← AACE Podcasts
AACE Podcasts
Episode 55: Sodium-Glucose Cotransporter 2 Inhibitors Should Be Avoided for the Inpatient Management of Hyperglycemia
00:25:25
Loading summary
Transcript58 lines
- [00:00]
A
Foreign.
- [00:06]
B
Welcome to ACE Podcasts. Thanks for tuning in as we elevate clinical endocrinology by taking deep dives into trends and topics that can help us improve our patient care and global health. Find the Latest episodes on aace.com podcasts and now let's meet the endocrine experts who will be talking with us today. Hello and welcome to this ACE podcast. My name is Vin Tangprecha. I'm the host of today's podcast and I'm also the editor in chief of ACE's official journal, Endocrine Practice. Today we have special guest Dr. Rifka Schulman Rosenbaum. She has written a paper on SGLT2 inhibitor use in the hospital and she published her article in the December issue of Endocrine Practice. And we have her today to discuss about her paper. I think it was online in December, actually.
- [01:02]
A
Yeah, I think it was. I was going to say I think it was online in December, but it was.
- [01:06]
B
Looks like April.
- [01:07]
A
Is it April? Yeah.
- [01:08]
B
Yes. So it was published online in December and now it's April 2024 issue. So thank you, Rifka, for joining us. Can you introduce yourself and let us know what you do at your institution and the roles you hold at Ace?
- [01:25]
A
Sure. Thank you, Dr. Tangprecha, for inviting me today. I'm really excited to talk about this article on this topic. As you mentioned, I'm Rifka Schulman Rosenbaum and I am the director of Inpatient Diabetes at Long Island Jewish Medical center, which is one of the tertiary centers for Northwell Health based in New York. And I'm also a professor at the Zucker School of Medicine at my institution. I've done very involved with ACE and currently serving as the chair elect for the annual meeting. So shout out for coming to our really excellent program in Orlando, May of 2025.
- [02:03]
B
Great. I'm going to be there for sure.
- [02:05]
A
Excellent.
- [02:07]
B
So I understand you teach fellows, residents and medical students in diabetes.
- [02:13]
A
Yes, I definitely do that. I am a clinical endocrinologist, so I see patients actively every day. I do mostly inpatient, but a little bit of outpatient as well. And I run our service in our hospital, which includes inpatient diabetes and also other inpatient endocrine topics. And I have a multidisciplinary team which includes nurse practitioners, diabetes educators, pharmacists, as well as our fellows, residents and students that rotate through. We are a really busy service. We see a lot of patients. We also have at our hospital, we are joint commission certified for advanced inpatient diabetes. So we're A diabetes center of excellence.
- [02:53]
B
So it sounds like you are maybe. Can I call you endohospitalist?
- [02:57]
A
Yeah, sure. I've heard that term thrown around and either inpatient endocrinologist or endocrine hospitalist for sure.
- [03:04]
B
So that's definitely a field that's growing within our own field. So you have a lot of expertise in management of inpatient diabetes, I assume?
- [03:14]
A
Yes. So I published a book called Diabetes Management in Hospitalized A Comprehensive Clinical Guide. It came out in January of this year, 2024. I'm the editor, and there are many excellent chapters from authors all over the country on every topic imaginable for inpatient diabetes.
- [03:33]
B
Looks very thick.
- [03:34]
A
It is thick. Took a lot of work.
- [03:37]
B
So this is a field that's been rapidly evolving. I mean, when I was a fellow inpatient diabetes was basically insulin and trying to tell people not to use sliding scale. Now it's more than that. I understand.
- [03:51]
A
Yeah, it's really grown so much. And like, from what you're saying, I recall also as a resident where you didn't really focus too much about diabetes in the hospital. Patients were continued on oral agents frequently. Sulfonylurea is no problem. Everything has changed so much, really. The field has really grown so much.
- [04:09]
B
So one of the things I hear often from the residents is that, oh, we don't want to change their regimen because they're going to be going home in a few days. Just keep it the same. Is that mantra still true?
- [04:20]
A
So you mean change what they're on at home or change what they're receiving in the hospital?
- [04:26]
B
They come in and they'll just start what they're on at home and no changes are made and they're discharged a few days later.
- [04:32]
A
So there may be patients where that is the right move because their A1C is really well controlled. They were doing fine at home, in which case, sure, I mean, you can, for the most part, continue what they're on in the hospital if those medications are appropriate for the hospital, which is a whole separate conversation. But then you have the patients that are not controlled well at home. Their A1C could be high or they could be having hypoglycemia all the time. And so this is a touch point where the inpatient endocrine team actually has the opportunity to really help people. And in my area in New York, there is a lack of appointments for access to outpatient endocrinology. It's at least a six months wait often to get in with new as an endocrinology patient. And so this is our opportunity to help patients adjust their regimens. Why do they need to wait another six or 12 months to see the endocrinologist in the office if I can help them right now? And so I spend a lot of time not only managing their insulin in the hospital, but also going through their outpatient plan and making sure that we don't need to escalate their regimen. Prescribe them a CGM, prescribe them a GLP1 or SGLT2 or whatever's appropriate.
- [05:40]
B
Yeah, I totally agree. I mean, they're in the hospital. You have all their glucoses. Why not make use of the data?
- [05:47]
A
Absolutely.
- [05:49]
B
Well, this is a great segue into SGLT2 inhibitors. Let's say you have a patient who's well controlled on SGLT2 inhibitor, and they come and get admitted for, let's say, six cellulitis or some other common internal medicine condition. What should we do?
- [06:06]
A
Well, that is a great question worthy of writing an article or perhaps two articles, because as you know, there was kind of a pro and a con series of articles on this topic. And to just give you a little bit of background, our major organizations, including ace, you know, we focus on using insulin for hospitalized patients as the current standard of care. And that's coming after many studies that kind of show the importance of glycemic control, the importance of kind of moderate glycemic control, not overly controlling them, but not under controlling them. And so I think at this point, most hospital providers are understanding that insulin is a basic way to manage diabetes in the hospital. And then in the last few years, some studies had come out regarding DPP4 inhibitors in the hospital, and some positive data came out showing that they may be appropriate for certain subset of patients with mild hyperglycemia. So not every patient has to be managed with insulin. If it's mild hyperglycemia, which could be someone who's on an oral agent at home, they're in the hospital, they might not need to get insulin they could receive. For example, linagliptin was the agent that we added onto our formulary. Now, the SGLT2 question that comes up because, well, if DPP4Is are good for certain patients in the hospital, maybe SGLT2s are, because firstly, this class of drugs has so many benefits in general. Right. So we're using them for our patients, not only for the glucose control, but for benefits with CHF and CAD and CKD and all the other reasons why we're outpatient prescribing SGLT2 inhibitors. The other Nice thing about the class is that they don't cause too much hypoglycemia. So. So or really minimal, just like DPP4 inhibitors. So it is the natural question to ask of maybe this would be the next big thing to use for hyperglycemia for milder cases in the hospital. And that is kind of like the underlying reason why we are having this conversation. And to start off, SGLT2 inhibitors are a little bit different from DPP4 inhibitors. Okay. And without getting into the nitty gritty of the mechanisms, there is a well known side effect of DKA or euglycemic DKA with SGLT2 inhibitors that we're going to spend a little time talking about. So while in the outpatient setting, euglycemic DKA is relatively rare, Right. We prescribe them all the time. We don't have all of our patients heading to the hospital. And you glycemic DKA and as giving you the perspective of as an outpatient endocrinology doctor, which I do once a week. You know, I do mostly hospital, but for the most part I prescribe SGLT 2 numbers all the time. And I, I have let's say one patient that I could think of who did develop eugocemic DKA recently, but most majority of the patients that have not. It is very different from the perspective of an inpatient endocrinologist. I get consults for SGLT2 induced DKA or EDKA every single week, at least one, if not more. Okay. So I think it's important to pay attention to the lens through which you are trying to assess the frequency of the problem because on the outpatient side it's very different from the inpatient side. So then why is that? Well, firstly, there are certain risk factors that we know about for SGLT2 induced DKA or EDKA. Those include fasting, right? Because you have a buildup of ketones or low PO intake or very low carb diet. Also high stress situations, patients going for surgeries, which is like high stress situations. So far, if you may notice all the things that happen in the hospital I've just listed, right, People come to the hospital, they get made NPO for who knows how long, waiting for procedures or surgeries or because they're sick and vomiting and they can't eat and a lot of them have procedures and surgeries. The other thing is insulin is actually protective against dka. So for some patients who are long term on insulin plus an SGLT2, if you remove that insulin, that actually raises the risk of dka. And unfortunately, a lot of times insulin gets held when they first hit the door to the hospital until, you know, endocrine gets called or the primary teams think to order it. So there's that issue issue as well. And the other thing is the FDA put out recommendations to hold SGLT2 inhibitors three to four days before any surgery or procedure, and that's to mitigate the risk of DKA or edka. So by holding the medicine, usually you're doing okay when you have your procedure, but what about using them in the hospital? Right. And so the patients made NPO, you have three, four days in advance to prepare for the NPO? Absolutely not. Like, we all know the NPOs, they come at you really quick. So there's a lot of concern surrounding this issue. For patients who are in hospital.
- [11:02]
B
It sounds like it. Also, if you have someone who's not on SGLT2 inhibitor, you just eliminate that as a potential cause of ketoacidosis. Right. Because there are many. Like you mentioned, there's so many causes of acidosis in the hospital. Hospital. And if you just have someone not on it, then you can't blame the drug. Right? Is that one?
- [11:21]
A
Yeah, certainly. I mean, yeah, patients are sick. So if this is a drug that has a risk for dka, metabolic acidosis, if there's another option that we can use that doesn't have those risks, that is definitely something to think about. And we are comfortable using insulin, and we're even comfortable for the mild cases using DPP4 inhibitors. So there has to be a good enough reason to use the SGLT2 inhibitor to take that risk. And just to expand it a little bit more, SGLT2s, they can cause some orthostatic hypotension, some hypovolemia issues. If you have patients who are already possibly having issues with hypotension, dizziness, and, you know, the risk of falls in the hospital is obviously a big issue. So while the other thing is this is really understudied for the glycemic control aspect. And. And I'll talk more about that soon. There are definitely studies about CHF, but for using SGLT2s for inpatient glycemic control, we really don't have data on that. So you kind of have to extrapolate from the information that we do have. The other thing I wanted to mention is the risk of genital urinary infections. And so while the more common thing would be mycotic infections that are not as serious, but UTIs are possible with SGLT2 inhibitors. And sometimes up to a quarter of hospitalized patients have Foleys in place. Right. And so having a drug that potentially increases the risk of uti, or patients with indwelling Foleys, the risk is even higher. So there's a number of different concerns.
- [12:54]
B
So let's go back to our hypothetical patient. They come in, they have cellulitis, they have type 2 diabetes, they're on SGLT2 inhibitor plus metformin. What are you changing them to now? They're just. They're going to be in the hospital for a few days for antibiotics and go home.
- [13:09]
A
Yeah, so that would be a good patient for inhibitor because they're just on, like, assuming their A1C is good. Right. Because you could have a patient like that with an A1C of 10, in which case, you know, they need insulin, they need an upgraded outpatient regimen. But for someone, let's say with an A1C of 6.5, and they're coming in on a metformin and SGLT2 and their glucose looks fine on admission, you could manage them with a DPP4 inhibitor with a correctional scale on the side. Sometimes we even do, like a little bit of basal insulin with the DPP4 inhibitor.
- [13:40]
B
Stop the metformin, too.
- [13:41]
A
I mean, look, in my institution, this is what we do is either we do insulin or we do DPP4 inhibitors. Again, this is for glycemic control. We can get to talking about CHF in a minute. And I think that's an important differentiating point, because the focus of my article was talking about why I would avoid using SGLT2 inhibitors for managing inpatient hyperglycemia. But glycemic control is a different indication than CHF management. And there have been some pretty strong randomized controlled trials showing support of using SGLT2s inpatient or close to discharge for heart failure management. And so I'm not speaking against that. I'm talking about glycemic control. Now, there still are risks of using SGLT2s for the CHF indication. Right. There's still the similar risks that we just talked about, but then at least you know, firstly that there's a goal like that there could be improved heart failure outcomes. So there's a reason to take that risk. But then you have to really risk mitigate. So you need to have some sort of process in place for the hospital, or you should have a process to make sure that patients are started on them for heart failure at the right time. And the right time would be when they're clinically stable when they're not going for procedures or surgeries, when they're eating and not NPO and all those other factors. Factors. So it's actually really complex. And unless you have some like hospital protocols or EMR trick that help this process, it could be really confusing and actually dangerous for providers.
- [15:20]
B
So we're kind of talking about someone who comes in the hospital, their A1C is 10 and they're hyperglycemic. No, don't do SGL2 inhibitor. Do probably insulin.
- [15:32]
A
It sounds like that's an insulin. Yep.
- [15:34]
B
Stuff the SGLT2 probably right.
- [15:37]
A
Well inpatient for sure. And then depending on the situation, like if that patient has an indication for the SGLT2 and there's no contraindication, then I'll add the insulin, let's say and resume the SGLT2 at discharge. Right. Once they go home, they could go back on it if there's no reason to stop it per se.
- [15:57]
B
So they're, they come in, they have heart failure. Once their heart failure taken care of, maybe do that.
- [16:03]
A
Yeah, exactly. Like start it closer to when they go home. Because we don't want to have clinical inertia where people that could benefit from an SGLT2 for their heart failure, like maybe that they won't get started on it. So you know, it could get started like really close to discharge or at the time of discharge. And there is data from the CHF literature supporting that. And that's why it's kind of cool that our two articles, the pro and the con, they actually don't even disagree with each other because the pro article says you should use SGLT2 inhibitors in the hospital for heart failure. And the Khan article says you should not use SGLT2 inhibitors for glycemic control. So they actually don't truly conflict with each other at all. You can kind of bring it all.
- [16:47]
B
Together for our audience there. This is, there are two companion papers. One is a pro and one's a con. And I think we're all on the same page that unless there's a real reason don't use the SGL2 inhibitor. And real reason would be someone who has heart failure. But that is one of my follow up questions. Has there been a study where someone who is in acute heart failure, does SGLT2 inhibitor help in that situation? Not the stable ones that are about to leave the hospital, ones that are acutely with chf. Is there a role for that?
- [17:22]
A
I mean there's been a few studies in the CHF literature, some of them focused on closer to discharge, some of them were a little bit earlier in the stay. The interesting thing is not all of them looked at glucose metrics and such. They more focused on heart failure metrics. So I don't have all the like glucose and ketone measures and all that for all those articles. But it is interesting to know that the benefits of the SGLT 2 towards the CHF does come pretty quickly. Like in one of the trials at the 15 day followup they already had improved measures of CHF and so that would be the reason to consider starting it like while they are still hospitalized but stable to make sure they get the full benefit of the CHF therapy. But yeah, I mean we don't have data on the glycemic control indication for using SGLT2 in the hospital and because it's not standard of care, we don't have a lot of data on it anyway. And a lot of people will cite numbers from outpatient studies about the risk of DKA and say, oh, look how rare DKA is with SGLT2S for these outpatient studies. But you really can't extrapolate that to inpatient.
- [18:38]
B
What about the infection? Do we have any data showing that's more increased risk of the mycotic genital infections?
- [18:45]
A
Yeah, so again, because it hasn't been used regularly, inpatient hasn't been formally studied, we don't have data on that. It's all based on outpatient. So this is all kind of using our best judgment with what we have now. But ideally there would be trials specifically looking at inpatient SGLT2 for glycemic control. And the other thing we don't have is for the kidney indication. Right. That's the third major FDA indication for SGLT2s is for CKD with albuminuria. And so we don't have data on using it for the kidney indication in the inpatient setting. Really just the, the CHF is the main thing that we could say.
- [19:24]
B
I was also surprised in your paper that SGLT2 inhibitor was no better than DPP4 inhibitor. I thought it would have been much better. Although you're probably going to tell me no data.
- [19:36]
A
Well, there was a retrospective study comparing the two and so there was no significant difference in the glucose metrics. And so if you don't have any benefit of the SGLT2 over the DPP4 in controlling the glucose, and you do have some major safety concerns for the SGLT2 class where DPP4 inhibitors are really pretty benign and very safe. So we use them for the Appropriate patients in the hospital. So there wasn't really a reason to push for trying out the SGLT2s. But again, lack of data.
- [20:10]
B
Yes. What do you think the next big study should? I mean I don't know if there are you aware one that's coming out or what should be done next.
- [20:19]
A
I think we now have a lot of data on CHF indication for inpatient SGLT2. So the thing we would benefit from would be a study on inpatient use of SGLT2 specifically for glycemic control. That's the thing we don't have.
- [20:33]
B
What would that be compared to like basal bolus or what do you think the comparative would be?
- [20:38]
A
Yeah, I mean I should mention that there was one rct, the DAPA hospital trial that looked at post cardiac surgery patients compared basal bolus versus basal bolus plus an SGLT2 and there was no difference in the glucose measures. It didn't reduce the insulin requirements. And while there was no outright dka, there was severe ketosis more frequently in the group with the SGLT2. And one thing about that study is that these patients were heavily insulinized, right. Because of their post cardiac surgery. So they were on insulin drips and then they were on like full on basal bolus plans with the SGLT2. And insulin is protective against DKA. So in that study, you know, you're comparing basal bolus versus basal bolus with SGLT2 and there was severe ketosis but not outright DKA. If you had a study, I'd be curious to know without the full on basal bolus plan. I guess there was a retrospective study with DPP4s but maybe of more like a prospective kind of study comparing the two and really honing in on rates of DKA and NPO and all the kind of risk factors and as well as like the, you know, data on the Foley catheters and the urinary tract infections and all of that.
- [22:04]
B
And I guess when you're sick, I mean you're catabolic and I mean insulin and you're insulin resistant. So maybe insulin is the way to go.
- [22:14]
A
Yeah, it works. I mean as long as you know how to dose it safely, it, it definitely works. Great.
- [22:21]
B
So we're about to run out of time, but if you could summarize to our audience what are the key take home messages from your paper.
- [22:29]
A
So my paper, which is sodium glucose CO transporter 2 inhibitor should be avoided for the inpatient management of hyperglycemia is the con article in the set of pro con articles which focused on the reasons why I would avoid using SGLT2s for inpatient management for hyperglycemia. The main kind of focus of the arguments would be the concern regarding DKA and euglycemic dka, which is a higher risk in the inpatient setting compared to the outpatient related to NPOS procedures, poor PO intake, holding of home insulin, a lot of risk factors combined with potential risks of urinary tract infections in the setting of Foley catheters and other patients that have urinary issues who are hospitalized, as well as orthostatic hypotension. All these inpatient factors that make me concerned for using SGLT2s in the hospital. Also having seen countless patients with EDKA in the setting of SGLT2 as an inpatient endocrinologist, that this is actually a really common kind of consult. It's a common endocrine consult. So any of us who work in the hospital can attest to this is real. This is not like a theoretical concern. This is something we're seeing every day. So until we have better data that might support it, and maybe we never will, I mean, maybe it's just not the right thing to do. Until then, basal bolus insulin for the pretty hyperglycemic patients and then consideration for DPP4 for the mild hyperglycemia patients seems a safer way to go. And just to kind of summarize for the PRO article that there is some strong data towards using SGLT2S inpatient for CHF indication, but once a patient stabilized and once they're past that higher risk state, closer to discharge. So I definitely don't want to discourage that for patients who need heart failure management.
- [24:25]
B
Well, thanks so much for writing this article and talking to us today. Do you want to give any shout outs to your other co authors?
- [24:32]
A
Sure. Yes. So Benjamin Cohen was my fellow when we wrote the article, who's now an attending endocrinologist in New Jersey, and Yael Toby Harris, who is endocrinologist and chief of our division at Northwell Health, they were really helpful and instrumental in writing our article. So thank you guys.
- [24:51]
B
Great. I'm glad they all were able to contribute. Thanks so much and I learned so much and I hope that the audience can continue to learn from you at our annual meetings that are coming up in 20, 25 and 26. Thank you.
- [25:04]
A
See you in Orlando.
- [25:05]
B
See you there. Thanks for listening to another great ACE podcast. Join us for another episode@aace.com podcasts and help us in our mission to elevate clinical endocrinology Together we are ac.