Chronic Kidney Disease Podcast

Chronic Kidney Disease and SGLT2 Inhibitors: Usage, Monitoring, and Possible Side Effects

Matthew A. Sparks, MD; Brendon Neuen, MD


August 11, 2022

This transcript has been edited for clarity.

Matthew Sparks, MD: Hi. I'm Dr Matthew Sparks and welcome to Medscape's InDiscussion series on chronic kidney disease. This is episode 1, and today we'll be discussing the proper use of SGLT2 inhibitors in primary care to prevent chronic kidney disease. We'll be discussing how to start them, how to monitor them, their side effects, and every detail you ever wanted to know. First, let me introduce my guest, Dr Brendon Neuen, who is a renal advanced trainee and research fellow at the George Institute for Global Health in Sydney, Australia. Welcome to InDiscussion.

Brendon Neuen, MD: Thanks, Matt. It's a real pleasure to be here. Really excited to speak to you today.

Sparks: Well, this is an area that I'm very excited about. In the nephrology world, we call it flozination. That is the initiation of an SLGT2 inhibitor in a person. And in order to become an expert flozinator, I thought I'd bring a good friend and colleague, Dr Brendon Neuen, whom I've known since he was a medical student. Now, you are a fellow and the only fellow that we'll be talking to during this series. But even at that career stage, you have really shown yourself to be one of the leading experts in this area. Can you tell us a little bit about yourself and how that occurred at such a young age?

Neuen: I've been incredibly lucky throughout my career. I did my master's at the University of Oxford and trained in clinical trials in epidemiology, and I also did my PhD around the same time with an interest in diabetic kidney disease. And just as I was starting my PhD, the CANVAS program was being reported; that was the cardiovascular outcome trial of canagliflozin, the SGLT2 inhibitor. I was heavily involved in the reporting of much of the results from CANVAS and then subsequently the CREDENCE trial as well, and have a strong interest in clinical trials to assess kidney disease progression and particularly preventing cardiovascular events. Now I'm involved in a number of other trials and in the CKD-EPI consortium in Boston, and hopefully we'll have a chance to spend some more time in the United States and visit you down in Durham.

Sparks: Oh, we'd love to have you. This is really a great topic, and flozination is a topic that I really like to talk about. I think there are a lot of misconceptions and a lot of issues that we need to talk about. Our goal is to get people comfortable prescribing these medications. If we're going to make a dent in this CKD epidemic that's currently happening and diminish the number of people who need dialysis or transplantation, this is the group of individuals that we need to empower to give these medications. First, you mentioned a little bit about some of the clinical trials that you are involved in. Can you talk a little bit about your conflicts of interest so that listeners understand?

Neuen: I've been involved in the reporting of the CANVAS and CREDENCE trials. They were sponsored by Janssen, and I served on the steering committee of the FINE-CKD trial, which is evaluating the effect of finerenone in nondiabetic kidney disease and that's sponsored by Bayer, and I also served on advisory boards of other companies, including those that make SGLT2 inhibitors.

Sparks: Well, let's dive right in. I think the very first topic I want to talk about is urinary tract infections in SGLT2 inhibitors. This seems to be the number one thing that people worry about. What is the reason for it and what are the data surrounding genital mycotic infections and also urinary tract infections in general?

Neuen: This is really interesting, Matt, and it's one of the most common misconceptions that we encounter when we speak to our clinical colleagues. When you look at the clinical trial data from CREDENCE and DAPA CKD — they're the renal outcome trials and the cardiovascular outcome trials [respectively] in type 2 diabetes — so, collectively, [in] about 40,000 people, there was no evidence that SGLT2 inhibitors increase the risk of urinary tract infections. Now, when you pool all the data together, including the heart failure trials, you get a very marginal increase, about 5% or so in that mark. That was in Natalie Staplin's meta-analysis published late last year. That risk is negligible in comparison to mycotic genital infections. The first important point, I think, is to distinguish between UTIs versus mycotic genital infections. For UTIs, there's probably little to no increased risk based on the randomized evidence, but a three- to four-fold-increased risk of mycotic genital infections, predominantly in the diabetic population. That's probably related to the glycosuric effect of these agents, and it doesn't seem to be as common in nondiabetic populations. When you look at CREDENCE — that was the renal outcome trial that I was involved in with canagliflozin — when we looked at mycotic genital infections, so thrush and that kind of thing, apart from the three- to four-fold increase, most of those events were pretty mild and patients were either able to continue treatment or discontinue temporarily while it was treated and then resume canagliflozin treatment. In most instances, they're pretty mild, but it does bring up an important point, and that is that we should always inform patients about this and emphasize the importance of personal hygiene to reduce the risk of mycotic infections. Generally, in people with type 2 diabetes, it doesn't seem to be a problem, and at least in DAPA CKD in the nondiabetic population.

Sparks: There are two questions I want to ask, and the first one is about urinary tract infections. What were the exclusion criteria for the clinical trials, because these were patients with very low risk for urinary tract infections, so I guess people were excluded who had a history of urinary tract infections?

Neuen: In CREDENCE and DAPA CKD, there were no explicit exclusion criteria for urinary tract infection. If you had a urinary tract infection before, it didn't mean that you couldn't get into the study. Now, there are no data collected on people with recurrent urinary tract infections, so we don't know. But based on the inclusion criteria of those trials, there would be people with recurrent infections who were included in the trials. There's no evidence to suggest that they're at elevated risk at this stage. Previous history of urinary tract infections is not an exclusion criterion or a reason to not use an SGLT2 inhibitor. Now, you might ask about people with suprapubic catheters or other complex anatomy. They haven't really been studied in any detail, so it's a bit hard to know or say with any certainty. But certainly, for the general population, if you have a woman with diabetes who's had a urinary tract infection before, that's not necessarily an indication to not prescribe an SGLT2 inhibitor.

Sparks: And just to tell our listeners, what exactly is a genital mycotic infection?

Neuen: That's things like thrush — generally the absolute incidence is higher in women than in men, as you'd expect, and they are generally treated with topical antifungal treatments, if necessary, but in most instances are pretty mild in severity.

Sparks: Just to summarize, urinary tract infections were not exclusion criteria. They have not been seen, just run-of-the-mill urine extraction function as an adverse event signal in randomized clinical trials. However, if you pool all the data, there may be a small marginal increase, not a lot. The real risk is genital mycotic infections, for which all patients should receive counseling about proper hygiene. Next is diabetic ketoacidosis and euglycemic ketoacidosis. I want to wrap that into discussion of type 1 diabetes. So why are patients more susceptible to this? Which patients [are susceptible] and how should it be monitored?

Neuen: I want to cover a couple of things: why it happens and who is at risk and what can we potentially do about it. The reason is that these drugs, this class of agents, they modestly increase your ketone levels because they alter the insulin-to-glucagon ratio by promoting glycosuria. Now, in most instances, that's not a problem. That modest increase in ketones might actually be a good thing because we know that in situations of stress, cardiac myocytes preferably use ketones as a really efficient energy source, but when that tips over and you get ketoacidosis, that's when it potentially becomes a problem. That's why it occurs. Now who's at increased risk of ketoacidosis? Generally, it's been observed that people who have long-standing diabetes, who have been on insulin for a long time, are at most risk of ketoacidosis. The so-called burnt-out pancreas, long-standing diabetic patient. Those patients might be at increased risk, and they seem to be at increased risk in the time of intercurrent illness. They might be fasting for a surgical procedure, or they might be unwell with an acute intercurrent illness like vomiting or diarrhea. Based on that kind of observation, I think it's not unreasonable for us, as practitioners in clinical practice, to inform patients that if they're unwell or not eating and drinking or fasting, then they should temporarily withhold their SGLT2 inhibitor. That's what I do in my own practice and that's certainly important around the time of surgery, and now our anesthesia colleagues are generally quite aware of that. Having said all of this, that advice might change over time. The reason I say this is because the risk is very, very low in the general diabetic population. The incidence in the trials is approximately 0.1 events per 1000 patient-years or 0.1 to 2 per 1000 patient-years. These agents are now being studied, as you might be aware, even in hospitalized COVID patients. And there was a trial of dapagliflozin in hospitalized patients with COVID, which showed that the drug was overall generally very well tolerated. It's still being tested in the recovery trial as a treatment for COVID. There's the risk that if we discontinue treatments in hospitalized patients, some of these patients may not get back on to treatment and have progression of kidney disease or worsening cardiovascular disease as a result. There's a balance to the sick day rules. Having said all that, I think it's reasonable to have some advice about sick day rules, and that's what I generally do now. I think that's going to change over time, though.

Sparks: And the last two things are type 1 diabetes and I think we should also mention the keto diet.

Neuen: Certainly, in type 1 diabetes, I wouldn't use an SGLT2 inhibitor because the benefit-risk profile is completely different from that in your general patient with type 2 diabetes or diabetic kidney disease. In those circumstances, I don't do it. I wouldn't recommend it unless you had the close involvement of an endocrinologist. And then in terms of the keto diet or variations on that, it potentially increases the risk of ketoacidosis as well. If you have a patient with diabetes who is undertaking that form of diet, I probably wouldn't, just out of caution, use the SGLT2 inhibitor because of the potential for an increased risk of ketoacidosis.

Sparks: One thing I always like to do is tell learners and trainees, what is that risk of ketoacidosis in a type 1 diabetic? What are the data for that? What percent of patients, in one of those clinical trials that were done in the early years of SGLT2 inhibitors in type 1 diabetes, develop ketoacidosis?

Neuen: There is certainly much less data, as you'd expect, in type 1 than in type 2. There are ongoing renal outcome trials that will include a small proportion of people with type 1 diabetes. EMPA-KIDNEY has some type 1 diabetic [participants]. But certainly, at least if you look at the type 2 diabetes trials, there's at least a two- to three-fold-increased risk. That absolute risk is very, very small. In type 1 diabetes, that absolute risk is larger. It's not like that 0.1 per 1000 patient-years. It might be substantially higher than that depending on the patient and how they manage their insulin and so forth.

Sparks: I do know of one individual, who is a nephrologist, who has type 1 diabetes and is on an SGLT2 inhibitor. Monitoring ketones is really important for that. The next area is something that gets a lot of attention, and I think we really need to dispel some of the myths on this and that is limb amputation. And you mentioned that you were involved in the CANVAS trial and that's the study that really sparked this controversy surrounding it. Can you tell us a little bit about where we currently stand with limb amputation and how someone with peripheral arterial disease should be evaluated for the potential use of SGLT2 inhibitors?

Neuen: I think this is an important question, Matt, because people with peripheral artery disease are at much higher risk of cardiovascular events. The absolute benefits of treatment are even larger in people with peripheral artery disease than in people without peripheral artery disease. If you have peripheral artery disease, that's a signal that you're at very high risk of cardiovascular complications and would benefit substantially in absolute terms from an SGLT2 inhibitor. The concern that you referred to is that in the CANVAS program, which was the cardiovascular outcome trial of canagliflozin, there was an increased risk of amputation observed in that trial. That was unexpected. And in the CREDENCE trial which followed, there was no increased risk of amputation observed with canagliflozin, despite patients in CREDENCE who had diabetic kidney disease being at much higher risk of amputation. Now, when you compare that across all the trials conducted to date with other agents within the class, no other trial has observed an increased risk of amputations overall or in the individual trials. We did a very detailed analysis of CANVAS and CREDENCE to try and find out if there were any patients who had increased risk and why might this be. We couldn't find any explanation as to why we observed an increased risk of amputations, and considering the number of outcomes that have been collected across the trials and across the agents within the class, there is a reasonable chance that one finding in one trial was probably a chance finding. You never know for certain, but that's looking more and more likely, and it is really reassuring when you look at the other trials that there's no — absolutely no — increased risk observed whatsoever. I think for the clinical question of the patient with diabetes and perhaps kidney disease with peripheral vascular disease, those patients are at substantially increased risk of cardiovascular events. We've shown, in the analyses of the trials that have been conducted, that the absolute benefits of treatment are larger in those patients. They should be prioritized certainly. Once that signal was observed in the CANVAS program, it was mandated that patients should have their limbs inspected at a clinic visit and have appropriate foot care. That's part of routine clinical care anyway and important to emphasize. But nonetheless, that is important, I think, and worth doing in general.

Sparks: This is a really important point. I want to just recap that. First off, in the United States, we have the FDA, and the FDA did place a black box warning on canagliflozin when CANVAS came out. However, it was removed in 2020. So just to be clear to our listeners, the black box warning for foot amputations [with canagliflozin] has now been removed by the United States FDA. However, I think any time you have a warning like that, it is hard to get it out of the minds of physicians and clinicians. And I still see it today where if someone had a prior amputation — which I do agree, that's a very high risk individual — they're not starting them on an SGLT2 inhibitor. I just want to hear from you a very succinct recommendation. If you have a patient with prior amputation, what should you do?

Neuen: To use the modern terminology, Matt, we need to flozinate … we need to flozinate those patients. Outpatients like that are at substantially increased risk of cardiovascular complications, and treatment with an SGLT2 inhibitor reduces their risk of heart failure by 30%, reduces their risk of major adverse cardiovascular events by about 15%-20%, and reduces their risk of needing dialysis by about 30%. The benefits are substantial. In relative terms, the absolute benefits are even larger because they're at increased risk of all those complications.

Sparks: Well said. Now, what about someone with an active lower limb wound like a diabetic foot ulcer or osteomyelitis? Would you start an SGLT2 inhibitor in that scenario if needed, for kidney and cardiovascular disease? Would you hold someone's SGLT2 inhibitor in that scenario?

Neuen: Yeah, look, I think that's an important subtle point. I don't think there's a right or wrong answer to this particularly. But if patients are hospitalized because they have osteomyelitis, they might have other intercurrent illnesses going on. I generally try and have people stable before I put them on an SGLT2 inhibitor, in terms of the other acute illnesses. If you look at cardiovascular medicine and, you know, we're not that dissimilar in in nephrology, there are a lot of proven treatments. But the biggest challenge for us as clinicians is getting everyone on treatment who deserves and needs treatment. So, I want to caution against delaying treatment, because we know that we're not very good, as clinicians, in getting people onto therapy that's shown in registries. But at the same time, I think it's not unreasonable, if people are unwell, to make sure that their infections resolve, they're more comfortable, more mobile. I don't think there's necessarily a rush, but you don't want people to miss the boat at the same time.

Sparks: Let's talk about classification of these medications. And this is something that I think we, in the field of medicine in general, need to really tackle. And I think it's also hindering our usage of these agents. Should they be classified as diabetic drugs? Should we reclassify them in a different way? In that same discussion, how effective are they at lowering glucose? I think the most interesting thing, if you look at a nondiabetic population, there's not even a risk of hypoglycemia. Talk a little about that and can we increase their usage if we stop talking about them in terms of diabetic drugs?

Neuen: Well, 100%, these are not glucose-lowering agents. I know I'm preaching to the choir here with you, Matt. These drugs have now shown significant benefits for patient-level kidney outcomes in people with heart failure, with and without diabetes. That's patients with heart failure with reduced and preserved ejection fraction, diabetic and nondiabetic kidney disease with consistent benefits, regardless of diabetes. Because of that, I think these medications are for all of us to use: endocrinologists, nephrologists, cardiologists. And it's important that no single specialty owns this drug because they have such clear benefits for patient-level kidney outcomes. With respect to the glucose-lowering effect specifically within nephrology, they're actually pretty average glucose-lowering drugs in the patients that we typically see in clinic. One of the characteristic aspects of this class of medication is that their glycosuric effect is entirely dependent on GFR [glomerular filtration rate]. If GFR falls, they actually become ineffective glucose-lowering agents. Despite that, they're at least as effective at preventing kidney failure, the need for dialysis, and heart failure across the spectrum of GFR, even with a starting GFR of 20. So clearly, they have benefits that go far beyond their glycosuric effect. And I think it's important that all specialties have a knowledge of SGLT2 inhibitors so that we can prescribe them to the patients that we see. And we're seeing all the same types of patients just from a slightly different lens. The patient with heart failure with preserved ejection fraction and some mild CKD seeing a cardiologist is not dissimilar to the types of people we see in renal clinic, and it's important that we all appreciate that and really empower everyone to use these agents as effectively as possible.

Sparks: We are getting to the end of this podcast, and it's been absolutely phenomenal. I'll call this The Flozination episode, and I want to go ahead and just wrap up our key points. We talked a lot about some side effects. And I want to point this out: these are still rare side effects, and these are very safe medications. We've also dispelled some myths. Let's talk about those. Diabetic ketoacidosis is a risk. In most patients, it's a low risk. We should talked about sick day rules; however, that paradigm is coming under question because of the very low risk of ketoacidosis in somebody who, for instance, is fasting for surgery. We talked about urinary tract infection and how the studies have not shown any difference in just run-of-the-mill urinary tract infections. We talked about genital mycotic infections — that risk is real but can be surpassed by proper hygiene. That's an important point. We talked about reclassifying these agents as good drugs to preserve kidney function, health, and cardiovascular disease. And importantly, we talked about limb amputation and a signal that was seen in only one clinical trial, the CANVAS trial, which resulted in a black box warning from the FDA that has since been removed, and we talked about how those patients with peripheral arterial disease are really the ones that have very high rates of kidney failure and heart failure and cardiovascular disease. And so, we need to look at those patients very closely. Those are the areas that we discussed. We also talked about all the great things about these drugs. What are the benefits of these drugs?

Neuen: In people with chronic kidney disease, they reduce the risk of dialysis by 30%-35% on top of standard of care. That includes renin angiotensin system blockade. They reduce the risk of hospitalization for heart failure by about 30%-40%, irrespective of the presence or absence of diabetes and in people with heart failure with reduced or preserved ejection fraction. And overall, when you look at meta-analyses across the trials, they also reduce cardiovascular death and extend life expectancy. So [they have] very clear benefits on patient-level kidney and cardiovascular outcomes.

Sparks: Okay, great. Today we've had the pleasure of talking with Dr Brendon Neuen, discussing the proper use of SGLT2 inhibitors in primary care to prevent chronic kidney disease. Thank you so much for joining us for this episode. I have thoroughly enjoyed it and can't wait to see you back for episode 2. This is Matt Sparks for Medscape InDiscussion.


SGLT2 Inhibitors in Primary Care: 'All Hands on Deck' for Improving Heart Failure Outcomes

Canagliflozin and Cardiovascular and Renal Events in Type 2 Diabetes

Canagliflozin and Renal Outcomes in Type 2 Diabetes and Nephropathy

CKD Epidemiology Collaboration CKD-EPI consortium

Effect of Finerenone on Chronic Kidney Disease Outcomes in Type 2 Diabetes

Net Effects of Sodium-Glucose Co-Transporter-2 Inhibition in Different Patient Groups: A Meta-Analysis of Large Placebo-Controlled Randomized Trials

EMPA-KIDNEY (The Study of Heart and Kidney Protection With Empagliflozin)

FDA Removes Boxed Warning About Risk of Leg and Foot Amputations for the Diabetes Medicine Canagliflozin (Invokana, Invokamet, Invokamet XR)

Follow Medscape on Facebook, Twitter, Instagram, and YouTube


Comments on Medscape are moderated and should be professional in tone and on topic. You must declare any conflicts of interest related to your comments and responses. Please see our Commenting Guide for further information. We reserve the right to remove posts at our sole discretion.