NEW NKF-USA KDOQI GUIDELINES: DIABETES IN CHRONIC KIDNEY DISEASE |
||||
Katherine Tuttle, Spokane, USA
Robert G. Nelson, Phoenix, USA |
||||
Chair:
Garabed Eknoyan, Houston, USA
|
||||
Francesco Locatelli, Lecco, Italy |
||||
|
|
Slide 1
I wanted to thank the organising committee for inviting us to speak here today. I’d like to take exception of one thing that Gary said, I came from Phoenix and the weather was 20 degrees centigrade warmer there than it is here, so I find it quite balmy in this room.
Slide 2
We had a large group of people work with us over the past 2 years to develop a set of guidelines for diabetes in CKD and I’d like to begin by acknowledging those people for two reasons, one is obviously to acknowledge their hard work but the other is to show that this was an International group of people to the limits that our budget would allow. You’ll notice that we had Rudy Bilous here from Newcastle and we also had Christoph Wanner from Germany and Pablo Ashner from Latin America. So in addition to the North American contingent we had an International group as well.
Slide 3
Now, one can never overstate the importance of the problem of diabetic kidney disease and it is possible to look at this from 3 different dimensions. From the perspective of diabetes it’s one of the most frequent vascular complications accounting for 30% of the vascular complications in type I and perhaps a little bit more in the type II diabetic patient. From the perspective of kidney disease it’s the single most common cause of end-stage renal disease in developed countries and it’s rapidly becoming the most common cause in the developing world as well. In the USRDS data we say that the incidence of diabetic ESRD is about 45%, that’s a general population estimate. If we look at some of the other ethnic groups, some of the specific ethnic groups, we see it is much higher. For example, in the American Indian population that I’ve worked with for the past 20 years about 93% of the end-stage renal disease in those groups are attributable to diabetes. Of course, we should not forget the issue of cardiovascular disease, chronic kidney disease is a risk promoter and amplifier so that much of the excess cardiovascular disease in diabetes occurs in a subset of patients that have some diabetic kidney disease.
Slide 4
Now just to give you a sense of the magnitude of the problem here are some data from the USRDS showing the geographic variation in the incidence of diabetic ESRD per million Caucasian population. You can see that the darker areas are the higher incidence, the lighter areas being the lower incidence. Remember this is 1992 and you see that much of the end-stage renal disease attributable to diabetes was found in the South Western United States with some also in the Mid Atlantic region and up in the upper Mid West. That’s 1992.
Slide 5
This is what it looked like 10 years later. The same sort of numbers here again, now we’re seeing blue all across the country, here even in the red states as some might say but you have very high incidence of end-stage renal disease attributable to diabetes and the counts keep going up. If one looks at the counts on an annual basis, you can see here on the left panel of this slide that the numbers continue to relentless climb exceeding that of hypertension glomerulonephritis and cystic kidney disease and the same is true for the incidence rates of these diseases. Interestingly enough in the last couple of years we’ve seen an actual plateauing of these rates giving us a little hope that perhaps something that we’re doing is helpful.
Slide 6
To look at that a little more in detail we need to split the data by ethnic groups and if you look at this slide here looking at younger people and the incidence of diabetic ESRD by age, race and ethnicity, you’ll see that in blacks the incidence continues to climb very rapidly both in the older and in the younger groups but in the whites we’ve actually seen a decline over a period of time and this was reported a couple of years ago at the American Society at Nephrology meetings.
Slide 7
Similarly we see that the rates seem to be climbing in American Indians, as well as Asian populations and in Hispanics. Interestingly enough for someone who works with the American Indian populations in the South Western United States we’re beginning to see declines in that group of people both from publications by the Indian health services and in the very populations that I work with we’ve been seeing since 1990 a decline in the incidence of end-stage renal disease attributable to diabetes.
Slide 8
Now as Gary said earlier guidelines are statements divided into two different categories, we’ve got the guidelines which are based on a consensus within our workgroup that the strength of the evidence was sufficient to make a definitive statement about the particular clinical practice. Then we’ve got the clinical practice recommendations where we’ve got some evidence but it may not be sufficient to be able to say this is definitive clinical practice and it may just be an opinion of the workgroup members and so we differentiate it between those two in an effort to give a somewhat comprehensive guideline for people to review.
Slide 9
In addition to that it gives us a place to jump off in terms of new research recommendations. One of the things that we’ve done in this set of guidelines, which incidentally aren’t published yet but which should be out later this year probably in December, we have a very robust section of research recommendations for diabetes and CKD. Of course, when you do evidence based guidelines you need to grade the quality of the evidence and what we’ve used has been 3 categorisations here A, B and C. A and B generally are found or typically are found in the full guidelines, the clinical practice recommendations have C ratings here. These ratings are based both on applicability to people with diabetes and CKD and on the methodologic quality of the clinical paper. Was it a clinical trial, was it a cross-sectional observational study? These sorts of questions. Having said that we’ll move on to the first guideline here. We had 5 guidelines, one of them was a diagnostic guideline and the other 4 were management guidelines and then we had 4 clinical practice recommendations. I’ll spend a few moments just talking about the first two guidelines and then I’ll hand the mike over to Cathy Tuttle.
Slide 10
This is the first guideline which is about the screening for diabetic kidney disease. In each of the guidelines we asked 6 questions which are why, who, when, what, how and where. We went through that process in each of the guideline statements. So the reason for doing these is to identify people that have diabetic kidney disease or CKD of other causes in the setting of diabetes. We want to screen people for that reason. We need to screen these people at a particular time, the recommendation of the American Diabetes Association is the one that we adopted which was that type I should be screened after 5 years and then annually thereafter. The type II diabetics being screened annually from the diagnosis simply for the reason that many of these people may develop diabetes without knowing it and you really don’t know the duration of diabetes in that group of people, so you need to be screening them early. Urine albumin/creatinine ratios are an easy, cheap and convenient way to test these people. We don’t need to be doing it over night or 24 hour urine collections, so the ACR is just fine at least 2 positive samples within 3 months are acquired to make this screening diagnosis. Combined with measurement of albumin to identify kidney disease we then need to stage that kidney disease, so we need to look at the estimated GFR which is based on the serum creatinine concentration and some other variables using the MDRD equation in adults and the Schwartz or the Counahan-Barratt formula in children one can calculate this quite easily. In many of the laboratories with whom I work now we’re getting these published on the laboratory slip along with the serum creatinine value. If you don’t have that available to you, you can go to the National Kidney Foundation site where they have a GFR calculator that will give you the result. Of course, all of this can be done in an outpatient setting, so it doesn’t require a nephrologist, it doesn’t even require a physician, it can be done by nurse practitioners, nurses, other people as well. So it’s a simple easy affordable test that can be broadly applied.
Slide 11
Now what do we have when we have those tests? Well, what we’re trying to do is identify people with diabetic kidney disease. Remember that particularly in the type II diabetic you may see a substantial number of people particularly in older populations that have non-diabetic causes of their kidney disease, so how do we differentiate the two? Well the only way to really differentiate the 2 is to do a kidney biopsy but we can’t do that on everybody obviously and we’re not recommending that in any sense of the word. What we want to do is be able to identify most of the people, most of the time and to do that if they’ve got a urine albumin/creatinine ratio above 300 or macroalbuminuria that person in most cases is going to have diabetic kidney disease. I’ll touch on the cases where it wouldn’t happen in a few moments. If they’ve got microalbuminuria, we get a little more of a problem. We want to be able to see the presence of retinopathy in these patients that will make us feel more comfortable that this is related to diabetic kidney disease. In the type I diabetic patient if they’ve got at least 10 years of duration, remember we know the duration with better certainty in those than in the type II, then we can say even if they don’t have retinopathy that this is most likely attributable to diabetic kidney disease.
Slide 12
Of course, there are a number of situations where it may not be diabetic kidney disease and the factors that might tip us off to that are illustrated on this slide and I won’t go into these in detail but certainly if you’ve got any of these particular findings in a patient with diabetes that has elevated albumin excretion, it might be reasonable for you based on clinical findings to go ahead and look for other causes of kidney disease other than diabetes. That’s not to say that it isn’t diabetes but one should have clinical suspicion at that point.
Slide 13
Now remembering that we diagnose kidney disease with the albumin and we stage it with the GFR, the reason we want to do the staging is that the different stages of kidney disease shown here on the left from 1 through 5, we have different actions that are applied. So that it’s important for us to be able to identify these stages to the extent that’s possible so that we can apply the appropriate therapy for those patients.
A couple of caveats here one is that diabetic kidney disease is a hyper filtering disease, so it’s not uncommon to see a patient that has a GFR say of 90 that was 180 two years ago and if that’s the case, that’s been a 50% reduction, well, it’s been a halving of their kidney function and even if their GFR is 90 that suggests they might have quite profound kidney disease.
Slide 14
Now, let’s move onto guideline 2 where we’re talking about the treatment of hyperglycaemia and this is where we got some help from the people from the American Diabetes Association. The workgroup member that developed, headed this guideline for us is actually a member of the ADA as well and one of their chief people. We agreed with the idea that the target haemoglobin A1c for persons with diabetes should be less than 7.0 irrespective of whether they have the presence or absence of CKD. Now I should mention that I agree with the person over here that raised the issue about cut off points and this is where I think education is going to be important. We can’t just reach this guideline on this level, we need to see if we can get people even below this level if we can. Drugs are helpful in doing this, insulins, often ureas and the others here are useful agents in lowering the A1c in these patients. We do want to avoid metformin if their serum creatinine is above 1.5 in men and 1.4 in women for the problems associated with that drug specifically with lactic acidosis.
Slide 15
Now I’ll just show you a brief bit of evidence to support our recommendation. We have data from type I and type II diabetes. I arbitrarily chose data for type I. These are the DCCT results. As you recall, this was a ten year study in which about 1440 patients were followed and divided into 2 groups randomly, half receiving intensive insulin therapy, the other half receiving conventional therapy. In an effort to reduce the glycated haemoglobin level in the intensively treated group to a substantially lower level than in the conventional group to see what would happen to these patients.
Slide 16
Subsequent to that study is closure in 1993. The patients were then followed till February of 2005 to see what effect non-randomisation at that point would have on these patients. You can see that over the 8 year follow up of EDIC within the first 2 years the differences in glycated haemoglobin which averaged about 2% in the DCCT were eliminated, so for the following 6 years then all of the subjects had approximately the same average level of glycated haemoglobin. Interestingly enough despite that you can see that the intensively treated group continued to have lower prevalence of elevated albumin excretion even 8 years after the end of the DCCT and if you calculate the cumulative incidence in these patients, you see that the intensively treated group had a 59% reduction in the frequency of elevated albumin excretion relative to the conventionally treated group years after the end of randomisation. So I think that it’s pretty clear at this point that improvements in glycaemic control have a beneficial effect on these patients and we should certainly seek to do that even in our patients that have chronic kidney disease.
Slide 17
I have two more slides here, one is the effect of intensive treatment. This is again from the EDIC study looking at the effect of intensive treatment versus conventional treatment on the development of cardiovascular end points. This is any first cardiovascular end point on the left. There was a 42% reduction in the patients in any of these CVD outcomes. If we looked specifically at myocardial infarction, death or stroke that was about a 57% reduction in those patients. So it’s pretty clear that not only do we reduce the frequency of kidney complications but we reduce the frequency of cardiovascular complications by reduction in the level of glycaemia in these patients and the effect of intensive therapy on outcomes, cardiovascular outcomes we think was partly mediated by reduction in the incidence of diabetic nephropathy.
Slide 18
My last slide here. There are some precautions any time you treat a person that has chronic kidney disease you need to think of some of these things. I mentioned lactic acidosis already with metformin. There are some increased risks of hypoglycaemia due to the reasons outlined here and that means that you need to consider reducing your insulin requirements, decreasing sulfonylurea doses unless you use something like glipizide which is the preferred agent the same as 2-4 nateglinide with repaglinide being the one that is preferred in the CKD patient and I’ll end by just saying that thiazolidinediones are associated with a certain amount of fluid retention something that becomes critically important in your patients with CKD and that needs to be kept in mind as well. I’ll pass the microphone over to Cathy at this point.
Slide 19
Well, good afternoon I’ll be picking up where Doctor Nelson left off. I too would like to thank the organisers for the opportunity to present this body of work to you.
Slide 20
This is just one of the studies looking at a comparison between the ACE inhibitor enalapril and telmisartan for the treatment of hypertension in a type II population with early or stage I CKD and basically with the outcome being change in GFR you can see that virtually there is no difference.
Slide 21
Also this issue about class of drugs really should be dealt with in the context of recognising that if you look at trials that include diabetic, as well as some non-diabetic populations that in fact, patients with CKD require multiple medications usually 3 or more and what we want to emphasise is that an inhibitor of the renin-angiotensin system in the form of an ACE inhibitor or angiotensin receptor blocker and a diuretic should be part of the regimen and then more likely than not these patients will need additional medications usually beta-blockers, calcium channel blockers and a variety of other types of medication that are available in most countries.
Slide 22
Guideline 4 addresses the treatment of dyslipidemia and diabetes and CKD and here our guideline states that in persons with diabetes and CKD stages 1-4 that we’re recommending an LDL cholesterol goal of less than 100 but to adopt the adult treatment panel 3 recommendation that less than 70 is a therapeutic option because of the high cardiovascular risk in this population.
Slide 23
The Tonelli study in the pravastatin pooling project did show a survival benefit in diabetes and CKD stage 1-3 but the limitation of this study is that it’s a post talk analysis but still very valuable considering the large number of patients that were included in randomised trials and cardiovascular disease.
Slide 24
We think that there are some very important research questions that come that arise from these very different conclusions. If statins benefit persons with diabetes and chronic kidney disease, at what stage is the benefit lost and why does the effect appear to change over the course of CKD? This is a key data slide from the pravastatin pooling project study showing that individuals who have both diabetes and CKD predictably have the highest rate of events shown in the placebo group but they also had the largest absolute risk reduction in response to pravastatin therapy.
Slide 25
Contrast this with the results of the 4D trial done in patients with type II diabetes on hemodiayslsis where the primary end point, which was a compositive cardiovascular events showed no benefit.
Slide 26
Guideline 5 addresses nutritional management. In this guideline we emphasise that persons with diabetes and CKD stages 1-4 should achieve a dietary protein intake of approximately the recommended daily allowance of protein, which is 0.8 g/kg body weight per day and this typically is about 10% of daily calories. Of course, this will necessitate balancing other macronutrients and as we discussed in the context of other risk factors, if there’s a higher carbohydrate intake, then more complex and low glycaemic forms are preferred and that the fat intake has to be increased, it should not be in the form of saturated fat or cholesterol but perhaps some of the healthy fats such as omega 3s and monounsaturated fats.
Slide 27
There is epidemiologic evidence of harm from high protein diets in the target population. There are now several studies in the literature, this is just one selected example from the NHANE study showing that this is people with type II diabetes who also had hypertension and this is the most representative group since most people with type II diabetes do have hypertension. There’s a 3 and a half fold increased risk of developing microalbuminuria, if protein intake is 20% versus the lowest quintile of intake of about 12%.
Slide 28
There is clinical trial evidence of benefit. This is reducing from the usual intake, which is an excess of the recommended daily allowance really not to a low protein level but actually down to the recommended daily allowance that rates of end-stage renal disease and deaths were substantially reduced. This was in a population of type I diabetes with CKD stage 2 and these people were on good drug therapy. Specifically 90% were receiving ACE inhibitors and had good control of hypertension. So that diet in addition to the drug therapies clearly adds benefit.
Slide 29
This is an update to the nutrition recommendations from the NKF K/DOQI hypertension guidelines. The only thing that we changed is we specified that in diabetes that even at CKD stages 1-2 should be kept at the 0.8 level or at the RDA level based on this evidence that I showed you. The previous hypertension guidelines had recommended a slightly higher intake based on the DASH diet. One thing that we pointed out in these guidelines is that the DASH protein recommendations emphasise vegetable sources of protein and in fact, epidemiologic studies such as INTERMAP have shown no benefit of higher animal protein intake on blood pressure and that the benefit is confined to vegetable sources of protein and at least small studies suggest that the vegetable proteins are renal kidney sparing, so that if a higher protein intake is, if a patient eats a higher protein intake, then vegetable sources are preferred whether or not there’s diabetes and in diabetes we believe that the 0.8 level is most appropriate.
Slide 30
I’d now like to turn briefly to a discussion of the clinical practice recommendations and to remind you these were areas where we thought there were very important issues in clinical practice but where the evidence base was insufficient to make these guideline statements.
Slide 31
And it is based on some data like this from RENAAL that showed that in type II diabetics with macroalbuminuria and hypertension who were treated with losartan that a reduction in albuminuria, for example, about a 50% reduction at 6 months correlated almost 1:1 with about a 50% reduction in end-stage renal disease events and then similar data were reported about the same time for cardiovascular events, particularly heart failure but also pretty good data from cardiovascular end points in general.
Slide 32
This is a very strong hypothesis but an alternate explanation could be that albuminuria goes down more easily in patients who aren’t as sick. So we would like to recommend in a research recommendation that actually albuminuria be the target of treatment and then if it can be shown that albuminuria reduction targeted by specific therapy improves outcomes, then that would be stronger than a clinical practice recommendation.
Slide 33
The diabetes guideline is unique in that we recommend addressing multiple risk factors. In the real world of the clinic, as Doctor De Luca told us earlier, we actually have to incorporate management of all of these risk factors into the same patient and we want to emphasise that addressing multiple risk factors is key to optimising outcomes and that part of this should also be to recommend that patients try to maintain a normal body weight considering the emerging literature on the role of obesity in both initiation and progression of kidney diseases including diabetic kidney disease.
Slide 34
There’s really only one study that we found in the specific target population, this was the Steno trial of type II diabetics with microalbuminuria where they targeted glycaemia, lipids and blood pressure in an integrated program and in addition to drug therapies they emphasised life-style modification including diet, smoking cessation, aspirin, multiple vitamins and weight control.
Slide 35
But in any case the risk factors were better in the intensively managed patients and even though they didn’t achieve all the goals, incrementally these risk factor modifications produced large improvements in outcomes, about a 50% reduction in cardiovascular events.
Slide 36
And about 60% reductions in progression of nephropathy, retinopathy, autonomic neuropathy we’re uncertain about what happened with peripheral neuropathy but by and large these patients clearly appeared to benefit from the perspective of very important complications including heart, kidney and eye outcomes.
Slide 37
Finally, as Doctor Nelson talked about earlier there are populations who have a particularly high risk and while the treatment of these individuals follows that of the general population we want to highlight that high risk populations should be a group that we focus on for both screening and intervention. As Doctor Eknoyan mentioned earlier, there are global implications of diabetes and kidney disease which is the most common cause of chronic kidney disease through the developed and developing world and in many countries end-stage renal disease programs are virtually non-existent and so prevention really is the most effective means to deal with this problem and here population based intervention such as those that have been used, for example, in countries with high prevalence of HIV maybe a reasonable approach to trying to deal with this in some of these areas.
Slide 38
Also for the first time in the KDOQI guidelines we dealt with the issue of pregnancy. Because with the onset of type II diabetes occurring in younger and younger people, young women have now the opportunity to develop diabetic kidney disease within their child bearing years and we thought that this was a very important special population and that specialists in kidney disease, as well as pregnancy, should co-manage these patients.
While we recognise how important renin-angiotensin system inhibition is for diabetic kidney disease, we also want to warn people about the risks of these drugs particularly for foetal malformations and that at the earliest indication of pregnancy these drugs should be stopped and probably even when pregnancy is planned.
We also emphasise using insulin to control the hyperglycaemia in women with diabetes and CKD. There are some studies looking at oral agents in diabetic women but none of them includes CKD patients and because there are some drug precautions the workgroup felt that the more conservative approach and use of insulin at this time should be the recommended drug therapy.
Slide 39
Finally, in order to achieve the management of multiple risk factors, particularly those that, particularly with the importance of life-style modification because if we’re really honest about diabetic kidney disease, this is largely driven by life-style and to address life-style is really at the core of dealing with the epidemic along with good medical therapies that we think self-management strategies that help improve health behaviours are also important in an area that we should target for future research, particularly the importance of monitoring and treatment of glycaemia in this diabetes and CKD population, nutrition, exercise and strategies that will help people to adhere to these complex medical regimens.
Slide 40
I’ll just conclude with this slide just emphasising that we think life-style modification even in the patients who have developed kidney disease is still beneficial and as for other health conditions that the common good of regular physical exercise, maintaining a healthy body weight and moderation of alcohol intake is important to improve the care of the diabetic patient with CKD, as well as for the general population. Thank you.
Slide 41
Chairman: Thank you very much. I think now these two speeches are open for discussion. Can you join her Doctor Nelson?
Question: David Hansel U.K. You continue to promote the guidelines on stopping Metformin beyond a certain level of creatinine, I’d like to question that. I don’t think you actually quoted the evidence level for that on your slide. I’d like to question it. One you quoted it around a specific value for creatinine. That’s what we used to do. I would have thought with new guidelines you’d think about relating that to eGFR, that’s the first point. My second point on that would be that again the recommendations for this came out before we knew that metformin was actually beneficial in reducing mortality in obese diabetics and the evidence for this is anecdotal and certainly never occurred in lower dose metformin and I just feel you haven’t had a chance to maybe reconsider these guidelines.
Prof Tuttle: This is something that we did discuss at length and as you can imagine, the scope of this evidence review was very large and we felt that we, at least in the United States that we had to stick with the FDA labelling on the drug, so that’s why we didn’t convert the serum creatinine to eGFRs because we actually really don’t know what the level is based on what you’ve said. The opinion of the workgroup was that even though the complication is rare, it is potentially fatal and there are many other drug options and while there is some evidence from the UKPDS on the value of metformin in obese patients in CVD, the evidence for that wasn’t considered really strong and there’s emerging evidence, for example, on TZDs going the same direction and we think that that area of specific therapies that are organ protective in the treatment of hyperglycaemia, the analogy of renin-angiotensin system inhibition in hypertension now applied to specific drugs used for the treatment of hyperglycaemia that maybe organ protective is basically a nascent field and we weren’t really going to comment on that until we have some studies that show that TZDs or metformin and so forth are both safe and effective in that regard.
Prof Nelson : If I could just mention one specific comment with regards to the metformin used too that wasn’t actually a guideline statement for metformin that was something that was part of the write up that went with it and we actually produced a number of research recommendations out of this that address some of these very issues. The guideline itself was specifically with regards to the A1c level.
Prof Tuttle: Yes exactly, that was part of the rationale and that’s the only guideline, if you’ll notice where we didn’t give a specific drug recommendation and I gave you the reasons why.
Question: Thanks very much for giving us a peak at this work in progress. You’ve really done a wonderful job. But I do have a question about the way in which you present the target blood pressure. You say less than 130/80. You’ve heard previous speakers say that because of the dispersion effect, this means that the mean blood pressure in the population who achieves this target will have blood pressures as low as 110/60. Now this may not be a problem but – looking at a very large observational database in diabetics it’s now clearly shown I believe that there’s a J curve, the J curve has been re-invented in a way in diabetics and I’m very concerned that instead of giving a ‘less than’ it would have been preferable to give ranges to give a sense to the treating physician that there is a lower level that is associated with excess mortality in diabetics.
Prof Tuttle: That also is in the rational section on the blood pressure guideline. We ended up leaving it as a ‘less than’ recommendation because we actually don’t know what the floor is. So it’s kind of like the metformin question we did address that, there is a whole section on what should be achieved levels of blood pressure and we ended up concluding, based on some of the evidence you’ve talked about and others, that less than 110 systolic is not recommended but we didn’t feel it was strong enough to put it up there in the statement itself but when you see the guideline and you read the section on targets for blood pressure control, it says that we think that there’s evidence of harm with a systolic below 110. Then to get back to the question earlier about targets and people getting in the right range, I’ll just give you my opinion on it. I understand exactly what you’re talking about. I think that it’s an educational issue and we have a section in our guidelines called implementation issues and I think that is at least as a practical manner where we can begin to address those things that what we really mean is a range of 110-130 so what we want to see is the systolic blood pressure in the 110s and 120s and I think the guidelines have to be simple almost sound bite in order to get the message out there but then we need to follow up with the detail of what we mean by that and I think that’s where we have a large role in the educational process.
Question: It depends on what outcome you want to prevent, for strokes it goes all the way down to 90 but for acute myocardial infarction in national data, the excess mortality began at a diastolic around 75, so it’s not this extreme lowering, it’s the fact that well within your thresholds you maybe looking at the potential, I’m not saying it’s real because all we have is an observational study we don’t have a randomised controlled trial.
Prof Tuttle: And we have to look at data from different sources and come up with the most reasonable integration of that data.
Chairman: Thank you very much Cathy and Rob. I’m afraid we’re over time and they need to get this room ready. Doctor Will has a question that he wants to pose to you personally and for the rest of you I’d like to thank you together with my Chairman for coming, participating and the questions that were asked and the lessons of where we can be with global guidelines, how the problems of implementing guidelines that were presented by Doctor De Nicola and then finally, new guidelines that are coming in diabetes. Thank you very much.