Nephrol Dial Transplant (2010) 25: 1998–2004
doi: 10.1093/ndt/gfp779
Advance Access publication 25 January 2010
The ‘blood group O problem’ in kidney transplantation—time
to change?
Petra Glander1, Klemens Budde1, Danilo Schmidt1, T. Florian Fuller2, Markus Giessing3,
Hans-Hellmut Neumayer1 and Lutz Liefeldt1
1
Department of Nephrology, Charité—Universitaetsmedizin Berlin, Campus Charité Mitte, Berlin, Germany, 2Department of Urology,
Charité—Universitaetsmedizin Berlin, Campus Charité Mitte, Berlin, Germany and 3Department of Urology, Heinrich Heine
University, Düsseldorf, Germany
Correspondence and offprint requests to: Lutz Liefeldt; E-mail: [email protected]
Abstract
Background. Patients with blood group O have disadvantages in the allocation of deceased donor organs in the
Eurotransplant Kidney Allocation System and fewer
ABO-compatible living donors. In order to investigate
the consequences of this dilemma, we analysed the outcome of patients with blood group O in our transplantation
programme.
Methods. A single-centre analysis of 1186 waitlisted patients for first deceased donor kidney transplantations between 1996 and 2008 was performed, and the mechanisms
of blood group-dependant differences for graft and recipient outcome were assessed.
Results. Median follow-up time until death or end of observation for all waitlisted patients was 66 months (range,
0–158 months) and for 589 recipients of a kidney graft was
61 months (range, 0–158 months). Patients with blood
group O had significantly longer waiting times for deceased donor kidney grafts, compared to non-group O recipients (median waiting time, 85 vs 59 months). As a
consequence, blood group O patients had an increased risk
for death without transplantation (13.1% for O patients vs
9.6% for non-O patients; P < 0.05). Despite a good human
leukocyte antigen match, graft outcome tended to be worse
in O recipients; 14.1% (95% CI, 8.2–19.9%) of all O
kidneys from deceased donors were transplanted into
non-O recipients, leading to the accumulation of O recipients on the waiting list.
Conclusions. The export of blood group O donor kidneys
to other blood groups leads to longer waiting times, to a
higher death rate and to accumulation of blood group O
patients on the waiting list, which will further aggravate
the problem in the future. Our results should prompt further research on the issues associated with blood group O.
Current allocation systems and living donor kidney exchange programmes should be re-evaluated to address this
problem.
Keywords: allocation; blood group O; kidney transplantation; outcome;
waiting time
Introduction
End-stage renal disease (ESRD) is associated with high
morbidity and mortality. It has been shown that renal transplantation improves life expectancy as well as quality of
life [1–3]. Due to the shortage of deceased donors, kidneys
have to be allocated in a fair and transparent way to patients on the waiting list. Most allocation systems use parameters for a successful outcome in order to optimally
utilize the scarce organs. One important goal is to balance
the best human leukocyte antigen (HLA) matching vs increasing ischaemia times. The rationale for this is the
knowledge of superior outcomes of zero-mismatched kidney grafts [4] and for patients with primary graft function
[5]. A fair allocation system should also consider the waiting time of waitlisted patients, as longer waiting times increase morbidity and mortality and are associated with
inferior outcomes after transplantation [6]. The Eurotransplant Kidney Allocation System favours good HLA matching and, in order to achieve this, blood group O organs are
allocated to non-O recipients in the case of a zero mismatch HLA constellation [7]. Additionally, O kidneys
are allocated to recipients of other blood groups within
special programmes (e.g. Acceptable Mismatch Programme,
Eurotransplant Senior Programme), combined organ
transplantations and repeated kidney transplantations. Altogether, this results in a substantial drain of O kidneys,
which results in a lower chance for patients with blood
group O on the waiting list to receive a kidney transplant.
In the last decade, the deceased donor shortage and
longer waiting times have led to increasing numbers of
living donor kidney transplantations worldwide. Again,
group O patients have a lower chance of a blood
group-compatible donor, which is an additional disadvantage for these patients. Due to biological barriers, a
patient with blood group O is disadvantaged in receiving
a graft from a living donor and is potentially disadvantaged in receiving a deceased donor organ in a timely
manner due to the current Eurotransplant kidney allocation policy.
© The Author 2010. Published by Oxford University Press on behalf of ERA-EDTA. All rights reserved.
For Permissions, please e-mail: [email protected]
‘Blood group O problem’ in kidney transplantation
1999
We, therefore, hypothesized that longer waiting times
may lead to worse outcomes for waitlisted patients with
blood group O. This has not been analysed in a large cohort of patients to date, as transplant registries do not usually include the outcome of waitlisted patients. Other
databases only analyse dialysis patients or waitlisted patients without follow-up after transplantation or removal
from the waiting list. The examination of outcome during
the entire transplant experience as a whole (from the start
of renal replacement therapy, during the time spent on the
waiting list and following transplantation until graft failure
or death) should provide a more comprehensive view of
the situation. In order to investigate our hypothesis, we decided to perform a single-centre analysis in a large cohort
of waitlisted patients with complete follow-up over a 13year period.
centre, the patient was censored at the date of transfer. The end of the
observation period was 31 March 2009.
Endpoints included death on the waiting list, graft survival and patient
survival after the start of renal replacement therapy and after kidney transplantation.
We computed survival rates according to the Kaplan–Meier method
with censoring at the end of the observation time. Frequencies were compared using the chi-squared test. Numerical values were tested using the
Wilcoxon rank-sum test (two groups) or by Kruskal–Wallis test (more
than two groups).
In univariate regression analyses, several covariates were tested (waiting time, ≤60 vs >60 months; transplantation period, 1996–1999, 2000–
2004 and 2005–2008, respectively; donor type, living vs deceased; age at
transplantation, ≤45 vs >45 years; blood group, O vs non-O; cold ischemia time, ≤12 vs >12 h; histocompatibility, zero mismatch vs ≥1 mismatch). Cox multivariable regression analysis for graft survival was
performed for weighing of competing covariates. Data were 100% complete for all variables except for cold ischaemia time (94.4%). A P-value
of <0.05 was considered to be statistically significant. All statistical analyses were performed using SPSS for Windows release 16.0.0 (SPSS Inc.,
Chicago, IL, USA).
Materials and methods
A retrospective single-centre analysis was performed using data from a
web-based electronic patient record system, TBase. Since 1996, all patients evaluated for kidney transplantation, waitlisted, transplanted and/
or treated in our transplant centre were prospectively entered into the database [8,9]. For this analysis, data were retrieved from all adult (>16
years at the start of renal replacement therapy, >18 years at transplantation) patients awaiting a first kidney graft or who were transplanted between 1996 and 2008.
The current Eurotransplant Kidney Allocation System was implemented in 1996, which provided another reason to start data retrieval
from that year. Allocation within the Eurotransplant Senior Programme
(started by 1 January 1999) was excluded. All data were verified for
plausibility by an independent transplant coordinator. In order to reach
complete follow-up, all patients with incomplete data were contacted
between January and March 2009. In the case of a transfer to another
Results
The study population is shown in Figure 1. Between 1996
and 2008, 997 patients underwent renal replacement therapy for the first time. A total of 589 patients received a
first kidney graft, including 189 patients who started maintenance dialysis before 1996. Complete follow-up information was available for all patients, except for 24
waitlisted patients who were censored at the time of transfer to another transplant centre.
Death on the waiting list remained a serious threat for all
patients, especially in the context of long waiting times
First RRT
01.01.1996 - 31.12.2008
n=997
(Including pre-emptive KTX: n=43)
RRT-start before 1996
n=189
Waitlisted patients for first KTX
01.01.1996 – 31.12.2008
n=1186
No KTX during follow-up
01.01.1996-31.03.2009
n=597
First KTX during follow-up
01.01.1996-31.03.2009
n=589 (living donors: 214)
n=106
Transfer
to other
centres
n=24
Removal
from WL
n=48
(24 alive)
n=24
n=38
Death
Lost to
follow-up
n=130
n=0
Graft
loss
n=78
(64 alive)
n=14
Death
n=52
Fig. 1. Flowchart of study population [median follow-up of all 997 patients with start of renal replacement therapy (RRT) between 1996 and 2008 is 66
months (0–158 months); median follow-up of all 589 recipients of a kidney graft is 61 months (0–158 months)]; KTX, kidney transplantation; WL,
waiting list.
2000
P. Glander et al.
A
1,0
Hazard
0,8
0,6
0,4
0,2
0,0
0 12 24 36 48 60 72 84 96 108120132144
Time (months)
B
0,5
Hazard
0,4
0,3
0,2
0,1
0,0
0 12 24 36 48 60 72 84 96 108120132144
Time (months)
Fig. 2. Cumulative hazard for death of all candidates for first kidney
transplantation after initiation of renal replacement therapy between
1996 and 2007 (end of observation: 31 March 2009): (A) grey line, all
patients receiving an allograft; red line, waitlisted but still not grafted
patients (log rank: P < 0.00001); (B) grey line, ESRD patients with
blood groups A, B or AB; red line, blood group O patients (log rank:
P = 0.28).
(Figure 2). During the first 5 years, 9.6 waitlisted patients
per 100 patient-years had died, with a substantial further increase to 16.3 during the next 5 years. Patients with blood
group O had identical survival during the first 5 years after
the start of renal replacement therapy. However, beyond 6
years, survival curves started to separate between groups
(Figure 2B). Despite similar baseline characteristics (Table 1), patients with blood group O had a tendency towards
inferior long-term survival. Approximately 25% (95% CI,
17.8–32.2%) of patients with blood group O had died within 10 years after the start of renal replacement therapy, compared to 20% (95% CI, 15.5–24.5%) for non-O patients (n.
s.).
Therefore, the outcome of waitlisted in patients was investigated in greater detail with respect to removal from
the waiting list (Table 2). Patients with blood group O were
more likely to die on the waiting list than all other patients.
The total number of first kidney transplantation for all
waitlisted patients demonstrated that those with blood
group O had a significantly lower chance of any type of
graft, either from a living donor or a deceased donor. As
a consequence, the rate of first kidney transplantations in
blood group O recipients was by far the lowest. Patients
with blood group O also had a 74% higher risk of being
removed from the waiting list instead of being transplanted
than patients of other blood groups.
Data from all deceased donor kidney transplantations
were then analysed, the receipt of which was influenced
by the Eurotransplant allocation practice (Table 3). Patients with blood group O had significantly longer waiting
times compared to all other groups (median waiting times,
85 vs 59 months). As expected, patients with blood group
AB had the shortest waiting time. Longer waiting times
were associated with higher rates of zero-mismatch allocation and better HLA matching in our cohort. However, ischaemia times were not associated with blood group.
Patients with blood group A had similar degrees of HLA
match and comparable ischaemia times, but with significantly shorter median waiting times.
In order to investigate the reason for the observed longer
waiting times for O recipients, the ABO non-identical allocation of deceased donor kidneys in our transplant
programme was investigated. The proportion of patients
with blood group O was almost identical in waitlisted patients (365/997, 36.6%) and deceased donor organs (135/
375, 36.0%). However, patients with blood group O received only 30.9% of all kidneys (116/375). In total, 19/
135 (14.1%; 95% CI, 8.2–19.9%) donor kidneys from blood
group O donors were allocated to non-O recipients due to
the current allocation policy. Group AB and B recipients
Table 1. Baseline characteristics of the study population
Age at start of RRT (years)
Male gender (%)
Cause of ESRD [number (%)]
GN
DM
Polycystic
Nephrosclerosis
Other
Unspecified
Blood group O
(n = 450)
Non-O blood groups
(n = 736)
P-value
42.9 ± 12.6
63.3
43.8 ± 11.9
63.6
0.24
0.93
166 (36.7)
25 (5.5)
64 (14.2)
38 (8.4)
74 (16.4)
83 (18.4)
287 (39.0)
63 (8.6)
104 (14.1)
68 (9.2)
105 (14.3)
109 (14.8)
0.47
0.055
0.75
0.64
0.30
0.10
‘Blood group O problem’ in kidney transplantation
2001
Table 2. Outcomes of patients with ESRD on the waiting list between 1996 and 2008 (end of observation period: 31 March 2009)
Patient blood group
Removal due to death before kidney
transplantation
Removal from waiting list due to illness
Kidney transplantations (all)
Living donor kidney transplantation
Deceased donor kidney transplantation
Removals per 100 KTX
*
O
Non-O
A
B
AB
Number (%)
59/450 (13.1)*
71/736 (9.6)
50/481 (10.4)
16/192 (8.3)
5/63 (7.9)
Number
Number
Number
Number
n
6/450 (1.3)
174/450 (38.7)**
58/450 (12.9)**
116/450 (25.8)**
37.3***
18/736 (2.4)
415/736 (56.4)
156/736 (21.2)
259/736 (35.2)
21.4
10/481 (2.1)
263/481 (54.7)
105/481 (21.8)
158/481 (32.8)
22.8
7/192 (3.6)
109/192 (56.8)
39/192 (20.3)
70/192 (36.5)
21.1
1/63 (1.6)
43/63 (68.3)
12/63 (19.0)
31/63 (49.2)
13.9
(%)
(%)
(%)
(%)
P < 0.05 (one-sided exact Fisher’s test, O vs non-O patients).
P < 0.001 (chi-squared test, O vs non-O).
P < 0.005 (chi-squared test, O vs non-O).
**
***
Table 3. Waiting times and histocompatibility of first kidney transplantations between 1 January 1996 and 31 March 2009 (adult recipients, deceased
donors, n = 375)
Blood group
Waiting time (months)
Mean ± SD
Median (range)
Number of mismatches
Mean ± SD
Median (range)
Zero mismatch allocation
Number (%)
Cold ischemia time (h)
Mean ± SD
Median (range)
O (n = 116)
Non-O (n = 259)
A (n = 158)
B (n = 70)
AB (n = 31)
79.6 ± 37.2
85.0 (6–239)
57.5 ± 32.5
59.0 (0–282)
61.8 ± 33.9
71.0 (0–282)
58.9 ± 29.8
56.5 (2–173)
32.9 ± 17.4
37.0 (0–84)
1.70 ± 1.48
2 (0–5)
1.96 ±1.55
2 (0–6)
1.66 ± 1.50
2 (0–6)
2.34 ± 1.55
3 (0–6)
2.65 ± 1.43
3 (0–5)
40 (34.5)
76 (29.5)
56 (35.7)
16 (22.9)
4 (12.9)
Level of significance
P < 0.00001a
P < 0.001a
P = 0.025b
n.s.a
13.9 ± 6.6
13 (3.5–45.1)
13.7 ± 5.9
12.8 (3.5–34.0)
13.3 ± 5.4
12.5 (4.5–30.2)
13.7 ± 6.3
12.3 (4.5–28.8)
15.8 ± 7.3
15.0 (3.5–34.0)
a
Kruskal–Wallis test.
Chi-square test for crosstabs with more than two categories.
b
profited by ‘excess’ allocation of ABO non-identical
kidneys [blood group AB, 26 + 5 (19.2%) kidneys; blood
group B, 61 + 9 kidneys (14.7%)]. For group A recipients,
the ‘deficit’ (n = 2) and ‘excess’ (n = 7) were almost balanced [153 + 5 (3.3%) kidneys]. As a result, at the end of
the observation period, 47% of patients on our waiting list
were non-grafted O patients (Figure 3).
The question of whether the disadvantage of longer waiting times might be compensated by better outcomes after
kidney transplantation due to better histocompatibility
was then investigated. Although statistically not significant,
a tendency towards inferior patient survival was observed,
and there was a strong trend (P = 0.076) towards inferior
graft survival in O recipients compared to non-O recipients
(Figure 4). The combined analysis of the survival of living
and deceased donor kidney grafts revealed the same trend
(data not shown). In another sensitivity analysis, we compared deceased donor kidney recipients of blood groups O
and A that differed solely with respect to the waiting time.
This analysis revealed significantly better graft survival in A
recipients.
Finally, a multivariate analysis was performed on predictors for graft outcome after transplantation. Since the
donor category (living vs deceased donor) was the strongest predictor for graft (P < 0.001) and patient survival
(P < 0.05) in our cohort, further analyses were restricted
to recipients of deceased donor kidneys. In the unadjusted analysis for graft survival, the only significant covariate was histocompatibility, whereas blood group (O vs
non-O) produced only a trend towards worse graft outcome. Cold ischaemia time (≤ 12 vs > 12 h) and waiting
time (≤ 60 vs > 60 months) were not associated with
worse graft outcome (Table 4). With a cutoff at 84
months, which reflects the median waiting time for blood
group O patients in our cohort, the impact of waiting
time on the graft outcome reached statistical significance
(P = 0.024; hazard, 1.79; 95% CI, 1.08–2.95; data not
shown). Adjusted for the impact of cold ischaemia time
and histocompatibility in a proportional hazard model,
the hazard for graft failure increased by 68% (95% CI,
1.04–2.72) in blood group O recipients.
Discussion
Data from this large single-centre analysis, with a long and
complete follow-up, demonstrate that patients with blood
group O had significantly longer median waiting times
than all other patients (85 vs 59 months) and a significantly higher risk of death while they were on the waiting list.
In combination with fewer living donor transplantations,
this results in a lower rate of kidney transplantation in
2002
P. Glander et al.
B
A
A
1,0
36.6%
36.0%
C
16.3%
6.9%
Patient Survival
5.9%
16.0%
0,8
40.8%
41.4%
0,6
0,4
D
0,2
30.9%
47.0%
42.1%
36.0%
0,0
0
8.3%
12 24 36 48 60 72 84 96 108 120
Time (months)
18.7%
13.6%
B
1,0
3.3%
BG O
0,8
BG B
BG AB
Fig. 3. Accumulation of blood group O recipients on the kidney waiting
list: the distribution of blood groups among waitlisted patients (A) and
deceased kidney donors (B) is approximately similar; the O recipients
do not receive all O grafts (C), leading to their accumulation on the
waiting list at the end of observation (D).
O recipients. As a consequence, these patients accumulate
on the waiting list mainly because a sizable proportion
(14.1%) of blood group O deceased donor kidneys is allocated to non-O recipients. The observed accumulation
of O recipients will further aggravate the problem in the
future.
Long waiting times for a deceased donor kidney allograft may have serious consequences for the prognosis
of patients with ESRD, may influence expenditures of
the health systems and are associated with a lower quality
of life. As a consequence, every effort has to be made to
shorten waiting times. The Eurotransplant Kidney Allocation System incorporates waiting time into the current allocation algorithm but, as evidenced by our analysis, the
current allocation policy results in substantial blood
group-dependant differences. Longer waiting times inevitably lead to a higher mortality. Thus, the higher waiting
list mortality of blood group O patients reported here could
be expected, given the longer waiting times for blood
group O patients.
Unfortunately, longer waiting times are not only associated with higher waiting list mortality and morbidity, but
may also lead to inferior outcomes after transplantation. A
large analysis from the United States Renal Data System
Registry reported a highly significant negative impact of
long waiting times on outcomes after kidney transplanta-
Graft Survival
BG A
0,6
0,4
0,2
0,0
0
12 24 36 48 60 72 84 96 108 120
Time (months)
Fig. 4. Outcome after a first deceased donor KTX 1996–2008 according
to recipient blood group (red line, O recipients; grey line, non-O
recipients; blue line, A recipients): (A) patient survival not censored for
graft loss (n.s.); (B) graft survival not censored for death (log rank O vs
non-O recipients: P = 0.076; O vs A recipients: P = 0.043).
tion [6]. According to this analysis, waiting time was the
strongest modifiable risk factor for the outcome after kidney transplantation. The good HLA match of O recipients
in our cohort did not compensate for waiting time-associated inferior outcomes. Even though our analysis confirmed the positive effects of zero mismatch allocation in
the multivariate analysis, we found a tendency towards
worse patient and graft survival for patients with blood
group O. This implies that a longer waiting time on dialysis is responsible for an inferior health status (e.g. with
more calcification) at the time of transplantation, leading
to inferior long-term outcomes after transplantation.
Our ‘holistic’ approach to analysing the whole experience, from start of renal replacement therapy until graft
failure, has provided the first evidence of a serious disad-
‘Blood group O problem’ in kidney transplantation
2003
Table 4. Proportional hazard model of determinants predicting loss of deceased donor grafts after the first month
Covariate
Log-rank test
Unadjusted hazard (95% CI)
Adjusted hazard (95% CI)
Waiting time (≤60 vs >60 months)
Recipient blood group (O vs non-O)
Cold ischemia time (≤12 vs >12 h)
Mismatches (0 vs ≥1)
P
P
P
P
1.30
1.53
1.36
0.58
–
1.68 (1.04–2.72)
1.59 (0.96–2.65)
0.47 (0.27–0.84)
=
=
=
=
0.272
0.076
0.222
0.048
vantage for blood group O patients, given the longer
waiting times under the current Eurotransplant allocation
policy.
The main reason for longer waiting times of O recipients
is the transfer of blood group O organs to zero mismatch
non-O recipients. The results of our analysis with respect
to waiting times, as well as the transfer of grafts between
blood groups, are in accordance with the respective analyses in the German database BQS [10]. In 2007, O recipients received only about 85% of all O kidneys, whereas
non-O recipients received an ‘excess allocation’ (A,
+7%; B, +14%; AB, +27%). Given the fact that identical
numbers of donors and recipients have blood group O, the
constant drain of O kidneys has led to an accumulation of
O recipients on the waiting list. Again, our results are in
accordance with the Eurotransplant kidney waiting list,
which currently contains more than 51% O patients
(5478/10 687 by the end of 2008). Compared to 2004
(5740/11 960, 48.0%), the number of O patients on the
waiting list has grown substantially. With a constant loss
of O kidneys, the problem will get worse in the future
and, even if the allocation of O kidneys could immediately
be restricted to O recipients, it would take years to get a
balance, as only 37% (1187/3207) of kidneys transplanted
in 2008 were from blood group O donors [11].
The limitation of our study is its single-centre nature, a
rather small cohort of 1186 patients and a limited followup, especially after transplantation. The advantages of this
study are the ‘holistic’ approach in analysing patients from
the start of renal replacement therapy until death with or
without a kidney graft and the complete follow-up. Although we did not see statistical significance with respect
to all outcome measures in our cohort, we strongly believe
that it was necessary to report the strong trends observed
before they reached statistical significance. We also believe that a larger cohort or a longer follow-up would provide significant results with respect to hard endpoints,
such as patient and graft survival. Therefore, the aim of
the study was to highlight this important issue for our patients with blood group O and to start a discussion on a
better (not only allocation) policy.
In order to bridge the gap between blood group O and
the other blood groups with respect to waiting time (and
prognosis), as well as shortening waiting times for all patients, irrespective of their blood group, various measures
are needed:
(1) It is obvious that an increase in post-mortem organ donation rates would positively affect patients of all blood
groups who were waitlisted and not only for kidney
transplantation. This matter, plus the fact that mainte-
(0.81–2.08)
(0.95–2.45)
(0.83–2.22)
(0.33–0.99)
nance renal (organ) replacement therapy is an enormous
burden for our health systems, makes it a political issue.
(2) Allocation of kidneys from blood group O deceased
donors to recipients from other blood groups should
be prevented as much as possible. A debate on ‘the
blood group O problem’ is overdue and a change of
current Eurotransplant allocation practice should be
discussed. The prioritization of zero-mismatch allocation should be re-evaluated by balancing this against
higher death rates in the context of long waiting times
and more effective immunosuppression in a larger cohort such as the Eurotransplant kidney waiting list.
(3) Given the current shortage of deceased donors, living
donor kidney transplantation is one suitable option to
achieve higher transplantation rates and shorter waiting
times for all patients. Living donation kidney transplantation including the utilization of ABO-incompatible [12] as well as crossover programmes [13] and
acceptance of altruistic kidney donation [14] should
be encouraged, especially for blood group O recipients.
Conclusion
In summary, policy makers and transplant physicians are
faced with a serious imbalance on kidney transplant waiting lists, threatening the goal of fair and equal access to
deceased donor transplantation. In order to restore this balance, the issues associated with blood group O should be
discussed, with the aim of obtaining an urgent change in
the current situation.
References
1. Wolfe RA, Ashby VB, Milford EL et al. Comparison of mortality in
all patients on dialysis, patients on dialysis awaiting transplantation,
and recipients of a first cadaveric transplant. N Engl J Med 1999;
341: 1725–1730
2. Kimmel PL, Cohen SD, Weisbord SD. Quality of life in patients with
end-stage renal disease treated with hemodialysis: survival is not
enough! J Nephrol 2008; 21: S54–S58
3. Griva K, Jayasena D, Davenport A et al. Illness and treatment cognitions and health related quality of life in end stage renal disease.
Br J Health Psychol 2009; 14: 17–34
4. Opelz G, Döhler B. Effect of human leukocyte antigen compatibility on kidney graft survival: comparative analysis of two decades.
Transplantation 2007; 84: 137–143
5. Quiroga I, McShane P, Koo DD et al. Major effects of delayed graft
function and cold ischaemia time on renal allograft survival. Nephrol
Dial Transplant 2006; 21: 1689–1696
6. Meier-Kriesche HU, Kaplan B. Waiting time on dialysis as the strongest modifiable risk factor for renal transplant outcomes: a paired donor kidney analysis. Transplantation 2002; 74: 1377–1381
7. Eurotransplant International Foundation. http://www.eurotransplant.nl.
2004
8. Schröter K, Lindemann G, Fritsche L. TBase2—a web-based electronic patient record. Fundamenta Informaticae 43. Amsterdam:
IOS Press, pp. 343–353
9. Lindemann G, Fritsche LWeb-based patient records—the design of
TBase2. In: Köckerling Bruch, Schug-Pa Bouchard (eds). New Aspects of High Technology in Medicine 2000, pp. 409–414
10. BQS-Outcome 2008. http://www.bqs-outcome.de.
11. Eurotransplant International Foundation. Statistics. Annual Report.
http://www.eurotransplant.nl/?id=annual_report.
12. Tydén G, Donauer J, Wadström J et al. Implementation of a Protocol for ABO-incompatible kidney transplantation—a three-center
H. Ekberg et al.
experience with 60 consecutive transplantations. Transplantation
2007; 83: 1153–1155
13. de Klerk M, Witvliet MD, Haase-Kromwijk BJ et al. A highly
efficient living donor kidney exchange program for both blood
type and crossmatch incompatible donor-recipient combinations.
Transplantation 2006; 82: 1616–1620
14. Rees MA, Kopke JE, Pelletier RP et al. A nonsimultaneous, extended,
altruistic-donor chain. N Engl J Med 2009; 360: 1096–1101
Received for publication: 27.9.09; Accepted in revised form: 28.12.09
Nephrol Dial Transplant (2010) 25: 2004–2010
doi: 10.1093/ndt/gfp778
Advance Access publication 26 January 2010
Cyclosporine, tacrolimus and sirolimus retain their distinct toxicity
profiles despite low doses in the Symphony study
Henrik Ekberg1, Corrado Bernasconi2, Jana Nöldeke2, Alexander Yussim3, Lars Mjörnstedt4,
Uģur Erken5, Markus Ketteler6 and Pavel Navrátil7
1
Department of Nephrology and Transplantation, Lund University, University Hospital, Malmö, Sweden, 2Pharmaceuticals Division,
F. Hoffmann-La Roche Ltd, Basel, Switzerland, 3Transplantation Department, Rabin Medical Center, Petah Tikva Israel, 4Department
of Transplantation and Liver Surgery, Sahlgrenska University Hospital, Gothenburg, Sweden, 5Transplantation Center, Çukurova
University of Medicine, Yuregir, Adana, Turkey, 6III Medizinische Klinik (Division of Nephrology), Klinikum Coburg, Coburg,
Germany and 7Department of Urology, University Hospital Hradec Králové, Hradec Králové, Czech Republic
Correspondence and offprint requests to: Henrik Ekberg; E-mail: [email protected]
Abstract
Background. Reducing side effects of immunosuppressive regimens has become a priority in transplantation
medicine because of the large number of patients and
grafts that succumb to infection in the short term and cardiovascular disease in the long term. The Symphony
study was a 12-month prospective, randomized, open-label,
multi-centre, four parallel arm study that aimed to evaluate
the safety and efficacy of low-dose immunosuppressive
regimens compared with a standard-dose regimen in renal
transplant recipients. This sub-analysis focuses on specific
toxicities observed with the low-dose regimens.
Methods. Adult patients (n = 1645) scheduled to undergo
renal transplantation received low-dose cyclosporine (CsA),
tacrolimus (Tac) or sirolimus (SRL) in addition to daclizumab induction or standard-dose cyclosporine without
induction. All patients received mycophenolate mofetil
and corticosteroids. We evaluated the incidence of adverse
events (AEs), tested specific group differences and assessed the relationship of selected AEs with drug levels.
Results. The four arms had similar incidences of AEs, but
serious AEs were more common with low-dose SRL and
led to more discontinuations. Infections were the most
common AEs, with the highest incidence in the standarddose CsA group, in particular, cytomegalovirus (CMV)
infections. Low-dose Tac had the most reports of new-on-
set diabetes, leucopenia and diarrhoea. Low-dose SRL
negatively influenced triglycerides, wound healing, lymphocele and anaemia. We found only weak relationships
between specific AEs and drug levels.
Conclusions. Despite the low doses, CsA, Tac and SRL
retained distinct and different toxicity profiles. These findings may be of relevance for tailoring specific immunosuppressive regimens to patients with particular needs.
Keywords: calcineurin inhibitors; cyclosporine; drug toxicity;
mycophenolate mofetil; sirolimus
Introduction
In renal transplantation, chronic immunosuppression is associated with considerable risks, in particular, related to infections and cardiovascular diseases, which are the predominant
causes of death in those with a functioning graft [1]. Therefore, reducing toxicities has become a priority.
So far, most efforts have been focused on minimization of calcineurin inhibitors (CNIs). This drug class,
which includes cyclosporine A (CsA) and tacrolimus
(Tac), is associated with nephrotoxicity, hypertension and
hyperlipidaemia [2–8]. Compared to CsA, Tac is asso-
© The Author 2010. Published by Oxford University Press on behalf of ERA-EDTA. All rights reserved.
For Permissions, please e-mail: [email protected]