5.7.3. Data description and analysis
Incidence
Children and adolescents
CKD in the pre-ESRD phases in childhood and adolescence in
the above mentioned population-based registry in Italy (Italkid) have shown an
incidence rate of CKD (defined as a GFR <75 ml/min per 1.74 m2) of 12.1 cases per million of the age-related population (pmarp) (Ardissino et
al., 2004). In the Swedish study (considering children and adolescents in the 6
months-16 years range) the corresponding figure was
7.7 cases /year pmarp (<30 ml/min per 1.74 m2 ) (Esbjorner
et al, 1997). In 2005 the incidence rate of RRT for ESRD in children aged 0-14
was 7.1 patients pmarp (Table 5.7.3).
Adults
CKD incidence in adults has been studied very little (Obrador et al, 2002) and is marred with
difficulties (Winkelmayer et al, 2005). No data
has been published in Europe. In the Atherosclerosis Risk in Communities (ARIC)
Study in the USA, a study which enrolled individuals 45-64 years old, the
incidence rate of 3-5 CKD was 7.8 per 1000 patients x years (Kurella et al, 2005). Similarly scarce are
the data concerning changes in CKD over time, whilst the correlation with
changes in the incidence of ESRD are mainly based on NHANES surveys and on a
survey made in Norway. In the USA, the incidence of ESRD appears to be
increasing faster than that observed for CKD (Hsu et al, 2004). Indeed, 9 new cases of
ESRD developed in 1983 for every 1000 prevalent patients with CKD in 1978. By
comparison, 16 cases of ESRD had developed in 1996 for every 1000 patients with
CKD in 1991. A similar finding in terms of the relative stability of CKD versus
a marked increase in ESRD was noticed in a second study that examined NHANES
data (Coresh
et al,2005).
In Europe similar data are available only in Norway. Prevalence of 1-5 CKD in Norway was 10.2% which is similar to the current
prevalence in the USA (11%). This contrasts with ESRD incidence rates which are
three times higher in the United States compared to Norway. In a recent
comparative study (Hallan et al, 2006a) the relative risk for progression from
CKD stages 3 or 4 to ESRD in US white patients compared to Norwegian patients
was 2.5. This was only modestly modified by adjustment for age, gender, and
diabetes. Age and GFR at the beginning of dialysis were similar, hypertension
and cardiovascular mortality in the populations were comparable, but US white
patients were referred later to a nephrologist and had higher prevalence of
obesity and diabetes.
In 2005 the incidence rates of RRT for ESRD ranged from 57
patients pmp in Estonia to 205 patients pmp in Portugal (Table 5.7.4 and Table
5.7.5). Just like prevalence, the incidence rate of RRT for ESRD in 2005
steeply increased with age and was higher in males than in females (Table
5.7.3). Whereas incidence rates in Canada (160 pmp in 2005) are similar to
those in Europe, incidence rates in US whites (286 pmp in 2005) (U.S. Renal
Data System, USRDS, 2007) are 1.5-3 times as high as in Europe. This is partly
due to a higher US incidence rate of diabetic ESRD.
Table 5.7.3. Incidence of RRT over
the 1992-2005 period per age group, gender and cause of renal failure.
Table 5.7.4. Incidence of RRT over
the 1992-2005 period per country
Table 5.7.5. Incidence of RRT over
the 2000-2005 period per country.
Trends over time
There is no information on the trends of the CKD incidence
rate in children/adolescents or adults.
In the 1992-2005 period the incidence rate of RRT
increased by more than 50% (Table 5.7.3). This was primarily due to the
incidence rate in patients over 65 years of age which had more than doubled.
Whereas the incidence of RRT for diabetic and hypertensive ESRD became twice as
high, the incidence of RRT for ESRD due to glomerulonephritis/
glomerulosclerosis remained stable.
Although there are considerable differences in absolute
incidence rates of RRT for ESRD across countries, there was a consistent
increase in incidence rates in virtually all Member States at least until 2002.
This was mainly driven by an increase in the incidence rates of RRT for
diabetic and hypertensive ESRD (Stengel et al, 2003; van Dijk et al, 2005) After 2002, the
incidence rates have tended to stabilize. The stabilization, or even decrease
in some causes of ESRD, have prompted some investigators to suggest that
treatment strategies for the prevention of ESRD have finally started to bear
fruit (Gansevoort
et al, 2004; Stewart
et al, 2006).
The
increasing incidence rates together with improvements in survival of RRT
patients (Jager and van Dijk 2007) resulted in a concomitant increase in the
prevalence of RRT over the past decades that is posing a still increasing
economic burden on Member States.
Socioeconomic variation in incidence
As for prevalence the incidence rate of RRT was higher in
socially deprived areas of the UK than it was in other areas (UK Renal
Registry, 2003).
Prevalence
For the 27 EU Member countries, national surveys on the
prevalence of CKD among adults are available for 12 countries. The data for the
United Kingdom (UK) and Ireland and part of the Italian data were based on
information derived from general practitioners data-bases. Data from other
countries are based on population samples representative of the general
population of those countries.
There is still a paucity of data on CKD. Data in Europe
have been gathered both by using medical databases (Ireland, England, Italy) or population surveys. Data from medical databases overestimate the prevalence of
diseases and this is apparent also in the data collected in this review. Indeed
the prevalence of stages 3-5 CKD was 1.3 to 1.5 times higher in medical
databases in Ireland, England and Italy than in population based studies in Belgium,
Netherlands, Spain, Italy, Iceland, Norway. The prevalence of stage 3-5 CKD
appears reasonably similar across EU countries and more frequent in females
than in males. The higher prevalence on CKD in females, which flies in the face
of ESRD statistics where men are disproportionally affected (Table 5.7.7), may
depend on the fact that the GFR, as estimated by the MDRD equation (Levey et
al, 1999), is lower in females than in males. Furthermore, the performance of
this equation in people with normal or mildly impaired GFR may be sub-optimal
(Lin et al, 2003).
Children and adolescents
Information on CKD in the pre-ESRD phases for children and
adolescents is very limited (Warady and Chadha, 2007; Ardissino et al 2003). Data in a
population-based registry in Italy including all people <20 years reported a
prevalence of CKD (defined as a GFR <75 ml/min per 1.74 m2) of 74.7 cases pmarp (Ardissino
et al, 2003). In
a survey in Sweden in a more restricted age-range (6 months-16 years) and
applying a lower GFR cut-off for defining CKD (<30 ml/min per 1.74 m2 ), the corresponding figure was 21 cases pmarp (Esbjorner et al, 1997).
The prevalence of ESRD undergoing RRT in children (<20
years) in Europe is about 60 cases pmarp (Van der Heijden et al, 2004). In 2005
the prevalence of RRT in the 0-14 years age group was 43 per million of
age related population (Table 5.7.7).
Adults
The prevalence of CKD by stages (as defined in Tables
5.7..1 and 5.7..2) in The Netherlands (De Zeeuw et al, 2005) and in Spain (Garcia-Lopez and Epirce, 2007) is shown in Figure 5.7.2.
Figure 5.7.2. Prevalence of chronic kidney disease per stage in two
EU Countries
Prevalence of stages 3-5 CKD - i.e. the stages showing a
higher risk for CV complications and for evolution to ESRD (Go et al, 2004) - in
population-based studies ranges from 3.57% (Norway) (Hallan et al, 2006) to 7.2% (Germany) (Meisinger et al, 2006) in males and from 6.2%
(Italy) (Cirillo
et al, 2006)
to 10.2% (Iceland) (Viktorsdottir et al, 2005) in females (Figure 5.7.3), while higher
figures are reported in medical databases (Stevens et al, 2007). Stage 3-5 CKD
prevalence was either similar in males and females (Germany, Italy) or higher in females (Belgium (Van Biesen et al, 2007) , England (Stevens et al, 2007), Iceland (Viktorsdottir et al, 2005), Norway (Hallan et al, 2006; Hallan et al, 2006). In all countries where this information was available the
prevalence of stage 3-5 CKD increased with age (Figure 5.7.4).
Figure 5.7.3. Prevalence of stages
3-5 of chronic kidney disease in selected EUGLOREH countries.
Figure 5.7.4. Sex and age specific prevalence of stages 3-5 of
chronic kidney disease by age and sex.
In 2005 the crude prevalence of RRT for ESRD varied from
321 patients pmp in Romania to 1057 patients pmp in Germany (Tables 5.7.6 A and
B). It increased with age and was more than 50% higher in males compared to
females (Table 5.7.7). Prevalence of RRT in Europeans is lower than that in US whites (1209 pmp) and in Canadians (1003 pmp) (U.S. Renal Data System, USRDS ,2007).
Table 5.7.6A. Prevalence of RRT over the period 1992-2005 by
country.
Table 5.7.6B. Prevalence of RRT over the period 2000-2005 by country.
Table 5.7.7. Prevalence of RRT over the period 1992-2005 by age
group, gender and cause of renal failure.
Trends over time
There is still very scarce information on time-trends of
CKD. The most solid source of information remain periodic surveys made in the USA (NHANES III and NHANES IV performed between 1988-1994 and between 1999-2004
respectively). In those surveys, the prevalence of stage 1-5 CKD rose from
14.5% (NHANES III) to 16.8% (NHANES IV) (Tonelli et al, 2006; Zoccali and
DeJong, 2007), while
the prevalence of stages 3-5 remained almost unmodified at about 6%. None of
these data are available in European countries.
In Europe over the period 1992-2005 (Table
5.7.7) the overall adjusted prevalence of RRT for ESRD increased from 480 to
807 patients per million population (pmp). This was due to a 40% increase in
the 15-64 age group and a more than 130% increase in the 65+ age group. In the
0-14 age group, however, the prevalence remained stable throughout the period.
Mortality
A recent meta-analysis has shown that the risk of
mortality in CKD rises exponentially with decreasing GFR (USRDS 2007 ).
Mortality in ESRD patients is very high. Five-year mortality rates in incident
RRT patients are 52% in all patients, and 21%, 32% and 72% for patients aged
0-14, 15-64 and over 65 years of age respectively (Table 5.7.8). Five-year
mortality in patients on dialysis is almost five times as high as that after
kidney transplantation: 60% and 13%, respectively.
Mortality on RRT is lower in Europe compared to the US (Sixth Annual Report
UK Renal Registry, 2003; Warady and Chadha, 2007; ERA-EDTA Registry, 2005). Also within Europe there are considerable differences in patient survival (van Dijk et al, 2007).
International studies to investigate the reasons for different outcomes in
haemodialysis patients are in progress (Young et al, 2000).
Table 5.7.8. 90-day, one-, two- and
five-year mortality rates in incident RRT patients per age group, gender and
cause of renal failure.
Trends over time
Previous analyses have shown that compared to patients
starting dialysis in the cohort 1980-1984, dialysis patients in the more recent
cohorts had a 6% (cohort 1990-1994) and 12% (cohort 1995-1999) lower risk of
death. The mortality risk reduction in transplant recipients was much higher:
32% and 56%, respectively (Van Dijk et al, 2001).
In conclusion, as previously emphasised, there is still a
paucity of data on CKD. Data in Europe have been gathered both by using medical
databases (Ireland, England, Italy) or population surveys. Data from medical
databases overestimate the prevalence of diseases and this was apparent also in
the data collected in this review. Indeed, the prevalence of stage 3-5 CKD was
1.3 to 1.5 times higher in medical databases in Ireland, England and Italy than
in population based studies in Belgium, Netherlands, Spain, Italy, Iceland,
Norway. The prevalence of stage 3-5 CKD appears reasonably similar across EU
countries and more frequent in females than in males. The higher prevalence on
CKD in females, which flies in the face of ESRD statistics where men are
disproportionally affected (Table 5.7.7), may depend on the fact that the GFR,
as estimated by the MDRD equation (Levey et al, 1999), is lower in females than in males.
Furthermore, the performance of this equation in people with normal or mildly
impaired GFR may be sub-optimal (Lin et al, 2003).
Although there are considerable differences in absolute
incidence rates of RRT for ESRD across countries, there was a consistent
increase in incidence rates in virtually all Member States at least until 2002.
This was mainly driven by an increase in the incidence rates of RRT for
diabetic and hypertensive ESRD (Stengel et al, 2003; van Dijk et al, 2005) After 2002 the incidence rates have
tended to stabilize. The stabilization, or even decrease in some causes of
ESRD, have prompted some investigators to suggest that treatment strategies for
the prevention of ESRD have finally started to bear fruit (Gansevoort et al, 2004; Stewart et al, 2006). The increasing incidence
rates together with improvements in survival of RRT patients (Jager and van
Dijk 2007) resulted in a concomitant increase in the prevalence of RRT over the
past decades that is posing a still increasing economic burden on Member
States.
Mortality in ESRD patients is still very high. Although
the better survival of transplant recipients is, at least in part, due to
selection bias, the survival of RRT patients could be considerably improved at
reduced costs by increasing organ donation rates.
