Anemia normalization in patients with type 2 diabetes and chronic kidney disease:

results of the NEPHRODIAB2 randomized trial

Emmanuel Villar
a,
, Michel Livre
b,c
, Michle Kessler
d
, Vincent Lematre
e
, Eric Alamartine
f
,
Michel Rodier
g
, Maud Franois
h
, Philippe Zaoui
i
, Olivier Moranne
j
, Gabriel Choukroun
k
,
Abdallah Guerraoui
l
, Anne Jolivot
m
, Grard Janin
n
, Bernard Branger
o
, Anne-Elisabeth Heng
p
,
Catherine Boudray
q
, Alvine Bissery
r
, Muriel Rabilloud
c,r
, Claire Pouteil-Noble
a,c
a
Hospices Civils de Lyon, Department of Nephrology, Lyon Sud Hospital, Pierre Benite, France
b
Hospices Civils de Lyon, Department of Clinical Pharmacology, Lyon, France
c
Universit de Lyon 1, Villeurbanne, France
d
Centre Hospitalier et Universitaire de Nancy, Department of Nephrology, Vandoeuvre les Nancy, France
e
Centre Hospitalier de Valenciennes, Department of Nephrology, Valenciennes, France
f
Centre Hospitalier et Universitaire de St Etienne, Department of Nephrology, St Priest en Jarez, France
g
Centre Hospitalier et Universitaire de Nmes, Department of Endocrinology, Nmes, France
h
Centre Hospitalier et Universitaire de Tours, Department of Nephrology, Tours, France
i
Centre Hospitalier et Universitaire de Grenoble, Department of Nephrology, La Tronche, France
j
Centre Hospitalier et Universitaire de Lille, Department of Nephrology, Lille, France
k
Centre Hospitalier et Universitaire d'Amiens, Department of Nephrology and Institut National de la Sant et de la Recherche Mdicale ERI-12, Universit Jules Verne, Amiens, France
l
Centre Hospitalier de Vichy, Department of Nephrology, Vichy, France
m
Hospices Civils de Lyon, Department of Nephrology, E. Herriot Hospital, Lyon, France
n
Centre Hospitalier de Macon, Department of Nephrology, Macon, France
o
Centre Hospitalier et Universitaire de Nmes, Department of Nephrology, Nmes, France
p
Centre Hospitalier et Universitaire de Clermont Ferrand, Department of Nephrology, Clermont Ferrand, France
q
Centre Hospitalier de Bourg en Bresse, Department of Nephrology, Fleuriat, France
r
Hospices Civils de Lyon, Department of Biostatistic, Lyon, France
a b s t r a c t a r t i c l e i n f o
Article history:
Received 4 January 2011
received in revised form 22 February 2011
accepted 8 March 2011
Available online 20 May 2011
Keywords:
Anemia
Chronic kidney disease
Erythropoiesis-stimulating agent
Glomerular ltration rate
Type 2 diabetes
Statements of the Problem: Correction of anemia in type 2 diabetes (T2DM) patients with chronic kidney
disease stages 34 may slow the decline of kidney function but may increase cardiovascular risk through
higher hematocrit. The NEPHRODIAB2 study was designed to assess efcacy and safety of complete
hemoglobin (Hb) normalization in these patients.
Methods: We randomly assigned 89 T2DM patients with an estimated glomerular ltration rate (eGFR;
abbreviated 175 Modication of Diet in Renal Disease formula) of 25 to 60 ml/min per 1.73 m
2
and moderate
anemia (Hb, 100129 g/l) to a target Hb value in subnormal range (110129g/l, group 1, n=43) or normal
range (130149 g/l, group 2, n=46). The primary end point was eGFR decline after 2 years of follow-up.
Secondary end points included iron and erythropoietin dosage, quality of life (Medical Outcomes Study 36-
item Short-Form Health Survey scores) and adverse events.
Results: Six months after randomization, the mean Hb levels were b120 g/l in group 1 and N130 g/l in group 2
(Pb.05 at 6, 12, 18 and 24 months). Blood pressure, 24-h proteinuria and HbA1c did not differ during follow-
up (PN.05). Two-year declines in eGFR were 8.712.2 in group 1 and 5.17.8 ml/min per 1.73 m
2
in
group 2 (P=.29). Mean weekly use of erythropoietin was 7.811.6 g in group 1 and 30.133.6 g in group 2
(Pb.0001). There was no signicant difference regarding Medical Outcomes Study 36-item Short-FormHealth
Survey score change or adverse event occurrence.
Conclusions: In this trial, normalization of Hb level in T2DM patients with chronic kidney disease was safe but
did not signicantly slow renal function decline and increased treatment cost due to erythropoietin use.
2011 Elsevier Inc. All rights reserved.
1. Introduction
Type 2 diabetes (T2DM) is now the main cause of end-stage renal
disease in industrialized countries (Foley & Collins, 2007; Stengel
et al., 2003; Villar, Chang, & McDonald, 2007). Despite survival
Journal of Diabetes and Its Complications 25 (2011) 237243
ClinicalTrials.gov number: NCT00279084.
Corresponding author. Centre Hospitalier Lyon-Sud, Service de Nphrologie,
Dialyse et Transplantation Rnale, 69495 Pierre Bnite Cedex, France. Tel.: +33 4 72 67
87 00; fax: + 33 4 72 67 87 10.
E-mail address: emmanuel.villar@chu-lyon.fr (E. Villar).
1056-8727/$ see front matter 2011 Elsevier Inc. All rights reserved.
doi:10.1016/j.jdiacomp.2011.03.003
Contents lists available at ScienceDirect
Journal of Diabetes and Its Complications
j our nal homepage: WWW. JDCJOURNAL. COM
improvement among T2DM patients on dialysis over the last decades
(Sorensen, Mathiesen, Heaf, & Feldt-Rasmussen, 2007; Villar et al.,
2007), this condition remains associated with poor outcomes. In
nondialysis-dependent T2DMpatients with associated chronic kidney
disease (CKD), level of evidence A strategies that slow the decline of
kidney function and improve cardiovascular risk are blood pressure
lowering using angiotensin-II receptor blockers (ARB) (Brenner et al.,
2001; Lewis et al., 2001; Parving et al., 2001) and intensive glycemic
control (Skyler et al., 2009).
Anemia is a common consequence of CKD due to erythropoietin
deciency. In CKD patients, glomerular ltration rate (GFR) under
45 ml/min per 1.73 m
2
is the mean threshold for detection of anemia
dened as hemoglobin (Hb) level b110 g/l (Moranne et al., 2009). In
T2DMpatients with associated CKD, anemia is more severe and occurs
earlier in the course of CKD (Ravanan, Spiro, Mathieson, & Smith,
2007; Thomas et al., 2004).
Observational studies suggested that anemia correction with an
erythropoietin-stimulating agent (ESA) in predialysis care may
slow the decline of kidney function (Jungers et al., 2001) and
improve CKD patient cardiovascular risk (Hayashi et al., 2000) and
quality of life (Alexander, Kewalramani, Agodoa, & Globe, 2007).
Since 1999, international guidelines on anemia have recommended
an Hb target of 110 g/l in predialysis CKD patients (Hrl et al.,
2007; Levin & Rocco, 2007), based on evidence of a better quality
of life.
Nevertheless, higher Hb levels may increase in T2DM patients the
risk for cardiovascular events through higher hematocrit and higher
blood viscosity in the context of diabetic micro- and macrovascular
disease (Orasanu & Plutzky, 2009). Due to these medical specicities,
we therefore designed the NEPHRODIAB2 randomized study to assess
the effect of complete Hb normalization on renal function decline and
cardiovascular safety in patients with T2DM and associated CKD
stages 34, when compared to recommended Hb level (Hrl et al.,
2007; Levin & Rocco, 2007).
2. Materials and methods
2.1. Study design
NEPHRODIAB2 was a multicenter, randomized, open-label,
parallel group study involving 15 nephrology centers in France.
The study was performed in accordance with the Declaration of
Helsinki and Good Clinical Practices. The study protocol was
approved by a local ethical review board (Comit de Protection
des Personnes Lyon A). The trial was registered at ClinicalTrials.gov
(NCT00279084). Patients were randomized centrally into two
treatment groups by using a block-size randomization procedure
stratied by center and estimated GFR [eGFR; b45 and 45 ml/min
per 1.73 m
2
, abbreviated 175 Modication of Diet in Renal Disease
(MDRD) formula]. Group 1 patients were assigned to a subnormal
Hb level range (110129 g/l) and group 2 to a normal Hb level range
(130149 g/l).
All patients gave written consent before inclusion. The trial began
in February 2004 and ended in May 2008, 2 years after inclusion of the
last patient.
The NEPHRODIAB2 study enrolled adult patients (1880 years old)
with T2DM, moderate anemia (Hb level: 100129 g/L) and CKD
stage 24 (eGFR of 2560 ml/min per 1.73 m
2
, abbreviated 175-
MDRD formula).
Major exclusion criteria were neoplasia, history of solid organ
transplantation, acute medical or surgical condition in the 2 months
before inclusion, myocardial infarction or unstable angina or stroke in
the 6 months before inclusion, contraindication to iron or ESA therapy
or inability to complete the Medical Outcomes Study 36-item Short-
Form Health Survey (SF36) quality of life questionnaire.
Patients who fullled the inclusion and exclusion criteria were
randomly assigned to receive iron and/or ESA therapy for targeting Hb
value in subnormal range (110129 g/l, group 1) or normal range
(130149 g/l, group 2). Iron and ESA therapy modality was under the
responsibility of each investigator (type of iron and ESA was free). ESA
therapy should not be started unless serum ferritin was N200 g/l.
All patients had clinical and biological assessment at 1, 2, 3, 6, 12,
18 and 24 months after inclusion. The study was scheduled to end
after each patient had completed 24 months of follow-up.
2.2. Study end points
The primary end point was the change from baseline in eGFR at 24
months using abbreviated 175-MDRD formula. Secondary end points
were as follows: (i) mean daily iron use and mean weekly ESA use, (ii)
quality of life assessed by the SF36 (score for each subscale range from
0 to 100, with higher scores indicating better quality of life) and (iii)
critical events. Critical events included death from all causes,
myocardial infarction, stroke, peripheral arterial disease, heart failure
leading to hospitalization, new diagnosis of cancer (except non-
melanoma skin cancer), bacterial infection, chronic renal replacement
therapy and symptomatic deep vein thrombosis. All these events were
adjudicated by an event committee blinded to the assigned Hb range.
2.3. Statistical analyses
Sample size calculation for NEPHRODIAB2 was based on the
primary outcome variable, eGFR change. The expected mean change
in eGFR from baseline to 24-month follow-up was 10 ml/min per
1.73 m
2
per 24 months in group 1 and 8 ml/min per 1.73 m
2
per
24 months in group 2. With a sample size of at least 102 patients
per treatment arm, NEPHRODIAB2 would have had 90% power to
show this difference by study group.
All quantitative values are reported as meanS.D. or median
value. All qualitative values are reported as number and percentage.
All primary and secondary end point analyses were based on the
intent-to-treat principle. The primary end point (change in eGFR)
was analyzed by the Student's t test. An eGFR of 10 ml/min per
1.73 m
2
was attributed to patients who required renal replacement
therapy at each visit after rst dialysis. Analyses were reprocessed
with eGFR value of 5 and 0 ml/min per 1.73 m
2
to test
reproducibility of the results. Lastly, a linear mixed-effects model
was used to compare the eGFR variation slopes between random-
ization groups.
For analyses of secondary end points, we used the Student's t test
for quantitative variables, and the
2
test for qualitative variables. We
compared SF36 scores within each randomized group at inclusion,
6 month and 24 months by using analysis of variance. We compared
SF36 scores between randomized groups at inclusion, 6 months and
24 months by using Student's t test.
Iron therapy was expressed in milligrams. ESA use was expressed
as darbepoetin equivalent dose (1 g of darbepoetin equals 200 UI of
epoetin ).
All statistical analyses were performed with STATA 10.0 statistical
software. P value less than .05 was considered statistically signicant.
2.4. Role of the funding source and study committees
NEPHRODIAB2 was mainly funded by an institutional grant
(Programme Hospitalier de Recherche Clinique) in 2002 and by a
grant from the Socit de Nphrologie. The trial was designed,
implemented and overseen by a Study Steering Committee that was
independent from these funding sources and from two other
industrial sponsors, Amgen France and Roche France. Sponsors have
not been involved in study design, recruitment of participating
centers, monitoring centers, data collection, data cleaning and
238 E. Villar et al. / Journal of Diabetes and Its Complications 25 (2011) 237243
analysis plan. The Steering Committee was granted access to all study
data. Interpretation of all study data was done independently
of sponsors.
An Event Validation Committee was constituted and validated all
clinical and adverse events during the study.
The clinical research organizer was the Hospices Civils de Lyon.
89 patients enrolled
43 assigned
to subnormal Hb level range.
Group1: 110 129 g/L
46 assigned
to normal Hb level range.
Group 2: 130 149 g/L
32 completed 24-month follow-up.
5 died
5 required dialysis
1 withdrew before termination
38 completed 24-month follow-up.
4 died
2 required dialysis
2 withdrew before termination
Fig. 1. Enrolment and outcomes.
Table 1
Patient characteristics at randomization
Group 1, 110129 g/l (n=43) Group 2, 130149 g/l (n=46) P
Hb (g/l), meanS.D. 114.78.10 114.210.6 .80
Serum ferritin (g/l), meanS.D. 191235 141129 .22
Age (years), meanS.D. 65.29.1 68.57.6 .07
Male sex, no. (%) 28 (65.1) 28 (60.9) .68
Original nephropathy, no. (%) .10
Diabetic 27 (62.8) 23 (50.0)
Vascular 11 (25.6) 21 (45.6)
Other 5 (11.6) 2 (4.3)
Renal biopsy, no. (%) 9 (21.0) 7 (15.2) .48
Body mass index (kg/m
2
), meanS.D. 31.55.3 31.55.6 1
Cardiovascular history, no. (%)
Hypertension 43 (100) 45 (97.8) .97
Myocardial infarction 3 (7.0) 6 (13.0) .34
Angina pectoris 4 (9.3) 4 (8.7) .92
Stroke 4 (9.3) 3 (6.5) .63
Peripheral vascular disease 10 (23.3) 11 (23.9) .94
Diabetic retinopathy, no. (%) 17 (39.5) 15 (32.6) .50
Neoplasm history, no. (%) 2 (4.6) 5 (10.9) .28
Chronic bacterial infectious history, no. (%) 1 (2.3) 0 (0) .48
Blood pressure (mmHg), meanS.D.
Systolic 146.818.9 145.719.0 .78
Diastolic 75.911.9 74.711.1 .62
No. of antihypertensive agent, meanS.D. 3.10.9 3.31.0 .32
Antihypertensive agent, no. (%)
ACEI 24 (55.8) 27 (58.7) .78
ARB 21 (48.8) 22 (47.8) .92
ACEI and/or ARB 42 (97.7) 41 (89.1) .11
Renal function, meanS.D.
Serum creatinine (mol/l) 190.452.8 183.459.1 .56
Cockcroft and Gault (ml/min per 1.73 m
2
) 37.09.1 36.69.4 .83
MDRD (175 formula; ml/min per 1.73 m
2
) 29.612.8 31.814.3 .49
Proteinuria (g/24 h) 2.52.8 1.72.4 .16
Proteinuria/creatininuria ratio (g/mmol) 3.13.6 2.04.9 .23
HbA1c (%), meanS.D. 7.51.3 7.61.4 .73
Treatment by insulin, no. (%) 25 (58.1) 32 (69.6) .26
LDL-cholesterol in (mmol/l), meanS.D. 2.91.1 2.91.2 1
HMG CoA reductase inhibitor use, no. (%) 24 (55.8) 28 (60.9) .63
Creactive protein in (mg/l), meanS.D. 7.08.8 8.816.0 .51
HMG CoA reductase, hydroxy-mthyl-glutaryl-coenzyme A rductase (i.e., statin).
239 E. Villar et al. / Journal of Diabetes and Its Complications 25 (2011) 237243
3. Results
3.1. Patient disposition and baseline characteristics
Patient disposition is shown in Fig. 1. Due to lower rates of renal
function decline than expected, the study had to be stopped before
recruiting the planned number of patients. Eighty-nine patients were
randomly assigned to the Hb target groups (group 1, 43 patients;
group 2, 46 patients). Patient characteristics did not differ at inclusion
(Table 1).
In average, patients were 65 years old, had serum creatinine
around 180190 mol/l corresponding to an eGFR of 3032 ml/min
per 1.73 m
2
using 175-MDRD formula, had overt proteinuria higher
than 2 g/24 h and had moderate anemia (Hb values around 114 g/l).
All patients but one had history of hypertension. The vast majority
were treated by an angiotensin receptor inhibitor (ARB) and/or an
angiotensin-converting enzyme inhibitor (ACEI). In average, 3.2
antihypertensive agents were used to achieve a mean blood pressure
around 146/76 mmHg in both groups. One quarter of the patients had
history of peripheral vascular disease.
The mean glycosylated Hb was 7.5%7.6%, and two thirds of the
patients received insulin.
Three patients withdrew their written consent after inclusion and
did not nish the study.
3.2. Anemia correction
Mean Hb levels were higher than 130 g/l in group 2 at 6 to
24 months, while they remained lower than 120 g/l in group 1 over
the study period (Fig. 2). Hb differed signicantly between the groups
at 6 (+12.6 g/l in mean in group 2), 12 (+16.3 g/l), 18 (15.6 g/l) and
24 months (+10.6 g/l) (all Pb.05).
Over the study period, the mean daily iron dosage was
69.791.4 mg in group 1 and 91.6100.3 mg in group 2 (P=.28).
Usual iron route administration was per os in all treated patients.
Height patients have had occasionally intravenous iron administra-
tion during the study period.
Mean weekly ESA dosage was 7.811.6 g (=15581314 UI) in
group 1 and 30.133.6 g (=60286729 UI) in group 2 (Pb.0001).
3.3. Renal function
The eGFR decline is shown in Fig. 3. Two-year declines in eGFR
were 8.712.2 in group 1 and 5.17.8 ml/min per 1.73 m
2
in
group 2 (P=.22). Changing eGFR values assigned to patients who
started dialysis (10, 5 or 0 ml/min per 1.73 m
2
) did not modify the
main result.
When adjusted on baseline eGFR, 24-h proteinuria and systolic
blood pressure, eGFR change remained not statistically different
between randomization groups (P=.41). When eGFR changes were
compared by using linear mixed-effects model, the difference was not
statistically signicant either (P=.31).
3.4. Blood pressure, proteinuria and glycosylated Hb
Blood pressure (Fig. 4), 24-h proteinuria (Table 2), antihyperten-
sive agent use (Table 2) and HbA1c (Table 2) did not vary signicantly
between groups during follow-up (all PN.05).
3.5. Quality of life
SF36 scores are shown in Fig. 5. There was no signicant difference
at baseline and at 6 and 24 months between randomization groups
(all PN.05). In each randomization group, SF36 score changes over the
study period were not signicantly different (all PN.05).
3.6. Safety
There was no signicant difference between the groups for the
incidence of critical events (Table 3). In group 1, causes of death were
cardiovascular disease in 1, infectious disease in 1, neoplasia in 1 and
unknown in 2 patients. In group 2, causes of death were neoplasia in 2
and unknown in 2 patients.
100
110
120
130
140
150
0 3 6 9 12 15 18 21 24
Months after randomization
H
e
m
o
g
l
o
b
i
n

g
/
L
Group 1 Group 2
Fig. 2. Hb level (difference between groups: Pb.05 at 6, 12, 18 and 24 months).
5
10
15
20
25
30
35
40
45
M
D
R
D

m
L
/
m
i
n
/
1
.
7
3
m
2
50
0 3 6 9 12 15 18 21 24
Months after randomization
Group 1 Group 2
Fig. 3. Renal function decline (P=.29).
50
60
70
80
90
100
110
120
130
140
150
160
170
180
0 3 6 9 12 15 18 21 24
Months after randomization
B
l
o
o
d

p
r
e
s
s
u
r
e

m
m
H
g
Group 1 Group 2 Group 1 Group 2
Fig. 4. Blood pressure evolution (all PN.05).
240 E. Villar et al. / Journal of Diabetes and Its Complications 25 (2011) 237243
4. Discussion
This randomized trial did not provide argument to fully normalize
Hb level in T2DM patients with associated CKD stages 34. Despite a
nondeleterious impact on blood pressure level or adverse events, this
study did not demonstrate that increasing Hb level over 130 g/l in
those patients changed signicantly CKDprogression rate or quality of
life when compared to 110130 g/l Hb level target. The mean dose of
ESA was fourfold higher in the group with high Hb level than in the
group with low Hb level.
These results were in line with the results of previously published
randomized trials (Drueke et al., 2006; Gouva, Nikolopoulos,
Ioannidis, & Siamopoulos, 2004; Macdougall, Temple, & Kwan, 2007;
Pfeffer et al., 2009; Ritz et al., 2007; Rossert et al., 2006; Singh et al.,
2006). NEPHRODIAB2 was designed to include only T2DM patients.
The hypothesis was that the potential benet of anemia normalization
on renal function decline through decrease of peripheral ischemia
could have been counterbalanced by an increased risk of cardiovas-
cular events through higher hematocrit and blood viscosity in these
patients with diabetic microvascular and macrovascular complica-
tions (Beckman, Creager, & Libby, 2002). The specic vascular
characteristics of T2DM justied a specic trial of anemia treatment
in CKD patients.
Previous studies that compared outcomes by Hb levels and ESAuse
are summarized in Table 4 (Drueke et al., 2006; Gouva et al., 2004;
Macdougall et al., 2007; Pfeffer et al., 2009; Ritz et al., 2007; Rossert
et al., 2006; Singh et al., 2006). Two studies using subcutaneous
epoetin alpha (Macdougall et al., 2007; Rossert et al., 2006) were
discontinued early in 2002 due to an emerging risk of epoetin-
induced pure red cell aplasia that contraindicated this route of epoetin
administration (Casadevall et al., 2002). Only the recently published
The Trial to Reduce Cardiovascular Events with Aranesp Therapy
(TREAT) Study reported by Pfeffer et al. (2009) focused specically on
T2DMpatients with CKD. There was no signicant difference between
the groups regarding the primary end points: composite cardiovas-
cular end point or renal composite end point (death or end-stage
renal disease). Rate of strokes was signicantly higher in the
darbepoetin group compared to the placebo group (5.0% vs. 2.6%).
Authors concluded that the risk of darbepoetin in T2DM patients with
CKD outweighed the potential benets. In NEPHRODIAB2, we did not
nd difference in adverse event rates between the groups, but our
study was underpowered to detect such a difference.
With the exception of the study reported by Gouva et al. (2004) in
88 patients in 1999, none of these studies showed a benet of higher
Hb level target in reducing the rate of renal function decrease or the
rate of cardiovascular event. With the exception of the TREAT Study
(Pfeffer et al., 2009), all studies included patients with a mean GFR
Table 2
24-h proteinuria, antihypertensive agent use and HbA1c evolutions over the
study period
Inclusion 12 months 24 months
24-h proteinuria (g/24 h)
Group 1 (110129 g/l) 2.52.8 3.13.6 2.23.5
Group 2 (130149 g/l) 1.72.4 2.04.9 1.41.8
P .16 .23 .19
Antihypertensive agent, meanS.D.
Group 1 (110129 g/l) 3.10.9 3.21.0 3.11.0
Group 2 (130149 g/l) 3.31.0 3.31.0 3.20.9
P .32 .57 .65
ACEI and/or ARB (%)
Group 1 (110129 g/l) 97.7% 94.8% 92.3%
Group 2 (130149 g/l) 89.1% 93.0% 89.2%
P .11 .72 .63
HbA1c (%)
Group 1 (110129 g/l) 7.51.3 7.41.5 7.41.6
Group 2 (130149 g/l) 7.6 %1.4 7.6%1.3 7.5%1.5
P .73 .50 .76
0
20
40
60
80
BP GH MH PF ER PR SF VT BP GH MH PF ER RP SF VT
Group 1 (110 129 g/L) Group 2 (130 149 g/L)
Inclusion 6
th
month 24
th
month
Fig. 5. SF36 scores evolutions by randomization group. BP, body pain; GH, general health; MH, mental health; PF, physical function; ER, emotional role; PR, physical role; SF, social
function; VT, vitality. All PN.05 within each group at inclusion, at 6 and 24 months, and between groups at inclusion, 6 and 24 months.
Table 3
Adverse events by randomization group (all PN.05)
Group 1 (110129 g/l) Group 2 (130149 g/l)
Death 4 5
Myocardial infarction 1 1
Stroke 1 0
Peripheral vascular disease 4 2
Heart failure 3 4
Cancer 3 3
Bacterial infection 6 4
Chronic RRT (dialysis) 5 2
Deep vein thrombosis 0 1
Total 27 22
RRT, renal replacement therapy.
241 E. Villar et al. / Journal of Diabetes and Its Complications 25 (2011) 237243
below30 ml/min per 1.73 m
2
, and one can ask if Hb normalization has
a potential benecial effect at this CKD stage both on cardiovascular
risk and on renal function. NEPHRODIAB2 study was therefore
designed to include CKD patients with associated T2DM and mean
eGFR N30 ml/min per 1.73 m
2
. Randomized patients were represen-
tative of CKD patients with associated T2DM: they were predomi-
nantly male and obese, all but one patient had a history of
hypertension, the vast majority was treated with ACEI and/or ARB,
23% had history of peripheral vascular disease, their renal involve-
ment included macroproteinuria and one third had associated
diabetic retinopathy.
Regarding renal function, previous studies underlined that in
T2DM patients with CKD stages 34, its decline was expected to be
4.5 to 5 ml/min per 1.73 m
2
per year (Estacio et al., 2000), (Leehey
et al., 2005) whereas in general population older than 40 years,
natural renal function decline is 1 ml/min per 1.73 m
2
per year. In
the NEPHRODIAB2 cohort, despite an average proteinuria above the
recommended 1 g/24 h in both groups over the study period and
treatment with ACEI and/or ARB, renal function decline was lower
than expected: 3.33.2 ml/min per 1.73 m
2
per year in average.
This may indicate that beyond reninangiotensin system blockers,
aggressive treatment and close monitoring are useful in controlling
renal involvement in T2DM with CKD.
Interestingly, in the present study, there was only a slightly and
nonsignicant increase in daily iron dosages in the group with high
Hb level, whereas weekly ESA dosage was fourfold higher in this
group than in the group with low Hb level over the study period.
Those results indicated that a 10-g/l increase in Hb level in T2DM
patients with CKD, from120 to 130 g/l in mean, required a huge gap in
ESA treatment and consequently in related costs.
Our study has limitations. GFR was not measured but estimated by
using abbreviated MDRD equation. Even if we cannot exclude bias in
GFR decline evaluation, especially underestimation of GFR (Chudle-
igh, Dunseath, & Evans, 2007; Chudleigh, Dunseath, & Peter, 2008;
Chudleigh, Ollerton, & Dunseath, 2008) and underestimation of GFR
decline (Beauvieux et al., 2007; Rigalleau et al., 2008), randomization
distributed equally factors that biased GFR evaluation between the
compared groups. The comparison of the groups was therefore
unbiased, and we did not nd difference between groups regarding
CKD progression rate whatever the estimator and the statistical
method we used. Inclusions in the study were discontinued early
because renal function decline was overall lower than that was
expected. The study was therefore underpowered to statistically
demonstrate signicant the difference we observed. Based on the
observed decline in eGFR in the NEPHRODIAB2 control group, with a
0.9 power and a .05 risk, such a trial should include 672 patients to
display a signicant difference. Strength of the present study was that
it included only patients with T2DM and that the control group was
treated as recommended by international guidelines (Hb level
between 110 and 120 g/l; Hrl et al., 2007; Levin & Rocco, 2007).
In conclusion, when compared to recommended Hb level target
(110120 g/l), our study did not showa signicant reduction in rate of
renal function decline with normalization of Hb level over 130 g/l in
CKD patients with associated T2DM and mild anemia but underlined
the increase of treatment cost due to ESA use.
Acknowledgments
We gratefully acknowledge all study participants, especially the
nephrologists and professionals who included patients and collected
data. We gratefully acknowledge Ms. J. Gillet for data monitoring.
Members of the Steering Committee: Prof. Eric Alamartine (St
Etienne), Prof. Michle Kessler (Nancy), Dr. Vincent Lematre
(Valenciennes), Dr. Michel Livre (Lyon), Prof. Claire Pouteil-Noble
(Lyon), Dr. Emmanuel Villar (Lyon).
Members of the Event Validation Committee: Dr. Alexandre Belot
(Pediatric Nephrology, Lyon), Dr. Sabrina Franois (Internal Medicine,
Lyon), Dr. Patrice Nony (Cardiology, Lyon).
Clinical research organizer: Hospices Civils de Lyon.
Conict of interest statement: None declared. Roles of funding
sources are described in the body of the article, in the Materials and
methods section.
The results presented in this paper have not been published
previously in whole or part, except in abstract form at the World
Congress of Nephrology 2009 in Milano, Italia (abstract M561), and at
the 11th Joint Congress of the Socit de Nphrologie and the Socit
Francophone de Dialyse 2009 in Toulouse, France (abstract CO006).
References
Alexander, M., Kewalramani, R., Agodoa, I., & Globe, D. (2007). Association of anemia
correction with health related quality of life in patients not on dialysis. Current
Medical Research and Opinion, 23, 29973008.
Beauvieux, M. C., Le Moigne, F., Lasseur, C., et al. (2007). New predictive equations
improve monitoring of kidney function in patients with diabetes. Diabetes Care, 30,
19881994.
Beckman, J. A., Creager, M. A., & Libby, P. (2002). Diabetes and atherosclerosis:
epidemiology, pathophysiology, and management. JAMA, 287, 25702581.
Table 4
Previous ESA-randomized trials in CKD patients
Study Population
eGFR at
inclusion Hb target Primary end point
Main result: primary
end point
Renal function
decline Adverse events
TREAT
(Pfeffer et al., 2009)
T2DM,
n=4038 patients
34 ml/min
per 1.73 m
2
Hb N90 g/l or
Hb N130 g/l
Death or
cardiovascular
event or dialysis
No difference No difference Higher stroke rate
in the group with
higher Hb level
ACORD
(Ritz et al., 2007)
T1DM and T2DM,
n=172 patients
45 ml/min
per 1.73 m
2
Hb 105115 g/l or
Hb 130150 g/l
Left ventricular
mass reduction
No difference No difference No difference
Gouva et al. (2004) No diabetes patient,
n=88 patients
24 ml/min
per 1.73 m
2
Hb N90 g/l or
Hb N130 g/l
Doubling of
creatinine, or
dialysis, or death
Higher Hb level was
associated with lower
rate of primary end point
Idem No difference
Rossert et al. (2006) 100 (25.6%) patients
with diabetic nephropathy
among 390 patients
19 ml/min
per 1.73 m
2
Hb 110120 g/l or
Hb 130150 g/l
eGFR decline No difference No difference No difference
Macdougall et al. (2007) 44 (22.3%) diabetic patients
among 197 patients
20 ml/min
per 1.73 m
2
Hb N90 g/l or
Hb N110 g/l
Left ventricular
mass reduction
No difference No difference No difference
CREATE
(Drueke et al., 2006)
157 (26%) diabetic patients
among 603 patients
25 ml/min
per 1.73 m
2
Hb 105115 g/l or
Hb 130150 g/l
Cardiovascular
event
No difference No difference No difference
CHOIR (Singh et al., 2006) 699 (49.4%) patients with
diabetic nephropathy
among 1416 patients
27 ml/min
per 1.73 m
2
Hb 105110 g/l or
Hb 130135 g/l
Cardiovascular
event
No difference No difference No difference
242 E. Villar et al. / Journal of Diabetes and Its Complications 25 (2011) 237243
Brenner, B. M., Cooper, M. E., de Zeeuw, D., et al. (2001). Effects of losartan on renal and
cardiovascular outcomes in patients with type 2 diabetes and nephropathy. New
England Journal of Medicine, 345, 861869.
Casadevall, N., Nataf, J., Viron, B., et al. (2002). Pure red cell aplasia and
antierythropoietin antibodies in patients treated with recombinant erythropoietin.
New England Journal of Medicine, 346, 469475.
Chudleigh, R. A., Dunseath, G., Evans, J. N., et al. (2007). How reliable is estimation of
glomerular ltration rate at diagnosis of type 2 diabetes? Diabetes Care, 30,
300305.
Chudleigh, R. A., Dunseath, G., Peter, R., et al. (2008). Inuence of body weight on the
performance of glomerular ltration rate estimators in subjects with type 2
diabetes. Diabetes Care, 31, 4749.
Chudleigh, R. A., Ollerton, R. L., Dunseath, G., et al. (2008). Performance of the revised
175 Modication of Diet in Renal Disease equation in patients with type 2
diabetes. Diabetologia, 51, 17141718.
Drueke, T. B., Locatelli, F., Clyne, N., et al. (2006). Normalization of hemoglobin level in
patients with chronic kidney disease and anemia. New England Journal of Medicine,
355, 20712084.
Estacio, R. O., Jeffers, B. W., Gifford, N., & Schrier, R. W. (2000). Effect of blood pressure
control on diabetic microvascular complications in patients with hypertension and
type 2 diabetes. Diabetes Care, 23(Suppl 2), B54B64.
Foley, R. N., & Collins, A. J. (2007). End-stage renal disease in the United States: An
update from the United States Renal Data System. Journal of the American Society of
Nephrology, 18, 26442648.
Gouva, C., Nikolopoulos, P., Ioannidis, J. P., & Siamopoulos, K. C. (2004). Treating anemia
early in renal failure patients slows the decline of renal function: A randomized
controlled trial. Kidney International, 66, 753760.
Hayashi, T., Suzuki, A., Shoji, T., et al. (2000). Cardiovascular effect of normalizing the
hematocrit level during erythropoietin therapy in predialysis patients with chronic
renal failure. American Journal of Kidney Diseases, 35, 250256.
Hrl, W. H., Vanrenterghem, Y., Aljama, P., et al. (2007). OPTA: Optimal treatment of
anaemia in patients with chronic kidney disease (CKD). Nephrology, Dialysis,
Transplantation, 22(Supp 3), 2026.
Jungers, P., Choukroun, G., Oualim, Z., Robino, C., Nguyen, A. T., & Man, N. K. (2001).
Benecial inuence of recombinant human erythropoietin therapy on the rate of
progression of chronic renal failure in predialysis patients. Nephrology, Dialysis,
Transplantation, 16, 307312.
Leehey, D. J., Kramer, H. J., Daoud, T. M., Chatha, M. P., & Isreb, M. A. (2005). Progression
of kidney disease in type 2 diabetes Beyond blood pressure control: an
observational study. BMC Nephrol, 6, 8.
Levin, A., & Rocco, M. (2007). KDOQI Clinical Practice Guideline and Clinical Practice
Recommendations for anemia in chronic kidney disease: 2007 update of
hemoglobin target. American Journal of Kidney Diseases, 50, 471530.
Lewis, E. J., Hunsicker, L. G., Clarke, W. R., et al. (2001). Renoprotective effect of the
angiotensin-receptor antagonist irbesartan in patients with nephropathy due to
type 2 diabetes. New England Journal of Medicine, 345, 851860.
Macdougall, I. C., Temple, R. M., & Kwan, J. T. (2007). Is early treatment of anaemia with
epoetin alpha benecial to predialysis chronic kidney disease patients? Results of a
multicentre, open-label, prospective, randomized, comparative group trial.
Nephrology, Dialysis, Transplantation, 22, 784793.
Moranne, O., Froissart, M., Rossert, J., et al. (2009). Timing of onset of CKD-related
metabolic complications. Journal of the American Society of Nephrology, 20,
164171.
Orasanu, G., & Plutzky, J. (2009). The pathologic continuum of diabetic vascular disease.
Journal of the American College of Cardiology, 53(Suppl 5), 3542.
Parving, H. H., Lehneert, H., Brchner-Mortensen, J., et al. (2001). The effect of
irbesartan on the development of diabetic nephropathy in patients with type 2
diabetes. New England Journal of Medicine, 345, 870878.
Pfeffer, M. A., Burdmann, E. A., Chen, C. Y., et al. (2009). A trial of darbepoetin alfa in type
2 diabetes and chronic kidney disease. New England Journal of Medicine, 361,
20192032.
Ravanan, R., Spiro, J. R., Mathieson, P. W., & Smith, R. M. (2007). Impact of diabetes on
haemoglobin levels in renal disease. Diabetologia, 50, 2631.
Rigalleau, V., Lasseur, C., Raffaitin, C., et al. (2008). Progression-related bias in the
monitoring of kidney function in patients with diabetes and chronic kidney disease.
Metabolism, 57, 14281433.
Ritz, E., Laville, M., Bilous, R. W., et al. (2007). Target level for hemoglobin correction in
patients with diabetes and CKD: Primary results of the Anemia Correction in
Diabetes (ACORD) Study. American Journal of Kidney Diseases, 49, 194207.
Rossert, J., Levin, A., Roger, S. D., et al. (2006). Effect of early correction of anemia on the
progression of CKD. American Journal of Kidney Diseases, 47, 738750.
Singh, A. K., Szczech, L., Tang, K. L., et al. (2006). Correction of anemia with epoetin
alpha in chronic kidney disease. New England Journal of Medicine, 355,
20852098.
Skyler, J. S., Bergenstal, R., Bonow, R. O., et al. (2009). Intensive glycemic control and the
prevention of cardiovascular events: implications of the ACCORD, ADVANCE and
VA diabetes trials. A position statement of the American Diabetes Association and a
scientic statement of the American College of Cardiology Foundation and the
American Heart Association. Journal of the American College of Cardiology, 53,
298304.
Sorensen, V. R., Mathiesen, E. R., Heaf, J., & Feldt-Rasmussen, B. (2007). Improved
survival rate in patients with diabetes and end-stage renal disease in Denmark.
Diabetologia, 50, 922929.
Stengel, B., Billon, S., van Dijk, P. C., et al. (2003). Trends in the incidence of renal
replacement therapy for end-stage renal disease in Europe, 1990-1999. Nephrology,
Dialysis, Transplantation, 18, 18241833.
Thomas, M. C., MacIsaac, R. J., Tsalamandris, C., et al. (2004). The burden of anaemia in
type 2 diabetes and the role of nephropathy: A cross-sectional audit. Nephrol
DialTransplant, 19, 17921797.
Villar, E., Chang, S. H., & McDonald, S. P. (2007). Incidences, treatments, outcomes, and
gender effect on survival in end-stage renal disease patients by diabetic status in
Australia and New Zealand (1995-2005). Diabetes Care, 30, 30703076.
243 E. Villar et al. / Journal of Diabetes and Its Complications 25 (2011) 237243