j o u r n a l o f s u r g i c a l r e s e a r c h 1 8 0 ( 2 0 1 3 ) 1 7 6 e1 8 2

Available online at www.sciencedirect.com

journal homepage: www.JournalofSurgicalResearch.com

Hypothermic machine perfusion in deceased donor kidney

transplantation: a systematic review

Vincent W.T. Lam, MBBS, MS, FRACS,a,b,* Jerome M. Laurence, MBChB, MRCS, FRACS, PhD,a,b
Arthur J. Richardson, MBBS, FRACS,a,b Henry C.C. Pleass, MD, FRCS, FRACS,a,b,c
and Richard D.M. Allen, MBBS, FRACSa,b,c
a
Discipline of Surgery, Sydney Medical School, University of Sydney, New South Wales, Australia
b
Department of Surgery, Westmead Hospital, Sydney, Australia
c
Transplantation Services, Royal Prince Alfred Hospital, Sydney, Australia

article info abstract

Article history: Background: Hypothermic machine perfusion (HMP) of kidneys is intended to mitigate the
Received 30 June 2012 deleterious effects of cold storage on organ quality, particularly when the cold ischemic
Received in revised form time is prolonged or the donor is otherwise marginal. The use of HMP has remained
5 October 2012 controversial; however, a number of randomized controlled trials (RCTs) have recently
Accepted 26 October 2012 been conducted to clarify its benefits.
Available online 15 November 2012 Methods: We undertook a systematic search of the Medline and Embase databases and of
the Cochrane Central Register of Controlled Trials. We included only RCTs in the meta-
Keywords: analysis. Outcomes analyzed were the incidence of delayed graft function (DGF), primary
Kidney transplantation nonfunction (PNF), graft loss, and patient death at 1 y.
Renal transplantation Results: We identified seven RCT trials and subjected them to meta-analysis, including 1353
Deceased donor kidney transplant recipients. Hypothermic machine perfusion significantly reduced the
Machine perfusion incidence of DGF (risk ratio [RR] 0.83, 95% confidence interval [CI] 0.72e0.96). There was no
Hypothermic machine perfusion difference in the incidence of PNF (RR 0.78, 95% CI 0.36e1.68), graft loss at 1 y (RR 0.87, 95%
Delayed graft function CI 0.64e1.19), and patient death at 1 y (RR 0.91, 95% CI 0.60e1.37) between HMP and donor
Systematic review kidneys preserved using cold storage.
Meta-analysis Conclusions: There are few RCT comparing HMP and cold storage of kidneys in deceased
donor kidney transplantation. Although these studies are small and heterogeneous in
design, HMP appeared to be associated with a reduced incidence of DGF. No difference
in the incidence of PNF, graft loss, or patient death at 1 y could be demonstrated.
Crown Copyright ª 2013 Published by Elsevier Inc. All rights reserved.

1. Introduction ischemic injury to the graft before and during organ

Delayed graft function (DGF) and primary nonfunction (PNF)
are well-known complications after kidney transplantation.
The frequency of DGF and PNF in deceased donor kidney
transplantation ranges from 2% to 50% and 2% to 15%,
respectively [1]. Delayed graft function is usually the result of
procurement, and is further aggravated by the reperfusion
syndrome [2]. Optimization of graft quality under conditions
of prolonged ischemia thus represents the primary goal of
deceased donor kidney preservation. To date, the predomi-
nant organ preservation method of deceased donor kidneys
used by most centers is static cold storage (CS). Cold storage

* Corresponding author. Department of Surgery, Westmead Hospital, PO Box 533, Wentworthville, NSW 2145, Australia. Tel.: þ61 2 9845
7365; fax: þ61 2 9893 7440.
E-mail address: vincent.lam@sydney.edu (V.W.T. Lam).
0022-4804/$ e see front matter Crown Copyright ª 2013 Published by Elsevier Inc. All rights reserved.
http://dx.doi.org/10.1016/j.jss.2012.10.055
 j o u r n a l o f s u r g i c a l r e s e a r c h 1 8 0 ( 2 0 1 3 ) 1 7 6 e1 8 2 177

preservation is based on the suppression of metabolism by

hypothermia. Blood is removed by flushing the kidney and
replaced with an appropriate cooled preservation solution [3].
The concept of hypothermic machine perfusion (HMP) of
deceased donor kidneys, which had been developed before the
routine use of CS preservation, provides an attractive alter-
native [4]. Hypothermic machine perfusion is based on
a controlled continuous circulation of a perfusate, which,
although not oxygenated or refreshed during storage, has
been hypothesized to protect the deceased donor kidneys
from injuries related to ischemia and reperfusion [5]. This
protection may be particularly important when the cold
ischemic time is prolonged or the donor is otherwise marginal.
Several clinical studies have compared HMP with CS
preservation of deceased donor kidneys. Early reports showed
no significant difference in DGF or graft survival rates when
comparing HMP and CS preservation [6,7]. In 2003, Wight et al
[8] published a systematic review and meta-analysis, and
demonstrated a 20% reduction in DGF when HMP was used.
More recently, two randomized controlled trials (RCTs) using
modern immunosuppression were conducted and showed
contradictory results [9,10].
The primary aim of this study was to assess evidence for the
effect of HMP compared with CS of donor kidneys on delayed
graft function using a systematic review and meta-analysis of
the available RCTs. We assessed the impact of HMP on rates
of primary nonfunction, graft loss, and patient death at 1 y.

2. Methods

2.1. Search methods

We undertook a comprehensive search through October 2011,

including the Cochrane Central Register of Controlled Trials,
MEDLINE, and EMABSE, with additional reports identified
from hand-searching conference proceedings and reference
lists of included studies. Two reviewers (V.L. and J.L.) inde-
pendently performed search strategies. When there was more
than one report from an individual trial, all reports were
examined for any available data. The publication reporting the
most outcome data was designated the index publication, and
is referenced in this report. Figure 1 shows the search strategy.

2.2. Selection criteria and study outcomes

We included all RCTs of parallel design comparing HMP versus

CS of kidneys for deceased donor kidney transplantation,
irrespective of blinding, sample size, publication status (i.e.,
whether published as the full text or presented only as an
abstract at a conference), and language. We excluded quasi-
randomized trials and other study designs. Primary
outcomes analyzed were the incidence of DGF and graft loss at
1 y. Secondary outcomes sought were the incidence of PNF
and patient death at 1 y.
2.3. Data collection and analysis
Two reviewers (V.L. and J.L.) independently screened search
results using titles, abstracts, and, where necessary, the full
Fig. 1 e Flowchart showing the process of identifying
randomized controlled trials for inclusion in systematic
review. RCT [ randomized controlled trial.
paper version to identify eligible trials. Two reviewers
extracted data independently, using a standardized form.
We assessed methodological quality of trials according to
the method recommended by the Cochrane Collaboration
[11]. We analyzed data using Review Manager (RevMan,
Version 5.1.4; The Cochrane Collaboration, Copenhagen,
Denmark).
178 j o u r n a l o f s u r g i c a l r e s e a r c h 1 8 0 ( 2 0 1 3 ) 1 7 6 e1 8 2

Table 1 e Characteristics of trials included in review.

Study Year Donor Preservation HMP model CS perfusate Initial
type method immunosuppression
(BDD/NHBD) (HMP/CP)

Mozes et al [15] 1985 All BDD 93/94 Waters MOX-100 Euro-Collin solution 38 AZA and steroids
120 ALG
29 CYC
Heil et al [17] 1987 All BDD 27/27 Waters MOX-100 Euro-Collin solution NS
Halloran and 1987 All BDD 91/90 Waters MOX-100 Euro-Collin solution CsA and steroids
Aprile [16] AZA and steroids
ALG, AZA  CsA and
steroids
Jaffers and 1989 All BDD 136/66 Waters MOX-100 Euro-Collin solution CsA
Banowsky [18] 10 kidneys with
Gambro machine
Kwiatkowski 2009 All BDD 37/37 Waters MOX-100 NS CsA, AZA, and steroids
et al [19]
Moers et al [9] 2009 588 BDD 336/336 LifePort NS Majority CsA or Tac-based
84 NHBD Kidney therapy
Transporter machines
Watson et al [10] 2010 All NHBD 45/45 LifePort University of Basliximab, Tac, MMF, and
Kidney Wisconsin solution steroids
Transporter machines

AZA ¼ azathioprine; ALG ¼ antilymphocyte globulin; BDD ¼ brain death donor; CsA ¼ cyclosporine; MMF ¼ mycophenolate sodium; NS ¼ not
stated; Tac ¼ tacrolimus.

2.4. Quantitative data synthesis and data analysis
We compared interventions with data synthesis using the
random effects method of DerSimonian and Laird [12]. We
expressed results for dichotomous outcomes as the risk ratio
(RR) with 95% confidence intervals (CIs), with values <1
favoring HMP.
We assessed heterogeneity among trials using inspection of
forest plots and Cochran’s Q test, and calculating the I2 statistic
(P < 0.05 and > 50%, respectively, considered evidence of
significant heterogeneity) to measure the proportion of total
variation resulting from heterogeneity beyond chance [13].
We performed subgroup analysis for the primary and
secondary outcomes to explore important clinical differences
among trials that might be expected to alter the magnitude of
treatment effects. Subgroups, defined a priori, were publica-
tion year (before 2000 versus after 2000), type of deceased
donor (heart-beating donor [HBD] versus noneheart-beating
donor [NHBD]) and specific immunosuppression regimen
used. We chose these subgroups because older immunosup-
pression regimens and NHBD have been shown to be asso-
ciated with poorer outcomes after kidney transplantation
[1,14]. In addition, we assessed publication bias by using
funnel plots of the log risk ratio.
3. Results
3.1. Systematic review and meta-analysis
3.1.1. Included trials
We included seven randomized controlled trials comparing
HMP with CS of donor kidneys for deceased donor kidney
transplantation, including a total of 1353 donor kidneys and
1353 kidney transplant recipients, in the review [9,10,15e19].
Table 1 depicts the characteristics of these trials. Four of
the seven trials were published before 1990, whereas the
remaining three were published after 2000. Five of the seven
trials included only HBD. The two most recent studies included
12.5% and 100% of NHBD [9,10]. Five trials used the Waters
MOX-100 Transport Unit (Waters Instruments Inc., Rochester,
NY); the remaining two used the LifePort Kidney Transporter
machines (Organ Recovery Systems, Chicago, IL) for HMP. Four
early trials used Euro-Collin solution for CS perfusate, whereas
one used University of Wisconsin solution. Two trials did not
specify the CS perfusate. No specific initial immunosuppres-
sion regimen was used among the seven trials.
3.1.2. Quality of included trials
There was some deficiency of reporting of methods for most
trials. Table 2 depicts the risk of bias summary. Sequence
generation was adequate in two trials and unclear in the
remaining five. Allocation concealment was judged to be
adequate to minimize selection bias in two trials, unclear in
four, and inadequate in one. Blinding was judged to be
adequate to prevent bias for objective outcomes in two trials,
unclear in four, and inadequate in one. Blinding methodology
was judged to be adequate to prevent bias for outcomes in two
trials and unclear in five. The quality of outcome data
reporting was adequate in two trials, unclear in two, and
inadequate in three. There was no selective reporting of
outcomes in two trials; in four trials this was unclear, and in
one trial selective reporting was evident. Other bias was
categorized as unclear in all seven trials.
3.1.3. Outcomes
All seven trials reported the incidence of DGF after trans-
plantation. Five trials defined DGF as the requirement of
 j o u r n a l o f s u r g i c a l r e s e a r c h 1 8 0 ( 2 0 1 3 ) 1 7 6 e1 8 2
Table 2 e Risk of bias summary: Review authors’ judgments about each risk of bias item for each included study.
Random Allocation Blinding of
sequence concealment participants
generation (selection and personnel
(selection bias) (performance
bias) bias)
Halloran and
Aprile [16]
Heil et al [17]
Jaffers and
Banowsky [18]
Kwiatkowski
et al [19]
Moers et al [9]
Mozes et al [15]
Watson et al [10]
(Color version of Table is available online.)
dialysis within 1 wk of transplantation. One trial defined DGF
as the requirement of dialysis after transplantation with no
specific time period, and one other trial did not provide
a definition of DGF. The incidence of DGF was significantly
reduced in the HMP group compared with the CS group
(RR ¼ 0.83; 95% CI 0.72e0.96; P ¼ 0.01) (Fig. 2A). There was no
evidence of heterogeneity (I2 ¼ 0% and P ¼ 0.50). Funnel plots
were also symmetrical, suggesting the absence of publication
bias (Fig. 3A).
Only three trials reported the incidence of PNF after
transplantation. The definition of PNF was different in all
three studies: namely, “never attaining plasma creatinine
<500 mmol/L and/or always requiring dialysis”; or “perma-
nent lack of function of the allograft from the time of
transplantation”; or “a graft that failed to provide 1-mo
dialysis-free survival excluding losses attributable to rejec-
tion or vascular thrombosis.” Although there was a trend
favoring the use of HMP, the incidence of PNF was not
significantly different between HMP and CS groups
(RR ¼ 0.76; 95% CI 0.45e1.28; P ¼ 0.30) (Fig. 2B). Heterogeneity
was evident but not statistically significant among the trials
(I2 ¼ 37% and P ¼ 0.21). For PNF, we did not assess publication
bias because of the small number of trials available for
analysis.
Six trials reported the incidence of graft loss at 1 y.
The incidence of graft loss at 1 y was not significantly different
between HMP and CS groups (RR ¼ 0.87; 95% CI 0.64e1.19;
P ¼ 0.39) (Fig. 2C). There was moderate although not statisti-
cally significant heterogeneity among the trials (I2 ¼ 42% and
P ¼ 0.12). There was no funnel plot asymmetry suggesting the
absence of publication bias (Fig. 3B).
179
Blinding of Incomplete Selective Other
outcome outcome data reporting bias
assessment (attrition (reporting
(detection bias) bias)
bias)
Four trials reported the incidence of patient death at 1 y.
Although there was a trend favoring the use of HMP, the
incidence of patient death at 1 y was not significantly different
between HMP and CS groups (RR ¼ 0.64; 95% CI 0.60e1.37; P ¼
0.64) (Fig. 2D). There was no evidence of heterogeneity among
the trials (I2 ¼ 0% and P ¼ 0.41). For patient death at 1 y, we did
not assess publication bias because of the small number of
trials included for analysis.
3.1.4. Subgroup analyses
W performed subgroup analyses for publication year and type
of deceased donor, but not for specific immunosuppression
regimen used, because there were insufficient data for anal-
ysis of this variable. Two outcomes were reported sufficiently
frequently to allow investigation: namely, DGF and graft loss
at 1 y. There was no evidence of a difference for DGF and graft
loss at 1 y on the basis of publication year or type of deceased
donor (Table 3).
4. Discussion
Cold storage is the most widely used preservation method in
kidney transplantation today. This meta-analysis of seven
randomized controlled trials involving 1353 kidney transplant
recipients showed that whereas HMP of deceased donor
kidneys significantly reduced the incidence of DGF, it had no
effect on PNF, graft loss at 1 y, and patient death at 1 y for both
HBD and NHBD. These results are consistent with a previous
systematic review in 2003 showing that HMP reduced the
incidence of DGF but had no difference on 1-y graft survival
180 j o u r n a l o f s u r g i c a l r e s e a r c h 1 8 0 ( 2 0 1 3 ) 1 7 6 e1 8 2

Fig. 2 e Forest plots illustrating the meta-analysis of outcomes in HMP versus CS of donor kidneys for deceased donor
kidney transplantation. The outcomes analyzed were (A) DGF, (B) PNF, (C) graft loss at 1 y, (D) patient death at 1 y. (Color
version of figure is available online.)

[8]. In addition, a US registry database analysis of 85,734
transplant kidneys also showed a significant reduction in DGF
rate of 19.6% in the HMP group compared with 27.6% in the CS
group [20].
The rationale for using HMP is based on the concept of
improving donor organ quality by both providing organs with
nutrients and simultaneously eliminating toxic metabolic
cellular products through continuous perfusate flow [3]. In
theory, HMP would protect the donor kidneys from injuries
related to ischemia or reperfusion, and thus reduce the inci-
dence of DGF. Reducing the incidence of DGF would be ex-
pected to lead to less use of dialysis postoperatively as well as
a shorter length of hospital stay [2]. Although DGF has been
shown to be a risk factor for graft failure after kidney trans-
plantation [2], this meta-analysis did not show improvement
in graft loss at 1 y for recipients of organs preserved using
HMP. The lack of improvement in graft survival found in these
studies may be a consequence of a relatively short follow-up
horizon. However, there is precedence for this finding in
data showing similar medium-term graft survival for HBD and
NHBD kidneys despite higher rates of DGF in recipients of
NHBD kidneys [21e23].
To date, because demand far exceeds the supply, NHBD
has become an important source of additional donor kidney
throughout the world. Organs from NHBD are subjected to
a period of warm ischemia, which affects post-transplant
graft function adversely [24]. Previous studies have sug-
gested that HMP of NHBD kidneys results in better early
function and improved graft survival compared with kidneys
preserved by CS [25,26]. However, two recent RCTs showed
 j o u r n a l o f s u r g i c a l r e s e a r c h 1 8 0 ( 2 0 1 3 ) 1 7 6 e1 8 2
Fig. 3 e Funnel plots of comparison of HMP versus CS of
donor kidneys for deceased donor kidney transplantation
for (A) DGF and (B) graft loss at 1 year. SE [ standard
error. (Color version of figure is available online.)
contradictory results. Watson et al [10] reported no superiority
of HMP over CS for NHBD kidneys, whereas Jochmans et al [27]
reported that HMP was associated with a reduced incidence of
DGF and better early graft function up to 1 mo after trans-
plantation [27]. The perceived different outcome of these two
studies might be related to the differences in study design and
statistical approach. Another important difference between
studies is the use of antithymocyte globulin, which was
administered to about 15% of recipients in the study by Joch-
mans et al, whereas it was not given to patients in the study by
Watson et al. Nevertheless, this meta-analysis demonstrated
no benefit of HMP over CS in the incidence of DGF on subgroup
analysis.
Table 3 e Subgroup analysis of publication year and type of deceased donor for outcome of DGF and graft loss at 1 y.
Potential bias Subgroup strata
Number
of trials
Publication year <2000 4 0.82
2000 3 0.82
Type of deceased HBD 6 0.80
donor NHBD 2 0.91
* P value for test of interaction.
181
A number of factors may have confounded the interpre-
tation of this meta-analysis. First, most of the studies
analyzed included a relatively small number of patients, and
the methodological quality was poor or uncertain. Second,
because the studies included span over 25 y, there have been
substantial changes in surgical techniques, perfusate, and
immunosuppression regimens. Even for the two most
recently published RCTs, the timings of the start of HMP were
different. In the trial by Moers et al [9], donor kidneys under-
went an initial period of CS of several hours before starting
HMP in the perfusion laboratory, whereas donor kidneys
underwent immediate perfusion after procurement in the
trial by Watson et al [10]. The results of these two trials were
contradictory, and it was proposed that the benefit of HMP
may require immediate rather than delayed use of HMP in
deceased donor kidneys [28]. Despite this, we observed rela-
tively little evidence of heterogeneity in statistical measures
in this meta-analysis. Patterns of care and practices in kidney
transplantation are so variable that the heterogeneity
observed in clinical trials mimics the conditions under which
HMP would be implemented in the real world. In addition,
there was no difference in DGF on subgroup analyses
according to publication year, which suggests no clear
temporal trend in the risk of this complication. In this study,
we have included only data from RCT. In contrast, an earlier
systematic review and meta-analysis combined the results of
RCT with data from other nonrandomized studies [8]. Because
we included no data from nonrandomized studies, there is
less potential for bias associated with other types of study
design, and the findings are therefore more robust.
Hypothermic machine perfusion is associated with
increased direct costs to the transplant program, principally
because of the capital costs of the machine and the disposables
required to perfuse each organ. It has been suggested that
these costs may be substantially or totally defrayed by reduced
requirement for hospital stay and dialysis associated with
a reduced incidence of DGF and improved graft survival [29].
One recent economic evaluationebased data from an interna-
tional RCT showed that HMP reduced the risk of DGF and graft
failure at lower costs than CS. The cost savings was $86,750
per life-year gained and the incremental cost-utility ratio
was e$496,223 per quality-adjusted life-year gained in favor of
HMP. The authors concluded that life-years and quality-
adjusted life-years could be gained while reducing costs when
HMP of deceased donor kidneys was performed [30].
The meta-analysis demonstrates that hypothermic
machine perfusion of deceased donor kidneys may be
DGF Graft loss at 1 y
RR (95% CI) P value* No of trials RR (95% CI) P value*
(0.67e1.00) 0.97 3 0.90 (0.71e1.14) 0.54
(0.66e1.01) 3 0.78 (0.52e1.16)
(0.67e0.94) 0.40 Insufficient data
(0.70e1.17)
182 j o u r n a l o f s u r g i c a l r e s e a r c h 1 8 0 ( 2 0 1 3 ) 1 7 6 e1 8 2
associated with a reduced incidence of delayed graft function
after kidney transplantation, and therefore may offer a signifi-
cant clinical benefit for recipients of deceased donor kidneys.
references
[1] Perico N, Cattaneo D, Sayegh MH, et al. Delayed graft
function in kidney transplantation. Lancet 2004;364:1814.
[2] Yarlagadda SG, Coca SG, Formica RN Jr, et al. Association
between delayed graft function and allograft and patient
survival: a systematic review and meta-analysis. Nephrol
Dial Transplant 2009;24:1039.
[3] Timsit MO, Tullius SG. Hypothermic kidney preservation:
a remembrance of the past in the future? Curr Opin Organ
Transplant 2011;16:162.
[4] Belzer FO, Ashby BS, Dunphy JE. 24-hour and 72-hour
preservation of canine kidneys. Lancet 1967;2:536.
[5] Yuan X, Theruvath AJ, Ge X, et al. Machine perfusion or cold
storage in organ transplantation: indication, mechanisms,
and future perspectives. Transpl Int 2010;23:561.
[6] Clark EA, Terasaki PI, Opelz G, et al. Cadaver-kidney
transplant failures at one month. N Engl J Med 1974;291:1099.
[7] van der Vliet JA, Vroemen JP, Cohen B, et al. Preservation of
cadaveric kidneys. Cold storage or machine perfusion? Arch
Surg 1983;118:1166.
[8] Wight JP, Chilcott JB, Holmes MW, et al. Pulsatile machine
perfusion vs. cold storage of kidneys for transplantation:
a rapid and systematic review. Clin Transplant 2003;17:293.
[9] Moers C, Smits JM, Maathuis MH, et al. Machine perfusion or
cold storage in deceased-donor kidney transplantation. N
Engl J Med 2009;360:7.
[10] Watson CJ, Wells AC, Roberts RJ, et al. Cold machine
perfusion versus static cold storage of kidneys donated after
cardiac death: a UK multicenter randomized controlled trial.
Am J Transplant 2010;10:1991.
[11] Higgins JP, Green S. Cochrane handbook for systematic
reviews of intervention, vol. 5.1.0 [updated March 2011]. In:
The Cochrane Collaboration, 2011. Available from: http://
www.cochrane.org/training/cochrane-handbook.
[12] DerSimonian R, Laird N. Meta-analysis in clinical trials.
Control Clin Trials 1986;7(3):177.
[13] Higgins JP, Thompson SG, Deeks JJ, et al. Measuring
inconsistency in meta-analyses. BMJ 2003;327:557.
[14] Hoogland ER, Snoeijs MG, van Heurn LW. DCD kidney
transplantation: results and measures to improve outcome.
Curr Opin Organ Transplant 2010;15:177.
[15] Mozes M, Finch W, Reckard C, et al. comparison of cold
storage and machine perfusion in the preservation of
cadaver kidneys. Transplant Proc 1985;17:1474.
[16] Halloran P, Aprile M. A randomized prospective trial of cold
storage versus pulsatile perfusion for cadaver kidney
preservation. Transplantation 1987;43:827.
[17] Heil JE, Canafax DM, Sutherland DE, et al. A controlled
comparison of kidney preservation by two methods:
machine perfusion and cold storage. Transplant Proc 1987;
19:2046.
[18] Jaffers GJ, Banowsky LH. The absence of a deleterious effect
of mechanical kidney preservation in the era of cyclosporine.
Transplantation 1989;47:734.
[19] Kwiatkowski A, Wszola M, Kosieradzki M, et al. The early
and long term function and survival of kidney allografts
stored before transplantation by hypothermic pulsatile
perfusion. A prospective randomized study. Ann Transplant
2009;14:14.
[20] Schold JD, Kaplan B, Howard RJ, et al. Are we frozen in
time? Analysis of the utilization and efficacy of pulsatile
perfusion in renal transplantation. Am J Transplant 2005;5:
1681.
[21] Gok MA, Buckley PE, Shenton BK, et al. Long-term renal
function in kidneys from non-heart-beating donors: a single-
center experience. Transplantation 2002;74:664.
[22] Keizer KM, de Fijter JW, Haase-Kromwijk BJ, et al. Non-
heart-beating donor kidneys in the Netherlands: allocation
and outcome of transplantation. Transplantation 2005;79:
1195.
[23] Nicholson ML, Metcalfe MS, White SA, et al. A comparison of
the results of renal transplantation from non-heart-beating,
conventional cadaveric, and living donors. Kidney Int 2000;
58:2585.
[24] Shiroki R, Hoshinaga K, Higuchi T, et al. Prolonged warm
ischemia affects long-term prognosis of kidney transplant
allografts from non-heart-beating donors. Transplant Proc
1998;30:111.
[25] Moustafellos P, Hadjianastassiou V, Roy D, et al. The
influence of pulsatile preservation in kidney transplantation
from non-heart-beating donors. Transplant Proc 2007;39:
1323.
[26] Plata-Munoz JJ, Muthusamy A, Quiroga I, et al. Impact of
pulsatile perfusion on postoperative outcome of kidneys
from controlled donors after cardiac death. Transpl Int 2008;
21:899.
[27] Jochmans I, Moers C, Smits JM, et al. Machine perfusion
versus cold storage for the preservation of kidneys donated
after cardiac death: a multicenter, randomized, controlled
trial. Ann Surg 2010;252:756.
[28] Balfoussia D, Yerrakalva D, Hamaoui K, et al. Advances in
machine perfusion graft viability assessment in kidney, liver,
pancreas, lung, and heart transplant. Exp Clin Transplant
2012;10:87.
[29] Bond M, Pitt M, Akoh J, et al. The effectiveness and cost-
effectiveness of methods of storing donated kidneys from
deceased donors: a systematic review and economic model.
Health Technol Assess 2009;13. iii, xi, 1.
[30] Groen H, Moers C, Smits JM, et al. Cost-effectiveness of
hypothermic machine preservation versus static cold
storage in renal transplantation. Am J Transplant 2012;12:
1824.