Reducing Kidney and CV Complications in Diabetes: 'Room for Improvement'

Interview With Peter Rossing

Anne L. Peters, MD; Peter Rossing, MD, DMSc


July 26, 2019

This transcript has been edited for clarity.

Anne L. Peters, MD: Hello. I'm Anne Peters from the University of Southern California, and I'm joined today by Peter Rossing from the Steno Diabetes Center in Copenhagen. Peter, you gave the Edwin Bierman Lecture at the American Diabetes Association (ADA) this year called "Linking Kidney and Cardiovascular Complications—Impact on Prognosis and Treatment." Could you go through a little of what you talked about and what you think is clinically important in terms of biomarkers and new clinical treatments for reducing cardiovascular and kidney disease risks?

Terrible Triad: CKD, CVD, and Diabetes

Peter Rossing, MD, DMSc: It was really a pleasure, and I am very happy that there has been a lot of focus on both cardiovascular and renal complications at ADA. It's a very important topic. I talked about how having diabetes does not have a very good prognosis, but kidney disease on top of that will increase the risk for cardiovascular complications, both myocardial infarction and heart failure. Therefore, it's important to look out for kidney disease. It's a kind of modifier, and you have this triad of adverse diseases: diabetes, kidney, and cardiovascular.

We have been looking at this epidemiologic link, but we can use a lot of biomarkers to predict kidney or cardiovascular disease that are shared by these two complications of diabetes. Of course, glucose is important. Uric acid was debated at the ADA meeting as a potential risk modifier. We are involved in a trial trying to target uric acid. But I think what we, as clinicians, can do right now is to measure urinary albumin excretion rate and estimate renal function in our patients, because if albuminuria goes up or estimated glomerular filtration rate (eGFR) goes down, then the risk for mortality or cardiovascular and renal events increases critically.

We can do something about it, but we need to measure [these biomarkers]. From recent data, this is not happening—at least not in Europe or in the United States. We need to improve this. In addition to urinary albumin excretion rate and eGFR, which are the markers we can use in the clinic today, I also talked about some potential markers for the future. We talked about markers of fibrosis because I think fibrosis is a shared biomarker for both kidney and cardiovascular disease, and I also suggested that we should target fibrosis in treating cardiovascular and renal complications.

We have worked with blocking the renin-angiotensin system with angiotensin-converting enzyme (ACE) inhibitors and angiotensin II receptor blockers (ARBs), but also using aldosterone blockers might be a good thing to do.[1] Yesterday I gave a talk about finerenone, a new compound that blocks aldosterone that may not be associated with potassium problems. It is being tested in large outcomes trials right now (FIGARO-DKD and FIDELIO-DKD). That will be interesting to see. Biomarkers can be used to indicate cardio/renal risk and also to give us some advice on new treatment options.

24-Hour Urine or Spot Test?

Peters: When you say measurement of albuminuria, do you mean a urine spot test or a 24-hour collection? Some people argue one way or the other. What should clinicians be doing?

Rossing: For many years we swore that you needed to do 24-hour urine collection, that that was the best estimate you could get for evaluation. But in everyday practice, we only do spot urine testing. It's reliable, has a good predictive value for kidney and cardiovascular outcome, and it's much easier for the patient to do. That is what we recommend. You can do 24-hour collections in the research setting and also evaluate sodium and protein intake and so on. You might get a better assessment. But usually when we talk about creatinine clearance, we say that measuring creatinine and collecting urine for creatinine excretion is no good because nobody can collect urine reliably for 24 hours.

I don't care [which test]. Collect the easiest available urine sample and test it for albumin. For most, that is a morning spot urine or a random spot urine. If that is normal, I think you are well on your way, and if it's abnormal, then you have a person at risk and you can do something about it. You should repeat the spot test because there is day-to-day variability, but don't order 24-hour urines in everyday clinic. That is too complicated and will not be done. The urine albumin-to-creatinine ratio takes care of this collection issue and is much more handy in the clinic.

Clinical Use of Albumin-to- Creatinine Ratio

Peters: Do you follow the ratio with treatment to show that it improves?

Rossing: Yes. We follow that because there are a lot of data showing that if you have a reduction in albuminuria, then you are doing something good. You are lowering the risk for both cardiovascular and renal outcomes in the long term. Just recently there was a publication of approximately 30 intervention studies and over a million observational datapoints linking short-term changes in albuminuria with long-term benefits in hard outcomes like renal failure or cardiovascular events.[2] It's not just about finding patients at risk; it's also about monitoring whether your intervention is successful.

Peters: If somebody was on an ACE inhibitor or an ARB and you didn't see a change in their albumin-to-creatinine ratio and their blood pressure could tolerate it, would that tell you that you should increase the dose?

Rossing: Yes, we worked previously with increasing doses beyond what is good for lowering blood pressure, and we saw additional effects on albuminuria, but we have no long-term data suggesting that that really translates into better [outcomes]. I think what we would probably rather do today is to put people on an ACE inhibitor or an ARB, and then, at least for patients with type 2 diabetes, use a sodium-glucose cotransporter 2 (SGLT2) inhibitor or glucagon-like peptide-1 (GLP-1) receptor agonist. For the moment, the best data are with the SGLT2 inhibitors. The CREDENCE study[3] with canagliflozin showed that you reduce renal outcomes in patients with proteinuria and type 2 diabetes.

Now we're beginning to discuss using SGLT2 inhibitors in type 1 diabetes, but it's very early days. To me it would be more important if it could affect the cardiovascular and renal outcome than lower glucose a little bit in the type 1 population. But for type 2 diabetes, we're getting more and more data that the SGLT2 inhibitors and also the GLP-1 receptor agonists have good effects in patients with persistent proteinuria.

SGLT2 Inhibitor Use in Renal Impairment

Peters: Yes, I've been really impressed by the SGLT2 inhibitor data coming out, and the CREDENCE [results were impressive]. I was thrilled to see that you could use them at a lower eGFR because that has been my concern as I treat patients, because I want to use agents safely. I'm a little nervous about patients with lower eGFRs in general. It was pretty impressive that you could use it all the way down to an eGFR of 30 mL/min. This may be a game changer.

Rossing: I think you are right. The labels are still saying, at least in my country, that if you are below 60 mL/min you cannot start SGLT2 inhibitors, and if you are below 45 mL/min and on an SGLT2 inhibitor, then you should stop it. But as you say, the studies go down to around to an eGFR of 30 mL/min and there is no increase in side effects, no increase in acute kidney injury. The kidney and cardiovascular protective effect also seems to be preserved in this range. The reason for the eGFR thresholds levels is more related to the lack of effect on lowering glucose than on side effects. Therefore, with more data we get in this range, I think that will change soon. I think some countries have already changed the label so that SGLT2 inhibitors can be used down to 30 mL/min or so, and labels will probably change in many other countries also going forward.

GLP-1 Receptor Agonist

Peters: It's very interesting. What about the GLP-1 receptor agonist? How do you see the role? I know they had some benefit, and it is seemingly less, but we really have not done head-to-head trials.

Rossing: We have no head-to-head trials in the renal or cardiovascular arena. We have just started a study called FLOW with one of the GLP-1 receptor agonists. It's a large outcome study in patients with type 2 diabetes and kidney disease to see if you can prevent hard renal outcomes in this population with a GLP-1 receptor agonist. These drugs have a cardiovascular protective effect in the population with a lower eGFR. There are also data at least on albuminuria and maybe on eGFR as well, that they have good effects in this population at high risk for both renal and cardiovascular problems.

How Would You Treat?

Peters: Let's say you see a 65-year-old patient with type 2 diabetes and albuminuria who is on an ACE inhibitor and a statin. What is the first thing you think or worry about? Their LDL cholesterol? Their blood pressure? Adding in more drugs? How do you assess that person?

Rossing: That is a good question. From the Steno-2 trial,[4] we learned that multiple interventions are extremely important in such a person. We need to address lipids, blood pressure, and glucose. We need to address smoking if that is an issue, and a healthy lifestyle is important as well. There were data from Joslin[5] showing that over the long term, this may impact the kidney in a beneficial way. If the patient was on a statin and an ACE inhibitor, I would probably consider a SGLT2 inhibitor if there was existing cardiovascular disease or a GLP-1 receptor agonist, particularly if there is desire for weight loss.

We need to consider these new drugs. Unfortunately there were some data from the United States by Mikhail Kosiborod, published in Circulation, showing that only approximately 8% of patients with type 2 diabetes are treated according to the guidelines.[6] Maybe that is just because the data were collected while the guidelines were being developed, so hopefully it would improve with a new survey. But there is room for improvement, and we as clinicians should address all of the risk factors because it really impacts survival, cardiovascular events, and renal events in our patients.

Peters: Yes, I encourage people to follow the current guidelines,[7] because after metformin, step two is to determine whether a patient has preexisting cardiovascular disease, chronic kidney disease, or heart failure, and then treat those patients differently perhaps, but also more aggressively for all of these risk factors.

Thank you very much for talking with us today.

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.