This transcript has been edited for clarity.
Today, I want to talk about using sodium-glucose cotransporter 2 (SGLT2) inhibitors in people with type 1 diabetes. Early on, I was a big fan, and I liked them because my patients liked them. They found they had less glycemic variability. They noted some weight loss and some A1c reduction, and they found their diabetes was easier to manage.
However, after using them in people with type 1 diabetes for a few months, I began to notice that people were developing diabetic ketoacidosis (DKA). I almost never have people develop DKA in my practice in affluent Beverly Hills, which is where I could use the SGLT2 inhibitors.
I ended up running into my colleague and friend, Dr Irl Hirsch, at a meeting in San Diego. After comparing notes, we realized that the increase in rates of DKA we both were seeing was due to the use of SGLT2 inhibitors.
We, along with several other coauthors, published the first case series on DKA caused by SGLT2 inhibitors. After this publication, and the ensuing change in the package insert, I developed a specific protocol for the off-label use of SGLT2 inhibitors in my own patients.
This was a very detailed protocol. I had people monitor their ketones. I educated them about what to do. We talked about DKA prevention. We talked about how and when to stop the SGLT2 inhibitor, and I reinforced this over and over again. Additionally, all patients had 24/7 access to me so they could reach out if they were at all concerned about this.
I found that my protocol actually worked really well on patients on multiple daily insulin injections. I didn’t have anybody go into DKA on multiple daily insulin injections. No matter how hard I tried, I still saw SGLT2-induced DKA in my patients on insulin pumps. This means patients on pumps alone as well as patients on automated insulin delivery systems.
What I wanted to do next was to see if I could prevent DKA in people on pumps if I used injected basal insulin in combination with their pump systems. That way, they would always have some basal insulin that would help prevent DKA. I wanted to start trying this, but then COVID happened and I stopped starting SGLT2 inhibitors in people with type 1 diabetes because I didn’t want to risk causing DKA in a pandemic.
However, over time, the data have amassed on the benefits of SGLT2 inhibitors in heart failure and CKD. I really wanted to use these agents for people with type 1 diabetes for their non-glycemic benefits, and this became a strong consideration. There’s a large amount of research going on about this, as well as the development of continuous ketone sensors.
I really am in favor of using SGLT2 inhibitors for these non-glycemic reasons as long as we can do it safely. However, we now have other agents, such as glucagon-like peptide-1 (GLP-1) receptor agonists, and nonsteroidal selective mineralocorticoid receptor antagonists, such as finerenone, for the treatment of chronic kidney disease (CKD) and heart failure.
Although these agents are also not well studied in people with type 1 diabetes, they do not seem to cause DKA and may be safer in this population. Of course, it seems highly unlikely that there will ever be head-to-head comparisons done, such as with an SGLT2 inhibitor compared with a GLP-1 receptor agonist, but we can certainly opt for off-label uses of these agents as well.
Regarding the use of continuous ketone monitoring, this may be helpful. I don’t know what it’s going to be like because I’ve never done it, but my big issue is that my patients get into trouble when devices fail, not when they work.
One of my stories of a patient who went into DKA and was on an SGLT2 inhibitor was a patient who was getting married. The night before the wedding was the rehearsal dinner, and she was very busy with wedding plans, out-of-town guests, her family, and everything else, and her insulin pump had an occlusion, her sensor failed, and she really didn’t think much about the fact that her glucose levels had become very high.
She went into DKA, pretty severe DKA, and was hospitalized in the intensive care unit at 10 PM the night before her wedding. Now, as you can imagine, we worked really hard to get her out of the hospital, still in mild DKA by the next morning, so she could have her wedding. That caused a large amount of diabetes trauma, and even though she obviously survived it, it was really upsetting to her that this happened.
I really want to prevent people from going into DKA on SGLT2 inhibitors if it’s at all possible. One of my worries about sensors is the issues that I have with people always having them available. As we all know, it can be hard to ensure that people will always have their sensors.
Even glucose sensors have their issues. They can fall off. They can become unusually unstable. It can be difficult to be sure that someone has a continuous supply. I can have problems with prior authorizations or refills. It’s difficult outside of a clinical trial to always be sure people have the supplies they need, so that worries me for people who would rely on a ketone sensor, who may have that sensor not work.
Finally, there is my underserved population in East Los Angeles with type 1 diabetes. They have much higher rates of diabetic kidney disease than patients I see in other settings. These are the people who need my help the most, but they have less access to healthcare. They have less access to someone who can help them if their ketones are positive.
They have lower levels of health literacy, and most of my patients don’t even have access to ketone test strips. For many, I can’t even use an RAAS inhibitor because I can’t deal with the hyperkalemia that may ensue. Those are people where it’s tough to use agents to help slow progression of their kidney disease, and I need to find other, safer options for them.
Thinking about it, wouldn’t it be safer to use a GLP-1 receptor agonist in these people where you’re not going to increase the risk for DKA, and even generic liraglutide, which may actually provide nonglycemic benefits? My patients in this population, even with type 1 diabetes, tend to be overweight or obese, and I think we need to think of other solutions, safer solutions, for a population where access issues are really significant.
It’s been an interesting journey, and my understanding of these agents only grows. I want all of my patients to have the best possible, safest care to help deal with the progression of diabetic kidney disease, heart failure, and other nonglycemic benefits. Thank you.
COMMENTARY
Preventing DKA in Patients With T1D on SGLT2 Inhibitors
Anne L. Peters, MD
DISCLOSURES
| December 23, 2024This transcript has been edited for clarity.
Today, I want to talk about using sodium-glucose cotransporter 2 (SGLT2) inhibitors in people with type 1 diabetes. Early on, I was a big fan, and I liked them because my patients liked them. They found they had less glycemic variability. They noted some weight loss and some A1c reduction, and they found their diabetes was easier to manage.
However, after using them in people with type 1 diabetes for a few months, I began to notice that people were developing diabetic ketoacidosis (DKA). I almost never have people develop DKA in my practice in affluent Beverly Hills, which is where I could use the SGLT2 inhibitors.
I ended up running into my colleague and friend, Dr Irl Hirsch, at a meeting in San Diego. After comparing notes, we realized that the increase in rates of DKA we both were seeing was due to the use of SGLT2 inhibitors.
We, along with several other coauthors, published the first case series on DKA caused by SGLT2 inhibitors. After this publication, and the ensuing change in the package insert, I developed a specific protocol for the off-label use of SGLT2 inhibitors in my own patients.
This was a very detailed protocol. I had people monitor their ketones. I educated them about what to do. We talked about DKA prevention. We talked about how and when to stop the SGLT2 inhibitor, and I reinforced this over and over again. Additionally, all patients had 24/7 access to me so they could reach out if they were at all concerned about this.
I found that my protocol actually worked really well on patients on multiple daily insulin injections. I didn’t have anybody go into DKA on multiple daily insulin injections. No matter how hard I tried, I still saw SGLT2-induced DKA in my patients on insulin pumps. This means patients on pumps alone as well as patients on automated insulin delivery systems.
What I wanted to do next was to see if I could prevent DKA in people on pumps if I used injected basal insulin in combination with their pump systems. That way, they would always have some basal insulin that would help prevent DKA. I wanted to start trying this, but then COVID happened and I stopped starting SGLT2 inhibitors in people with type 1 diabetes because I didn’t want to risk causing DKA in a pandemic.
However, over time, the data have amassed on the benefits of SGLT2 inhibitors in heart failure and CKD. I really wanted to use these agents for people with type 1 diabetes for their non-glycemic benefits, and this became a strong consideration. There’s a large amount of research going on about this, as well as the development of continuous ketone sensors.
I really am in favor of using SGLT2 inhibitors for these non-glycemic reasons as long as we can do it safely. However, we now have other agents, such as glucagon-like peptide-1 (GLP-1) receptor agonists, and nonsteroidal selective mineralocorticoid receptor antagonists, such as finerenone, for the treatment of chronic kidney disease (CKD) and heart failure.
Although these agents are also not well studied in people with type 1 diabetes, they do not seem to cause DKA and may be safer in this population. Of course, it seems highly unlikely that there will ever be head-to-head comparisons done, such as with an SGLT2 inhibitor compared with a GLP-1 receptor agonist, but we can certainly opt for off-label uses of these agents as well.
Regarding the use of continuous ketone monitoring, this may be helpful. I don’t know what it’s going to be like because I’ve never done it, but my big issue is that my patients get into trouble when devices fail, not when they work.
One of my stories of a patient who went into DKA and was on an SGLT2 inhibitor was a patient who was getting married. The night before the wedding was the rehearsal dinner, and she was very busy with wedding plans, out-of-town guests, her family, and everything else, and her insulin pump had an occlusion, her sensor failed, and she really didn’t think much about the fact that her glucose levels had become very high.
She went into DKA, pretty severe DKA, and was hospitalized in the intensive care unit at 10 PM the night before her wedding. Now, as you can imagine, we worked really hard to get her out of the hospital, still in mild DKA by the next morning, so she could have her wedding. That caused a large amount of diabetes trauma, and even though she obviously survived it, it was really upsetting to her that this happened.
I really want to prevent people from going into DKA on SGLT2 inhibitors if it’s at all possible. One of my worries about sensors is the issues that I have with people always having them available. As we all know, it can be hard to ensure that people will always have their sensors.
Even glucose sensors have their issues. They can fall off. They can become unusually unstable. It can be difficult to be sure that someone has a continuous supply. I can have problems with prior authorizations or refills. It’s difficult outside of a clinical trial to always be sure people have the supplies they need, so that worries me for people who would rely on a ketone sensor, who may have that sensor not work.
Finally, there is my underserved population in East Los Angeles with type 1 diabetes. They have much higher rates of diabetic kidney disease than patients I see in other settings. These are the people who need my help the most, but they have less access to healthcare. They have less access to someone who can help them if their ketones are positive.
They have lower levels of health literacy, and most of my patients don’t even have access to ketone test strips. For many, I can’t even use an RAAS inhibitor because I can’t deal with the hyperkalemia that may ensue. Those are people where it’s tough to use agents to help slow progression of their kidney disease, and I need to find other, safer options for them.
Thinking about it, wouldn’t it be safer to use a GLP-1 receptor agonist in these people where you’re not going to increase the risk for DKA, and even generic liraglutide, which may actually provide nonglycemic benefits? My patients in this population, even with type 1 diabetes, tend to be overweight or obese, and I think we need to think of other solutions, safer solutions, for a population where access issues are really significant.
It’s been an interesting journey, and my understanding of these agents only grows. I want all of my patients to have the best possible, safest care to help deal with the progression of diabetic kidney disease, heart failure, and other nonglycemic benefits. Thank you.
Any views expressed above are the author's own and do not necessarily reflect the views of WebMD or Medscape.
TOP PICKS FOR YOU