Methylprednisolone Therapy in Deceased Donors Reduces Inflammation in the Donor Liver and Improves Outcome After Liver Transplantation
←
→
Page content transcription
If your browser does not render page correctly, please read the page content below
ORIGINAL ARTICLES
Methylprednisolone Therapy in Deceased Donors Reduces
Inflammation in the Donor Liver and Improves Outcome
After Liver Transplantation
A Prospective Randomized Controlled Trial
Katja Kotsch, PhD,* Frank Ulrich, MD,† Anja Reutzel-Selke, PhD,† Andreas Pascher, MD, PhD,†
W. Faber,† P. Warnick,† S. Hoffman,† M. Francuski, MD,† C. Kunert,* O. Kuecuek,‡
G. Schumacher, MD, PhD,† C. Wesslau, MD,‡ A. Lun, MD, PhD,* S. Kohler, MD,† S. Weiss, MD,†
S. G. Tullius, MD, PhD,§ P. Neuhaus, MD, PhD,† and Johann Pratschke, MD, PhD†
soluble interleukins, monocyte chemotactic protein-1, interleukin-2,
Objective: To investigate potential beneficial effects of donor treat-
interleukin-6, tumor necrosis factor-␣, and inducible protein-10 was
ment with methylprednisolone on organ function and outcome after
observed. Methylprednisolone treatment resulted in a significant
liver transplantation.
downregulation of intercellular adhesion molecule-1, tumor necrosis
Summary Background Data: It is proven experimentally and
factor-␣, major histocompatibility complex class II, Fas-ligand,
clinically that the brain death of the donor leads to increased levels
inducible protein-10, and CD68 intragraft mRNA expression. Sig-
of inflammatory cytokines and is followed by an intensified isch-
nificantly ameliorated ischemia/reperfusion injury in the posttrans-
emia/reperfusion injury after organ transplantation. In experiments,
plant course was accompanied by a decreased incidence of acute
donor treatment with steroids successfully diminished these effects
rejection.
and led to better organ function after transplantation.
Conclusions: Our present study verifies the protective effect of
Methods: To investigate whether methylprednisolone treatment of
methylprednisolone treatment in deceased donor liver transplanta-
the deceased donor is applicable to attenuate brain death-associated
tion, suggesting it as a potential therapeutical approach.
damage in clinical liver transplantation we conducted a prospective
randomized treatment-versus-control study in 100 deceased donors. (Ann Surg 2008;248: 1042–1050)
Donor treatment (n ⫽ 50) consisted of 250 mg methylprednisolone
at the time of consent for organ donation and a subsequent infusion
of 100 mg/h until recovery of organs. A liver biopsy was taken
immediately after laparotomy and blood samples were obtained after
brain death diagnosis and before organ recovery. Cytokines were
assessed by real-time reverse transcriptase-polymerase chain reac-
A dvances in immunosuppressive protocols, organ preser-
vation, and perioperative management have significantly
improved patient and graft survival after liver transplantation.
tion. Soluble serum cytokines were measured by cytometric bead Although it is the treatment of choice for end-stage liver
array system. disease, primary nonfunction (PNF) or delayed graft function
Results: After methylprednisolone treatment, steroid plasma levels are still a challenge in a considerable number of recipients.1
were significantly higher (P ⬍ 0.05), and a significant decrease in Extended donor criteria and use of suboptimal organs have
highlighted the relevance of donor-associated risk factors,
including age, comorbidities, or steatosis of the graft. But,
From the *Institute of Medical Immunology, Charité, Universitätsmedizin apart from the general characteristics of the deceased donor
Berlin, Berlin, Germany; †Department of General, Visceral and Trans-
plantation Surgery, Charité, Campus Virchow Clinical Center, Univer- and the important role of ischemia/reperfusion (I/R) injury,
sitätsmedizin Berlin, Berlin, Germany; ‡Deutsche Stiftung Organtrans- there is growing evidence that brain death (BD) and its
plantation, Berlin, Germany; and §Division of Transplant Surgery, complex pathophysiological changes have a significant influ-
Brigham and Women’s Hospital, Harvard Medical School, Boston, ence on graft function and survival.2– 4 This is supported by
Massachusetts. the fact that kidney allografts from unrelated living donors5
The study was supported by DFG Grants PR578 2-2 and PR 578 2-3.
The first two authors contributed equally to this report. with a poor HLA matching demonstrated superior graft func-
Reprints: Johann Pratschke, MD, PhD, Department of General, Visceral and tion and survival in comparison with deceased donation.6,7
Transplantation Surgery, Charité, Campus Virchow Clinical Center, BD as a central nervous catastrophe is a dynamic
Augustenburger Platz 1, 13353 Berlin, Germany. E-mail: johann. process associated with major hemodynamic, hormonal, and
pratschke@charite.de.
Copyright © 2008 by Lippincott Williams & Wilkins immunologic alterations. Increasing ischemia of the brain
ISSN: 0003-4932/08/24806-1042 stem leads to a sympathetic stimulation or the so-called
DOI: 10.1097/SLA.0b013e318190e70c “autonomic storm” guided by a massive release of cat-
1042 Annals of Surgery • Volume 248, Number 6, December 2008Annals of Surgery • Volume 248, Number 6, December 2008 Methylprednisolone Reduces Inflammation in Donor Liver
echolamines, tachycardia, arterial hypertension, and vasocon- cated to recipients, which were not listed in our center
striction.8 Later on, a decrease in sympathetic activation and (Department of General, Visceral and Transplantation sur-
cardiac output ends in a state of organ hypoperfusion, which gery, Universitätsmedizin Berlin, Charité, Berlin) and which
affects morphology and quality of organs.9 –12 Hormonal were shipped elsewhere were not included in the study due to
changes include a decrease in cortisol, free triiodthyronine an insufficient follow-up after transplantation. Due to these
(FT3), free thyroxine (FT4), and insulin levels.13,14 The effect limitations, only 100 donors out of more than 450 suitable
of hormonal resuscitation in brain-dead donors may result in donors were included in the study. No dropouts were ob-
a stabilization of the donor with more organs suitable for served after randomization.
transplantation.15,16 The impact of immunologic activation is
proven by numerous experimental reports, demonstrating an Sampling of Tissues and Laboratory
induction of cytokines, adhesion molecules, and major histo- Parameters
compatibility complex class II (MHC class II) expres- During donor surgery, a liver biopsy was taken imme-
sion.17–21 Graft exposition to organ injury caused or pro- diately after laparotomy before further surgical manipula-
moted by BD leads to an increased vulnerability of ischemia tions. Serum samples in the donor were obtained at the time
and to following reperfusion injury and a higher percentage of consent for organ donation (TP1) and immediately before
of graft dysfunction and acute or chronic rejection episodes in the start of donor surgery (TP2). In the recipients, biopsies
kidneys.22–24 Recently, we were able to demonstrate this were performed on demand to confirm clinically observed
causal relationship after clinical liver transplantation.25 In rejection episodes and routinely after 6 months. Recipient
comparison with living donors, BD was associated with a biopsies were graded by the same pathologist in a blinded
significant upregulation of inflammatory cytokines leading to fashion according to the Banff criteria.
a deteriorated I/R injury, impaired early graft function, and In the donor, the measurement of steroid and hormone
higher rates of PNF and acute rejection episodes. levels (FT3, FT4) were measured in samples taken at TP1 and
Several authors have investigated potential treatment TP2. Lactate, international normalized ration, albumin, bili-
options for BD-associated graft injury experimentally in vivo. rubin, alanine transaminase (ALT), aspartate transaminase
Strategies included reduction in inflammation and immune (AST), alkaline phosphatase (AP), gamma glutamyl transpep-
response but also induction of protective proteins. Applica- tidase (␥GT), and glutamate dehydrogenase in the recipient
tion of catecholamines, glucocorticoids, protease inhibitors, was measured routinely on a daily basis up to 10 days after
or inducers of heme oxygenase 1 (HO-1) had a significant transplantation at the Institute for Clinical Chemistry of the
effect on inflammation, I/R injury, graft rejection, and sur- Virchow Clinic, Charité, Universitätsmedizin Berlin. The
vival.26 –32 Moreover, clinical reports are still limited to initial bile flow after transplantation was assessed by placing
retrospective analysis of catecholamine application in kidney a T-tube in the bile duct for the first 5 days. This study was
transplantation.33,34 To evaluate the effect of methylpred- approved by the ethics committee of the Charité, Universitä-
nisolone treatment in deceased donors and its implications on tsmedizin Berlin, Germany.
I/R injury and early graft function we designed a prospective
randomized trial in brain-dead liver donors. Sample Preparation and mRNA Preparation
Liver tissue biopsy samples were removed at the indi-
MATERIALS AND METHODS cated time point, immediately snap-frozen in liquid nitrogen,
and stored at ⫺80°C until processing. Before analysis, thaw-
Patients ing tissues were transferred in 700 L guanidinium isothio-
From 2003 through 2006 in this single center study, cyanate solution and homogenized using a Mixer Mill
100 deceased organ donors were prospectively randomized in (Qiagen, Hilden, Germany). RNA extraction was performed
2 groups in the organ procurement area Berlin-Brandenburg with the Stratagene Mini Kit according to the manufacturer’s
by the Deutsche Stiftung Organtransplantation. The first instructions (Stratagene, Amsterdam, The Netherlands). Sam-
group of deceased donors received treatment with methyl- ples were tested for genomic DNA contamination, and if
prednisolone (group 1, n ⫽ 50) and the second group re- tested positive, they were excluded from the study.
mained without methylprednisolone treatment (group 2, n ⫽
50). Donor treatment consisted of 250 mg methylpred- Quantification of mRNA Expression
nisolone i.v. at time of consent for organ donation in a Complementary DNA synthesis and real-time reverse
blinded fashion and thereafter 100 mg/h i.v. continuously transcriptase-polymerase chain reaction of nonamplified
until organ recovery. Randomization was performed with mRNA were performed as recently described.24 The hypo-
sealed envelopes, the recipient, the pathologist, the immunol- xanthine-guanine-phosphoribosyl-transferase housekeeping
ogist, the donor, and the recipient surgeon were blinded. In gene was used as an internal standard in the comparative
this trial, donor management was highly standardized and threshold cycle method (2⫺⌬Ct). Primers and probes for CD3,
protocolized according to the guidelines for donor treatment tumor necrosis factor ␣ (TNF-␣), transforming growth factor
of the Deutsche Stiftung Organtransplantation Berlin Bran-  (TGF-), intercellular adhesion molecule 1 (ICAM-1),
denburg. All deceased donors receiving steroids outside the interferon gamma inducible protein 10 (IP-10), fas ligand
study during admission or intensive care and donors with (FasL), heme oxygenase 1 (HO-1), and the major histocom-
infections, defined as the requirement of antibiotic treatment, patibility complex (MHC) class I chain-related protein A
were excluded from randomization. Donors with livers allo- (MICA) were designed and evaluated at the Institute of
© 2008 Lippincott Williams & Wilkins 1043Kotsch et al Annals of Surgery • Volume 248, Number 6, December 2008
Medical Immunology, Charité Universitätsmedizin Berlin, TABLE 1. Donor Characteristics
using Primer Express software (Applied Biosystems, Darm-
stadt, Germany). The markers CD68, CD69, CD80, MHC Group 1 Group 2
Variable (n ⴝ 50) (n ⴝ 50)
class II (HLA-DRB), toll-like receptors 2 and 4 (TLR2 and
TLR4) as well as CCL19 and CCL21, were purchased as Donor gender (N ⫽ M:F) 23:27 30:20
“Assays on Demand” (Applied Biosystems) and were applied Donor age (yrs ⫾ SD) 48 ⫾ 16 47 ⫾ 17
according to the manufacturer’s instructions. Gene expres- ICU (h ⫾ SD) 105 ⫾ 76 110 ⫾ 103
sion was quantified using the ABI PRISM 7500 Sequence *Donor infection (yes/no) 0/50 0/50
Detection System (Applied Biosystems, Darmstadt, Germany). Cardiac arrest (yes/no) 3/50 0/50
The amplification took place in a two-step polymerase chain Hypotensive period (yes/no) 5/45 0/50
reaction (40 cycles; 15-second denaturation step 关95°C兴 and Preservation time (min ⫾ SD) 634 ⫾ 121 601 ⫾ 153
1-minute annealing/extension step 关60°C兴). The mean Ct values *Defined as infection of the donor requiring antibiotic treatment.
for the housekeeping gene and the genes of interest were
calculated from double determinations and samples were con-
sidered negative if the Ct values exceeded 40 cycles. TABLE 2. Recipient Characteristics
Cytometric Bead Array System Group 1 Group 2
Variable (n ⴝ 50) (n ⴝ 50)
Blood samples were obtained immediately after consent
to organ donation and immediately before donor surgery. Sam- Recipient gender (M:F) 31:19 29:21
ples were collected in ethylenediamine tetraacetic acid-contain- Recipient age (years ⫾ SD) 44 ⫾ 14 43 ⫾ 13
ing tubes and stored at ⫺70°C until processing. Serum samples Cold ischemia time (min) 634 ⫾ 121 601 ⫾ 153
were analyzed for the concentration of soluble markers, includ- Meld score (average) 14.8 14.1
ing IL-2R␣, IL-6, IL-8, and TNF-␣ by Immulite (DPC Buehl- Etiology end-stage liver disease
mann GmbH, Salzburg, Austria). The markers, including IL-2, HCC 9 10
IP-10, IFNy, monocyte chemoattractant protein-1 (MCP-1), and HCV 8 5
IL-10 were assessed by the Cytometric Bead Array System from HBV 2 3
Becton Dickinson (Becton Dickinson, Heidelberg, Germany) Alcoholic cirrhosis 12 11
according to the company’s instruction. Cryptogenic 6 5
Amyloidosis 1 —
Statistical Analysis Oxalosis 1 —
Data are expressed as mean ⫾ SD in text or tables and Budd-Chiari syndrome 1 1
presented as mean ⫾ SEM in figures. For single comparisons, Primary biliary cirrhosis 2 2
normally distributed data were analyzed using unpaired, two- Autoimmune 1 2
tailed t test, non-normally distributed data were evaluated by Primary sclerosing cholangitis 2 3
nonparametric Mann–Whitney test. Categorical variables Other 5 8
were expressed as number (ratio/percent) and compared using ESLD indicates end-stage liver disease.
the chi square test or Fisher exact test. P ⬍ 0.05 (2-sided) was
considered statistically significant.
A power calculation to detect a minimum 50% differ-
ence in the AST level between the 2 groups at day 1 after I/R Model of End Stage Liver Disease
injury (primary end point), with a Type 1 error (a) of 0.05 and No statistical differences were observed in patients
80% power, suggested that n ⫽ 40/group were required to receiving either a graft from treated or untreated donors with
demonstrate significant differences. regard to the recipient age, donor gender, and diagnosis
(Table 2). The recipient status according to the Model of
RESULTS End-Stage Liver disease score did not reveal significant
differences between the groups (MELD score average was
Donor Characteristics 14.8 in group 1 vs. 14.1 in group 2). The time for vascular
There were no statistical differences between both groups anastomosis, defined as warm ischemia time, was comparable
regarding donor age, donor gender, hypotension, cardiac resus- in both groups (43 ⫾ 11 minutes in group 1 vs. 45 ⫾ 9
citation, and perioperative cardiac arrests as well as the fluids minutes in group 2).
(including infusions, albumin, etc.) and catecholamines applied The immunosuppressive regimen was in both groups
in both groups (Table 1). The preservation time, including cold based on tacrolimus or cyclosporin A in addition with my-
and warm ischemia time was equal in both donor groups (634 ⫾ cophenolate mofetil and steroids. None of the patients re-
121 minutes in group 1 vs. 601 ⫾ 153 minutes in group 2). The ceived an induction therapy. There were no significant dif-
time spent at the intensive care unit was comparable in both ferences in the immunosuppressive regimen between both
groups (105 ⫾ 76 hours in group 1 vs. 110 ⫾ 103 hours in group groups, over short and long term, especially the dosage of
2). The weight of the transplanted livers did not differ signifi- steroids did not differ significantly. The rate of HCV reinfec-
cantly between both groups (1756 ⫾ 145 g in group 1 vs. tions after liver transplantation was comparable in both
1699 ⫾ 183 g in group 2). groups 3 months after transplantation.
1044 © 2008 Lippincott Williams & WilkinsAnnals of Surgery • Volume 248, Number 6, December 2008 Methylprednisolone Reduces Inflammation in Donor Liver
Drug Application TABLE 3. Cytokines and Markers Determined by RT-PCR
The total amount of fluids (including infusions, albu-
Donor
min, etc.) applied was comparable in both groups. The aver- Treatment Mean (2-DCt) SE P (Mann-Whitney)
age treatment time was 855 ⫾ 137 minutes, which did not
differ significantly from the untreated patient group (789 ⫾ CD3 No 0.0042 0.0015 0.8867
111 minutes). On average, 1650 ⫾ 142 mg methylpred- Yes 0.0028 0.0005
nisolone was applied per donor in the treatment group. The CD68 No 12.35 5.85 0.0309
application of methylprednisolone led to higher steroid levels Yes 9.00 1.06
in group 1 before organ recovery (group 1, TP1 735 ⫾ 204 CD69 No 0.0229 0.0032 0.2519
pg/mL vs. TP2 1238 ⫾ 343 pg/mL, P ⬍ 0.05). In contrast, Yes 0.0329 0.0051
serum steroid levels in the untreated group demonstrated a CD80 No 0.0041 0.0008 0.0047
decrease until organ recovery (group 2, TP1 678 ⫾ 255 Yes 0.0017 0.0003
pg/mL vs. TP2 222 ⫾ 45 pg/mL, P ⬍ 0.05). No complica- TNF-␣ No 0.0067 0.0009 ⬍0.0001
tions associated with methylprednisolone infusion were ob- Yes 0.0025 0.0005
served. In both groups, insulin was applied if glucose levels TGF- No 5.11 1.12 0.7902
reached over 250 mg/dL. The assessment of FT3 and FT4 did Yes 4.67 0.73
not show significant differences at any time point (data not ICAM No 0.6044 0.1028 0.0017
shown). Yes 0.2677 0.0736
IP10 No 0.0445 0.0117 ⬍0.0001
Yes 0.0105 0.0035
Reduced Intragraft mRNA Expression of FasL No 0.0075 0.0023 ⬍0.0001
Selected Genes After Therapy in Brain-Dead Yes 0.0013 0.0002
Donors HO-1 No 0.4574 0.0897 0.2872
To evaluate the influence of methylprednisolone appli- Yes 0.3200 0.0487
cation on the inflammatory status of the graft, we determined MICA No 0.0228 0.0051 0.0157
selected markers for adhesion, antigen presentation, migra- Yes 0.0381 0.0031
tion, apoptosis, and lymphocyte infiltration. In general, the DRB No 0.5657 0.2073 0.0226
analysis revealed that mRNA expression levels of most in- Yes 0.5393 0.1283
vestigated genes were downregulated after methylpred- CCL19 No 1.39 0.28 0.6628
nisolone treatment compared to untreated donors with the Yes 1.05 0.23
exception of the stress-inducible molecule MICA (Table 3). CCL21 No 2.07 0.42 0.2708
Interestingly, significant differences were detectable for Yes 2.68 0.49
TNF-␣, the chemokine IP-10, and FasL (P ⬍ 0.0001, Table TLR2 No 0.0236 0.0047 0.6657
3). Additionally, the costimulatory antigen CD80, IL-6, and Yes 0.0173 0.0029
ICAM-1 showed further decreased mRNA expression levels TLR4 No 0.0191 0.0030 0.4096
in the treated group (P ⬍ 0.01) and reduced gene expression Yes 0.0151 0.0019
was detectable for MHC class II antigen (DRB1) and CD68, IL-6 No 0.155 0.76 0.0092
a marker for macrophages (P ⬍ 0.05). Surprisingly, the Yes 0.051 0.042
analyzed mRNA expression did not display any statistical
significance for CD3, CD69, TGF-, TLR2, TLR4, CCL19/
21, and HO-1 (Table 3).
Initial Graft Function and Rejection Rates
To investigate whether the decreased immune activa-
Reduced Levels of Soluble Cytokines After tion within the graft due to donor treatment with steroids
Steroid Therapy in Brain-Dead Donors might have an impact on allograft outcome, AST and ALT
By investigating soluble cytokines in the serum, the values were determined as markers for I/R injury. The results
analysis revealed that initially before the onset of the meth- illustrate that both serum parameters were low for the treated
ylprednisolone application, levels in both groups were com- organ donor group in comparison with the untreated donors
parable and did not differ significantly (TP1, Fig. 1, 2, and 3). in the first 10 days posttransplantation (AST day 1, group
However, significantly lower concentrations for the inflam- 1: 327 ⫾ 147 vs. group 2: 1470 ⫾ 314 IU/L; ALT day 1,
matory markers IL-6, TNF␣, IL-2, and IL-2R␣ were ob- group 1: 461 ⫾ 110 vs. group 2: 758 ⫾ 213 IU/L; P ⬍ 0.05;
served in the methylprednisolone treatment group in compar- AST day 10, group 1: 31 ⫾ 5 vs. group 2: 41 ⫾ 7 IU/L; ALT
ison with those in untreated donors immediately before organ day 10, group 1: 75 ⫾ 12 vs. group 2: 115 ⫾ 15 IU/L; P ⬍
recovery (TP2, P ⬍ 0.01, Fig. 1 and 2). Furthermore, in- 0.05). Additionally, yGT and AP as markers for biliary injury
creased levels for the chemoattractive chemokines, including were significantly lower after 10 days in group 1 (yGT, group
MCP-1 and IP-10 were observed for the untreated group 1: 135 ⫾ 25 vs. group 2: 236 ⫾ 35 IU/L; AP, group
(TP2, P ⬍ 0.01, Fig. 3). In contrast, no significant differences 1: 127 ⫾ 19 vs. group 2: 157 ⫾ 20 IU/L; P ⬍ 0.05). Total
could be detected for the chemokines IL-10, IL-8, IL-10, and bilirubin as a marker for liver function was significantly
IFN␥ (data not shown). lower in group 1 at postoperative day 10 and after 6 months
© 2008 Lippincott Williams & Wilkins 1045Kotsch et al Annals of Surgery • Volume 248, Number 6, December 2008 FIGURE 1. Serum concentration of soluble markers IL-6 and IL-2R␣ in groups 1 and 2 at the time point of brain death (TP1) and immedi- ately before organ recovery (TP2, **P ⬍ 0.005). FIGURE 2. Serum concentration of soluble markers IL-2 and TNF␣ in groups 1 and 2 at the time point of brain death (TP1) and immedi- ately before organ recovery (TP2, **P ⬍ 0.005). FIGURE 3. Serum concentration of soluble markers MCP-1 and IP-10 in groups 1 and 2 at the time point of brain death (TP1) and im- mediately before organ recovery (TP2, **P ⬍ 0.005). (bilirubin mg/dL: day 10: group 1: 2.3 ⫾ 0.4 vs. group 2: 4.9 ⫾ 1.0; min assessed at days 3 and 10 did not differ significantly at month 6: group 1: 0.6 ⫾ 0.06 vs. group 2: 1.03 ⫾ 0.08; between both groups (data not shown). Organs from un- P ⬍ 0.05). International normalized ration, lactate, and albu- treated brain-dead donors experienced a higher rate of biop- 1046 © 2008 Lippincott Williams & Wilkins
Annals of Surgery • Volume 248, Number 6, December 2008 Methylprednisolone Reduces Inflammation in Donor Liver
sy-proven acute rejections between grades 1 and 3 within the soluble P-selectin glycoprotein ligand was able to demon-
first 6 months after transplantation (group 1, n ⫽ 11, 22% vs. strate superior rat renal isograft or allograft function and a
group 2, n ⫽ 19, 36%; P ⬍ 0.05). There was a tendency reduced incidence of acute or chronic rejection.27,28 Protease
toward more severe rejections in the untreated group (group inhibitors such as urinary trypsin inhibitor decreased leuko-
1: n ⫽ 6, grade 1a, n ⫽ 4, grade 1b, n ⫽ 1, grade 2b; group cyte recruitment and release of cytokines and proved its
2: n ⫽ 9, grade 1a, n ⫽ 4, grade 1b, n ⫽ 5, grade 2a, n ⫽ 1, ability to restore hepatic tissue flow in the liver of hypoten-
grade 2b). The bile flow at the fifth day was comparable in sive BD rats.32 Another strategy is based on the cytoprotec-
both groups (group 1: 144 ⫾ 22 mL vs. group 2: 108 ⫾ 26 tive and antioxidative properties of stress proteins such as
mL). There was no significant difference in the occurrence of metalloprotoporphyrin cobalt protoporphyrin, an inducer of
PNF (group 1, n ⫽ 2 vs. group 2, n ⫽ 3). We observed a HO-1.30
higher rate of ischemic-type biliary lesions diagnosed by In our present study, we performed donor therapy with
endoscopic retrograde cholangiography in group 2 in com- methylprednisolone since this option offers the advantage of
parison with that in group 1 (n ⫽ 2 vs. n ⫽ 5), although this counteracting immunologic and also hormonal alterations
result did not reach statistical significance. caused by BD. Additionally, its application is easy and safe in
humans. In vivo, rat kidney allografts and pig hearts demon-
strated a favorable outcome and function after treat-
DISCUSSION ment.26,27,42 Clinical investigations on decreased basal and
To evaluate the effect of donor treatment with methyl- stimulated plasma cortisol in BD organ donors are supple-
prednisolone and its influence on I/R injury and early graft mented by experimental data in rabbits, demonstrating im-
function after liver transplantation we designed a prospective proved blood flow and reduced Kupffer cell activation after
randomized trial in 100 BD donors. Methylprednisolone therapy methylprednisolone administration.13,31 After methylpred-
resulted in a significant decrease in intragraft and serum cytokine nisolone treatment, starting at an average of 15 hours before
expression as well as in improved liver function. Besides an
the recovery of organs, we were able to show significantly
ameliorated I/R injury, reduced rates of acute rejection were also
higher steroid levels in the treatment group. According to
observed during the clinical follow-up.
previous in vivo and clinical studies,20,25,36 we were able to
Effects of BD on donor livers in vivo are characterized
detect elevated serum levels of cytokines in the BD donor,
by elevated liver enzymes, vacuolization, and necrosis of
which upon treatment with methylprednisolone was able to
hepatocytes and a decrease in bile production and sinusoidal
reduce in the present study.
perfusion.35,36 Longer periods of cold ischemia aggravate the
injury and result in a severe reduction of survival. Besides The important role of transcriptional profiles of livers
direct organ injury, induction of cytokines, endothelial anti- from deceased organ donors was supported by Colombo et al,
gens, and adhesion molecules lead to a higher vulnerability showing a significant upregulation of immune- and stress-
with regard to I/R injury, alloreactivity, and chronic graft response gene categories, including acute-phase response,
dysfunction.37 Recently, we have been able to demonstrate complement activation, and response to oxidative stress.43
the effect of BD on mRNA cytokine expression in a clinical The importance of preapoptotic conditions after BD was
LT study, illustrating a significant increase in IL-6, IL-10, reported by van der Hoeven et al, illustrating experimental
TNF-␣, TGF-, and Mip-1␣.25 Cellular infiltration (CD3⫹ data about increased apoptosis of hepatocytes by caspase-3
and CD25⫹ cells), CD41⫹ deposits of platelets, and increased activity and deoxynucleotidyl transferase-mediated dUTP-
MHC class II induction were found in immunohistology. biotin nick end labeling staining.44 Additionally, upregulation
These findings correlated with elevated liver enzymes and of Fas and FasL indicated an activation of the death receptor-
bilirubin levels and an increased rate of PNF and acute mediated pathway. It has been shown that TLR4 plays a role
rejection episodes. On the basis of these data, we evaluated in hepatic I/R in murine and rat models. The Kupffer cell is
the possibility of clinical donor treatment to ameliorate BD- the primary cellular mediator of I/R in the liver. Kupffer cells
associated inflammation. This inflammatory response is am- are activated by a large amount of endotoxin released through
plified by multiple interactions of the abovementioned com- portal circulation after I/R; this causes maturation of the
ponents and promotes migration of polymorphonuclear Kupffer cells and a significant increase in TNF-␣ and TLR4
leukocytes, T cells, and macrophages into donor organs.17,18 mRNA production.45,46 However, methylprednisolone treat-
Several authors demonstrated the impact of these findings for ment of the BD donor resulted in a significant downregulation
organ function and rejection rates in experimental organ of inflammatory mediators, including TNF-␣, the apoptosis-
transplantation.21,22,24,35,38 – 40 The concept of donor treat- related gene FasL, IP-10, and ICAM1 in liver biopsy speci-
ment was shown clinically and experimentally to improve mens. Additionally, markers of alloreactivity, including
kidney graft function, with a more rapid decrease in serum MHC class II and CD80 illustrated also a marked decrease,
creatinine and less delayed graft function.33,34 Experimen- but no influence on TLR2 and TLR4 gene expression could
tally, the amelioration of BD in rats helped to explain these be detected. On comparing deceased and LD livers before
clinical effects, showing reduced monocyte infiltration and transplantation, Jassem et al noticed significantly higher lev-
downregulated expression of proinflammatory cytokines and els of CD3⫹ lymphocytes, CD68⫹ macrophages, and FasL
MHC class II, whereas HO-1 expression and its antioxidative staining in deceased donors.47 Surprisingly, we could not
properties were upregulated.29,41 Additionally, antagonizing observe enhanced T cell infiltration characterized by CD3
adhesion molecules P-, E-, and L-selectin and recombinant mRNA expression analysis. Furthermore, mRNA transcripts
© 2008 Lippincott Williams & Wilkins 1047Kotsch et al Annals of Surgery • Volume 248, Number 6, December 2008
of the MHC-related antigen MICA were induced in the or gradual increase of intracranial pressure on myocardial structure and
steroid group. It can be speculated with the application of function. Circulation. 1993;87:230 –239.
9. Power BM, Van Heerden PV. The physiological changes associated with
methylprednisolone, counter-regulatory mechanisms are ac- brain death– current concepts and implications for treatment of the brain
tivated, leading to partial immunologic activation. However, dead organ donor. Anaesth Intensive Care. 1995;23:26 –36.
after kidney transplantation, elevated transcription levels of 10. Nagareda T, Kinoshita Y, Tanaka A, et al. Clinicopathology of kidneys
the stress-inducible non-HLA ligand MICA were associated from brain-dead patients treated with vasopressin and epinephrine.
with significantly better long-term outcome after 12 months Kidney Int. 1993;43:1363–1370.
11. Avlonitis VS, Fisher AJ, Kirby JA, et al. Pulmonary transplantation: the role
(unpublished data). The beneficial effects in our trial were of brain death in donor lung injury. Transplantation. 2003;75:1928 –1933.
overwhelming, and it is difficult to estimate sufficiently the 12. Salim A, Martin M, Brown C, et al. Complications of brain death:
role of partial immunactivation as it is observed with elevated frequency and impact on organ retrieval. Am Surg. 2006;72:377–381.
MICA levels in combination with Kupffer cell activation in 13. Dimopoulou I, Tsagarakis S, Anthi A, et al. High prevalence of de-
this study. creased cortisol reserve in brain-dead potential organ donors. Crit Care
Med. 2003;31:1113–1117.
The combination of the abovementioned mechanisms, 14. Novitzky D, Cooper DK, Rosendale JD, et al. Hormonal therapy of the
including inflammation, cellular infiltration, platelet aggrega- brain-dead organ donor: experimental and clinical studies. Transplanta-
tion, and activation of apoptotic pathways result not only in tion. 2006;82:1396 –1401.
acute damage with higher vulnerability to ischemia and acute 15. Rosendale JD, Kauffman HM, McBride MA, et al. Aggressive pharma-
cologic donor management results in more transplanted organs. Trans-
rejection but also in persisting alterations, leading to chronic plantation. 2003;75:482– 487.
graft dysfunction in organs susceptible to chronic rejection. 16. Salim A, Martin M, Brown C, et al. Using thyroid hormone in brain-
This is reflected by our clinical follow-up, since the methyl- dead donors to maximize the number of organs available for transplan-
prednisolone treatment group demonstrated lower levels of tation. Clin Transplant. 2007;21:405– 409.
hepatic transaminases and bilirubin and had less acute rejec- 17. Kusaka M, Pratschke J, Wilhelm MJ, et al. Activation of inflammatory
mediators in rat renal isografts by donor brain death. Transplantation.
tion episodes. 2000;69:405– 410.
Our clinical data are supported by United Network for 18. van der Hoeven JA, Molema G, Ter Horst GJ, et al. Relationship
Organ Sharing data, reporting on the experience of hormonal between duration of brain death and hemodynamic (in)stability on
resuscitation in 10,292 consecutive deceased donors with a progressive dysfunction and increased immunologic activation of donor
methylprednisolone bolus and infusion of vasopressin and kidneys. Kidney Int. 2003;64:1874 –1882.
19. van der Hoeven JA, Ploeg RJ, Postema F, et al. Induction of organ
thyroid hormone, which resulted in an increase in trans- dysfunction and up-regulation of inflammatory markers in the liver and
planted organs.15 The same treatment protocol led to less kidneys of hypotensive brain dead rats: a model to study marginal organ
graft dysfunction and improved early survival in 4543 heart donors. Transplantation. 1999;68:1884 –1890.
recipients.48 20. Olinga P, van der Hoeven JA, Merema MT, et al. The influence of brain
Confirming experimental data about the positive effects of death on liver function. Liver Int. 2005;25:109 –116.
21. Wilhelm MJ, Pratschke J, Beato F, et al. Activation of the heart by donor
treatment with steroids in BD donors, our present study proved brain death accelerates acute rejection after transplantation. Circulation.
its efficacy in the context of clinical liver transplantation. In- 2000;102:2426 –2433.
flammation and I/R injury were significantly reduced by our 22. Pratschke J, Wilhelm MJ, Kusaka M, et al. Accelerated rejection of renal
treatment protocol, leading to favorable graft function. However, allografts from brain-dead donors. Ann Surg. 2000;232:263–271.
it could be speculated that especially the recipients of marginal 23. Pratschke J, Wilhelm MJ, Laskowski I, et al. Influence of donor brain
death on chronic rejection of renal transplants in rats. J Am Soc Nephrol.
donor livers might benefit from donor treatment, which should 2001;12:2474 –2481.
be examined in further follow-up studies. 24. Sanchez-Fructuoso AI, Prats D, Marques M, et al. Does donor brain
death influence acute vascular rejection in the kidney transplant? Trans-
plantation. 2004;78:142–146.
REFERENCES 25. Weiss S, Kotsch K, Francuski M, et al. Brain death activates donor
1. Johnson SR, Alexopoulos S, Curry M, et al. Primary nonfunction (PNF) organs and is associated with a worse I/R injury after liver transplanta-
in the MELD Era: an SRTR database analysis. Am J Transplant. tion. Am J Transplant. 2007;7:1584 –1593.
2007;7:1003–1009. 26. McLean KM, Duffy JY, Pandalai PK, et al. Glucocorticoids alter the
2. Pratschke J, Wilhelm MJ, Kusaka M, et al. Brain death and its influence balance between pro–and anti-inflammatory mediators in the myocar-
on donor organ quality and outcome after transplantation. Transplanta- dium in a porcine model of brain death. J Heart Lung Transplant.
tion. 1999;67:343–348. 2007;26:78 – 84.
3. Pratschke J, Neuhaus P, Tullius SG. What can be learned from brain- 27. Pratschke J, Kofla G, Wilhelm MJ, et al. Improvements in early behavior
death models? Transpl Int. 2005;18:15–21. of rat kidney allografts after treatment of the brain-dead donor. Ann
4. Bos EM, Leuvenink HG, van Goor H, et al. Kidney grafts from brain Surg. 2001;234:732–740.
dead donors: Inferior quality or opportunity for improvement? Kidney 28. Gasser M, Waaga AM, Kist-Van Holthe JE, et al. Normalization of brain
Int. 2007;72:797– 805. death-induced injury to rat renal allografts by recombinant soluble
5. Coleman TR, Westenfelder C, Togel FE, et al. Cytoprotective doses of P-selectin glycoprotein ligand. J Am Soc Nephrol. 2002;13:1937–1945.
erythropoietin or carbamylated erythropoietin have markedly different 29. Gottmann U, Brinkkoetter PT, Bechtler M, et al. Effect of pre-treatment
procoagulant and vasoactive activities. Proc Natl Acad Sci USA. 2006; with catecholamines on cold preservation and ischemia/reperfusion-
103:5965–5970. injury in rats. Kidney Int. 2006;70:321–328.
6. Terasaki PI, Cecka JM, Gjertson DW, et al. High survival rates of kidney 30. Kotsch K, Francuski M, Pascher A, et al. Improved long-term graft
transplants from spousal and living unrelated donors. N Engl J Med. survival after HO-1 induction in brain-dead donors. Am J Transplant.
1995;333:333–336. 2006;6:477– 486.
7. Jassem W, Koo DD, Cerundolo L, et al. Cadaveric versus living-donor 31. Ogawa K, Ito Y, Takahashi T, et al. Effects of cortisol administration on
livers: differences in inflammatory markers after transplantation. Trans- hepatic circulation during brain death in rabbits. Surgery. 2002;131:
plantation. 2003;76:1599 –1603. 450 – 462.
8. Shivalkar B, Van Loon J, Wieland W, et al. Variable effects of explosive 32. Itabashi K, Ito Y, Takahashi T, et al. Protective effects of urinary trypsin
1048 © 2008 Lippincott Williams & WilkinsAnnals of Surgery • Volume 248, Number 6, December 2008 Methylprednisolone Reduces Inflammation in Donor Liver
inhibitor (UTI) on hepatic microvasculature in hypotensive brain-dead donor. In these livers, I think the beneficial effects overwhelm
rats. Eur Surg Res. 2002;34:330 –338.
33. Schnuelle P, Lorenz D, Mueller A, et al. Donor catecholamine use
the negative effects. If we transplant an organ where steroids
reduces acute allograft rejection and improves graft survival after ca- are flushed out, we have the inflammatory effects in these
daveric renal transplantation. Kidney Int. 1999;56:738 –746. organs. Alternatively, you must also be aware that we apply
34. Schnuelle P, Yard BA, Braun C, et al. Impact of donor dopamine on
immediate graft function after kidney transplantation. Am J Transplant. steroids anyway after transplantation. I think most centers,
2004;4:419 – 426. and not all centers, apply between 500 and 1000 mg of
35. Van der Hoeven JA, Lindell S, van Schilfgaarde R, et al. Donor brain steroids before reperfusion. If there is a negative effect of
death reduces survival after transplantation in rat livers preserved for 20 hr.
Transplantation. 2001;72:1632–1636. steroids on regeneration, I think this would be more impor-
36. Okamoto S, Corso CO, Leiderer R, et al. Role of hypotension in tant. However, I am convinced that the positive effects are
brain-death associated impairment of liver microcirculation and viabil- much more important than the negative effects, which re-
ity. Transpl Int. 2000;13:428 – 435.
37. Schuurs TA, Gerbens F, van der Hoeven JA, et al. Distinct transcrip- main, I think, questionable.
tional changes in donor kidneys upon brain death induction in rats: Your second question is that how it influences postop-
insights in the processes of brain death. Am J Transplant. 2004;4:1972– erative biliary complications. Well, I think with biliary leak-
1981.
38. Zweers N, Petersen AH, van der Hoeven JA, et al. Donor brain death age you always blame the surgeon. We cannot blame organ
aggravates chronic rejection after lung transplantation in rats. Trans- inflammation for that, but there is a difference. The structure
plantation. 2004;78:1251–1258. of the bile duct is the most problematic with reperfusion and,
39. Koudstaal LG, ’t Hart NA, van den Berg A, et al. Brain death causes
structural and inflammatory changes in donor intestine. Transplant Proc. of course, has the most problems with microcirculation. The
2005;37:448 – 449. reperfusion is better with the antiflammatory effect where the
40. Contreras JL, Eckstein C, Smyth CA, et al. Brain death significantly adhesion molecules are reduced. Thus, ischemia is reduced in
reduces isolated pancreatic islet yields and functionality in vitro and in
vivo after transplantation in rats. Diabetes. 2003;52:2935–2942. the bile duct. We see fewer ischemia-type biliary lesions. So,
41. Schaub M, Ploetz CJ, Gerbaulet D, et al. Effect of dopamine on right now, with this group there is a tendency but it is not
inflammatory status in kidneys of brain-dead rats. Transplantation.
2004;77:1333–1340.
significant.
42. Lyons JM, Pearl JM, McLean KM, et al. Glucocorticoid administration Your third question regarding why we excluded donors
reduces cardiac dysfunction after brain death in pigs. J Heart Lung who were treated with antibiotics. We did this because, with
Transplant. 2005;24:2249 –2254.
43. Colombo G, Gatti S, Turcatti F, et al. Alteration in the transcriptional
donors on antibiotics, you are always in danger of bias. If an
profile of livers from brain-dead organ donors. Transplantation. 2006; inflammation is caused by an infection, you have an artificial
82:69 –79. situation. Then you cannot determine what is caused by
44. Van Der Hoeven JA, Moshage H, Schuurs T, et al. Brain death induces
apoptosis in donor liver of the rat. Transplantation. 2003;76:1150 –1154. intensive care, by brain death, by cuticular means, or
45. Tsoulfas G, Takahashi Y, Ganster RW, et al. Activation of the lipopoly- simply by an infection. We benefitted by having many
saccharide signaling pathway in hepatic transplantation preservation donors, around 160 to 200 donors a year, so we could
injury. Transplantation. 2002;74:7–13.
46. Peng Y, Gong J, Liu C, et al. Expression of toll-like receptor 4 and exclude these donors from the study and did not lose much
MD-2 gene and protein in Kupffer cells after ischemia-reperfusion in rat time. We required about 2 and half years. Of course, it
liver graft. World J Gastroenterol. 2004;10:2890 –2893. would have been faster, but I think the answers are clearer
47. Jassem W, Koo DD, Cerundolo L, et al. Leukocyte infiltration and
inflammatory antigen expression in cadaveric and living-donor livers after excluding these donors.
before transplant. Transplantation. 2003;75:2001–2007.
48. Rosendale JD, Kauffman HM, McBride MA, et al. Hormonal resusci- H. JEEKEL: Quite some time ago, in the early 70s, in
tation yields more transplanted hearts, with improved early function. Maastricht, we carried out some experiments on donor treat-
Transplantation. 2003;75:1336 –1341.
ment. We used a kidney graft model in the rat, treated the
donor, and, in other experiments, treated the organ itself. We
applied immunosuppressive agents and, indeed, we found
Discussions prolonged survival and graft survival. We looked at the
H. TILANUS: There is a relationship between inflamma- mechanism, at that time there was perhaps not as much to
tion and regeneration. Is this donor, steroid treatment contra- look at, but did seem to find some influence on antigen-
indicated in split level or small graft transplantation? The presenting cells. We also looked at the graft versus host
immunosuppression regimen influenced postoperative biliary reaction from the organ reacting to the recipient. You looked
complications. Were there differences in biliary leakage and at mechanisms like the cytokine levels only. Did you ever
strictures between the 2 groups? A substantial number of look at the cellular level? Did you look at dendritic cells? Did
donors are treated with antibiotics before or up until donation you look at the organ itself or are you planning to do that?
and I did not understand why these donors were excluded
from the study. J. PRATSCHKE: This is, indeed, one of our problems. Of
course, it is not only 1 mechanism. In our former study, we
J. PRATSCHKE: The first question is hard to answer. looked for cells and in the donor organs the cellular infiltrates
Some publications report that steroids inhibit regeneration. before transplantation are higher. There is also a very good
On the other hand, we must be aware that we are treating the publication focusing on cellular infiltrates, brain-dead versus
© 2008 Lippincott Williams & Wilkins 1049Kotsch et al Annals of Surgery • Volume 248, Number 6, December 2008
non– brain-dead, in human livers that was published in an In answer to your third question regarding the mecha-
American journal by Yassim et al 2 or 3 years ago. They also nism of steroids, I agree with you that there are multiple
showed that cellular infiltrates are higher in these organs. mechanisms and I am aware that Kupffer cells could be
We can also say that the cellular infiltrates after treat- blocked by that. I think we must investigate the small details
ment are reduced, not only with cytokines but also with in experimental models to get the whole picture; therefore, I
cellular infiltrates, which makes sense because you are reduc- cannot definitively answer the question as to what the mech-
ing adhesion molecules, you are reducing the signaling; anisms really are. It was a very pragmatic, surgical approach
therefore, as a consequence, the cellular infiltrate should be to apply a drug that we know is potent and anti-inflammatory,
reduced. but still we cannot figure out which detailed mechanism it is
that produces the beneficial effect.
P-A. CLAVIEN: I have 2 questions. One relates to the
protocol you used. I would like to know which patients could As for your fourth question on thrombocytopenia, we
enter the study and which could not. What were the criteria? did not observe this in either the donor or the recipient. You
How did you decide who would or would not go into the know that, after application of immunosuppressive drugs,
study? Regarding the recipient, did you check protocol bi- thrombocytopenia is a common problem if you prescribe
opsy? Did you look at the biopsy? Was this done on protocol Celcept or induction therapy. In the organ itself, we could not
or was it retrospectively analyzed? observe any thrombocytopenia, but we did not look specifi-
The second question, also relating to the mechanism, I cally for this problem.
think everybody who works in this area is not sure what the
mechanism of this steroid is, and showing that there is a R. ADAM: My question relates to whether you had the
decrease in the steroid could be a surrogate observation, but opportunity to discriminate the beneficial effect of steroids
it has nothing to do with the effect. What we know about according to the quality of the organ. It would be very
steroids is that they block the Kupffer cells, which could also exciting to show any difference between an optimal donor
be one of the mechanisms. It has also been shown that the liver and a suboptimal or marginal liver. Did you look at the
leukocyte infiltration, almost to the platelet infiltration, could possible differential effect between the fat-infiltrated livers or
be responsible for injury. Did you look at thrombocytopenia elderly livers as opposed to optimal livers?
on the periphery to see if there is a decrease in platelets, or a
decrease in platelets after reperfusion in the patients who J. PRATSCHKE: It is a very good point. In general, you
received steroids? can say that all livers benefit from it. We had a couple of
livers in which we had never experienced such low levels of
J. PRATSCHKE: Your first question was why did we not transaminases, around 100 to 150. When I checked my data
include more patients in this study and how did we decide afterward, these organs had been pretreated. However, in
which patients to include. We did not use donors over 65 fatty livers, you often observe such a high level, 500 to 1000,
years of age because if we have an age-related problem, we but, even in this subanalysis, the fatty livers with steroids
introduce a bias. We are conducting a separate study where performed better than those without steroids.
we try to define the influence of age on the donor organ
quality. That is one of the reasons. Our donors are getting R. ADAM: Is the effect more marked in the marginal
older; around 50% are over 65 years. donors?
Of course, the biopsies were part of the protocol and
were taken at the time of organ harvest. Thus, it was not the J. PRATSCHKE: I think the effect is more marked because
biopsy after reperfusion, it was at the time of organ harvest you are closer to the border where the organ is not working
before the organ was packed and sent to the centers. The third any more, and perhaps, there is a safety margin where you
reason why we have only 50 in each group is that we only might still transplant. As a side effect, we also looked at the
used organs that were transplanted at our center, when we had kidneys that were transplanted. With this study, which has
the possibility of follow-up to assess reperfusion injury been going on for 2 years now, although the numbers are not
among other things. If you send these organs to other centers, yet high enough, these kidneys indicate a better initial func-
follow-up becomes a problem and usually you do not get tion with a lower rate of delayed graft function, so in other
complete data. organs, it also has an effect.
1050 © 2008 Lippincott Williams & WilkinsYou can also read
