RECOMMENDED AND ACHIEVED TREATMENT GOALS IN CKD. THE TABLE STUDY

Luca De Nicola, Naples, Italy

Chair: Garabed Eknoyan, Houston, USA

Francesco Locatelli, Lecco, Italy

Prof L. De Nicola Chair of Nephrology Second University of Naples Naples, Italy

---

Slide 1

Thank you Professor Eknoyan and Professor Locatelli. Dear colleagues and friends let me first thank the organising committee for the opportunity to be here and show you the results of the TABLE study, which is a study that I have the honour to coordinate in Italy. The aim of this study is to assess the achievement, the degree of implementation of clinical practice guidelines and try to improve achievement of a blood pressure target, cholesterol target, anaemia target in Italian renal clinics.

Slide 2

Well, let’s start from a slide that Professor Eknoyan showed last year here at the EDTA meeting. As all of you are aware, we are facing the growing epidemic of end-stage renal disease with a number of patients treated worldwide with the maintenance dialysis, which is expected to double in this decade.

Slide 3

However, due to the very high mortality rate which has been registered, at least in the United States, most likely entering a renal replacement therapy program may not be the worst outcome for CDK patients.

Slide 4

Indeed, studies in the United States have shown that the cardiovascular risk is 8-fold greater than the risk of reaching end-stage renal disease for these patients. More specifically, as shown by large studies recently published, mortality risk increases with decline of renal function and this happens since the very early stages of renal disease. Also what is important to note is that the risk is doubled in patients with chronic kidney disease and diabetes mellitus.

Slide 5

The cardiovascular problem is so important that a non-nephrological society such as the American Heart Association has now stated that the kidney disease is a major risk factor for development of cardiovascular disease. Sarnak and co-workers in this statement published 3 years ago in Circulation have also listed the potential determinants of cardiovascular risk, which are actually prevalent in CKD patients. They also make a difference between the traditional risk factors and non-traditional risk factors. The first are those classically described by the Framingham study, while non-traditional are those specifically related to renal disease per se. But what is important for the clinical management of CKD patients is that here we have unmodifiable risk factors such as age, gender, diabetes or family history of cardiovascular disease and some others that are modifiable. That means that we can influence the role of these factors by intervening with a more aggressive therapy. Mainly, these are hypertension, albuminuria, anaemia, dyslipidemia and also what is important to note is that most of these modifiable risk factors that are determinants of cardiovascular risk also are involved in the pathophysiology of progression of renal disease. So now we can define, at least most of them, as cardio-renal risk factors.

Slide 6

This definition is not useless because it actually can help us in our daily job because now we know that we can kill two birds with the same stone. I mean we can cure two diseases, heart disease and renal disease with the same therapy.

Slide 7

That’s actually what we have to remember every time we sit in front of our patients trying to improve the cardio-renal outcome by intensification of therapy as suggested by current clinical practice guidelines.

Slide 8

Now, the critical question arising here as mentioned by Professor Eknoyan is what is, after wide diffusion of guidelines, the level of implementation of guidelines in the real world of clinical practice? Well, this question actually is what is behind the TABLE study.

Slide 9

The first slides will be describing the cross-sectional phase of this study, which has been already published and in the second part I will show you the prospective phase. We enrolled patients with frank chronic kidney disease with a GFR less than 60. An important criterion was that these patients to be enrolled, required had a massive adequate nephrology follow-up, this means at least 1 year of follow-up in the same renal clinic. This is important because I believe that if we want to improve our management of CKD patients, we must start all the processes at home in our renal clinics, because nephrology luckily is still the reference of care for these patients. Obviously, we excluded patients undergoing renal replacement therapy, as well as patients with suspected or diagnosed acute renal failure. So we enrolled more than 1200 patients, regularly seen in the 26 Italian renal clinics either academic or public hospitals, spread out over the entire Italian country.

Slide 10

We published in the Journal of Nephrology and 3 months ago in KI the results of this first phase.

Slide 11

Now, let’s look at the basal clinical characteristics. As you can see, these characteristics identify a high risk cohort because these patients are old, one out of 3 has diabetes. Almost 60% has left ventricular hypertrophy and 1 patient out of 3 has a previous cardiovascular event. Also GFR characterises them as having advanced CKD and 40% of patients had a significant proteinuria that means proteinuria above 1 g/day. Nephrology follow-up prior to the study was of 3 years on average.

Slide 12

Here I show you the prevalence of risk factors, which are out of the levels recommended by clinical practice guidelines. Patients have been stratified by GFR level in stage 3, stage 4 and stage 5 of KDOQI. As you can see, the most prevalent uncontrolled risk factors are hypertension, dyslipidemia and anaemia. Specifically, hypertension that we define as a blood pressure level above the target of 130/80 was present in almost 90% of patients. Dyslipidemia that is a cholesterol level above 200mg/dL was present in 50-60% of patients and as expected, prevalence rates of anaemia rose with decline of GFR up to 50% in stage 5 of CKD. The prevalence of hypertension and anaemia most likely accounted for the high frequency of left ventricular hypertrophy detected in these patients. On the other hand what we found is that other risk factors such as smoking, alteration of calcium-phosphate metabolism and malnutrition had a very low prevalence rate.

Slide 13

Another further important finding was that probability for having in these patients an increased number of uncontrolled risk factors was dependent on the clinical characteristics of patients. In fact, while age and gender did not significantly influence the number of factors what we found is that diabetes and cardiovascular disease and the presence of a more advanced CKD were associated with the risk of having a higher number of uncontrolled risk factors. The meaning of this finding in my opinion is that nephrologists do not properly intensify therapy when the cardio-renal picture gets worse in these patients. This is actually a paradox because these high risk patients will gain the most benefit from intensification of therapy as shown in several papers and recommended by clinical practice guidelines.

Slide 14

So the next question is, how was the prescribed therapy?

Slide 15

Well, let’s start with hypertension. We found that most of these patients were hypertensive, this occurred despite the vast majority of these hypertensive patients were actually treated according to guidelines in terms of number of drugs and in terms of inhibition of rennin-angiotensin system, so where is the problem? The problem probably is that there was a less appropriate intervention on the main cause of hypertension in CKD, which is the extracellular volume expansion. Indeed, we found that only 18% of these patients were adherent to a low sodium diet, here measured by 24-hour urinary sodium excretion and despite this poor compliance to low sodium diet only 40% of patients were taking loop diuretics. But the problem is not only of frequency of prescription of loop diuretics indeed what we found is that most of these patients with advanced renal insufficiency were actually taking a placebo dose of furosemide.

Slide 16

Hypercholesterolemia was present in 48% of these patients and in spite of this we found that only less than 1 patient out of 3 was treated with statin.

Slide 17

The same held true for anaemia. The cumulative prevalence rate was 20% and again the vast majority of anaemic patients were left untreated.

Slide 18

So the next question is, is it possible to improve implementation of guidelines in the real world of renal clinics?

Slide 19

To do this, to answer this question, we built up a prospective phase, which is a goal-oriented protocol aimed at improving achievement of these 3 important cardio-renal targets, in which we do not give any pre-specified therapeutic algorithm to participating physicians but we only ask them to try to achieve the therapeutic target. Now, we are completing the third visit. I’ll show you the data of the active patients which are 639. We have now completed 2.3 years of follow-up. Two more visits and one more year of follow-up is expected in the next future.

Slide 20

Well, achievement of BP target improved by almost 60% in the prevalence rate of the target but still there is large room for improvement because we have still 80% of patients with blood pressure values above 130/80.

Slide 21

The improvement in achievement of target was likely not due to the slight increase in the number of drugs even though here you can see it is statistically significant but in my opinion it is not clinically significant but it was probably rather due to the larger use of furosemide, the increase in the dose of furosemide and above all the larger implementation of low sodium diet.

Slide 22

Also the cholesterol target improved.

Slide 23

Even though we still have 40% of patients out of target and this was obviously due to the larger use of statins but again there is room for improvement because here we have almost 60% of patients that are left untreated despite high cholesterol levels.

Slide 24

Now, at variance with the previous two targets we did not note any improvement in the achievement of the haemoglobin target. They actually slightly got worse.

Slide 25

This maybe due, at least in part, to the natural progression of renal disease but what is important, I mean what we want to study here is what are the areas for improvement?

Slide 26

So, let’s look at therapy that is something that we can modify. So these are the EPO prescriptions in these anaemic patients and as you can see, a lack of improvement in the haemoglobin target was associated with an increase in the prevalence of patients treated with erythropoietin. So most likely there was no main role for prescription of EPO, even though we still have 50% of patients that need to be treated.

Slide 27

Also the dose of EPO either epoietin or darbo poietin appeared adequate for these patients. The problem here emerging is that probably nephrologists change their way of thinking when they shift from dialysis to pre-dialysis clinics. The sense is that they overlooked the iron status in these patients.

Slide 28

Indeed, assessment of iron status by measuring ferritin and transferrine saturation levels was obtained in only 56% of patients and when we had these data, we found that only one patient out of 3 had normal iron status while 70% of these patients had either an absolute iron deficiency or a functional iron deficiency.

Slide 29

So what are the changes in the prevalence of targets and what are the possible areas of improvement suggested by this study? As I told you, we got the increase in the prevalence of BP target and maybe we must go for it by further increasing the number of prescriptions of furosemide trying to be stronger with our patients implementing low sodium diet because it works. Also, this is the most obvious part, we can improve the target of cholesterol in these patients by larger use of statins. We must be more careful in analysing also in these patients as we do with the dialysis patients. We have to monitor the iron stores in our patients.

Slide 30

So what are the conclusions? In non-dialysed CKD patients achievement of therapeutic targets for hypertension, anaemia and dyslipidemia is far to be adequate despite prolonged follow-up in renal clinics. Achievement of targets is less frequent, paradoxically it’s less frequent in patients carrying the highest cardio-renal risk such as diabetic patients, patients with previous cardiovascular disease and patients with more advanced renal failure but the good news is that we can make it, we can improve it and to do that we need to perform a more careful monitoring and intensify therapy in these patients. Thank you for your attention.

Slide 31

Chairman: Thank you very much Luca. Are there any questions? Yes.

Question: Iswill, Leeds U.K. this is obviously a very nice paper and the idea that you should use goal-orientation to achieve change is a perfectly plausible suggestion and you leave us with the impression that if only we would do the right thing then we would get the results and I’m afraid it just simply is not true. The simple translation of a target value given for blood pressure, cholesterol or haemoglobin is not a recipe for achieving the best outcomes. Because a recipe for a blood pressure less than 130/80 does not really tell you where to go with the blood pressure, when to stop treatment, when to back off. In fact as you have shown in many, many studies now if you use this technique, the mean of the outcome will be the upper limit of your goal orientation, so you’ve got 60% which is close to mean median of that outcome and the reason is that when people get to 130 they stop. 130/80 is the point at which the last drug should go in, it’s not the point at which you should stop. The hypertensive doctors seem to be entirely obsessed with the notion of a value less than, less than 140/90, less than 150 whatever and it’s a recipe, a systematic recipe for failure and exactly the same thing goes for cholesterol. If you stop treatment when the cholesterol gets to 5 mmol, then you will actually have a mean outcome of 5 and half your patients will actually not achieve the target and in this sense the guidelines are systematically responsible for underachievement because we haven’t understood that you can simply take the desirable limits and transfer them into a practical instruction for the clinic. So, I’m commenting on this as a general phenomenon but we really need to work out how to lower those values of exasperation.

Prof De Nicola: Well, thank you for the comment. What I can say is that what we asked all the participating physicians was not obviously to achieve the target but was to maintain the target because we already told them we were going to continue with the follow-up. So what will happen in the next 2 years when we analyse the data will be to understand what has been the best treatment to achieve and maintain the therapeutic target? Also I don’t think the nephrologists reach the target and after try to start to down titrate drugs. It is not usual, usually you go there and you stay there with the therapy. Also what is important to note is that what I showed is that we still have after 2 years 80% of patients not reaching the target but what is most impressive is that we have in our cohort more than 60% of patients, which is not at the target suggested for essential hypertension, 140/90. So there is a long way to go.

Question: Well, what we know from dispersion studies and distributions is that to get 85% of cholesterols under 5 you have to have a mean value of 4.3. If you want to get 85% of systolic blood pressures under 130, you will have to have a mean value of 122. If you want to get 85% of haemoglobins over 11, you have to have a mean value of 12.5. This is known actually, it is intractable, it can’t be changed and the guidelines don’t contain the information.  So the point of my comment is to say let’s not turn limits into practical instructions, let’s respect the dispersions and put those in the guidelines.

Prof De Nicola: Well, let’s clarify what is our goal, which is achieving the target 50% of patients.

Chairman: Professor Lameire.

Question: I have a comment. I’m over here. Good. So it is know from population studies in the States and in Europe that approximately in the general population 10% of the population has a GFR, an estimated GFR below 60 ml/min. So there are 50 million Italians something like that? So that means 5 million Italians are out there and taken care of not by nephrologists, so you may wonder what is going on in primary care in Italy, in Belgium, in the U.K. whatever. So do you have any idea what that means for nephrology practice? And I partially agree with the previous speaker that we should indeed, if we want to achieve 85% of the targets that does mean that if you say 11 is our minimum your mean should be 13 you see because the targets as defined by the guidelines should be considered as minimum values at least if they are based on evidence based medicine. So just coming back to the points of the question what is going on in the general practice in Italy with chronic kidney disease patients, do you have any idea?

Prof De Nicola: I have an idea because actually we did two more studies with primary care physicians which were published in AJKD and a few months ago appeared also a study in which we were collaborating with diabetologists and appeared in Diabetes Care and what emerges that all non-nephrologists figures are actually not measuring GFR, are actually treating CKD patients as essential hypertensives that’s what is occurring I believe. Coming back to the first question I want to make the point, that our goal is to reach target not in 100% of patients but in at least 50% of patients.

Question: What is the incentive you think that we should apply trying to stimulate the nephrologists or the GP or whoever takes care to do better? Is it financial oh well, I’m not sure that it is no.

Prof De Nicola: I’m not sure.

Question: In countries where achieving targets is related to reimbursement policies, the targets are more easily obtained than if you are not connecting it to money. So normally doctors always think about money and never speak about it but this is a financial problem also.

Prof De Nicola: I don’t know, the problem is that I don’t think that there is a difference whether you compare this data with other data I mean in other countries, so the problem is that there is a general attitude of the physician that is true that every time you sit in front of a very ill patient, you are scared of intensifying therapy because what we know is that under treatment of CKD patients is not limited to nephrologists but occurs among the cardiology community, it occurs among general practitioners. Also large studies done in essential hypertensives have shown that physicians are reluctant to intensify therapy and paradoxically these occur when patients are very ill.

Chairman: One last small question and a very, very quick answer.

Question: I’ll try Drueke from Paris. You showed very interesting mean values for the whole country of Italy. Of course, you had individual values for each centre. Would you show the centres with the best achievements and those with the least good achievements and tell the ones who have the least good achievements the other ones are much better than you are?

Prof De Nicola: So they’re going to kill me. No, actually telling the truth and you can get this information from the paper, when we analysed these data we obviously adjusted for the centre and there is no influence of the centre so also in my centre we have the fault.

Question: Yes but what was the variability of the problem?

Prof De Nicola: No there was not so much variability.

Chairman: Thank you.