Hello everyone, welcome to today's webinar. So this webinar is going to be talking about the operative care of patients using wearable diabetes devices, some tools and a strategy for their safe utilization, and we are hosted by the American Diabetes Association Clinical Centers and Programs Interest Group. My name is Paulina Cruz-Bravo, I'm an endocrinologist and associate professor of medicine at Washington University in St. Louis, I'm part of the Division of Endocrinology, Diabetes, and Lipid Research. So this is our today's agenda, we're going to be having a roundtable discussion with our panelists and at the end there's going to be a question and answer segment. So for the questions, we're going to be taking those at the end as I mentioned and please put them in the Q&A box so that we can make sure we're answering them and keep track of them. The chat box is going to be for announcements and links that we are going to be sharing with you. And I also have a few announcements, the future webinar is going to be during Diabetes World Day on November 14th and the title is Hyperglycemic Crisis, Putting Recommendations into Practice and you can see the speakers on the screen for that webinar and the link to register will be shared in the chat. And finally, this is the Diabetes Pro Community Forum for members so the discussion from today's webinar and other topics can continue there. So these are our today's panelists, we have a great group and it's my pleasure to introduce them to you today. So Dr. Errol Hirsch, he's a professor of medicine in the Division of Metabolism, Endocrinology, and Nutrition, Department of Medicine at the University of Washington School of Medicine. So Dr. Hirsch has a very long career, he got his MACA degree from University of Missouri School of Medicine, then did his residency at University of Miami and his fellowship in endocrinology at Washington University in St. Louis. He has a very productive career over the years focusing on studying insulin therapy and technologies for type 1 and type 2 diabetes and more recently he's interested in glycemic data management from continuous glucose monitoring and its relationship to traditional diabetes biomarkers and how they can predict microvascular disease and diabetes complications. He also has done incredible work advocating for patients so that they can have access to insulin and technology to improve their care, has over 300 publications, numerous books, has been part of numerous clinical trials for both type 1 and type 2 diabetes and this year he actually retired sadly from clinical medicine but he continues his work as a researcher. Our second panelist today is Dr. Troy Wiles. He is a professor of anesthesiology in the Department of Anesthesiology at the University of Nebraska Medical Center and he serves as the vice chair of Preparative Medicine Quality and Safety at the University of Nebraska Medical Center. He focuses his clinical work in preparative medicine especially management of patients with diabetes and has been instrumental in developing and implementing management protocols at both Washington University and University of Nebraska. And our third speaker is Dr. Janet McGill. She's a professor of medicine in the Division of Endocrinology, Metabolism and Lipid Research at Washington University School of Medicine. Dr. McGill also has a very long career, has been an active clinical investigator and has participated in over 200 clinical trials for both type 1 and type 2 diabetes as well as its complications. Additionally she has a particular interest in atypical diabetes and is a self-investigator for the RADIANT study, has numerous publications. He serves as editor for our Washington Manual Endocrine Specialty Council book and she has been awarded numerous awards including the 2022 Outstanding Physician-Clinician in Diabetes Award from the ADA. So those are our panelists for today. So I prepared some questions to guide the discussion. So full disclosure, we have been working on this topic for the past couple of years and we recognize that diabetes technology is a rapid evolving field, things are changing quickly and there is really an urgent need to have guidance and tools to help clinicians in this space of preoperative medicine so that patients can undergo procedures as safely as possible. Some of the discussion points you can find in a publication that we did on this topic and the link for that is going to be shared in the chat box and I also want to acknowledge our other co-authors that are not here, Dr. Alexis McKee, Dr. Hou Hsien and Dr. Gail Ringenberg who were also part of our paper. So to start with, I want to ask the first question to Dr. Hirsch. So currently, how common would you say is the use of continuous glucose monitors and insulin pumps in patients with diabetes? Well first of all, thank you for the invitation to participate today. I want to preface my comments by saying first of all, this is a topic that all of us who see patients, it comes up quite frequently and it comes up even more frequently now due to the explosion of diabetes technology. Now these numbers that we quoted in the paper because of how quickly this is exploding are already out of date. Let me give you some examples. It's estimated that in the U.S. between 25 and 30% of Americans are treated with insulin, which translates to about 6 million individuals. Now some estimates are closer to 8 million individuals treated with insulin and the use of these wearable treatment devices is expanding rapidly in both type 1 and type 2 diabetes. Remember that even Medicare now, not when we started writing this paper, but now Medicare and most insurance companies, for example, where I live in the Pacific Northwest, they will approve continuous glucose monitoring even for basal insulin alone with type 2 diabetes and the ADA also recommends this. When we look at type 1 diabetes using insulin pump therapy, it increased from 1% in 1995 to 53% in 2017, but this is an old number and it's very clear the number is even higher now. Within this type 1 population, CGM increased from 3% in 2006 to 38% in 2017. Now those 2017 numbers come from the T1D exchange. That was a long time ago. AID system usage is also rising rapidly. When Medtronic first came out, they had 123,000 patients with type 1 diabetes using their 670G AID system and that was just in the U.S., but registry data shows 33% of patients with type 1 diabetes were using AID in 2022. I can tell you in our clinic in our type 1 population in Seattle, we are north of 75% using pumps and almost all of them are using AID systems, so we can't keep up with these numbers in these registries that we're quoting here. AID has been shown to be safe and effective in patients with type 2 diabetes and the use is expected to increase in subpopulations of patients with type 2 diabetes requiring insulin therapy, pending additional data. I can tell you very recently, earlier this year, the Omnipod 5, for example, was approved for type 2 diabetes. We are completing a trial with type 2 diabetes with another pump. I just found out that there is a third AID system doing a study and then finally there are two companies that are developing specifically type 2 patch pumps, so pump therapy is likely going to explode in type 2 diabetes. My final comment for patients working in the hospital and especially going to the operating room, really important point, just within the last month, we now have two over-the-counter continuous glucose monitors approved, and so we are going to be seeing more and more pumps and more and more CGMs coming in, and that includes CGMs in people without diabetes or perhaps of the 90 million Americans with pre-diabetes. So suffice it to say the technology is just exploding. Yeah, thank you, Dr. Hirsch. Yeah, that's fascinating. Dr. McGill, so my next question is for you. Can you briefly describe what is the current state of wearable devices? So what type of devices are out there? How they connect to each other? Well, thank you, Paulina, and thanks, everyone, for joining. Thank you to ADA for hosting this webinar. This is not rocket science-level news for members of ADA. This is things you all know, that wearable devices include CGMs in insulin pumps. More and more often, the insulin pumps are automated insulin delivery systems that can be either hybrid closed loop or fully closed loop. The values from the CGM are needed to regulate insulin infusion rates via Bluetooth technology and an algorithm typically on the pump, sometimes on a phone. These systems are designed to avoid hyper and hypoglycemia in the real world and to achieve stricter glycemic control. The question is, will they work? Should they be continued? Will they work? Should they be discontinued? And which devices should be discontinued or perhaps even ignored or something else during surgery? Exactly. So continuing to how do we decide, should we continue the devices? Should we not? These are considerations that really need to be planned and should be incorporated to the preoperative planning of these procedures. So for Dr. Wiles, what are some important considerations in this case for continuous glucose monitors when patients undergo procedures? Thanks, Dr. Cruz. And again, thank you to all the attendees for taking time out and for the ADA for sponsoring this panel. So first, one of the things we advocate when it comes to CGMs in the peri-procedural period is that we should do whatever we can to not remove them from patients when they're coming in for procedures. And we recognize how disruptive this can be for patients and continuing their monitoring and treatment once they get out of surgery. And so we'd like to make sure that we don't remove them. And of course, this involves also trying to make sure that they're placed in a location where they're going to be compatible with the procedure and the positioning for that procedure when they come in for those cases. So one of the biggest challenges that speaks to our potential use of CGMs in the intraoperative settings is that the current data that we have suggests that we do not get uniformly reliable data across all patients during procedural care. So this may be related to different procedural energies that patients are exposed to during procedures, things like fluoroscopic imaging, electrocautery. We do need additional data to better understand why these CGM failures are occurring in these procedural settings. And another problem we have is I think it's unfortunately very easy for just any clinician to kind of expect that a device that works well in the outpatient setting is going to work just fine when a patient comes into the hospital setting and in for a procedure. And so I think we really have to educate that this currently isn't known to be the case and that we can't rely on these patients or on these devices during surgery. The other issue is that if we have accuracy issues of CGMs, we also have patients who are often anesthetized and can't voice any symptoms they may be experiencing due to dysglycemia. So another safety factor that really contributes. So for all these reasons, we really advocate that even if CGMs are in place on a patient, that really the cornerstone of perioperative monitoring for patients with diabetes is going to be point-of-care glucose testing for the time being. Now, of course, as we'll talk about in a few minutes, for patients who are going to be on automated systems during surgery, we do need to rely a little bit on having accurate CGM function and connectivity during procedures. However, again, we're going to advocate that this be accompanied by regular interoperative monitoring of glucose using point-of-care testing and not relying on CGMs to provide the sole monitoring for those patients during surgery. And finally, I think that we do allude to that it seems like it would be particularly useful to have more use of CGMs in perioperative care in the future should we show them to be more reliable. But another issue we discussed in the paper is just that we don't currently have technologies in place to allow clinicians caring for patients in inpatient and intraprocedural settings for them to see the actual glycemic data that are coming from these devices. That's another roadblock that we're going to need to work out. Yeah, so those are very important considerations. And you mentioned kind of like the fluoroscopy or kind of different energies that could be used during the procedures. And another thing that we discussed in the paper and mentioned was making sure the CGM is not like there is no like too much pressure in that area from where the patient is, how the patient is positioned, because as you said, they're like under anesthesia, so they're not really moving, and that could potentially also interfere later on with the measurements and it might not come back even after the procedure. It might still not work. Okay, so moving on to insulin pumps. For Dr. Hirsch, can you give us an overview, kind of the differences between the standalone insulin pumps and the AAV systems in this space of perioperative surgery and planning? Sure, this is really, I think, the meat of the paper, because this is just so critical to understand. First of all, right now, we have two different types of pump, tubed pump and tubeless pumps, and this becomes critical for those caring for patients during surgery to know about, because you don't want those tubes to get in the way. And as I mentioned earlier, we are going to have more tubeless patch pumps come on the market, specifically more for type 2 diabetes, but also type 1 in addition. But the big thing that differentiates the standalone pump or what we call open loop compared to the AID or the closed loop pump has to do with the basal delivery. In the open loop, which we've had really since the early 1980s, the basal sets are set, they're preset. Yes, we can have basal rates that we switch for exercise, and we can have alternate basal rates, but they're still set. On AID, these are modulating basals, and the basals go up or down based on the sensor, which communicates with the pump and the algorithm. These algorithms are all different. The way you would change all of the settings on these different pumps are all different. There is a wonderful non-branded website from the Barbara Davis Center in Denver I want to refer people to called pantherprogram.org, and how these pumps all work under the hood and how you can change these pumps are on that website. It's kept up to date as the technology changes. I use it a lot. I have our fellows use it, and I think it's a wonderful resource for everybody. But from our paper, I think it's important to note that, you know, what the question is, is that continuation of these AID systems, since more and more people are coming on these, continuation of these during surgery brings further considerations beyond those of standalone open-loop pumps. And it's particularly an understudied area. In our experience, these AID systems can be continued in most circumstances where maintaining a closed-loop standalone pump is appropriate if additional considerations are understood and accounted for. And these are the important points because an additional consideration is whether the AID system will be continued in AID mode, requiring accurate and integrated CGM function, or whether it would be prudent to switch to a non-AID mode, that is, switch it to an open-loop, and the reliability of the CGM input actually may be impacted by intraoperative factors, especially electrocautery. We actually did a study prior to the pandemic. We hurried it up when the pandemic started. We published in Diabetes Care showing that the electrocautery would actually make the sensor unusable. We were using Dexcom G6s, and we would just get question marks on the G6. Now, the good news was, as soon as the surgery was over and they got out of surgery, the sensor worked fine, but this is very relevant when dealing with an AID system that is dependent on that CGM data. So changes in an AID to an open-loop mode may be preferable in these circumstances. First, there are significant concerns about accurate CGM function during the procedure. Secondly, the CGM must be removed due to procedural site incompatibility. If the CGM is right there where the operation is going to take place, well, that's not going to work. Third, the CGM device will be exposed to intraoperative fluoroscopy or the electrocautery that's nearby. And how nearby? We didn't study that. What I can tell you is it didn't have to be very close. It was pretty much anywhere in our study. Safe use of AID system is dependent on an external third component, such as a smartphone or controller. And in some operating rooms, they're not allowed in and into the operating suite. And in that situation, consulting with the patient's diabetes clinician is strongly advised for such programming changes, especially since non-AID mode settings may not have been updated. And this, again, you can see this on the pantherprogram.org website, but as I think many of our audience knows, the only current AID program where the basal rates are used as part of the algorithm is the tandem control IQ. And so many of our patients, they have basal rates set, but we don't know if those basal rates work or not. And what I would do with patients is I would actually have them test the basal rates to make sure they're at least close if they're going to wear their pump in the surgery. It's noteworthy that while the first AID systems released in 2017, that was the Medtronic one, it was modified after FDA input to have met more frequent AID mode exits to enhance safety. And that really frustrated our patients, all of those exits. The real world outpatient data on the subsequent generation demonstrated that the AID mode is actually both quite safe and efficacious, requiring fewer exits from its proprietary autonomous algorithm. And this includes during surgery. So this is a lot of information. We're learning as we go. We need formal studies, obviously. And I would really like to invite and suggest that everybody on this webinar read the paper because this is actually, I think, the fundamental reason why we decided to write the paper because all of us were asking questions, scratching our heads, what's the best thing to do? And I think this summarizes our opinions quite well, even though we don't have formal randomized control data on these points. Yeah, so definitely. So a lot of unknowns as far as where the CGM if putting it farther away from the site of surgery is better to be studied further. But for the time being, that might be a reasonable thing to consider. So Dr. Wiles, so when you're thinking about continuing, discontinuing an insulin pump during a procedure, any other factors that are important in this decision? Absolutely. So first, I'm just gonna refer back to it, much of which Dr. Hirsch just spoke about, which he covered many of the factors quite well. And then of course, we're gonna definitely refer readers to look at our publication on this because we do have this fairly detailed table too. And I find that when the rubber hits the road, when it comes to evaluating whether or not a patient's device should be continued, it's really gonna be necessary to think about all the potential contraindications that may come up, not just the ones, a short list that one might easily remember. So I think that's what we've learned is you need to really evaluate all the factors. But in addition to what Dr. Hirsch has already spoken about, I would just add one more. And that is one that turns out to be the common theme that really defines whether or not procedures, the majority of procedures are good ones for compatibility with insulin infusion devices during the surgery. And that's gonna be procedure duration and invasiveness, which often are quite linked together. And so we've drawn this line at two hours. And again, this is kind of a risk benefit kind of comparison you might say, for patients having small procedures where they're going to have multiple potential switches between an insulin infusion device and an alternative therapy in a short period of time for a procedure that's not very physiologically stressful. That probably is a lot more risk than just continuing an insulin infusion device directly through the procedure. On the flip side, you have a patient who's having a longer procedure, which there's more opportunities for physiologic derangements during the procedures. Patients get cold, patients have different types of shock developed during procedures related to the invasiveness. In those procedures, it's gonna be important that we don't try to manage the patients with an insulin infusion device through that procedure, but instead replace them with an alternative regimen of insulin treatment that will get them through that period and they can be transitioned appropriately afterwards. And just one more one that we often see come up that is sometimes easily missed. It's really making sure that, even if a procedure is short, is the patient actually gonna be capable of managing that insulin infusion device system by themselves or with a loved one after the procedure? Because sometimes they may be out of the anesthetic, but they still have some significant sedation going on. They may have sensory impairments that have developed related to maybe surgery on the eyes or something like this, or surgery on the extremities that may make it hard for them to actually effectively manage these devices. So those are additional things that need to be thought about when thinking about whether a patient's gonna continue the device through the procedure. Thank you, Dr. Wiles. Yeah, so as you mentioned, like we created a table in the paper. So the readers could go there and kind of see what are some of these requirements. And as you mentioned, it can range from how complex the procedure is to if the patient capable of managing the pump afterwards. So Dr. Hirsch mentioned kind of like he sometimes will have patients test their settings just in case they have to go to a manual mode. Dr. McGill, any other things you tell your patients in preparation as far as adjustments to their insulin pump or any particular mode, like exercise mode, things like that? So these are all important considerations. So we, as diabetes providers, need to understand what the surgery is, what the location is, how long it's likely to take, what are the possible complications of the surgery, and if it appears to be too long, too invasive, higher risk, or if an insulin drip is planned or a hospital admission or an ICU admission, the pump should just come off and be handed to the family members before the surgery. And the CGM potentially also depending, CGMs might be in the way of blood pressure measurements, IVs, other things. So figure out what the surgery is and whether another delivery method will be in place anyway. Okay, so now we think a patient can continue the pump during surgery. Well, the next thing is to ask the patient. I've had patients absolutely insist, it's a short surgery, I should be out, I'd like to continue. And I've had patients tell me, no, I'd rather just take a shot. I don't want to worry about it. I don't want to worry about it getting lost. We've had losses, which are unfortunate. So figure out what the patient wants. Then the patient may need to be seen. We may need to look at their total daily dose, their basal infusion rate, and look at the settings on the pump. Because if they are reverting to either a manual mode, now some of the pumps have a continuation of the learned infusion rates. So the settings you might say don't matter that much, but I'm going to suggest that they might matter. So a patient who is set for 1.2 units an hour, but you find on the pump download that what they're getting for total basal is 15 units, well, their setting is way too high. So we need to reset the basal rate to be more in line with what the pump is actually delivering. And I've seen these discrepancies, huge discrepancies at times, and sometimes they're pretty close. So review the basal settings and review what the pump has determined that the patient will do well with. The next thing is to consider putting the pump, and this is an AID pump, in an activity mode or exercise mode or something that raises the target a little bit or use a temporary target so that the pump is not trying to get the blood sugar down to 110 or where we'd like to have it on the every day. So lots of considerations may require a visit, may need some additional understanding, but we also need to understand what the patient would like us to think about. So lots of things to check off during that visit before surgery. Yeah, so planning for this procedure is very, very important, trying to predict those possible complications or trying to have a plan in place even before this procedure takes place. Sometimes there is gonna be a different times during the procedure. The pump either has to be discontinued in the beginning or at some point later on, and we really want to highlight the importance of having an alternative regimen in place. And I would ask Dr. Hirsch if you can tell us why basal insulin is so important to continue in these patients. It's a really, really great question. You know, I've never seen data on when the most iatrogenic DKA happens, what month of the year, but I can tell you in my training, both in Miami that Paulina you introduced in my training, and then back in St. Louis where you are now when I was there with Dr. McGill in the late 1980s, we saw iatrogenic DKA when interns would stop the insulin drip from a DKA and there was no basal insulin on board. My guess is there may be a little blip in DKA in July and August with the new interns, but as any internal medicine resident learns often unfortunately the hard way, you have to have basal insulin in somebody who does not make any insulin. Now, with that being said, and as we see more patients with type two diabetes coming on pumps, theoretically, if that basal insulin is not provided, we're not going to go into full blown DKA, but I think it's important that this basal insulin is critical for preventing DKA and controlling blood glucose. And one other point about basal insulin in type two diabetes I want to make, and that's the fact that we are seeing this epidemic in the United States of type two diabetes in youth, in adolescents and young adults down to the age of 10. And as it turns out, the type two diabetes we see in that age group is much more aggressive and they lose their beta cell secretory capacity much faster than their parents and grandparents. Now, we don't yet know what their basal insulin is going to look like when they're in their forties and fifties, but suffice it to say, given what we now know about them in their teenage years, I suspect it will be much less than their parents and grandparents. And what that is going to do is put them more at a risk of DKA in a stressful situation like the hospital, especially if they're on an SGLT2 inhibitor. But then we get to the point about insulin in general and basal insulin in particular about how do we replace it for patients going through surgery? Well, there are really two ways. And the way that I have liked to do this over the years and decades is exactly what Janet said. And that is just use the basal insulin in the pump as a way to sort of guesstimate. And it wasn't quite as good when everybody was on set basal doses, but it was still good enough as a way to guesstimate what to do with replacing insulin subcutaneously. Now with the AID systems, you can look at it. And like Janet said, if somebody has it set at 1.2 units an hour and they're actually getting 15 units a day, and they're going to go on to a manual mode using the basal doses, well, in that case, those basal rates need to be dramatically reduced. With that being said, especially if you're in a situation where the provider, perhaps it's an anesthesiologist, without the ability to download the pump, without the knowledge how to even press the buttons to look to see if you can find basal doses in the pump itself by pressing the buttons, there's another way, which is quite acceptable since nobody has ever formally tested this, and that is just doing it based on weight. For example, somebody who's very insulin sensitive, a typical thin type 1, or maybe somebody with pancreatic diabetes could take 0.15 to 0.2 units per kilogram if you're converting that to a GAR gene, or you could convert that to NPH, which has gone obviously much quicker. For somebody who's more insulin resistant, the more typical type 2 patient could take a higher dose of 0.25 to 0.3 units per kilo. And then of course, if you're using it as GAR gene, you would repeat it every 24 hours. If you did this as NPH, you could repeat every 12 hours. As far as prandial insulin is concerned, it's just, unfortunately, it's more of a trial and error. The way we do it in our hospital in Seattle is we do it based on low, medium, or high correction dose based on their total daily insulin doses, is how we do prandial insulin and correction dose insulin. One of our authors on our paper, Tony Chang, he actually looked at this in type 2 diabetes, and we did surveys of academic hospitals around the United States. And what Tony found out is that every hospital does this differently when patients with type 2 diabetes are admitted to the hospital. So there's no standardization because we don't have data on how to best do this, whether you're just coming into the hospital or in the operating room itself. So we gave our guidelines in our paper. It's really based on the consensus of the authors, but I'll tell you, it would be nice if we had some data. Yeah, absolutely. Yeah, so way-based dosing is, in my experience, works, and as long as the weight is correct, so that's another thing to just make sure. It's a fast way to guesstimate a patient's total, either based on dose or total daily dose. And in these situations, the table in our paper could be a good guidance for people not as familiar with the technology or people that just have to start a dose quickly during the procedure or in that space. So another point that I want to also discuss today is the targets for glycemic control during these procedures. So, Dr. Walz, can you comment on this point about what kind of targets anesthesiologists are looking for, and what brands, treatment, and what levels? Absolutely. So I think one of the things we've learned in the last couple of decades in perioperative care is that while it is important to manage hyperglycemia in the perioperative period, that if we try to aim for excessively tight control in these patients, that we can increase the risk of dangerous hypoglycemia, which can have a very poor outcome. So now, today we're really focused more on low-risk procedures, as we've already described. This is really the in-scope population for patients who might continue their diabetes devices through surgery. And so actually, the latest advice regarding this, which actually came out about a month or two before our paper, which comes from the Society for Ambulatory Anesthesia, which is really the go-to diabetes treatment publication for anesthesiologists, is that when we're dealing with these low-risk, short procedures, it's appropriate to not try to lower the blood sugar below 250 milligrams per deciliter. So that's really what's being advocated for these shorter procedures now. And the rationale for this is that when we have these short periods of time where patients are in these facilities, you really increase, you don't have much time really to intervene. And you increase the chances of things like dose stacking when you're trying to correct hyperglycemia. And it can really increase the risk of hypoglycemia. So again, we'd like to treat it, but the reality is it doesn't work well. And so that's why it's kind of being advocated that we not treat above or below this threshold of 250. Of course, again, we talked about these same patients, the ones having those low-risk procedures, are going to be the same types of patients who are going to be candidates for continuing their devices through surgery. And so I think it's one thing we pointed out in the publication that is important just to think about from a cognitive perspective is that these same patients who were at the treatment threshold on a clinician basis, maybe 250, these devices are not going to tolerate that. They're aiming for a much tighter control. So it's just important for clinicians to be aware that these devices are going to be targeting these lower values and that that is appropriate as long as they're monitoring the blood sugars with POC checks as well. Yes. And then another kind of thing that we commonly as endocrinologists observe is that patients do get steroids, particularly dexamethasone, during procedures. Any advice in that regard how to mitigate the hyperglycemic effects of the steroids? Yeah. I'm sure probably most of the people in the audience have seen a patient who's gotten dexamethasone for part of their procedure and maybe has had some out-of-control blood sugars afterwards. So just to state the obvious, dexamethasone is an important part of the armamentarium that we use to fight against postoperative nausea and vomiting, which is a very important complication for patients. That being said, the advice in the recent guidelines that I referred to previously is to be careful about what dose we use in patients with diabetes. And specifically, they advocate using a smaller dose of four milligrams rather than larger doses, as this has been associated with lesser disturbances in glycemic control in the post-op period. I think something that's still going to be a challenge is that there are those patients who have more extreme responses and are more difficult to control blood sugar. And we don't really have great ways to operationalize who those patients are and for them to maybe have a higher threshold to consider giving dexamethasone. It's certainly going to be a risk-benefit decision in each patient. But clearly, we'd like to better understand who are these patients who are going to be hyper-responders, so we don't get them into trouble for unnecessary small birth benefits in PUNV prevention. Yeah. That's a great point of identifying that the patients do have diabetes and choosing the lower dose of dexamethasone. OK. Another challenge with this AAD system that was briefly mentioned was the discrepancy of the targets for anesthesiologists, which are higher than these systems operate on. Do you have any comments on this, Dr. McGill? Well, we obviously have lots of counterbalancing influences here. So the higher target is not there are no AAD systems in which we can effectively aim for a target that's much over even 160, 180 is pretty high. So that's a problem, a system that learns the patient and then will continue insulin infusion based on prior day's infusion rates. Then you add dexamethasone, which may not throw things out of whack right away, but certainly can within hours during the day and overnight, which will be a bit of a problem for patients. I think one of the considerations is we would like to know who is going to get steroids and who may not. And there may be a cost-benefit relationship for those fighting hyperglycemia with sort of unknown insulin requirements. There's no way we know what that insulin requirement is. So that's a potential problem. Even four milligrams is not a really small dose. It's still a somewhat effective dose. So I think perhaps that conversation should be had, but often it may not be because the anesthesiologist just is making decisions when they meet the patient, when they understand the procedure, and there's not really time to communicate. The other interesting issue is if patients elect to do injectable therapy, we need to provide it. Hopefully they have it for backup already, but not all perioperative units have Lantus. They may have regular insulin. They might have short-acting. They may not have the long-acting insulin that you think would be or a large, not Lantus, but insulin large, or the long-acting insulin that you think would be appropriate to get the patient through the surgery, through the night, and onto their pump the next day. So the patients need to have that available, take it either at home or take it even in the facility, but know what the dose, what dose they took. It would be nice to have that recorded somewhere that's, you know, home medicines are recorded. They don't really accommodate insulin doses very well. So lots of confusion here about how the insulin will be delivered, what the target is, what drugs will be used that may cause hyperglycemia in particular, and which insulin, and is the insulin available if the pump really needs to come off in the perioperative holding area. So lots of things that should be discussed in advance. Yeah, absolutely. And the other thing, as you were mentioning, these pumps really are not, are targeting at much lower blood glucose. So in case you have a patient with an insulin pump and the glucose has started rising above the level you would expect. So Dr. Wells, what are the things that, when would you decide to remove it or to treat the hyperglycemia and wait? Yeah, great question. So this is one of those things, and I'd like just to take this opportunity to highlight something, which we highlight in the paper, but also try to highlight whenever we educate clinicians about the perioperative use of these devices is that whenever you have unexpected dysglycemia, it's always important to be suspicious about whether there may be some sort of problem either with the system, whether it comes to the CGM, to the connection to an infusion device, to the infusion device itself, or to the patient, some other disturbance that they're experiencing related to medications they're receiving, the stress of the surgery, et cetera. So whenever there's dysglycemia, we always want people to think about that and think about maybe I need to abandon the initial plan that I had to continue this device through surgery and switch to an alternative plan that has better control for this dynamic situation and takes this unknown of how the system's behaving in intraoperative period out of the equation. So once the clinicians have determined that that is not what is going on, that maybe you're dealing with a little bit of mild hyperglycemia, which you may want to consider treating, clearly the major consideration is going to be, as we highlight in the table, is this a system that's functioning in an automated mode, in which case we would expect the system to react to this hyperglycemia and manage it effectively. And the role of the clinician in that regard would then be to continue their point of care glucose monitoring of the patient and ensure this is in fact occurring. And as long as it does, then to continue onwards. And if it doesn't, then again, to have those suspicions about device malfunction. Of course, if we have a device that's not functioning in an automated mode, then there will be no such reaction to this hyperglycemia that's noted in the intraoperative period. And so then the question becomes, would it be appropriate to augment the existing therapy that's occurring in a non-automated fashion with a small dose of insulin to lower a slightly elevated glucose into a more preferred target range? Yes. Yeah. So that's important to differentiate. If it's not an automated insulin device, then that is more relies more on the anesthesia theme of what treatment the patient receives. There are a couple of questions from the audience that I want to just make sure we answer. So one, the first one is regarding, like removing or suspending the insulin pump in a patient that is sedated under anesthesia and if the anesthesiology team should know how to suspend or disconnect the pump. So do your anesthesiology residents, do they get any training on disconnecting the pump or? Well, yeah, the fundamental reality is, is that due to the wide scope of different issues and things that we expected to train people on, this is not something that's going to be the common part of any armamentarium of the vast majority of anesthesiology clinicians. And so that's why we felt was important to provide reference information to clinicians that they could seek out in these very rare instances, or they would need to know how to discontinue a device like this in an urgent situation that wasn't otherwise planned. And so we did cover that in a table in our publication so that, you know, on a device by device basis, you know, basically what are the simple steps that can be taken that will basically discontinue the flow of insulin to that patient. So once you've taken care of that, then you can introduce another alternative insulin regimen to the patient. Great question. Yeah. So we don't expect anesthesiology to know how to operate the pump, but we actually would not recommend them dosing insulin through a patient insulin pump. We do have also like a flow chart in the paper that I think is a very useful tool that people can look at. And it kind of guides the different systems and the CGM through the different stages of care. So from pre-op, intra-op and post-op. So I think that's probably the best tool out of the paper for these situations. And then the second question, it's about if endocrine team sees all these patients before procedures to give instructions, or if the perioperative team at your institutions provide those instructions. Well, Earl, what goes on in your place? Ladies first. Well, this would be ideal, but this is not what happens. So our patients should let us know if they're going through surgery. Most do. And it ends up being an inbox task to figure out what to do. Some back and forth with inbox. Patients are not always equipped to come in for a visit before surgery. I think it's not a bad idea if they haven't had a visit recently, if we haven't had a pump download recently that we can look at, or if it's available because they're connected. I think it is a good idea to come in, but that doesn't happen all that often. I feel the pain of the perioperative teams trying to figure out what to do with this patient and what to do with the pump. Hopefully if it's clearly positioned out of the way, I had a patient who just put her pump in her thigh or buttock. She was getting a hysterectomy, nothing in the belly. So she got it moved out of the way. It worked very smoothly. We talked about it ahead of time and she was prepared, and it went okay. Other patients have shown up in the pre-op area with less preparation. I think they need to have a phone number, someone to contact for advice, what insulin dose, what makes sense. Then the next piece is, what do you have? Do you have NPH? What do you have if we're going to come up with insulin doses? So advanced planning is always better, doesn't always happen. Then we get last minute consults. Then we're trying to find people in pre-op and recovery areas that we're not familiar with, but we find them. Paulina, just very quickly on that. Just to add to what Janet had to say, we do have protocols at our institution, which actually are quite helpful. We have a hyperglycemia management team in the hospital, which to my knowledge, I'm pretty sure we're the only one in our area, maybe even in our state that has one. What I have been asking my patients really for the last five or 10 years is that if they do have to go into the hospital emergently, whether it's surgical or not, and there are any concerns, please let us know. We have an on-call team. The nurses can get it and then get in touch with us. Because my bigger concern, at least where I live, is that they are admitted where there's no endocrinologist on staff, there's no hyperglycemia management team, and it's not just the diabetes technology, it's just the diabetes in general. We're always happy to take a phone call. I can't tell you how many interesting phone calls I've had over the years, including, how do you administer IV glargine? I have a lot of very interesting phone calls. Dr. Walsh, do you want to comment? We have two minutes left. Sure. Just to add to what was already said, I think having an ideal state would be a lot of preoperative planning and a lot of ability on the day of the procedure to verify the plan that's put in place is viable and to carry it out and to adjust as needed. But it's really an implementation challenge to have that entirely in place across any institution or surgery care center. The more that you can have for your patient, I think, the better. We do have those things in place, but the more we can do for preoperative planning, the better. Again, I think one of the challenges is that these smaller procedures that we're really talking about so much are becoming so common and the flow is so quick that it's hard sometimes to get these patients the necessary preoperative planning. And we feel for our endocrinology colleagues, there's quite a lot of these patients coming through and sometimes we definitely see the logistical challenge of trying to interact on all of these patients when these procedures are becoming so common, these minor procedures. So, yeah, that's all I had to add. All right. That's great. So, thank you, everyone. This was a great discussion, I think. It's been recorded, so it'll be available for other people that could not join today. So, yeah, I just want to thank you all so much and enjoy the rest of your evening.

perioperative care

wearable diabetes devices

continuous glucose monitors

insulin pumps

automated insulin delivery

diabetic ketoacidosis

basal insulin

point-of-care monitoring

glycemic control