Nephrology Dialysis Transplantation 2007 22(Supplement 7):vii165-vii175; doi:10.1093/ndt/gfm336
© The Author [2007]. Published by Oxford University Press on behalf of ERA-EDTA. All rights reserved. For Permissions, please email: journals.permissions@oxfordjournals.org
Demography and management of childhood established renal failure in the UK (Chapter 13)
Malcolm Lewis1,
Joanne Shaw1,
Chris Reid2,
Jonathan Evans3,
Nicholas Webb1 and
Kate Verrier-Jones4
1Central Manchester & Manchester Childrens University Hospitals NHS Trust, 2Guys and St Thomas's; NHS Foundation Trust, 3Nottingham University Hospitals NHS Trust and 4University Hospital of South Wales NHS Trust
Correspondence and offprint requests to: Dr Malcolm A Lewis, Renal Office, Royal Manchester Children's; Hospital, Hospital Road, Pendlebury, Manchester M27 4HA, UK. Email: malcolm.lewis{at}cmmc.nhs.uk
 |
Abstract
|
|---|
The incidence and prevalence of ERF in children in the UK are
relatively static at 8.0 and 47.7 per million population under
the age of 15 years, respectively.
The prevalence of ERF in children from the South Asian community is almost three times that of the White population whilst the incidence is over three times that of the White population and similar to the increase seen in the adult population. The high incidence and prevalence are related to the high incidence of inherited diseases which cause ERF in the South Asian community.
ERF in children is more common in males than females (male to female ratio 1.54:1). This is due to a preponderance of males with renal dysplasia and obstructive uropathy causing ERF. For the South Asian patients, the gender ratio is 1:1 as the inherited diseases are mainly autosomal recessive.
Renal dysplasia is the single most common cause of ERF in childhood, followed closely by glomerular disorders and then obstructive uropathy.
The majority of prevalent paediatric ERF patients (76%) have a renal allograft. Of these, 28% are from living donations.
The proportion of patients from ethnic minority groups with a functioning allograft is significantly smaller than that in the White population (P < 0.0001). Despite this, the rate of living related donation is no higher in the ethnic minority population.
In prevalent patients PD is twice as commonly used as HD with the majority managed with automated PD. For patients at one year from starting RRT, 49% are on PD, 10% on HD and 41% have a transplant.
Keywords: aetiology; chronic kidney disease; demography; dialysis; end stage renal disease; epidemiology; ERF; established renal failure; ethnicity; incidence; management; prevalence; transplantation
|
Introduction
|
|---|
Knowledge of the demography of the ERF population is important
both for the planning of service provision and for the development
of preventative treatment programmes. This article covers the
demography of ERF in children in the UK and their current modality
of ERF treatment.
|
Paediatric ERF population
|
|---|
The paediatric arm of the Renal Registry currently holds data
on some 1800 patients who had ERF in childhood. A number of
these patients have died and many have been transferred to adult
units. The population of ERF patients being treated in paediatric
units on 1st April 2005 stood at 768. This is a small fall on
the number from 2004. The reasons for this probably lie with
incomplete data returns from three units, together with variability
of the population with the transfer of teenage patients to adult
units.
Table 13.1 shows the prevalent population by gender and ethnicity together with the numbers who were under 18 years of age and 15 years of age on 1st April 2005. As in previous Reports, there are about 20 young people over the age of 18 years remaining in paediatric units. These patients are transferred between the age of 18 and 20 years. There are no patients over the age of 20 years in the current cohort. Reasons for delayed transfer include the management of specific paediatric comorbidities and concerns over growth, development and education. The distribution of the population with regard to gender and ethnicity was unchanged from previous reports. There remains a predominance of males and just over 17% come from ethnic minority backgrounds.
Figure 13.1 shows the size of the population under the age of
15 years from 1986 to 2005. The apparent growth in this population
seen in 2004 has not been maintained but this will be due to
some missing data from units with incomplete submissions together
with some variability year on year in presentation rates. The
overall trend has been that of a slowing of the initially sharp
increase in the population. This is supported by the data on
incidence and prevalence presented subsequently.
The age distribution of the population over a number of years
is shown in
Tables 13.2 and
13.3. The former gives the customary
divisions by age and the latter shows the population divided
into four year age bands for ease of comparison. Though there
is year to year variability, the numbers have been fairly static
of late, clearly showing a cessation of the rapid population
growth seen after paediatric ERF treatment became available.
Figure 13.2 shows the data in
Table 13.3 graphically and clearly
shows that over recent years there has been no significant change
in the age distribution of the population.
The gender distribution of the paediatric ERF population is
shown in
Figure 13.3. Throughout the age range, males predominate
but there is a steady rise in the proportion of females in the
population with increasing age.
Of the ethnic minority patients, the vast majority are of South
Asian origin. The age and gender distributions of this cohort
are somewhat different to that of the White population. This
is secondary to the different causes of ERF in the South Asian
community and is dealt with in detail subsequently.
Table 13.4 shows the age distribution of the population according to ethnicity.
Although the difference in age distribution between the White
and ethnic minority populations does not reach statistical significance
the pattern is demonstrated in
Figure 13.4.
The difference in gender distribution between the White and
South Asian paediatric ERF populations is shown in
Figure 13.5 which contrasts the proportion of the population in each age
group who are male. In the under-the-age of 4-years group, 77%
of White patients group are male. Thereafter, there is a fall
in the proportion of males in the White population, with an
increase in the proportion of males in the South Asian population,
until in the young adults, both lie between 55 and 60%. There
were only five Asian patients under age 4 so these have been
removed from the graph.
|
Prevalence and take-on rate
|
|---|
Data on the UK population divided according to age and ethnic
background was taken from the Office for National Statistics’
website (
www.statistics.gov.uk). Data for this report are based
upon current population estimates which themselves are extrapolated
from the United Kingdom Census of 2001.
Table 13.5 shows the
prevalence of ERF per million childhood population for each
age group. These figures have changed little since previous
Reports [
1–6] as one might expect from the stable population
numbers.
Figure 13.6 shows this graphically, clearly demonstrating
the steady rise in prevalence with patient age until the fall
in the over 16-year-old group, secondary to transfer to adult
units. The figures for prevalence of ERF in the UK are comparable
with those presented in the USRDS and ANZDATA registries [
7,
8].
Whilst there is no mention of ethnicity in the most recent ANZDATA
report the USRDS report does give an ethnic breakdown but not
one which is specific to the paediatric age range. As the majority
of the patients are adults and there are varying rates of glomerulonephritis,
hypertensive and diabetic nephropathies amongst the different
adult ethnic groups, it is impossible to extrapolate this published
data to look at prevalence and ethnicity in children. As with
previous reports from the UK paediatric registry the prevalence
of ERF is much higher in the South Asian community, being almost
three times that of the White population, whilst the prevalence
of ERF in the Black population and those of other ethnic origins
is a little below that of the White community. This is demonstrated
in
Figure 13.7. The reasons for this distribution lie in the
varying causes of ERF with ethnicity and are discussed subsequently.
The take on rates of patients starting RRT has been assessed
looking at a 5-year period to even out the peaks and troughs
seen with annual data collection when relatively small numbers
are being analysed. This is demonstrated well by the undulant
picture shown by the ANZDATA incidence chart. Looking at take
on rate as a mean of consecutive 5-year periods, there is clearly
little change in the incidence of ERF in children. Overall,
the incidence of ERF in children in the UK is very similar to
that of the Australian, New Zealand and US cohorts. These data
are shown in 4-year age bands in
Table 13.6 and graphically
in
Figure 13.8. There is a nadir of presentation of ERF in the
4–8-year-old group following a peak in the first 4 years
of life with the presentation of many children with obstructive
uropathy and renal dysplasia. Following this there is a steady
rise in incidence as the number of patients with glomerular
diseases increases. As with the prevalence data, the take on
rate of new patients with ERF in the South Asian community far
outweighs that of the White community with an incidence per
million childhood population 3.7 times that of the White population
(
Figure 13.9). This incidence figure will, over a number of
years, lead to the proportion of the total population of children
with ERF coming from the South Asian community rising still
further. The distribution of the ethnic minority population
(and consequently the ethnic minority children with ERF) around
the UK is not evenly spread [
6]. This has significant implications
for resource management.
|
Causes of ERF in children
|
|---|
The causes of ERF have been analysed by looking at a total of
913 incident patients presenting with ERF before the age of
16 years, since the inception of the registry in 1996, for whom
a primary diagnosis was stated. Diagnoses have been grouped
into 12 bands. These are shown in
Table 13.7 with a further
breakdown of each of the groupings in
Tables 13.8–13.17
.
Renal dysplasia remains the single most common diagnostic group
comprising almost a quarter of the total cohort. There is a
male predominance in patients with renal dysplasia, and this
together with the male contingent with obstructive uropathy
from posterior urethral valves, accounts for the overall gender
distribution of the paediatric ERF population. The gender distribution
of each diagnostic group is shown in
Figure 13.10. Although
there is no explanation for this, a high incidence of renal
dysplasia in males has not only been noted in the UK registry
reports but also in the NAPRTCS report [
9]. Glomerular disease
follows closely behind renal dysplasia, accounting for 22% of
patients. Obstructive uropathy is the third most common cause
accounting for 15%.
The nature and distribution of the diseases causing ERF in childhood
have not changed significantly over the years that reports have
been generated by the Registry. However, this will be due to
the fact that a complete and expanding cohort has been used
to look at this distribution. Certainly the information provided
by ANZDATA suggests a similar distribution of causes if one
excludes the 15–20-year-age group, which appears to be
a complete cohort in ANZDATA and therefore dramatically expands
the band of patients with glomerulonephritides. The USRDS data
available does not give a specific diagnostic breakdown for
children. The NAPRTCS report is more difficult to interpret
as the analysis of transplant and dialysis patients is separate.
Certainly it appears that for White and Hispanic patients, renal
dysplasia leads in conjunction with obstructive uropathy. Unlike
the UK data, glomerular diseases causing ERF appear less frequent
in this population. This however, is offset by the high incidence
of glomerular diseases causing ERF in the Black population.
Differences in the patterns of primary pathology with ethnicity
in the UK population are dealt with subsequently.
To investigate whether there has been any change in the pattern of primary pathology causing ERF in children over the period the Registry has been collecting data, the distribution of diagnoses have been compared within the 12 main classifications in those patients presenting between 1996 (when data collection began) and 1999, with those patients presenting between 2002 and 2005. These data are shown in Table 13.18. There is no significant difference in the patterns of disease. The incidence of obstructive uropathy has fallen slightly and time will tell whether this is an ongoing trend. Reflux nephropathy has fallen and there has been a parallel rise in the incidence of ERF of uncertain aetiology. Knowing the difficulty in categorizing patients who present with small kidneys, either in or near ERF, it is possible that this simply represents variability in classification. The incidence of tubulo–interstitial diseases has risen. Again, only time will tell whether this is a true trend, however, it is something that may be expected given the rising South Asian population and the increased frequency of these pathologies in this ethnic group.
As alluded to above and published in previous reports from the
Registry, there is a significant difference in the pattern of
diseases causing ERF in different ethnic groups. This is shown
in
Table 13.19. Whilst for the White population renal dysplasia
predominates followed closely by glomerular diseases, in the
South Asian population glomerular diseases predominate with
a lower incidence of renal dysplasia. Tubulo–interstitial
disorders, metabolic diseases and congenital nephrotic syndrome
are much more common in the South Asian community. The overall
difference in the distribution of diseases between the White
and South Asian populations is highly significant (chi-square
= 40.2,
P < 0.0001). Interpretation of the distribution of
diseases in the Black population and those from other ethnic
backgrounds is more difficult because of the small numbers.
Black patients with glomerular diseases contribute over 50%
of the cohort and renal dysplasia is much less common with only
occasional cases of other disorders appearing. Certainly data
from NAPRTCS would suggest that this is not an unrepresentative
pattern of disease.
Much of the difference between the patterns of disease in the
South Asian patients compared to the White cohort can be explained
by the high incidence of autosomal recessive inherited disorders
in this population.
Table 13.20 shows the pattern of inheritance
of the primary cause of ERF in 913 patients presenting after
1996 and starting ERF before the age of 16 years for whom both
details of primary diagnosis and ethnicity were available. Overall,
190 patients (20.8%) had diseases with a clear inheritance link,
showing the major contribution of genetic problems to childhood
ERF. Of these, the vast majority (90.5%) were autosomal recessive
diseases with just a small number of dominant, sex linked and
mitochondrial disorders. These of course do not include patients
with diseases that probably do have a strong genetic component
that has not yet been clearly defined, such as isolated renal
dysplasia. The proportion of each ethnic group with inherited
disease as a cause of ERF is shown in
Figure 13.11. This clearly
shows the excess of inherited disease both in those of South
Asian origin and in those of ‘Other’ origin. Consanguineous
marriage is more common in both of these groups compared with
the White population. Although the small numbers of patients
in the ‘Other’ group make valid statistical analysis
difficult, the increased proportion of inherited disease in
the South Asian group compared with the White population is
very significant (
P < 0.0001, Fisher's exact test).
The age distribution of the paediatric ERF population is determined
by both the survival of patients and the age of presentation
with ERF. This in turn is often dependent upon the aetiology
of ERF. The effect of diagnosis upon the population age distribution
is shown in
Figures 13.12,
13.13 and
13.14 subsequently. For
each of these figures panel A shows the percentage of patients
in a designated diagnostic group presenting in each age group,
whilst panel B shows the percentage of patients in each age
group belonging to that diagnostic group. Thus, for patients
with renal dysplasia, 32% present with ERF in the first 4 years
of life and 32% present between the ages of 12 and 16 years
whilst the remaining third present in the intervening 8 years.
View larger version (49K):
[in this window]
[in a new window]
[Download PowerPoint slide]
|
Fig. 13.12 (A). Percentage of incident patients with renal dysplasia, obstructive uropathy, glomerular diseases and reflux nephropathy presenting in each age band. (B) Percentage of patients presenting in each age band with renal dysplasia.
|
|
View larger version (44K):
[in this window]
[in a new window]
[Download PowerPoint slide]
|
Fig. 13.13 (A). Percentage of incident patients with tubulo–interstitial diseases, metabolic diseases, congenital nephrosis and polycystic disease by age band. (B) Percentage of incident patients by age band with tubulo–interstitial or metabolic disease, congenital nephrosis or polycystic kidneys as the cause.
|
|
View larger version (44K):
[in this window]
[in a new window]
[Download PowerPoint slide]
|
Fig. 13.14 (A). Percentage of patients with renovascular diseases, malignant diseases, drug nephrotoxicity and uncertain aetiology presenting in each age band. (B) Percentage of incident patients in each age band with renovascular or malignant disease, drug nephrotoxicity as the cause or uncertain aetiology.
|
|
The proportion of patients with renal dysplasia as a cause of
ERF in each age group, account for 34% of those in the first
4 years of life but only 20% of those between the ages of 12
and 16 years because other causes of ERF have become more frequent
in this latter age group. The pattern for obstructive uropathy
is virtually identical to that for renal dysplasia. As with
renal dysplasia, virtually all patients will have been born
with their problem. The distributions of both these groups show
the combined effect of the severity of the initial problem and
the subsequent rate of decline of GFR with the stresses of growth
and hyper-perfusion glomerulopathy. Reflux nephropathy rarely
causes ERF in the first 8 years of life and just over one-third
present between 8 and 12 years of life with almost 60% entering
ERF between the age of 12 and 16 years. Even so, reflux nephropathy
only accounts for 11% of patients between the ages of 12 and
16 years with ERF. The addition of patients with renal dysplasia
and reflux nephropathy together leads to a block accounting
for a little under or over 30% of patients in each age group.
In both conditions there is a high incidence of vesico-ureteric
reflux and in both conditions there is likely to be congenital
renal dysplasia. In view of the reduced frequency of urinary
tract infections and clinical pyelonephritis in the older age
groups, hyper-perfusion glomerulopathy is likely to play a major
part in both conditions in determining the speed and timing
of the decline into ERF. It is most likely therefore that reflux
nephropathy and renal dysplasia share common origins. Glomerular
diseases are rare in early childhood and 75% of children with
these diseases will enter ERF beyond the age of 8 years. As
glomerular diseases are the most common cause of ERF in Black
children this explains the age distribution of this cohort.
Whilst this is a small group within the UK, this observation
is important with regard to the development of services in developing
countries. Those with a predominantly Black population where
consanguineous marriage is rare can expect their paediatric
ERF population to come from the older childhood groups. This
will limit the potential size of the paediatric unit, particularly
if transfer to adult services is at a much younger age than
is the norm in the UK and Europe.
The same data for the main disease groups with inherited diseases are shown in Figures 13.13 A and B. As one might expect, diseases such as congenital nephrotic syndrome and polycystic kidney disease peak in the first 4 years of life, whilst the tubular and metabolic disorders peak later in childhood.
The final four groups are shown in Figures 13.14 A and B. Numerically these very different conditions account for only a small percentage of patients, both overall and in any one age band.
|
Current treatment of paediatric ESRF patients
|
|---|
Of the 768 patients, data on modality on 1st April 2005 were
available for 684 (89%). The distribution of modalities has
changed little since previous reports with 76% of patients having
a functioning allograft and for the remainder, peritoneal dialysis
being a more common treatment than haemodialysis. For those
with allografts, over two-thirds have cadaveric grafts with
21% of the total population (28% of those with allografts) having
a graft from a living donor. For those on peritoneal dialysis
the vast majority are receiving automated PD with few centres
using CAPD (
Figure 13.15).
The proportion of engrafted patients, whose graft has come from
a living (usually related) donor, rather than a cadaveric donor,
is slowly but steadily increasing (
Figure 13.16). This, in the
face of a stable ERF population with a stable proportion whose
management is with an allograft, highlights the shortage of
suitable cadaveric organs, the need to use living donation to
maintain the proportion of engrafted patients and the change
in medical practice in the UK with a greater emphasis being
placed upon the benefits of living donation.
The distribution of RRT modalities divided according to ethnic
origin is shown in
Figure 13.17. Whilst 80% of White patients
have a functioning allograft only 63% of South Asian patients
and 42% of Black patients have one. These populations therefore
have proportionately larger numbers on dialysis. For all groups,
peritoneal dialysis is the most frequent dialysis modality employed.
The difference between ethnic groups in the distribution of
treatment modalities is significant (
P < 0.0001, Chi-squared
= 22.2). Part of the explanation for the lower transplantation
rates in ethnic minority groups is the lower rate of living
donation. Certainly the proportion of South Asian patients with
an allograft from a living donor is significantly lower than
the proportion of White patients with one (
P = 0.0466). This
difference loses its significance if all ethnic minorities are
compared with the White population. The ethnic minority population
have a different distribution of tissue types and blood groups
to the White population who form the vast majority of the donor
pool. In these circumstances it is inevitable that there will
be fewer offers of well matched cadaveric allografts for ethnic
minority patients than White patients. In these circumstances
only an increase in the number of live donors in the ethnic
minority groups will allow the proportion with a functioning
allograft in these groups to rise to that of the White population.
An important aspect of ERF management is treatment modality
change with time.
Figure 13.18 shows the distribution of patients
according to whether or not their treatment modality had changed
since the previous data collection in 2004. Clearly for the
majority there was no change. Just under 11% of the cohort had
had a change in treatment modality during the year whilst 77%
did not. The remainder were new patients with no previous annual
record.
For those who had had no change over the previous year, the
vast majority (84%) had a functioning allograft. Nine percent
were maintained on peritoneal dialysis and 7% on haemodialysis
(
Figure 13.19).
For those who changed treatment modality over the course of
the year the reason in most was because they were transplanted.
Sixty-one percent of this cohort received an allograft and the
distribution of these between patients on peritoneal and haemodialysis
was appropriate for the numbers on each modality. Nineteen percent
lost grafts and started dialysis, 75% of these started peritoneal
dialysis. Fourteen percent of the cohort moved from peritoneal
to haemodialysis whilst only 5% of the cohort moved in the opposite
direction. One patient recovered enough renal function to stop
dialysis (
Figure 13.20).
The distribution of RRT modalities in April 2005 of the 81 patients
starting ERF management during that year is shown in
Figure 13.21.
As expected, the single largest group accounting for 49% of
the cohort were those on peritoneal dialysis. Just 10% were
on haemodialysis whilst 41% had a functioning allograft. A proportion
of this latter group would have had pre-emptive grafts whilst
others will have received an allograft during the first year
as a second treatment modality.
|
Conclusions
|
|---|
The incidence and prevalence of ERF in children in the UK has
changed little over recent years. Similarly, analyses of the
causes of ERF in childhood shows little change over the past
decade. After an initial steep growth following the commencement
of RRT services for children in the UK, the size of the paediatric
ERF population is now relatively static. As with most paediatric
and adult RRT studies there is a male predominance. In the paediatric
population this is secondary to both the large proportion of
patients with posterior urethral valves as a cause of ERF and
the predominance of males with renal dysplasia as a cause of
ERF.
The striking data is the high incidence and prevalence of ERF in the South Asian community in the UK. This is in part due to a high incidence of autosomal recessive inherited diseases causing ERF in this population. This could potentially lead not only to further growth of the ERF population over the next two decades, but also to a change in the pattern of disease causing ERF in the UK childhood population in addition to equalization of the gender distribution of ERF.
The commonest RRT modality for children with ERF is transplantation, with 76% of the population having a functioning allograft. The paucity of cadaveric organs has led to an increase in the proportion of these patients with an allograft from a living donor. Living donation is less frequent in the South Asian community who by virtue of their tissue types and that of the cadaveric donor pool, are also less likely to receive a graft. This could lead to a growing number of patients on dialysis as the ethnic minority population grows. For those on dialysis the majority are managed with peritoneal dialysis and the vast majority of these patients receive APD rather than CAPD.
|
Acknowledgements
|
|---|
This report was reviewed, revised and approved by the Paediatric
Renal Registry subcommittee comprising: Dr Kate Verrier-Jones,
Dr Chris Reid, Dr Jonathan Evans, Dr Nicholas Webb, Dr Rodney
Gilbert and Dr Malcolm Lewis.
|
References
|
|---|
- Ansell O, Feest T, eds. UK Renal Registry Report 1999, Chapter 15. Bristol, UK: UK Renal Registry.
- Ansell O, Feest T, eds. UK Renal Registry Report 2000, Chapter 15. Bristol, UK: UK Renal Registry.
- Ansell O, Feest T, eds. UK Renal Registry Report 2002, Chapter 16. Bristol, UK: UK Renal Registry.
- Ansell O, Feest T, eds. UK Renal Registry Report 2003, Chapter 14. Bristol, UK: UK Renal Registry.
- Ansell O, Feest T, eds. UK Renal Registry Report 2004, Chapter 13. Bristol, UK: UK Renal Registry.
- Ansell O, Feest T, Williams A, Winearls C, eds. UK Renal Registry Report 2005, Chapter 18. Bristol, UK: UK Renal Registry.
- Data USRenal. System: USRDS 2006 Annual Data Report. (2006) Bethesda, MD. Atlas of end-stage renal disease in the United States. National institutes of Health, National Institute of Diabetes and Digestive and Kidney Diseases.
- Craig J, Henning P, McTaggart S, McDonald S, Chang S, Excell L. Paediatric Report. ANZDATA Registry Report 2006, 143-153. Adelaide, South Australia: Australia and New Zealand Dialysis and Transplant Registry.
- NAPRTCS 2006 Annual Report. North American Pediatric Renal Trials and Collaborative Studies (https://web.emmes.com/ped/annlrept/annlrept2006.pdf).
CiteULike 
Connotea 
Del.icio.us What's this?
Top
Abstract
Introduction
