Slide 1

Thank you very much. First of all I’d like to thank the organisers for the opportunity to share the paediatric experience. What I’d like to do is address the public health perspective and perhaps convince you at the end of this session that we need to address chronic kidney disease and prevention of chronic kidney disease in childhood.

Slide 2

So the objective that I’d like to fulfill is to review the risk factors for CKD originating in pregnancy and in childhood and then to go on to talk about the increased prematurity rate as one of the major factors for reduced nephron endowment and subsequent risk for CKD later in life. Furthermore I would like to address the ramifications of the epidemic of childhood obesity.

Slide 3

Unfortunately, South Western Ontario exhibits the highest prevalence of obesity in Canada that is associated with a steady increase of CKD. This slide demonstrate the increase of CKD patients in the adult population in the US. The experience in Canada is quite similar.

Slide 4

Generally, children are not perceived as being at risk for CKD. As was just shown, we witness a massive increase of CKD and while it was said that there has recently been some tailing off in the cases of new ESRD, I think we need to brace for much more in the future and I’ll tell you why. I will not be talking about genetic disposition because that’s not modifiable, however, I want to focus on low birth weight and prematurity and I want to talk about obesity during pregnancy and pre-pregnancy at conception. Then move on to the ever growing epidemic of childhood obesity. In our area we now have 1 in 4 children who are overweight or obese and its major association with hypertension which is a very well-known risk factor for CKD.

Slide 5

This slide has kindly been provided by Professor Ritz and elegantly summarizes the Barker hypothesis. It is the concept of having the same genes but a plasticity and the exposure to the conditions during pregnancy that possibly affect epigenetic modification that alter the risk for future CKD with the same genes. The Barker hypothesis was developed as a consequence of observations after the Battle of Arnheim and of the subsequent Dutch famine. Children who were born after the winter of 1044/45 and were exposed to the famine during the third trimester were at major risk for CKD. This lead to the idea those in early life humans are plastic, that they are affected by their intrauterine environment and that especially when they’re hit during that critical phase of rapid cell division that then that could have ramifications for the rest of their life.

Slide 6

The concept is based on the growing evidence that the nephron number at birth or even at 36 weeks is final and that no additional nephron formation takes place after birth. This was shown in a number of studies and I’m just showing one as an example that showed that when you have a low birth weight, you have a lower number of nephrons. However, these fewer nephrons become larger, resulting in a similar filtration surface whereas but the total number of filters is much less.

Slide 7

In that context, we have to be concerned about the the ever growing rate of prematurity. As you can see from this slide, it has risen from 6.7% in 1967 to 11.9% in 2001, In some places, such as Minneapolis for instance, we witness a stunning prematurity rate of 16.9%. Paediatric inpatient admissions are declining, however, there’s an ever increasing demand for new basinets for young children, especially for infants born with a very low birth weight.

Slide 8

What is also contributing is that we actually have been getting much better in achieving survival of these prematurely born infants. We face an increasing rate of prematurity on the one hand and improved survival of children born with very low birth weight on the other hand. This may result in a young generation where possibly up to 15% are at risk for future CKD. 

Slide 9

This is frequently cited slide from Dr. Jean Pierre Guignard from Lousanne, Switzerland, who has been making that point for some time but we have to really think about the fact that nephrogenesis stops at delivery and with that high rate of premature delivery you will result in a much lower nephron endowment. Thereafter it’s only really hyperplasia and this was subsequently addressed in a number of studies most noticeably in Julie Ingelfinger’s work.

Slide 10

When investigating prematurely born infants at an older age, these children exhibit a higher uric acid level and a higher blood pressure.

Slide 11

These data are from our centre. We analyzed 19,171 first singleton deliveries from one of the sites of Children’s Hospital, LHSC, where there are no high risk deliveries. In London, Ontario, we have approximately 6,400 births annually, and half of these take place at the Victoria Campus where I practice. This slide summarizes the average pre-pregnancy BMI over a 10 year period. The average BMI of the mothers at conception was 24.3 in 1997 and rose to 25.3 in 2007. Maternal age was not a factor, the median age remained unchanged at 28 years of life. You can see that over the last decade only 64% of the mothers had a normal BMI at conception. Increased maternal pre-pregnancy unfortunately negatively impacts the offspring.

Slide 12

These children are exposed to intrauterine hyperglycaemia and insulin resistance which results in an increased risk for congenital abnormalities. For instance here is the combined risk for cardiac and renal abnormalities at birth and you can see the 10-fold increase with regards to the body mass index at conception.

Slide 13

This very large Iranian study confirms those data on over 90,000 deliveries. It’s quite clearly shown that the risk of having a low birth weight for an obese mother is substantially increased. The same applies for preterm delivery and the same applies for caesarean section. Unfortunately, caesarean section is a risk factor for admission to the neonatal intensive care unit and some morbidity that also may result in an increased prevalence of CKD later on in life.

Slide 14

We could actually confirm these associations in our own database and you can see the risk for admission to the neonatal intensive care unit is associated with the BMI at conception. I will now move on from the pregnancy to older childhood to illustrate that these premature babies are at risk for CKD because of a low nephron endowment.

Slide 15

Moreover, they are also at risk for developing obesity later in life, as shown nicely here in Barker’s work again.

Slide 16

It has been known for some time that obesity is a major risk factor for CKD. This is from the Kaiser Permanente Study, conducted in California. 320.000 adult members were followed consecutively between 1964 and 1985. When revisiting the data in 2006, there were a stunning 1,500 patients who were on dialysis. When adjusting for blood pressure and for diabetes which are known risk factors for the development of CKD, and then taking a normal BMI at the early adulthood of between 18.5 and 25 as normal and a relative risk of 1, you can see that very clear relationship between BMI in young adulthood and risk for CKD. As the BMI goes up between 25 and 30 the risk for CKD doubles, between 30 and 35 the risk increases 3.5 fold etc. So the higher the baseline BMI of a young adult, the greater the risk for a CKD even when adjusting for blood pressure and diabetes mellitus.

Slide 17

In that view we were quite concerned when we looked at the Body Mass Indext z-score of some 5,000 paediatric nephrology patients seen in our centre. We noticed that their BMI z-score reflect the normal development of the BMI based on the US data, however, these children seen in our clinic were significantly overweight compared to the American reference population. We wondered whether that’s a feature of paediatric nephrology patients. Most patients seen in our clinic clearly are at risk for future CKD. Common diagnoses are reflux nephropathy and other congenital abnormalities of the urinary tract, IgA nephropathy, hypertension etc.

Slide 18

We have always been hampered by the fact that we did not have data from another country. We did not know whether this was a feature of paediatric nephrology patients per se, or whether the particularly poor Canadian diet was a special risk factor. Fortunately, there was actually one study that I conducted some time ago that allowed the inclusion of German and Canadian children. We analyzed the BMI z-scores of these children, and there was a very marked difference. We do not believe that the high BMI z-scores were related to the CKD but rather just a local variability. This points to a higher risk for CKD of Canadian children.

Slide 19

This is just the record of our own clinic, including 401 children referred for hypertension. I don’t think you need to be a statistician to see that not only are there is an increasing prevalence, but also that the BMI scores of these children are quite dramatic. [Pointing to one of the highest points in the graph]: This is a patient I’m following myself who’s a 10 year old–boy who has already metabolic syndrome, diabetes, hypertension, fatty liver degeneration and I wonder what will his life expectancy be?

Slide 20

This is from an unpublished report of our Province, Ontario, of the prevalence of overweight children and obese children. The second data row reflects our own catchment area. Out of the 500,000 that we have in our catchment area, some 114,000 children are overweight or obese.

Slide 21

This rate of childhood obesity is posing a huge public health risk for the future. I want to point out a few more details and finish off with this cross-sectional study that we’re currently conducting in our hospital. Basically every child that goes through our hospital gets height weight and blood pressure done and they are plotted on the appropriate American reference interval datasheets and because height weight and blood pressure are age dependent they need to be plotted as z-scores and we analysed this with the methods given here.

Slide 22

And this slide summarises what we found and that is that our children have on average 0.4 standard deviations higher BMI than the American children which are already believed to be the heaviest children in the world. That poses a huge risk for future CKD.

Slide 23

This is the summary on 7863 children. The height weight score was higher than the US. This was not surprising in view of the lower prevalence of Hispanics. However, the data on weight and BMI z-scores were astounding. Worryingly this was associated with a higher blood pressure z-score. When analyzing this cross-sectional cohort by birth year, we noticed the unfavourable development of the BMI z-score within this one generation: The current adolescents are much leaner than the younger children.

Slide 24

Children born after 2005 were excluded because you have to have at least a 2 year old to have reference data for the BMI. So we’re seeing this childhood obesity epidemic even in 1 generation and BMI z-scores are really age independent so anything above 0 would be heavier than the average and you can see that very clear and very concerning development. I dare predict that this obesity wave is only the tip of the iceberg and will result in future increase of CKD.

Slide 25

Let me spend the last few minutes on the Barker hypothesis. I assume that the audience is quite interested to find out whether we could actually confirm the Barker hypothesis. Interestingly, studies confirming the Barker hypothesis in childhood remain elusive. I show you data on 1,621 children and there was interestingly a close correlation between the systolic blood pressure and the BMI z-score which is known from adults but there was also a relationship with the pre-pregnancy BMI suggesting that maybe some environmental and some genetic factors may play a role but in view of the increase of pre-pregnancy BMI over the short period of time it cannot only be genetic. There was a very weak relationship with the gestation age but interestingly we found no correlation whatsoever with the birth weight and even when we used the multivariate analysis in adjusting for the BMI z-score we found no correlation with the birth weight. We also divided into adolescents and younger children and found no correlation with the birth weight. Only diastolic blood pressure correlated very weakly with the end gestational age which would be in keeping with the Barker hypothesis. So what are the conclusions from this? We suggest that the modifiable risk factors such as BMI are maybe a much more important risk factor for future CKD than the non-modifiable ones such as the low birth weight.

Slide 26

Really pointing to the fact of how important a normal blood pressure is for future kidney disease, I’m just showing you this paper published in JASN. I am showing the data for the females only, but the same applies for the males. If you have an optimal blood pressure below the 50th percentile and take that as a risk of 1, your risk for CKD increases as the blood pressure increases. Keeping in mind the significantly elevated blood pressure z-scores of the children I just showed you, I become very worried about the increased risk of CKD in this sizeable hospital patient cohort, based on blood pressure alone.

Slide 27

So to summarise, I think we need to brace for a much higher prevalence of CKD in the young generation. Prematurity is a clear risk factor for future CKD and is increasing. Risk factors for perinatal morbidity are associated with maternal obesity. I think it is important that we start to target that in view of the 35% of overweight mothers. Childhood obesity is clearly associated with risk for CKD. The children attending our hospital were taller and more obese than the American children and it is therefore conceivable that these children are at great risk for future CKD. It also is very interesting that the Barker hypothesis does not appear to be applicable for these children and that the modifiable risk factor of BMI is a more important risk factor. And as that is potentially modifiable, I think one really needs to stress the need for targeting healthier weights in our young generation. With that I’d like to finish and it would be very interesting to hear your questions. Thank you.

Slide 28

Chairman; Thank you Doctor Filler for this very interesting but rather chilling information on the current epidemic. May I come back to one statement you repeatedly made, this must be environmental it is not genetic. Perhaps it is a bit more subtle and I’d like to illustrate it with the experience in Leipzig. The communists kept excellent statistics, it was the only thing that they were good at about children’s body weight throughout the childhood. What they now find is that there is an explosive increase of body weight and type 2 diabetes but the median has not changed since the wall came down. But it starts at the 70th percentile indicating that there is a subset of individuals who are susceptible to these environmental changes of excess food and less physical activity. So there is an interplay apparently between genetics and environment.

Dr. Filler: Well, I would agree and there’s probably also epigenetics and we don’t really know what the ramifications of intrauterine exposure to hyperglycaemia is and that may switch on some thrifty genes and it’s probably important that we find out how to switch them off again.

Chairman: So often in medicine explanations are never simple.

Dr. Filler: Yes.

Chairman: Maybe I can have this question over here.

Question: Thank you for this very nice presentation. I have two questions. My first question, do you have albuminuria measured in your cohort and do you see any correlation between albuminuria in these very young children in association with obesity and with prematurity? The second question is about how to approach and measure nephron number to test this hypothesis of foetal programming of CKD. It would be very nice to approach that and do you see any way this could be approached at least for clinical research?

Dr. Filler: The two parts of the questions, let me answer the latter part first which is about how could we measure nephron endowment. The gold standard would be a PAH clearance which is not feasible and not practical. The issue is if a child is born with a low nephron endowment, within 1 year it will hyperfiltrate and may actually exhibit a normal GFR although that’s done with a much lower number of nephrons. What we’re currently doing in collaboration with McGill university in Montreal as we’re looking prospectively at whether we can use cystatin C, which is a marker that does not cross the placenta, in babies that are born at risk with renal dysplasia or prematurity and whether we could then correlate that with renal size and so on as a surrogate of that. I think this is still very premature work, this is hot of the press but we actually do find that very small amounts of cystatin C cross the placenta whereas beta-trace protein does not. So I think there’s work that needs to be done and ideally we need a surrogate for the effective renal plasma flow. With regards to the microalbuminuria that was not part of this study and I think I would agree it’s  a very useful tool but obviously in our health care system you have to defend every single thing that you do, that would really dramatically increase the cost. Our target is to have 10,000 children and we are almost there, so the costs for 10.000 determinatinos of microalbuminuria would be prohibitive.

Chairman: Thank you so much Doctor Filler you convinced us that cross talk, cross-departmental barriers are intellectually stimulating. This brings this session to a close. Thank you for your participation, thank you for your questions.