change, that bias towards action. Taking those steps to unfreeze the organization to be ready for change, prepared for it, uniting around a purpose, which was a pretty clear one this year, and then finding those solutions and the problem-solving. We’ll talk about a couple in the diabetes space – insulin management software, which we have at Glytec, and the use of that that you’ll hear from Dr. Pasquel with continuous glucose monitors.

So we’ll look through these briefly here. So the first one, seeing the system differently – here, in the past year, we were forced to pivot, embrace change. How can we set up that sort of embracing of change in the future for future projects? We know the importance. If we want to do change, you got to change the system. As we know from quality improvement, Dr. (inaudible) has said if you want to effect change, every system is designed to get the results it gets. So if you want to change, you have to change the system.

Really, we saw early on the importance of networking. New information was being flooded regularly. We had to learn from our colleagues. I think many around the country were talking to New York and the Northeast early on. We needed to hear from outside voices. And that’s how we should always be thinking when we want to see the system differently. Dr. Pasquel will share how he’s certainly allowed networking to really help accelerate change nationally as well. So always important to network, hear outside voices, and then have this bias towards action. This has been a model I’ve heard from a lot of my mentors. We’ve got to fail faster. We’ve got to learn from our mistakes, improve rapidly within a framework. Here’s a framework I’m sharing here, but also a framework – whether it’s the IHI model for improvement, Six Sigma, Lean Six Sigma – having a framework for improvement.

We can no longer – I think we’ve all recognized this past year – this is how we’ve always done it doesn’t work anymore. In healthcare and medicine, a need for constant improvement, continuous improvement. Talked about being mindful of that last line – I think we can all look back in the past year and saw things that we perhaps implemented a little too quickly and didn’t really have validity or data behind it. So really be thoughtful in our approach and understand the unintended consequences, too.

And then taking those steps to unfreeze in the organization as we’re preparing for change, getting that sentiment that change is necessary, having our team not operate in silos and all work together, that we can continue with the status quo. And we do that by uniting around a purpose. Here again, the purpose was pretty clear. But we always have a purpose that we can help unite our team around – patient safety, reduction of medical errors.

And then lastly, problem-solving. Again, we’ll talk through some technologies that we’ve felt have helped us solve some problems. But obviously, as a team, when you’re working through change, you’ll work through the solutions that you need to improve.

So let’s step back and talk about the challenge of COVID-19, and we’ll go through the challenge of COVID-19 with diabetes. Big picture, what we’ve all been thinking about and seeing – we’re, unfortunately, at a point here in our country – we’re over 540,000 deaths related to COVID-19. The job losses have been stemming. We know that employment was over 3% before the pandemic, reached highs of over 14%. Now, back down around 6%, but still over double where we were pre-pandemic. School closings have been a huge issue, a storyline all year. Thankfully, at this point now, it seems to be about three out of four schools have instituted some in-person learning, and that will hopefully continue safely. And all this has – we’ve all been aware of the inequities that this has amplified as well throughout the past year, and moving forward, how to rectify those inequities.

And then diving down closer to our theme today, diabetes. So we know diabetes is a primary risk factor for COVID-19 infection. Really, patients with diabetes wind up having worse outcomes with COVID-19 and tend to stay in the hospital longer with more complications. Particularly, stress hyperglycemia, high-dose steroids complicate blood sugar control in the hospital. And we know that these high blood sugars – hyperglycemia is associated with high risk of death. And then what we’ll really stress here and continue to stress is that aggressive glycemic management, even for those short hospital stays, is extremely important.

So looking at the lens through diabetes here, let’s understand briefly the status quo, where we’ve been around diabetes management in the hospital setting, recognizing that the prevalence is high, is pervasive. One in three Americans have glycemic control issues. Insulin is the standard of care to manage hyperglycemia in the hospital, but it is fraught with risks and errors. And then on the right there, 30-40% – in some hospitals we work with, up to 50% of patients during their hospital stay will have diabetes and/or hyperglycemia – stress hyperglycemia requiring insulin. So that’s a vast amount of patients that we’re managing regularly. In particular, during the pandemic, for patients that have hypoxemia, steroids have been the mainstay of treatment since the end of the summer. About 80% of patients in the hospital will have hypoxemia and qualify for steroids, and steroids contribute to the hyperglycemia as well.

Now we’ve had standards that are still in place, standards based on years of research. So one of the things that I alluded to at the beginning – 300,000-plus different studies this year related to COVID. So we’ve got to be mindful of the breadth of research and be mindful of ones that really drive change. So we have standards that should still be in play. We know that continuous IV insulin is the most effective method to reach glycemic targets and insulin infusions administered with valid written or computerized protocols.

Unfortunately, most hospitals in the US are still doing outdated protocols, and particularly I think this came to light during the pandemic, where protocols – I’m showing here protocols for IV insulin management that require multiple steps, that require, perhaps, calculations. So nurses that are managing COVID-19 pneumonia, managing sepsis, the complications related, and then glycemia management is on that list – the last thing we want to do is have our front-line workers have to worry about multiple steps, calculations when they’re dealing with all the other myriad problems that these patients were facing. Yet we know that a lot of health systems are dealing with protocols that have multi-step processes.

In addition, there’s a variety of other challenges that make it difficult to really safely get our patients under control and get their blood sugars under control. It’s often not a priority. Again, multiple other issues that our patients are facing, whether it was COVID-19, sepsis, acute myocardial infarction. So it’s not always a priority. Often, there’s poor teamwork collaboration, working in silos.

And one of the big things that I’ve seen – and I’ll echo a few times – is data. I would certainly reflect that your institution’s health system – what kind of data do you have around glycemic management? We know that a good portion of hospitals in the country don’t have any, or if they do have some, it may be one measure. They know their hyperglycemia rates, but don’t know too many other rates, and as mentioned about the pace of protocols, often using outdated tools. Today, we’re excited to talk about some of the technology that exists today and is emerging to be able to use.

So our team at Glytec worked with some national leaders early on in the pandemic to get a sense of the impact of blood sugar in patients with COVID-19. This is a study that we published early on in April, showing in red there patients with diabetes and/or stress hyperglycemia – because those are the group of patients that don’t have diabetes that have high blood sugars during their hospital stay. As you could see here, almost a fourfold higher risk for death in the hospital. Then we broke that group out to the diabetes and stress hyperglycemia group. So that bar in red, 41.7%, versus the bar in blue, the patients with normal blood sugar, you can see a seven-fold difference, a seven-fold higher mortality in those patients with stress hyperglycemia compared to patients with normal blood sugars during the hospital stay. So this is highlighted on the left side here that seven times’ higher mortality. That also led to a longer length of stay.

On the right side here, I wanted to highlight a study that Dr. Pasquel and I and other national leaders with Dr. Umpierrez, Dr. Klonoff did, published at the end of the year in Diabetes Care, where we looked at this target blood sugar opportunity window, recognizing that it can take a couple of days, two to three days, for patients in a non-critical care setting to get their blood sugars controlled. We saw about half the time that by that window, patients’ blood sugars were still high. And if they were still severely high in that 48- to 72-hour window, compared to people that have their blood sugars controlled during that target window, they had a seven times’ higher rate of mortality. So high blood sugar is related to worse outcomes. We have an opportunity to get those blood sugars down. And if they still remain high two to three days – risk for worse outcome.

These results aren’t surprising. This is similar to what we’ve seen with other conditions, but really highlighted by this large group of patients with COVID-19 we’ve been seeing the past year. But we’ve seen this with sepsis, coronary disease, pneumonia, stroke, critically ill. And we picked this target window, working with Dr. Pasquel on this, because we know from data that we can achieve control. In the ICU setting, we can achieve control in as little as six hours on IV insulin. In the non-ICU setting for most patients, we can achieve control within two to three days using a standard evidence-based subQ basal bolus regimen.

One of our coauthors, Dr. Umpierrez, says about this research that it suggests that blood sugar should be high on the order set for COVID-19 patients, irrespective of a preexisting diabetes diagnosis. Again, we saw that those patients without diabetes stress hyperglycemia had worse outcomes.

We can control blood sugars in the hospital, but for a variety of reasons, the study shows that it’s not happening or hasn’t been a priority for COVID-19 patients. Our findings strongly suggest that early intervention to bring blood glucose into a target range will reduce mortality rates in COVID-19 patients.

So I would think through how you’re managing these patients, and again, what’s applicable for COVID-19 is applicable for all of our patients with hyperglycemia issues. Think through your checklists. We know on our checklist for managing COVID-19 has been oxygen for those with low oxygen levels, proning patients – turning them on their stomach when their oxygen levels are low. We learned after a few months the importance of delaying ventilation, the use of remdesivir in certain situations. VTE blood clot prevention, steroids – all standards of care. Is glycemic management on your checklist? And if not, it should be. It needs to be. We owe it to our patients to get their blood sugars controlled during their hospital stays.

So again, review – patients with diabetes at risk for worse outcomes. 40% of COVID deaths have occurred, unfortunately, patients with diabetes. Again, steroid use, which is prevalent in patients with COVID-19 – again, those patients, particularly with hypoxemia, low oxygen levels, steroid use, contributes to hyperglycemia.

And then this last bucket, which will begin to tell the story that Dr. Pasquel will share with us, is that COVID did create a lot of challenges for our standard of care for critically ill patients – IV insulin. Preserving PPE was necessary early on and remains necessary when we face surges. The need to reduce nursing exposure led to places putting in things that were not standard. Often with IV insulin, you really need hourly blood sugar checks, and places were trying to reduce that. So other strategies that we could safely reduce that.

So just briefly, before I hand off to Dr. Pasquel, I did want to talk about our insulin management software that we have at Glytec and really the tools we built around that insulin management software in thinking about technology that we have available in the hospital today to manage. So in the center here, Glytec, our eGlycemic Management System. We work closely (audio cuts out; inaudible) this is a team-based need. Again, up to 50%, 40% of patients in the hospital with diabetes or hyperglycemia.

So on the top left there, Glucommander, that’s the name of our software – personalized insulin dosing decision support at the point of care. We operate on the top there in the cloud to make it as easy as we can be able to upgrade remotely. Moving on the right – I’ve already echoed and will echo again the importance of data. So we have our GlucoMetrics platform to get data back. And that GlucoSurveillance is a real-time tool, so not just retrospective data, but understanding in real time who would benefit from insulin. The SmartClick solution that we have helps make this seamless within your EMR. And then GlucoView is our alerting system to be mindful of certain alerts as we’re monitoring our patients.

Again, I wanted to echo the importance of data throughout all this. You can’t measure what you don’t know. If you’re not having data, you might not have any idea where you are around glycemic management. Truly, very important to track, trend, and respond in real time.

So the future for diabetes technology is here – this future that’s emerging. I talked briefly about our technology that we have to help safely manage insulin in the hospital. Glucose telemetry is emerging, and some recent great studies on that for patients that are on continuous glucose monitoring to perhaps track and trend really rapidly and predict the patients that will have highs and lows. Data analytics alluded to, insulin pens that are becoming more and more smart, and closed-loop systems being the future of really tying all this together really seamlessly.

I skipped CGM, that second one, but that’s what we’re going to really spend the next half here focusing on – really the great work that has been done by Dr. Pasquel and team and opportunity to use technology in the hospital. He’s got a great story to tell. So I’m going to hand off to him to tell his story – his case study of using CGM at his health system. I think you’ll hear echoes of what I’ve talked about, I would think, through a framework that can you use for change. You’ll hear a lot of those echoes as he goes through his case. So I’ll hand off to you, Francisco.

FRANCISCO PASQUEL: Hello, everyone. Thank you, Jordan. Thanks for the opportunity to share our experience during these difficult times. In the next 20, 25 minutes, I will share a little bit of what we have done at Grady to care for patients with COVID-19. Here are my disclosures.

As Jordan mentioned, it was very clear that patients with diabetes or stress hyperglycemia were having poor outcomes in the hospital. We were hearing everything that was happening in New York, and we were not seeing that yet, but it was obvious that endocrinologists had to step in to try to help manage patients with diabetes, and they are the ones that actually started using CGM in the hospital.

In the last few years, we’ve been working on strategies to individualize therapy in the hospital. For patients with type 2 diabetes, here on the left side in this figure, we have been thinking about how can we make this easier. How can we avoid that many finger sticks? How can we stratify patients? Now, we think that patients with mild hyperglycemia can be treated maybe with oral agents if they have the specific characteristics, some patients with basal insulin plus or minus oral agents, and those for more severe hyperglycemia with basal bolus. Obviously, for those with type 1 diabetes, we can continue an insulin pump if there’s a setup for that in the hospital, and obviously, basal bolus is the regimen of choice.

But on the other hand, as Jordan was mentioning, for critically ill patients, the targets are either within 100-180 or 140-180 according to different societies. But that is really difficult to achieve in this population, because patients are on steroids, patients are on vasopressors, patients receive medical nutrition therapy. So it’s really difficult to control and keep those patients in the right target, because we know that worse hyperglycemia is associated with worse outcomes in this patient. So we thought that we needed to do something to try to increase the number of patients using continuous insulin infusion to achieve targets.

And for patients with DKA, as you know – and we recently looked at DKA data from the Glytec database, and we’ve learned that patients with diabetes and DKA really have an increasingly high mortality compared to patients that do not have COVID-19, as you would expect, but the mortality can reach up to more than 45% in older patients with DKA. So it’s very important to control glucose levels in the ICU.

Really, continuous insulin infusion has been, in general, the way we do it. But as you can imagine, during the beginning of the pandemic, if you remember, there was no sanitizer. There were no masks. Wherever you went, there was no toilet paper. It was really a disaster. And you can see these responses – this is not just in the United States. This is across the globe, where there’s no personal protective equipment, people improvising how to care for patients. Then we started seeing some of our healthcare workers getting infected and some of them actually dying. That was really difficult to watch.

And then we heard that Dexcom and Abbott started responding to the pandemic and helping hospitals implement the use of CGM for remote monitoring so that we could monitor our patients, still see what their sugar levels looked like, but at the same time, decrease the exposure of our healthcare workers.

So based on this opportunity and other transformations in care that we were seeing, we started receiving protocols on different ways to treat DKA for patients with COVID-19. We started seeing the changes in CGM. So we started thinking that there is a need and probably a systematic way that we should assess those transformations in care. As Jordan was saying about the use of frameworks to try to respond to this disaster, we started thinking about the preparedness cycle to address these transitions. And when we got together with colleagues from the school of public health, obviously, the first response that we were thinking was response and mitigation, changes in CGM use, and the use of technology in patient response teams, e-consults, and understand what protocols are being implemented and how those protocols are being used, and then start learning.

We thought that we also needed to prepare for the recovery. We know that we’re switching to telemedicine, and then we’re going to probably go back to some sort of hybrid method. So we were kind of trying to predict what was going to happen. We thought about the need for surveillance instead of looking at patients longitudinally, and that’s what happened now. There’s a couple of grants at Emory looking at patients, and now we’re thinking about following patients with post-acute COVID syndrome to try to understand that. But this idea that we need to see what’s happening, we need to evaluate those changes, and we need to systematically study those changes to see if they’re actually working.

So we created a website to collect all those different transformations. You can access that. It’s still active. It’s covidindiabetes.org. We have a Twitter account where we share protocols and now literature related to the topic. During this experience, we started talking to other colleagues across the nation that were using CGM, and based on those ideas, the way CGM is being used, and based on our previous experience, this is kind of like the way that it’s been implemented in the hospital. So this could be done in the non-ICU setting or in the ICU setting.

But the main concept is that with the current technology, those sensors that you can place in a patient, you can actually get the data remotely if you are within six meters or 20 feet. In the case of the Libre, which is another sensor being used during the pandemic, you need to flash that intermittently to get that data. But the point is that if you have this smartphone, then you can obtain that data. You can send the data to the cloud, and you can have followers. You can send that data to the nursing station for telemetry. That data will go to the cloud for population health management, where you can get summarized data of glycemic control, and that is also being integrated into the EHR.

But there’s also a need to document. Because we can be using a CGM device, but where do we document that in the chart? So we thought that there was a need to document CGM values in the EHR and find some sort of way to make sure that the sensor values that we were using to care for our patients were accurate, because we didn’t have a lot of information, especially in the critical care setting.

So we got together with some colleagues from New York and other centers where there was experience with the use of CGM in the hospital, and we started actually thinking of how do we do this? How do we implement this emergently for patients with COVID-19? And we focused a little bit more on the two devices that do not need calibration, meaning that now patients can use those devices in the outpatient setting without the need for finger sticks – that actually can make clinical decisions or insulin adjustments. Those are the sensors that we were focusing on. That’s the Libre from Abbott and the G6 from Dexcom. And we put some recommendations and a little bit of information on how to use them, what to expect, what are the potential interactions or interferences, what is the duration, the range that you can – the ranges of glucose – how long does it take to work, etc., so that you could have a little bit of a guide on how to implement that.

Also, we started thinking about who are the most appropriate patients for this use? Obviously, those patients with the highest glucose fluctuations, high glycemic variability, maybe those patients with steroid-induced hyperglycemia, type 1 diabetes. And then we started thinking about the need for protocols to make sure that the sensor values that we are using, we’re confirming and making sure they are accurate.

But when we think about the use of CGM in the hospital, we thought that the main place where you will get the highest benefit is really in the ICU, because in the ICU, to be able to deliver insulin intravenously, you need to check point of care every hour. That’s what most protocols use is an every-hour glucose check, so you can get an insulin dose to adjust those glucose values to keep the glucose in this specific target.

In the non-ICU setting, we typically use subcutaneous insulin, and we check glucose levels before meals and our bedtime. So that’s about three or four point-of-care testing. But really, in the ICU, you are talking about 24 point of cares needed. So the nurse will go to the patient and do a finger stick or take a capillary or an arterial sample to do that point of care. That’s where we thought we needed to make a change.

And in our conversations with other centers, we saw that there was this protocol from Kathleen Dungan from the Ohio State University. We started this conversation on how they were doing it, and we actually invited them to our future conversations.

But just as a background, Citlalli – she’s my postdoc fellow. She’s an endocrinologist working with technology in the cardiac ICU. We took a look at our data to see how the sensors were performing. We are measuring continuous glucose monitoring in patients in the cardiac ICU to understand changes during stress. And we saw that the accuracy of the sensors were pretty good. But we saw that there were potential interferences in this setting. So in this figure, on the left side, we have the recovering sensors. So we have Patient 1 in the OR and the same patient 24 hours in the ICU. These are patients that just underwent coronary bypass. They are typically on vasopressors after surgery.

And as you can see, on the left side of the figure, if you can focus here on the blue lines, there’s a little bit of gaps. Those little gaps, those are for the sensor when the patient is undergoing surgery. So at the time of surgery, because of the electrocautery, we think that’s when this happens. So there’s a loss of signal, and there’s a potential interference with surgery. However, when those sensors recover after surgery, you can see that those sensors were pretty accurate in patients receiving vasopressors – meaning that a patient that is receiving vasopressors, you will think that the accuracy of the sensors will not work, but as long as the mapped values are OK, the sensors work pretty well.

On the right side, you can see that those interferences – the longer interferences in blue – some of those sensors did not recover by the end of surgery. So you can see that in the 24 hours following surgery, those sensors actually didn’t work, meaning that there’s an interference with that technology in the operating room. But at least we learned that if the sensor came working well from the OR, that sensor was accurate in patients with vasopressors.

So the next step was how can we integrate that with a hybrid protocol with point-of-care testing in a computerized algorithm for insulin infusion? And we also started thinking about EHR documentation to keep track of what we were doing in the hospital. Based on the conversations with Kathleen Dungan and Eileen Faulds from OSU and Shivani Agarwal from New York, it was clear that there was a need for inviting all the relevant stakeholders for making these changes, because it was going to be really difficult to implement this.

So we invited, from the beginning, endocrinologists. Eileen Faulds from OSU joined several of our Zoom meetings. We invited unit managers, safety officers, the ICU team, administration, bioengineering and lab team, pharmacy directors, nurse directors, nursing leadership, and IT. And we all started thinking, what is the best protocol? What do nurses need?

Actually, the changes to the protocol came from the nurses. One of the things that Jordan was mentioning is that they don’t have the time, and they don’t want to be calculating values. So there was a need to find a system to validate that in the EHR so that that could be helpful for nurses.

Essentially, the system works with a smartphone that is within 20 feet. There was the possibility of having the pump externalized outside the room, implementation of glucose telemetry. We needed EHR documentation in the use of the computerized algorithm. Fortunately, we have had this at Grady for a long time. We started some of the initial studies, and I had experience with Glucommander, since I was a fellow in one of the cardiac surgery studies that we did years ago. And we also thought that there was an opportunity to have remote monitoring by the endo team and our PharmDs. Based on the availability of this dashboards, these platforms, we also thought that that could be useful to follow a large group of patients.

So the technology setup – so here, on the left side, you can see the smartphone outside of the patient room. The pump was externalized, in some cases, following a protocol that was led by the PharmDs. You can see here a tablet in the nursing station where you can see the glucose values remotely and see the alarms.

Here’s data from our first patient and the different platforms where you can follow those glucose values. Actually, our first patient had a cardiac arrest and was resuscitated minutes later. You can see the loss of signal and the alarm that we received when that happened. We thought that that’s maybe a relevant signal of how we can see the changes that are happening in our patients. And we have been looking recently into CGM and cardiac arrest and see if we can predict loss of pulse. But this is how you can see it.

And then here is the EHR documentation. Here’s the protocol where we will enter the CGM value and compare that to a point-of-care value. As long as the CGM value is within that 20% range from the point-of-care value, that sensor was validating. So that is our validation criteria.

Here we have the protocol on the right side. So if the sensor was close enough, then the nurses could start checking point-of-care testing only every six hours, as opposed to every hour. And every hour, the nurse could enter the glucose values into the computerized algorithm for insulin adjustment.

Here on the left side is the app that you use to follow patients, and we have been following sometimes even up to six patients at the same time, where we will get the alarms when the patients are getting low or persistently high to see if we need to make any adjustments. And on the right side is how you can actually get the summarized data on the overall control of our patients. This is the ambulatory glucose profile that we use in the outpatient setting that can also apply to the inpatient setting.

This is some of our initial experience. Our first nine patients – here, you have a correlation between point-of-care testing and the first CGM value. So you can see that the correlation was pretty good, with a .78 R-squared, meaning that the values were pretty close. There was just one value that didn’t meet the validation criteria. So this means that most patients, when they started with their CGM, their CGM was ready to be used with the protocol.

Here’s an example of the patients that we followed. Here you can see – in blue, you have the CGM trends. In green are the point-of-care values. In gray are the insulin adjustments per hour. We started seeing some potential interferences in this patient. So you can see that there’s these little drops, these little points going down in blue. It’s a little hard to see on here, but these were moments where there were mechanical interferences with the use of CGM, and our protocol said that every time when your point-of-care or your CGM value is less than 100, we only used point-of-care. We started that from the beginning, and it was actually helpful, because at some times, you may have some interferences that you need to take care of or change the sensor.

Here’s more closely the example – hypothermia devices, for example, interfered with the CGM. If the patient was on pronation, on some of the protocols for COVID-19 patients, you could have also those potential interferences. But following the protocol, we were able to identify all of them, and the system simply will go back to point of care until we have again a validation.

So the advantages to this approach is – the first one, and the reason why we implemented this, is because we wanted to decrease the burden of care for nurses. This is associated with less PPE use. We are reducing about 60% the number of point of cares needed. We were able to maintain adequate glycemic control. We looked at the data from New York, and the timing range was about 40% in those patients in the ICU with written protocols. They do not have a computerized algorithm, and many of those ICU patients were treated with subcutaneous insulin. So we can achieve good glycemic control.

Obviously, we decreased the blood loss. We’re using CGM. We’re not having to get an arterial sample every hour. And if the patient is awake, you will have less finger sticks if we’re doing a finger stick.

We can also follow remotely the patients. We had the endocrinology team and PharmD following. We had also a telemetry system where nurses could see the values in the nursing station, and we had a documentation. So the documentation is extremely important in this process. And finally, as we said, the comprehensive performance metrics.

So in summary, we know that CGM can replace finger sticks in the outpatients, so that the new devices that are factory-calibrated – you don’t have to do a point of care all the time to make sure that the sensor is working. We still think that this is needed in the hospital, because there’s potential interferences, as I showed you, in the OR or in the ICU. There’s potential mechanical problems. There’s a lot of potential interactions with medications that we don’t know that are being investigated, and there’s a lot of interest about learning about those. We know that FDA is for now not objecting to the use of CGM, and they are interested in doing studies with YSI, which is the gold standard for glucose measurements, to make sure those sensors are accurate in a wide variety of patients.

The technology now is incredible. You can follow remotely. You can have followers – either family members, doctors – anyone can follow the patients in real-time, especially with the G6 that is available right now. As I showed you, we’re doing this in the US, but this has also been expanded in the UK and Canada – has been approved, or at least there’s no objection for the use in the ICU.

So we need to still be vigilant, because there’s potential interferences with CGM. We don’t think it’s ready to replace point of care, but we still think it’s very helpful, and I think it has helped our system right now. This is a protocol that is driven by nurses. There’s two leaders in the ICU that take care of it. They actually place the sensors themselves. They send those invitations to us to follow. But it’s been really incredible, and we think it’s been positive. So now the next steps are to actually evaluate data, looking at harder outcomes, and that’s what we’re working on right now.

So with that, I’ll finish here, and I think we have a minute for questions, and we will be happy to answer any questions.

CARBAJAL: Thank you, Dr. Pasquel and Dr. Messler, for a great presentation today. So we will go ahead and start the Q&A. It looks like there’s some questions in there. So the first question here is the correlation of high glucose with mortality is well known, but in this condition, do you have any outcome data that tight control changes outcome?

MESSLER: Yeah, great question. I could start answering it a little bit, and then if you’ve got anything to add, Dr. Pasquel. Yeah, great point. We have data in a lot of other conditions. There’s been certainly a lot of literature in COVID-19. As I mentioned, there’s 200,000 articles. I might have missed some.

But we do know there’s one around COVID-19 and outcomes over the summer out of Italy. They had a small group of patients. Sardu and others did publish a small paper looking at insulin management in the hospital and outcomes and had composite outcomes that were better in COVID-19 patients that had better glycemic control.

What we’re relaying here, as you point out, we know that there’s a correlation between high blood sugar and worse outcomes on admission. There’s a correlation with high blood sugars that persist during the hospital stay and worse outcomes. Is there clarity on controlling those blood sugars with insulin during the hospital stay and those patients having better outcomes? I think well-designed trials are needed, although we’ve extrapolated what we’ve seen with COVID, what we’ve seen in other studies. So I think we’ve had good base of data that the outcomes are there when you control blood sugars. Anything else to add?

PASQUEL: Yeah. So I think that’s a great question, because glucose is a marker of disease severity. If you’re sick, your glucose levels are high. So it may be just a marker, and it’s not necessarily mediating those poor outcomes in patients.

However, when you look back into data from years ago, there was a clear benefit in reduction in glucose levels. This is even before the Belgium studies with the tight glycemic control. In the beginning, the target is just bring it down below 200, and there was a reduction in outcomes in observational and in smaller clinical trials. Then Van den Berghe conducted those clinical trials in a single center where they did a close monitor, and there’s a lot of discussion about how it was designed based on the glucose that patients received from the beginning in parenteral nutrition. But in that study, there was a significant reduction in outcomes in patients that were randomized to a tight glucose target. However, when those results – we tried to reproduce by the NICE-SUGAR trial, a multi-center trial that was to finalize that question. They actually showed that being too aggressive, overly aggressive, could have worse consequences.

So the discussion is not so much about glycemic control or not. We know that higher glucose levels are worse. The discussion is more about the target. If you’re targeting less than 100 with a paper protocol, you’re putting the patient at a huge risk. However, if you look at the cardiac surgery community, they target lower glucose levels. There’s a previous study from Lazar, for example. They will go for 100-140, as opposed to 140 over 180.

But now, the recommendation is to not go to the overly tight target. In the ICU, we target 140-180 in medical patients, because we think that’s a safer range. But based on a study we did with cardiac surgery patients, when we used a computerized algorithm – we used Glucommander – our lower target patients, the non-diabetic patients, actually had less complications, and there was not a single episode of hypoglycemia. So I think the matter here is about getting better control, but removing the confounding factor of hypoglycemia. So if you can do it well, I think that’s a good service to your patient.

CARBAJAL: That’s great. Thank you both there. Our next question here is will recovery COVID-19 patients with diabetes have added challenges for control of BS?

MESSLER: Can you repeat that? I don’t think I heard it all.

CARBAJAL: Oops, I’m sorry. Will recovery COVID-19 patients who have diabetes have added challenges for control of blood sugar?

MESSLER: Ah, so asking about the long-term effects of COVID-19 patients with diabetes? I think that’s still unclear. I think we saw a prior coronavirus with MERS that there was a group of patients that developed diabetes from exposure to coronavirus. It did seem to persist, but it seemed to be maybe only a couple of years. So will there be new cases of diabetes from COVID-19? I think time will tell on some of that. There’s been some examples of COVID-19 contributing to hyperglycemia and perhaps persistent diabetes, but still more research needed.

PASQUEL: Yeah, and I would like to add to that, that there has been a lot of interest in this potential role of SARS-CoV-2 on entering the pancreatic beta cell and increasing the risk of DKA in patients or causing diabetes. And there’s some signals that that may be true. But there’s an important paper in Cell Metabolism from Al Powers where they could not see a significant number of ACE2 receptors, for example, for the lesions from the virus that were initially hypothesized.

But there’s a registry called CoviDiab. I think Dr. Rubino from the UK is leading that effort from Oxford, and they’re collecting information about patients with diabetes or new diabetes that occurred during COVID. It’s going to be hard to make a conclusion from that registry, but at least we’ll have some initial preliminary data.

But there’s a lot of interest from NIH in funding studies that look at this, and I think that’s something that people will be focused on now is understanding the risk of diabetes and sequelae related to COVID. They call it PACS, or post-acute COVID syndrome. So I think we’re going to start learning more about that. There’s already publications out there that I think diabetes is just one of the potential sequelae that could occur, but we’re not sure at this moment.

CARBAJAL: Next, we have – what were the obstacles, if any, did you experience with the lab approving your protocol?

PASQUEL: So at the beginning, more than anything is that this technology is not known by the community in general. This technology has been always kind of like a field for the endocrinologist. Just suddenly, with COVID, this became much more common. And also the fact that it’s much more affordable, especially for the Libre now, you can see it everywhere, and patients are having more access to it.

So the first thing is that, for example, ER doctors didn’t know about CGM. Critical care doctors did not know about CGM. The nurses did not know about CGM. So the first obstacle is that part, the knowledge. So coming into the ICU and show them the devices and giving them the opportunity to wear those devices, to understand how they work and how powerful the tool is, I think that’s important. So bringing the devices and showing them that this is what it is and show the data and how it works – I think you can gain a lot of buy-in with that.

And then involving all the stakeholders is extremely important. Because as doctors, we’re coming in with this device and saying we’re going to place it on a patient, and we’re going to start – that’s not going to work. Because really who’s taking care of the patient, is next to the patient at the bedside, is your nursing staff, and they’re the ones that are doing it. If you get the buy-in from them, then you may be able to implement the protocol if it’s useful for them.

Otherwise, we don’t really have the backup or the data to change practice right now. We’re continuing to work on getting meaningful data to see if this is actually helpful in the long run, decreasing clinical outcomes. So for now, it’s been a matter of implementing this emergently just to help get better control for our patients and reduce PPE waste and exposure.

CARBAJAL: Thanks, Dr. Pasquel. It looks like we have time for a few more here. So our next attendee question is how does your system interact with your EMR, your electronic medical record system?

MESSLER: For Glucommander, is that what you’re asking? Is that the question? Or for the CGM? I can ask about Glucommander first, and then you can talk about CGM, Francisco. But yeah, Glucommander, we’re EMR-agnostic. We had a terrific IT team. We worked with our systems IT team to make sure that our insulin management software in Glucommander basically appears to be a seamless interaction for the end user. So it’s in Epic and Cerner and other systems, and when you’re in a patient’s chart, it’ll be a tab that your end user can click on to access Glucommander. So really, a seamless integration. If that was more related to the protocol that Francisco did, I’ll defer the rest to him.

PASQUEL: So we’ve used Epic, and we’ve helped several other centers to create the documentation in Epic for the protocol for the documentation. That’s a critical part. Unfortunately, right now, it’s more manual, so there’s no continuous glucose monitoring data going in real time into the chart. That’s not available yet. That’s one of the questions – if we will be able to do that in the near future.

We use Epic, and we have an automated calculation of the ranges for glucose values so that the nurses don’t have to do the calculations, see where the glucose value is, and they will document. So they will document in the flow sheet. And that will also create a trend of CGM that overlaps with point of care in different areas of the chart. So that is really helpful, at least for now.

There’s a lot of interest in importing summarized data, like the AGP, as I showed you in the EHR. Abbott is working on that, and there’s an interest also from Dexcom. Juan Espinoza in California has implemented or has the EHR integration of those reports for the care of children with type 1 diabetes that come to the chart –so they are getting all that data integrated into EHR. So I think it’s not going to be complicated for the summarized data. I think that the part of the real-time data is a little bit more of a step that, hopefully, we will have in the near future if this actually works.

CARBAJAL: Finally, here, we’ll have time for one more. How do you see what you’re learning about CGM – or is what you’re learning about CGM in the acute care setting changing diabetes care in the ambulatory or primary care setting?

PASQUEL: I don’t think we will have an impact in the ambulatory setting right now. We’re using this example in our healthcare system to actually expand it to our primary care and say, hey, there’s opportunities for change. And we’re trying to expand the use of CGM for our diabetes clinic. We’re expanding right now. There’s a discharge program that we’re working with Abbott. Dexcom has been helping for the inpatient side.

But I think there’s an opportunity to start those discussions that this technology is extremely helpful, particularly for those higher-risk patients. So patients now with type 2 that are on multiple doses of insulin that have risk for hypoglycemia, they’re actually benefiting from that technology. And the technology is becoming more affordable. Now, this is integrated into the standards of care, so there’s standardized metrics for CGM interpretation. So this is going to change a lot in the near future. The uptake on the use of CGM is going up incredibly rapidly. We will be hearing about this technology a lot in the next few years. We think it’s positive, and it’s really changing the quality of life of our patients.

CARBAJAL: Well, that’s great to hear. It sounds like there’s a lot more to come there. That is all the time that we have for today, and I want to again thank Dr. Pasquel and Dr. Messler for an excellent presentation and to Glytec for sponsoring today’s webinar. To learn more about the content presented today, please go ahead and check out the resources section on your webinar console. Thank you again for joining us today, and we hope you have a great day.

MESSLER: Thanks. Bye.

PASQUEL: Thanks.

ECO #01125-A