|
|
|
 |
The Advancement of Clinical Practice in
Chronic Kidney Disease
 RPA Satellite Symposium
Washington, D.C., February, 2002. Panel Discussion |
|
|
|
|
|
|
Question index:
Questions received during this symposium have been paraphrased
and the answers submitted by the panelists are presented.
Is it beneficial to treat symptomatic cardiac disease in patients that are on dialysis?
Panelist Responses
(Back to question index)
Is it beneficial to treat symptomatic cardiac disease in patients that are on dialysis?
Dr. Patrick Parfrey: I'd like to ask Dr. Bolton a question: You were involved in the Normalization of Hematocrit Study, which involved patients with symptomatic cardiac disease who were on dialysis. There was an adverse outcome in that particular study. On the other hand we've got data coming from Silverberg which suggests that treatment of symptomatic heart failure in patients who have got anemia may be beneficial. So in some ways that conclusion is counterintuitive to the data that came from your study. Would you like to comment on that?
Dr. Kline Bolton: Yes. The study that was performed relative to normalization of hematocrit that was published in the New England Journal of Medicine was structured to look at a very high risk group of patients. The idea was that they already had bad congestive heart failure or ischemic heart disease and that a beneficial effect would be most apparent in this high risk group. So first of all they were atypical of the general dialysis population in many ways. The supposition was made that a normal hematocrit, since that's what we all walk around with, would be the ideal, and that the standard hematocrit at that time of approximately 30 to 33 would be used as a control. As we all know, the study was stopped as it approached the point of doing harm. It didn't reach statistical significance but it was stopped and there was a lot of concern as to what was going on. A lot of analysis has gone into trying to understand why this happened.
One of the outcomes of the study that I think is important is that it was a very rigorous study, it was prospective and randomized, it was well-monitored and it had an excellent statistical analysis. And what it showed was the same thing that we've talked about today. It did show that for any patient, whether they were in the normalization group or in the regular hematocrit group, that the higher the hematocrit the better their outcomes was. There seemed to be, for reasons that were unclear, an associated increased risk of death in being randomized to the high hematocrit group. The high hematocrit patients fared well.
A number of factors I think may be involved in addition to the fact that these were not a representative population and were high risk. One of those is that much more iron was required for the high hematocrit group and there's controversy as to whether iron is toxic, whether it could have had an effect on lipid peroxidation or on free radicals. Another was that there was an increased incidence of thrombosis of the vascular access. And so there was a possibility that this also led to some of the increased morbidity that was seen and mortality.
The hematocrits were about 45, 43 to 45 in this group of patients. And so when they were dialyzed and underwent ultrafiltration we don't know exactly how high the hematocrits got. It was possible that some of those got very high and played a role. And furthermore the Kt/V values were lower in the patients with a high hematocrit. The final factor was the fact that those patients with normalization had much larger doses of erythropoietin in order to get the hematocrits up and we don't know whether erythropoietin, which is a growth factor of itself and also has vasoactive properties, could have played a role in that. A lot of analysis was done but the riddle has not been solved and hopefully more representative samples of populations may help to deal with that.
(Back to question index)
What about looking at hematocrits in the 39% range and seeing if that might be better in the CKD population in terms of preventing people from ending up with LVH?
Audience member: There was a study by Nathan Levin, that looked at a population where the hematocrit at the onset of dialysis was 39% and then they looked at another population where it was lower. And the incidence of LVH was much higher in the lower hematocrit group. As a compromise, what about maybe doing a trial to look at hematocrits in that range, the 39% range and seeing if that might be better in the CKD population in preventing people from ending up with LVH since we already know that 74% percent of the patients that are going to start dialysis are going to start with LVH. And they don't start with hematocrits of 39%, I can tell you that. Maybe we should be looking at a hematocrit of 39%. I'm not saying 43%. But I'm saying like 39 or something along that range. What do you think about that?
Dr. Donal Reddan: I think that's a very important question. And keep in mind that the upper group target in the CHOIR study is going to be 13 to 13.5 g/dl hemoglobin which would correspond exactly to the hematocrit levels that you're speaking of. So maybe we'll know the answer then.
Dr. Patrick Parfrey: Can I just comment about that? In the trial that we're doing looking at normalization of hemoglobin in patients starting dialysis where we're targeting to a hemoglobin of between 13.5 and 14.5, it turns out that the hemoglobin that we're actually achieving is going to be about 13.2 and the hemoglobin in the control group where we're targeting between 10.5 and 11.5, the hemoglobin that we're actually achieving is below 11. So we'll answer your question, not the one we planned to answer.
(Back to question index)
The prevalence of anemia in renal patient populations is high; is that something that is encountered in cardiology populations?
Audience member: I have a question I'd like to ask Dr. Al-Ahmad and I guess the entire panel. When we started looking at the prevalence of anemia in these different patient populations we were really surprised to find how prevalent it actually was. Is that something that you encounter in cardiology populations and if so why, when you review the literature, is the cardiology and the renal literature so silent about it?
Dr. Amin Al-Ahmad: Absolutely. It's something we find quite a bit and particularly as patients are sicker and patients who have heart failure class III and IV certainly have a much higher incidence of anemia. I'm not sure why the literature is silent about it. I think that there's beginning to be interest in the cardiology community now after the analysis of SOLVD and after some of Don Silverberg's work. And I think that I can say that there's very likely to be a randomized controlled trial that's currently being planned to look at heart failure and treatment of anemia. Which is a little bit of a different question than just looking at it from a renal standpoint but I think an important one to answer.
(Back to question index)
Why don't we as nephrologists use quality of life as being an important outcome for clinical trials like the oncologists do?
Dr. Patrick Parfrey: One consistent observation that we find in trials of getting higher hemoglobins is the observation about improvement in quality of life. And if we're going to try and get better clinical outcomes in the moderately severe chronic renal disease population that Donal is going to study it might be that we need better sample sizes to achieve that outcome. So why do we not as nephrologists use quality of life as being an important outcome for clinical trials like the oncologists do, seeing that our disease is even more malignant than the cancer?
Dr. William McClellan: I don't in any way want to diminish the importance of quality of life as a parameter in the management of chronic kidney disease. As we all know chronic diseases extend over long periods of time and the burden of the chronic disease is often best captured by how you feel and how you live your life as much as whether you live your life to the full length that you otherwise would expect to. In other words, whether there's a decrement in life expectancy. Having acknowledged that, I think that we have always regarded renal replacement therapy as a life extending intervention, and so we have focused on longevity and survival to the exclusion of other things like social and mental and physical well-being and functioning. Having acknowledged that, I must say that the paucity of any clinical trials looking at therapeutic modalities in chronic kidney disease has really begged this question of whether we should be using hard or soft outcomes. I would think that a trial looking at the impact of anemia correction in chronic kidney disease patients with heart failure could very easily include as one of its primary outcomes quality of life. Having said that, you as well I know that the FDA and the people who are responsible for labeling issues would not be very impressed with that kind of analysis.
Dr. Patrick Parfrey: That's politics, right? I mean, the oncologists can prescribe erythropoietin to improve quality of life and we can't get it for kidney disease patients.
Dr. William McClellan: Well, I think the idea is, well, the oncology patient is going to die anyway so let's make them more comfortable, whereas in nephrology our task is survival and rehabilitation.
Dr. Kline Bolton: I think that part of what we have to do is to emphasize the fact that oncology patients are approached in this same way and yet our patients have a worse prognosis than the majority of the oncology patients that are being treated. So we really have a double standard in that if you have AIDS or rheumatoid arthritis or cancer, you're allowed to begin therapy when the hematocrit falls to 36. You don't have to wait until your hematocrit is 30. And you could treat until the hematocrit reaches 39. And so the question is why are renal patients being discriminated against which is basically what this is is discrimination.
Relative to the quality of life issue, Pat, I do feel strongly that that's important. In the normalization trial, there was a quality of life assessment and certain parameters within the quality of life assessment actually were statistically significant. Others were not. But I think when you begin to look at the issue of nutrition and exercise and cytokines and the other parameters that are involved in outcomes that quality of life and more exercise and better nutrition may well play a significant role and that we should be looking at quality of life assessment tools while we're doing the rest of the things.
Dr. Amin Al-Ahmad: I just wanted to add a quick comment, that in heart failure patients we often look at quality of life but in many heart failure studies there have been improvements in hemodynamics, improvements in quality of life with a worsening mortality with the therapy such as oral inotropic therapy which now nobody uses it because of that reason. So I think you do need the hard outcomes at least for the heart failure patients as well as the quality of life.
(Back to question index)
Why were there discrepancies between prevalence of CKD as estimated from NHANES and the recognized prevalence from the PAERI study?
Dr. Anton Schoolwerth: Dr. McClellan, you emphasized the discrepancies between prevalence as estimated from NHANES and the recognized prevalence from the PAERI study. Might we speculate on some of the reasons for the discrepancy? Were these creatinines not obtained? They were not looked at? They were not recognized for their significance? Do you have any data from that study you can share with us?
Dr. William McClellan: We're just now looking at the proportion of patient charts screened in which an actual creatinine was available. But we certainly expect it to be less than uniformly present in the chart. The second factor is, of course, the NHANES data that has been published is based on the derived GFR using the MDRD formula. And so extrapolating those prevalence estimates of four and a half to five percent for moderate and severe renal insufficiency to a comparable proportion of patients with a serum creatinine and a comparable level is somewhat fraught with hazard although it's a reasonable extrapolation. When you look at Camille Jones' data that were published a number of years ago and then just recently updated in a publication this year in Archives with Josef Coresh at Hopkins, they estimated that about three percent of the population had a serum creatinine, if they were male, of 1.6 or greater and female 1.4 or greater. So even using those numbers we're finding fewer patients. But not as severe deficiency as when you use the creatinine clearance. So underascertainment is our working hypothesis. We don't believe for a minute that the primary care population is healthier than the general population, or at least we hope not. But that of course is something that we'll know after we've got the data analyzed.
Dr. Patrick Parfrey: But it brings up some other questions because I was pretty impressed by the fact that in your data the family practices identified 2.3 percent of their population as having renal impairment, i.e., they probably found 50 percent. That's pretty good, considering that there's no effort to get out and have screening programs that we know to be efficacious. Identify people who have a raised serum creatinine that's not due to diabetes, not due to hypertension, is not associated with proteinuria, it may not be likely to progress. We don't even know what happens to those groups of patients.
(Back to question index)
Any thoughts on how to spread the word to our colleagues and the public to screen for and manage CKD better?
Dr. Anton Schoolwerth: Any thoughts on how to spread the word to our colleagues and the public to screen and manage better?
Dr. William McClellan: Probably the most important initiative right now along those lines is the National Kidney Disease Education Program which the NIDDK has inaugurated during the last year. Hopefully you'll be seeing educational messages both for patients in the general population as well as provider populations over the next year. I think the publication of the K/DOQI guidelines and the coming management guidelines that will be coming from the NKF on chronic kidney disease will give us an evidence based platform with which to talk to the primary care physician population as well as our nephrology colleagues.
Beyond educational efforts I think there's an intense interest in devising systematic reforms in the way we deliver care to this chronic disease population, particularly early renal insufficiency clinics. And this randomized trial that Pat has described in Canada using nurse clinicians and algorithm driven management of patients with chronic kidney disease supervised by nephrologists is a nice model and at least proof of principle has been supplied by Dr. Bolton and his colleagues from the University of Virginia.
So beyond education and beyond devising new models of care the final thing I think you're going to see is an increasing emphasis in the quality assessment and quality improvement activities in hospitals and perhaps even within the ESRD program asking the question, for example, of those patients we completed a form 2728 for when they begin dialysis, what proportion of them were anemic, what proportion of those who were anemic were on erythropoietin. How many of them came in with a mature access for dialysis? How many of them had been seen appropriately seen by a nephrologist prior to initiation of renal replacement therapy and was their management reflective of some of these disease modifying therapies that we discussed?
And if not perhaps bringing that information to the attention of the clinicians involved in that particular system of practice, hospital dialysis clinic group or whatever, might encourage them to improve care. So I think that all of those things ... education, better practice models and then an emphasis on quality assurance and improvement activities that have worked so well in dialysis are all going to have to happen if we're going to see improved care.
Dr. Kline Bolton: I think also that we're pretty far behind the curve in nephrology in terms of doing studies. And we don't know much about what the natural course of, chronic kidney disease, is. I mean, we have the CKD data showing 20 million at risk and then another 20 million people with some degree of chronic kidney disease. But we don't have a very good clue other than when they get GFRs of 30 or below what's going to happen. And I think that if we're going to have a convincing voice relative to working with our primary care colleagues that we need to have some of that information. And I think that the cohort study, the CRIC study from NIH that's currently underway and will be coming out in about five or six years, is going to be providing a great deal of important information that can be used for educational purposes or ammunition and what have you to help support these positions as well.
(Back to question index)
Is kidney disease a risk factor for cardiovascular disease globally?
Dr. Anton Schoolwerth: Dr. Parfrey, you presented some data that, as I interpret it, showed some contradictions. It may be more semantic although I think it isn't. Is kidney disease a risk factor for cardiovascular disease globally?
Dr. Patrick Parfrey: I think that hypothesis I'm suggesting is that renal insufficiency per se is not associated with a pro-atherogenic state, and that ... the incidence or the prevalence of disease is attributable to comorbid conditions like diabetes, hypertension, ect. And that conclusion is based on the language in those studies that I discussed. I think that the incidence rates of de novo heart failure are much higher than in the general population and the data concerning hypertrophy and the development of heart failure are such, that I think the evidence is better to say it's a cardiomyopathic state. Now, part of the conclusions you make from that information is not semantics. But what it means to me is that if the renal insufficiency state is not pro-atherogenic then it's really the diseases that are causing the renal insufficiency that we should be going after that allows us to identify high risk people just as Bill was saying and it allows us to put interventions into effect for those high risk people. Otherwise we go chasing around saying renal insufficiency has got atherogenic components to it that are over and above those diseases however, this is my hypothesis, it may not be true.
(Back to question index)
What about proteinuria and albuminuria, are they global risk factors?
Dr. Anton Schoolwerth: What about proteinuria. Albuminuria?
Dr. Patrick Parfrey: I think that proteinuria is probably generally reflective of a widespread endothelial dysfunction and that it's likely to be associated with adverse cardiac effects at the end of the day. We know that proteinuria is associated with progression of renal impairment and in the Framingham study it's associated with the increased risk of cardiac events.
Dr. William McClellan: So if you have a patient with proteinuria you need to engage in cardiovascular risk modification.
Dr. Kline Bolton: And it needs to be remembered that proteinuria is bad for the kidney. And anti-proteinuric regimens in the NKF PARADE initiative I think is extremely important in terms of outcomes for our patients.
(Back to question index)
Is renal insufficiency a risk factor for atherosclerotic disease?
Dr. Donal Reddan: I have a question for Dr. Parfrey as well. You've presented some pretty compelling data that chronic renal insufficiency may not necessarily be a risk factor for atherogenic diseases, maybe more a cardiomyopathic state. I'd like to point to the paper ... I think it was in JASN in January looking at NHANES II where NHANES II data was linked with National Death Statistics Data. I think Dr. Paul Muntner was the first author which suggested that chronic renal insufficiency as in a GFR less than 70 was associated with an increased risk of all cause death and also cardiovascular death. I'd just like to suggest that data to introduce a little bit of equipoise into the discussions. I know you're going to say that the National Health Statistics death data has a lot of problems and the cause of death is not very well ascertained. But I think it's important that there may be evidence on the other side that renal insufficiency very well may be a risk factor for atherosclerotic disease.
Dr. Patrick Parfrey: The data from NHANES about cardiac events is that renal insufficiency is not a risk factor. This is a paper that's coming out in Kidney International in the next two months written by Garg et al which confirms the Framingham information.
 |
 |
||||||||||
|
|||||||||||
 |
 |
