Slide 1

Chairman, colleagues, thank you for the invitation, it’s a great pleasure to be here in Stockholm talking to you about who to screen and how to screen.

Slide 2

As you all know, we already have several published guidelines on screening for CKD, so is there really more to discuss? I definitely would say yes because this is a highly debated topic. Some are very pro screening but others are stating that there are several formal criteria for a screening programme that is not fulfilled. So we’re looking into some of these problems and we’ll start by looking at these published guidelines and then we’ll look a little at the formal criteria for such screening.

Slide 3

Screening for CKD in diabetes patients has been established for a long time. We have quite detailed guidelines from the American Diabetes Association which very clearly focus on prevention, detection and treatment of diabetes nephropathy. So this is very CKD-specific you could say. When it comes to hypertension and screening for CKD, it’s a bit different. Both the British and the US guidelines mention CKD but here it’s more as CKD as an additional cardiovascular risk factor and as a cause of hypertension. Likewise we all know the KDOQI guidelines from 2002 which focus on classification of CKD so they suggest to test those with diabetes, hypertension and certain other risk groups and also those at age 60 plus.

Slide 4

The UK guidelines published in 2005 I guess, also have the same approach. They have slightly different groups that they think are at risk. But they focus on diabetes and hypertension of course also those with prevalent cardiovascular disease and those with prostate symptoms. So the guidelines are there and when it comes to the formal criteria for such screening programmes these were stated nearly 40 years ago by Wilson and Junger and they are still quite unchanged. CKD definitely fulfils many of these criteria but there are some that are still debated and that is about the progression rate in CKD patients, as currently defined and also related to that is, do we have a test that is suitable for a screening programme? And we will look more into this. Data from the Norwegian HUNT studies and also from the US and HANES study both show that the prevalence of CKD as currently defined is very high it’s around 10% and maybe it’s increasing in the US to 12-13%.

Slide 5

But anyway it’s very high and this is in contrast with the prevalence of ESRD which has been dramatically increasing during the last 30 years both the prevalence and incidence but still it’s a rare disease. For example, in Norway the prevalence of ESRD was 708 per million while the prevalence of CKD at the same time was 150 times higher. So this is a paradox. So it’s quite clear that very few of those currently classified as having CKD progressed to ESRD.

Slide 6

So who should we screen? Should you try to find a classical person with glomerulonephritis, a young or a middle aged man? Or should we focus on the obese diabetes patients and what about all those elderly persons with an estimated GFR of 50 or so?

Slide 7

First we could start with looking at the effectiveness of the different strategies suggested. This is again data from the HUNT II study which is a population based study from Norway. All adults were invited in a county and the participation rate was 71%, so it was a lot of people it was 65,000 people attending. So if we only decided to include those with known diabetes or hypertension in the screening programme, we would find 44% of all CKD cases in this population in the total adult population. We would only include 12%, so this would be quite an efficient way of screening. We have to screen 6 persons to find one case with CKD. That’s ok. If we expand these criteria, so we include more people we would, of course, find more CKD cases. Here we would find 74% of all cases but the number included is much higher so that efficiency of the screening strategy is much lower we have to screen 16 people to find one CKD case. Which brings us to the suggested guidelines, screening strategies. We see that quite surprisingly the UK guidelines will only find 61% of all CKD cases. The US guidelines will find more 89% and both of them are quite efficient. We have to screen 9 people to find one CKD case.

Slide 8

Some years ago the ISN suggested to screen everybody attending their general practitioner and then we would find by definition you could say, nearly 100% but then we would have to screen 21 people to find one case.
In this study we found that the best strategy here would be to include those with diabetes or hypertension or age above 55. We would find 93% of all CKD cases and the number needed to screen would still be rather low. However, other screening strategies, as recently summarised by Paul de Jong in his article can also be very interesting. Their approach A here is much the same as what I’ve just described. This includes the KDOQI guidelines, the UK guidelines and so on. But approach B is interesting and we will hear more about that in the next speech I guess. Here he includes only those people where the general practitioner has found a clinical reason for testing their kidney function. In approach C we include only those who have an increased albumin excretion rate and such alternative approaches as the B and C approach here could be interesting because it’s quite clear, as we will see in the next slide, that follow up of all CKD cases found in the general population as intended here is not very efficient.

Slide 9

Back to the HUNT study we see that in stage 3A those with a GFR of 45-60 only 0.4% of those CKD cases progressed to ESRD in 8 years. Even among those with CKD stage 3 B, GFR 30-45 it was only 1.3%. Only those with CKD stage 4 had a substantial risk and even here it was only 18%. Other studies have also shown that the risk of progression in CKD stage 3 is quite low.

Slide 10

Eriksen from Tromsa in Northern Norway published this study based on a large laboratory database, so this is much like approach B in the previous slide. Even here we see that the risk for progression in CKD stage 3 is not very different from the general population. For a 60 year old male the progression or the decline in GFR was minus 0.9 ml/min/year. It was slightly less in women and slightly more at higher ages.

Slide 11

So, how should we test those people that we decide to include in our screening programme?

Estimated GFR with all its limitations is in my view quite clear the backbone of all such testing but it’s also clear that we need something else to more accurately identify those rather few people who are at risk. This something else is very, very likely to be proteinuria in some way or another. We all as clinicians have a clear feeling that proteinuria is important and Doctor Taal and Brenner in 2 review articles in the Kidney International summarised this and they pointed out that proteinuria has emerged as the single most important marker of renal risk.

Slide 12

Proteinuria is not new, visual science for proteinuria have been tested for several hundreds of years and for the last decades we have been able to test quite specifically for albuminuria and now we are even able to detect hundreds of different proteins in the urine in the same sample and in a very short time. But the problem is that we still do not know very well how to use this information for practical risk stratification for individual patients, so in some way it’s still too much a business of fortunetelling from tea leaves or using a horoscope you could say. That we must change of course.

Slide 13

Recently some studies have been published that can help us to look more into this. This study by Ishani was the first to give us information of the importance of combining GFR and albuminuria. Ishani looked into to MRFIT study and they followed 12.000 men with an increased cardiovascular risk for 25 years. After adjusting for many important co-variants like age, sex, race, diabetes, hypertension, smoking, lipids and so on they found that at all GFR levels the risk was increased with higher albuminuric excretion rates. Vice versa at all levels of proteinuria there was an increased risk with lower GFR. For example those with a GFR below 60 and proteinuria 2+ or more, they had a 33 times increased risk compared to those with a normal GFR and negative dipstick.

Slide 14

However, we still do not know how to combine GFR and albuminuria in the most optimal way. Should we stratify for albuminuria at all GFR levels below 60 or only for stage 3A or 3B or even stage 4? Should we subdivide albuminuria into micro and macro albuminuria? Paul de Jong and Ron Gansevoort in this very recent NDT editorial suggested to include albuminuria for stage 3. British guidelines I know will suggest to use macroalbuminuria for stage 3 and stage 4 as well. However, we do not know how much this will increase the diagnostic value of the classification system that we have.

Slide 15

In the future it’s quite clear that testing for other blood proteins in the urine will become important tools for diagnosing high risk patients. Several sophisticated methods are now available for detecting hundreds of different proteins in the same sample. All these methods in some way do some kind of separation of the proteins and then introduce them into an analyser and the different methods have different pros and cons.

Slide 16

And for more details in this very interesting subject I would recommend this very informative review article by Danilo Fliser. They very clearly conclude that urinary proteomics will play an important role in our very near future but very importantly they state that the perfect single biomarker does not exist. Very probably we will have to use a panel of different defined and sequenced biomarkers.
The technology is now quite advanced and well known, so what remains is to focus on standardisation of these methods and doing clinical trials to see what kind of clinical utility we can have from this.

Slide 17

So to conclude it’s clear that screening for CKD is recommended by several authorities but in my view I would say that we are probably not yet in a position to advocate screening in a general population and that relates to the fact that we need to have some more information of what groups we should be screening, who and definitely we need more knowledge about screening tests, especially about how to optimise the combined use of estimated GFR and albuminuria. Also is there some value in other clinical or other traditional factors here? And hopefully we’ll also have new markers in the future that can help us more precisely to find those people at highest risk. Then finally a full screening programme should probably be tested as randomised controlled trials before it can be advocated for our colleagues and our authorities. Thank you.

Slide 18

Chairman: Thank you very much Doctor Hallan for this nice overview. If there are questions then please use the microphones which are at the back of the aisles and identify yourself. Yes there’s one question over there please.

Question: Thank you my name is – I’m from the Kalinska University Hospital. Thank you for a good presentation. In order to advocate a screening programme, it’s not sufficient to show that you identify patients at risk or with the disease, you also have to show that they benefit from this. So it’s not only defining the patients, you also have to assure that they gain something from being identified maybe they can be treated anyway.

Dr. Hallan: Yes I definitely agree on that one. Cost benefit analysis has been done for diabetes patients for screening for CKD in diabetes patients and also in hypertensive patients and it is quite clear that in this high risk group there is a good cost benefit. For the other groups it’s not yet been tested and that’s also part of the formal criteria that has to be fulfilled before you can recommend it. Yes.

Chairman: Doctor Zoccali please.

Prof. Zoccali: I appreciated very much your final statement, a full screening programme should be tested in a randomised clinical trial because I think that till now this has been an issue which has been largely overlooked. Would you expand on this and tell us which kind of randomised clinical trial you envisage to be tested in the near future hopefully?

Dr. Hallan: That’s a very hard question. I have no clear thoughts about that but probably you would have to randomise populations to be screened and not screened. It’s a very difficult question and difficult topic and for many of those screening programmes already running in other practices this has not been done but it definitely should be done.

Question: May I add to the points you made and especially also regarding the two questions that came forward right now. You emphasised that you want to screen in an attempt to find those subjects who progress to ESRD and then later on with an intervention study to prevent it of course. But you did not rely that much on the fact whether we could by such screening also prevent the patients to progress to cardiovascular events and doing studies in an interventional way to see whether we can prevent the cardiovascular problems which in the earlier phases of CKD of course, are the first ones to happen more than reaching that final stage of ESRD.

Dr. Hallan: Yes, that’s very true the most important thing for these patients is probably the risk of cardiovascular death. That’s much more likely to occur to them than starting renal replacement therapy. So that should be included in the cost benefit analysis. For the patients, of course, both are of importance but if you only look at the economic aspects here it could be that ok you save them from dieing and then they can progress to renal failure at a later stage.

Chairman: Ok if there are no more questions then once more thank you Doctor Hallan.