Evolution of Chronic Kidney Disease: Defining a Model of Care

World Congress of Nephrology
(ASN/ISN) Scientific Meeting October, 2001


Closing Comments, Questions and Answers

Panel Discussion

Dr. Singh
Dr. Kimmel
Dr. McClellan
Dr. Provenzano
Dr. Silverberg
Dr. Dickmeyer
Question index:
Questions received during this symposium have been paraphrased
and the answers submitted by the panelists are presented.


Why do patients with prerenal azotemia (e.g., CHF) develop low erythropoietin levels and anemia?
At what level of GFR should CMS (HCFA) reimburse for epoetin alfa treatment?
Is creatinine a valid surrogate for renal function in prerenal azotemia (e.g., CHF)?
In the KEEP study, were there subgroups of renal disease of interest?
Is improvement of ejection fraction with epoetin alfa meaningful, since this is expected with a fall in heart rate?

Why is the rate of patients entering dialysis who are on epoetin alfa increasing so slowly?

What are the pros and cons of biweekly dosing of epoetin alfa or its congeners?

Panelist Responses

(Back to question index)

Why do patients with prerenal azotemia (e.g., CHF) develop low erythropoietin levels and anemia?

Audience member:
I have a difficult time thinking pathophysiologically why someone who is strongly prerenal, which is basically what CHF patients' renal failure is from, which is evidenced by the fact that they get better after heart transplant, why they would be associated with a low serum erythropoietin level. It seems to me a fundamentally different disease process from say, diabetic nephropathy, where you have loss of renal mass. I wonder if you have measured erythropoietin levels in these patients to distinguish how much of the anemia that you see amongst CHF patients is from anemia of erythropoietin deficiency because of renal failure versus how much is erythropoietin resistance from chronic inflammation or perhaps anemia of chronic disease? The treatment of both would still be epoetin alfa probably, but I was just wondering if erythropoietin levels have even been measured in such patients?

Dr. Silverberg:
It's an excellent question. Volpe and others have shown in Italy that as heart failure gets worse, erythropoietin rises markedly. You can get levels of erythropoietin in heart failure that are 20 times normal. So how can it be that you get so much anemia? So there is certainly a group of patients where we have not measured this. But there is certainly a group of patients in whom the erythropoietin levels are extremely high in heart failure, and despite that, they still are anemic. Clearly, this is erythropoietin resistance, perhaps related to TNF alfa, which is known to interfere with erythropoietin.

There is, however, another group of patients where erythropoietin is low. I don't know how many of you are aware of it, but in diabetes with proteinuria, there is practically no erythropoietin in the blood and half the diabetics with proteinuria are anemic because their levels of erythropoietin are totally absent. They don't have it. It also goes out in the urine. There have been studies done by Vaziri and others, for example ( Vaziri ND et al, Am J Med. 1992 Jan;92(1):35-40.). There is a study from Jerusalem now Feinstein S et al, Am J Kidney Dis. 2001 Apr;37(4):736-42.) There are many studies.

So there seems to be two groups here, a group with a low hemoglobin from perhaps loss of erythropoietin, and a group with resistance. But take a look at your diabetics. You will find anemia very early on in their course. If they have proteinuria, they are anemic.

(Back to question index)

At what level of GFR should CMS (HCFA) reimburse for epoetin alfa treatment?

Audience member:
This is a question and also a comment directed to Dr. McClellan. The PAERI study shows that there is a very high prevalence of anemia in patients with chronic kidney disease as defined by creatinine clearance using Cockcroft-Gault. I think the MDRD GFR equation will actually give us an even larger prevalence and incidence of patients with anemia. Right now in the US, many patients are dependent upon CMS or HCFA for their coverage in the pre-ESRD population because as the "baby boomers" get older, the incidence of chronic kidney disease patients is also getting older.

What concerns me is that right now the people at CMS, which used to be called HCFA, are asking for advice. The physicians that work for CMS are asking for advice as to what level of renal failure to start treating anemia at. From my understanding, they want to cut off the treatment with a creatinine of 60 mL per minute. I think they're going to miss a lot of anemia by doing that. I think they are going to be doing a disservice to us. I was wondering if, Dr. McClellan, you could comment on this and if you felt perhaps it would be appropriate for American physicians to e-mail their congressman and perhaps e-mail HCFA regarding this concern.

Dr. McClellan:
I certainly can't speak to the policy environment because I haven't been involved in those discussions, unfortunately. Clearly, there is a strong linear correlation between the level of GFR and the prevalence and degree of anemia. In terms of influencing those discussions, it certainly would help to engage the RPA and more directly, your congressperson -- man or woman -- and to the folks at the Health Policy Branch at CMS. So to the extent that those decisions are being carried on at a national level, rather than at the local carrier level, that will certainly influence the decision-making process.

(Back to question index)

Is creatinine a valid surrogate for renal function in prerenal azotemia (e.g., CHF)?

Dr. Kimmel:
My question was similar to that raised before. It's a question that came up both in Dr. Silverberg's talk and your talk where much of the data are presented on the basis of serum creatinine. In several of your studies -- and I think Dr. Silverberg, in your study -- we don't have urinary protein concentrations or excretion rates. So the question of how much is prerenal azotemia because of heart failure is an important one. I wanted to ask you Bill if you could give us an estimate of whether there is a bias towards overestimating the presence of chronic renal insufficiency in your study or presentation. Or do you think that creatinine is a valid surrogate for both renal disease in the presence of congestive heart failure without what we normally think of as structural renal disease?

Dr. McClellan:
First, I need to clarify one of my -- the Medicare data that I showed you was actually based on the first creatinine measured during admission and the first hemoglobin measured during admission. So to the extent that these patients were decompensated volume overloaded, there is a bias towards a dilutional lowering of both the creatinine and the hemoglobin. That is to say that once they are stabilized, diuresed and at dry weight, you would expect the creatinine to actually go up and the hemoglobin to go up as well. That's observed in several early clinical trials, about a 10% increase in creatinine between admission discharge.

So the data that I showed tends to underestimate the presence of elevated serum creatinine and overestimate the degree of anemia. Do I think that the serum creatinine is an accurate -- or the Cockcroft-Gault GFR is an accurate estimate of the underlying renal function? Yes I do in heart failure, but it doesn't tell you whether the diminished GFR is due to prerenal or renal causes.

Dr. Kimmel:
So the implications of a decreased GFR, whether it's with more traditional renal diseases or mostly prerenal azotemia, you would say are the same?

Dr. McClellan:
At least in the observational studies, that's right.

Dr. Silverberg:

No one has ever verified the Cockcroft-Gault formula in heart failure, but we think it's okay. But there has been no verification of this.

(Back to question index)

In the KEEP study, were there subgroups of renal disease of interest?

Audience member:
A question to Dr. McClellan about the KEEP study and the prevalence of CKD in family members of patients with ESRD. Were there any data on the risk of CKD relevant to the cause of ESRD in the proband? If diabetes was at high prevalence in your area, this may give a different result vs. GN or hypertension. Was there any sub-grouping of patients? What's the risk in each group?

Dr. McClellan:
The 184 family members, of course, that I showed you data for, did not have end-stage renal disease. They came off the street for screening, and about half of them were diabetic. We did ask what the cause of renal failure was in the index case. In other words, the family member that led them to declare a family history of end-stage renal disease. The proportions of individuals with diabetic renal failure and hypertensive renal failure were very comparable to what we were seeing in the Georgia population or the Southeastern U.S, population. About one-third were hypertension and about 40% were diabetics. The rest were miscellaneous causes or unknown.

We haven't actually looked though at the influence or association between the reported index cases cause of end-stage renal disease and the presence of abnormalities in the family member. That will be an interesting thing to do, but we just haven't gotten around to doing it yet.

(Back to question index)

Is improvement of ejection fraction with epoetin alfa meaningful, since this is expected with a fall in heart rate?

Audience member:
I have a question for Dr. Silverberg. The improvement in ejection fraction was very impressive and interesting in your patients with heart failure whose anemia was corrected. But of course, as the heart rate slows, the ejection fraction -- the stroke volume -- increases. So I want to ask you whether, as a surrogate for cardiac output, if you multiply the ejection fraction by the heart rate, whether the cardiac output would increase as you corrected the anemia, indicating an improved function of the failing heart?

Dr. Silverberg:
Yes, you could do that. I would just like to make one point to you all. I don't know how many people in the panel would agree or not agree. When you see a patient with a creatinine that's elevated and a hemoglobin that's elevated, do you do an echo to see whether that patient has serious heart damage already? Very often you can't know. That patient may have a dilated heart or left ventricular hypertrophy. They may have, in fact, the beginning of heart failure and may need the beta blockers that are used and the ACE inhibitors that are used and the Aldactone (spironolactone) that is used. All these are part of the treatment of heart failure. But that patient sitting across from you at the table may have a very damaged heart. Unless you get an echo, you may miss that diagnosis. Do you do those? I think we're missing a lot of cases of damaged hearts. If you find them when they're on dialysis five years later it may be too late to intervene.

(Back to question index)

Why is the rate of patients entering dialysis that are on epoetin alfa increasing so slowly?

Dr. Singh:
I would like to ask the panel a question. If you look back over the past five years, I think it's true that there has been an improvement in the number of patients who start dialysis, in terms of the number of those patients that are on erythropoietin at the time of starting dialysis. But it's not been a very dramatic improvement. Can anyone, Dr. Provenzano, Dr. McClellan speculate as to why that is? Is it because this is an issue with reimbursement, an issue with access to care, or all of the above? Why do you think that's happened?

Dr. Provenzano:
I can try to answer the question, but there is evidence that everything you have said is true. Obviously, in a busy practice the logistics involved in administering a medication every week and specifically administering a medication intravenously is often preclusive. I think only in the last several years has information come out on the clinical importance of correcting anemia.

Certainly, you can go back 10 years with hemoglobins in the 10-11 range. There are arguments made that this doesn't impact outcomes, and "We will just wait until they go on dialysis and treat the anemia at that time." Now with increasing evidence -- particularly as was pointed out today -- of the overwhelming positive impact of correcting anemia, trying to deal with the logistics of correction has become more important.

What I have found is, that although there has been some increase in the use of epoetin alfa in CKD, I think data from five years ago suggested that only 30% of anemic patients presented to a nephrologist were treated. That has improved to 45-50% now. As I pointed out, working through clinics as far as drug administration, whether it's weekly and now maybe every other week or less frequently, use of physician extenders, disease management models, and various setups so that IV iron can be administrated, I think you're going to see an improvement in the logistics of treating anemia. So I am hoping that some of these barriers will be overcome.

I'm not sure out there -- maybe with a show of hands. Does dosing of erythropoietin weekly, is that an impediment to therapy? Do you view that as an impediment? Would less frequent dosing be something that you think would improve your patient compliance or logistics in your clinic of treating patients? One person says yes. All right.

If I could just make one other comment in regard to the CKD population. Our experience in working with a couple of national health plans is that the majority of PCPs, primary care physicians, do not know who to test for CKD with a serum creatinine. I think there is a huge opportunity for the nephrology community to educate the PCP community of number one, who should be tested, who should have a serum creatinine, and number two, when they should be referred to the nephrologists.

I think there is a perception out there that PCPs do not necessarily want to refer patients to nephrologists, but we're also hearing that the PCPs at least tell us, "Have that nephrologist educate us in regard to when these patients should be referred." So I guess the suggestion that I'm making is that there are many patients out there that have CKD. We don't know that they have CKD because they aren't being tested. There aren't clear guidelines as to when these patients should see the community nephrologist.

And taking off from that, a recent poster analyzing Medicare data on CKD patients shows still that the majority of cases are treated by nephrologists. The expectation of both family practitioners and internal medicine physicians is that the nephrologist will be treating this anemia. So there is some increasing comfort level with the nephrologist of treating the anemia, but certainly there has to be an improvement.

Dr. Silverberg:
Can I make a comment? If you want to see the patients who are going to be on dialysis, the worst patients, if you want to detect them, you don't have to go any further than your medical wards in the hospital. We looked at 1,700 admissions in December of last year to our medical wards. 17% were heart failure, most of them recurrent heart failure. Their average hemoglobin was 10, their average creatinine was 2.2.

The patients who are going to come onto dialysis are those patients who are currently lying in your hospital beds in medical wards and cardiac wards. Those are the patients that you've got to get. I don't know how you're going to get them, but what we did was we simply wandered up and down the wards and spoke to the doctors and said, "Send them to us, at least for the treatment of the anemia." We got them by the hundreds.

Dr. McClellan:
Let me amplify that point. Heart failure is the most common cause for admission to the hospital or discharge from the hospital in the Medicare population. There are well over 1,000,000 discharges a year for heart failure in that population. So just to amplify what Don has said, based on the Georgia data, about a third of them will have chronic kidney disease, and about 40% of them will be symptomatically anemic. So this is just one example of how labeling somebody heart failure, rather than chronic kidney disease with heart failure frames our management and often precludes effective interventions, simply because we're not thinking about the disease properly.

(Back to question index)

What are the pros and cons of biweekly dosing of epoetin alfa or its congeners?

Audience member:
I think in response to what Bob Provenzano just said, will once-a-week or twice-a-month make any difference, for example, in the dosing of drug to raise hemoglobin. I think that it's important to bear in mind what Dr. Dickmeyer was talking about, that basically chronic kidney disease is a lot more of a problem than just the management of anemia.

The management of blood pressure is clearly one of the biggest problems that we have in patients with chronic kidney disease. So it makes sense that if we can interact with a patient, at least our staff can interact with a patient more frequently, the better opportunity we have to encourage a patient to take their blood pressure medicine. So in that regard, I think there can be a cancellation effect of the convenience to a patient to get a drug twice a month, to have more interaction with his physician and his physician staff or her physician staff in interacting once-a-week. So I think there might be a cancellation effect and even a positive benefit if you look at the total picture.

Dr. McClellan:
You're absolutely right. We were very pleased to find in the 600 patients that have been studied on weekly therapy that the compliance rate was above 85%. The opportunity to interact with nursing staff certainly helps pull together all aspects of their care. I agree with you. I think you're going to see the tradeoff.

Dr. Singh:
So we'll end here. I'm sure you'll join Dr. Kimmel and myself in thanking our presenters this evening, Dr. Provenzano, Dr. McClellan, Dr. Silverberg and Dr. Dickmeyer. Thank you very much.