Liver transplantation for Budd–Chiari syndrome: A European study
on 248 patients from 51 centres*
Gilles Mentha1,*, Emiliano Giostra1, Pietro E. Majno1, Wolf O. Bechstein2, Peter Neuhaus3,
John O’Grady4, Raaj K. Praseedom5, Andrew K. Burroughs6, Yves P. Le Treut7,
Preben Kirkegaard8, Xavier Rogiers9, Bo-Goran Ericzon10, Krister Hockerstedt11, Rene ´ Adam12,
1Transplantation Unit, Department of Surgery, University Hospitals, 1211 Geneva 14, Switzerland
2Johann Wolfgang Goethe Universitat, Frankfurt, Germany
3La Charite ´ Humboldt University, Berlin, Germany
4King’s College Hospital, London, UK
5Addenbrooke’s Hospital, Cambridge, UK
6Royal Free Hospital, London, UK
7CHU Marseille, Marseille, France
8University Hospital Copenhagen, Copenhagen, Denmark
9Universitatskrankenhaus Eppendorf, Hamburg, Germany
10Huddinge University Hospital, Huddinge, Sweden
11University of Helsinki, Helsinki, Finland
12Hepatobiliary Center, Ho ˆpital Paul Brousse, Paris, France
13Medizinische Hochschule Hannover, Hannover, Germany
Background/Aims: The results of liver transplantation for Budd–Chiari syndrome (BCS) are poorly known and the
role and timing of the procedure are still controversial. The aim of this study was to investigate the results of
transplantation for BCS, focusing on overall outcome, on prognostic factors and on the impact of the underlying
Methods: An enquiry on 248 patients representing 84% of the patients transplanted for BCS in the European Liver
Transplantation Registry between 1988 and 1999.
Results: Of the 248 patients, 70.4% were female and 29.6% male. The mean age was 35.7 years. The overall actuarial
survival was 76% at 1 year, 71% at 5 years and 68% at 10 years. 77% of deaths occurred in the first 3 months: 47%
were due to infection and multiple organ failure, and 18% to graft failure or hepatic artery thrombosis. Late mortality
(O1 year) occurred in nine patients, due to BCS recurrence in four of them. The only pre-transplant predictors of
mortality on multivariate analysis (Cox) were impaired renal function and a history of a shunt.
Conclusions: Liver transplantation for BCS is an effective treatment, irrespective of the underlying cause, and should
be considered before renal failure occurs.
q 2005 European Association for the Study of the Liver. Published by Elsevier B.V. All rights reserved.
Keywords: Liver transplantation; Budd–Chiari; Prognostic factors; Recurrence; Myeloproliferative disease
Journal of Hepatology 44 (2006) 520–528
0168-8278/$32.00 q 2005 European Association for the Study of the Liver. Published by Elsevier B.V. All rights reserved.
Received 3 May 2005; received in revised form 8 December 2005; accepted 9 December 2005; available online 27 December 2005
*From the European liver transplantation association (ELTA) and the European liver transplant registry (ELTR).
*Corresponding author. Tel.: C41 22 37 27607; fax: C41 22 37 27600.
E-mail address: firstname.lastname@example.org (G. Mentha).
Budd-Chiari syndrome (BCS) is a clinical condition
caused by obstruction of the venous outflow of the liver.
The pathogenesis of BCS is complex: usually, an
acquired thrombotic stimulus associates with a genetic
clotting abnormality to produce occlusion of the hepatic
veins [1,2]. Several treatments are available for the
different stages and clinical manifestations of the disease,
and the role of orthotopic liver transplantation (LT) has
been controversial since it was performed for the first
time in 1974 . While LT was rapidly identified as the
only chance to survive the fulminant form of the disease
or the terminal stage of chronic progressive form of
BCS, studies from pioneering centres concluded that, in
most patients and irrespective of the underlying disease,
LT should be offered only after more conventional
therapy, such as anticoagulation, diuretics or porto-
systemic shunts, had failed [4–7]. However, the above-
mentioned series spanned long periods of time and
included patients operated at the beginning of the liver
transplantation era, with mortality and morbidity figures
higher than currently expected in a young patient
The aim of the present study was to report an updated
picture of LT for BCS, focusing on outcome, on
preoperative prognostic factors, on the impact of the
underlying disease, and on the effects of postoperative
anticoagulation on morbidity and disease recurrence.
2. Patients and methods
To identify the current outcome of LT for BCS the Committee of the
European liver transplantation association (ELTA) decided to start a BCS
project. The data collected in the European liver transplantation registry
(ELTR) served as a basis. A questionnaire was sent to all European centres
known to have transplanted BCS patients between 1988 and 1999. The
centres participated on a voluntary basis (Appendix A).
The questionnaire, intentionally limited to three pages to encourage
completion, contained 121 items (Appendix B).
According to the ELTR, from January 1st, 1988 to December 31st,
1999, 295 LT for BCS were performed in 66 centres. Completed
questionnaires were obtained for 248 patients from 51 centres (compliance:
77% of centres known to have performed LT for BCS, corresponding
to 84% of all procedures performed during the period), no patients were lost
to follow-up in the centres that responded, and all questionnaires were
The number of questionnaires ranged from 1 to 40 in each centre. Six
centres contributed each more than 10 patients, for a total of 115 patients
and 133 patients came from the remaining 45 centres.
Major items (Patient and graft survival, liver function, renal function,
anticoagulation, recurrence of BCS) were available for all patients, and
were cross-checked with ELTR data for accuracy and consistency.
Patients were classified in prognostic categories based on prothrom-
bin time, bilirubin, encephalopathy and ascites as described by Murad et
al. . All four items were available for 151 patients (61% of the 248
patients of the study population—missing data on mainly due to the
prothrombin time before anticoagulation at diagnosis, rather than at
transplantation). These patients did not differ from the 229 patients with
three values available (92% of the study population) and from the whole
population as for the main factors related to severity (creatinine,
3. Statistical methods
Analysis was performed with the SPSS statistical software
(SPSS Inc., Chicago, IL, USA). Descriptive statistics were
provided as meanGSD. The cumulative survival probability
was estimated by the method described by Kaplan and
Meier. Factors influencing survival (covariates) were
assessed by univariate analysis comparing the subgroups
(strata) using the log-rank test. A multivariate analysis was
performed with a Cox proportional-hazards model on all 248
patients, based on the factors that were significant by
univariate analysis. This model, proposed by Cox, is a
multiple regression model for analysis of censored survival
data. A model with the backward elimination method was
used to select the variables. For each variable, the regression
coefficient and its significance were given. For each
significant variable, a relative risk was estimated .
P-values !0.05 were considered statistically significant.
Pre-LT characteristics of the 248 patients are summar-
ized in Table 1.
A cause of BCS had been looked for in all patients, and
the main factors are summarized in Table 2. No patient had
protein S deficiency.
an emergency in 21%, and the urgency status was unknown
in 24% of the patients (this concerned the early cases in the
series when the urgency status was still undefined).
Histology was obtained before transplantation by
percutaneous liver biopsy in 95 patients, during an operation
in 43 patients and by the transjugular approach in six
patients. As six patients had both a percutaneous and an
operative liver biopsy, 138 patients (56%) had liver
histology before LT. Overall, details concerning the
histological diagnosis were available for 184 patients
(74%). Cirrhosis was present in 36 patients (20%), fibrosis
and congestion in 29 (16%), extensive hepatocellular
necrosis and sinusoid dilatation in 53 (29%), congestion
alone was the main histological sign in 57 (31%) and
hyperplastic liver nodules and sinusoid dilatation in eight
patients (4%). Hepatocellular carcinoma in the explanted
liver was found in three patients: one patient had cirrhosis, a
second extensive fibrosis, and nodular hyperplasia was the
prominent sign in the third patient. These three patients
were alive at the end of the study.
Pre-LT treatment included anticoagulants in 58% of the
patients (heparin35%, coumarins in 36%, both treatments in
16% and others 3%), and diuretics in 68%. A surgical porto
systemic shunt (SPSS) had been performed in 20% of the
patients, a transjugular intrahepatic porto systemic shunt
(TIPS) in 4%, a percutaneous angioplasty in 5%. An
intravascular lysis had been done in four patients
(streptokinase/urokinase or actilysis) and prostacyclin had
G. Mentha et al. / Journal of Hepatology 44 (2006) 520–528521
been given in 1 case. LT was performed within 2 weeks of a
shunt procedure (SPSS or TIPS) in 14 patients and within 1
month in 17 patients. In addition to occlusion of the hepatic
veins, pre-LT portal vein thrombosis was present in 19% of
the cases, thrombosis of the inferior vena cava in 16%, and
thrombosis of other veins in 10% (inferior mesenteric vein,
splenic vein, internal jugular vein, femoral vein) (Table 1).
At transplantation, the mean score described by Murad
 was 1.86 (G0.8) for the 151 patients in whom all
variables were available (Class III, with a poor outcome, is
defined by a score of 1.5 or higher): 69 in class III (47%), 76
in class II (51%) and six in class I (2%). Of the 229 patients
with at least three items (92% of the series), 69 were in class
III (46%), 76 in class II (50%) and nine in class I (4%) (the
missing item lowering the score of the patient). The mean
Murad score was 1.77 for the 48 patients with a SPSS, and
1.78 for the 10 patients receiving TIPS, and did not differ
from the mean score of the remaining patients (1.96).
After a median follow-up of 48 months (SD 42 months),
a total of 67 deaths (27% of the patients) were observed in
this study; 33 during the first month (49%) and nine between
1 and 8 years (13%) (Fig. 1). The main causes of early
deaths were sepsis and multiple organ failure (47%). Other
causes were graft dysfunction or hepatic artery thrombosis
(HAT) (19%), venous thrombosis (12%), cardiac compli-
cations (9%), brain damage (5%) and others (8%). Mortality
was high in patients whose transplantation was performed
shortly after a SPSS or a TIPS procedure (9/14 patients
within 2 weeks and 11/17 patients within 1 month:
Thirty-seven patients were retransplanted (15%), four of
them twice: 14 during the first post-transplant week for
primary non-function in 13 and for HAT in one patient,
15 between the second and the 52 week due to HAT
(six patients), portal vein thrombosis (three patients) and
other causes (six patients), eight patients after the first year
post-transplant due to chronic rejection in three, HAT in
three and two for other causes. Eleven of the 37 patients
Aetiology of Budd-Chiari syndrome
Protein C deficiency
Factor V Leiden
Pregnancy or post-partum
AT III deficiency
Paroxysmal nocturnal hemoglobinuria
Total O100% (10% of patients with two or more causes).
Pre-OLT clinical characteristics of patients transplanted for BCS
Prior variceal bleeding
Fulminant hepatic failure
Normal (creatinine %120 mmol/L)
Mildlyimpaired(creatinineO120 and%160 mmol/L)
Severely impaired (creatinine R160 mmol/L)
Surgical porto-systemic shunt
Transjugular intrahepatic shunt
Clinical presentation 95
AST pre-OLT (IU/L)
ALT pre-OLT (IU/L)
Renal function pre-OLT161
Vein thrombosis pre-OLT19
Delay from onset of symptoms to OLT (days)
Delay from onset of symptoms to diagnosis (days)
G. Mentha et al. / Journal of Hepatology 44 (2006) 520–528522
retransplanted died in the first 3 months after retransplanta-
For the entire series, the mean survival was 48.6G42.5
months. Actuarial survival was 75.6% at 1 year, 71.4% at 5
years, and 68.0% at 10 years (Fig. 2). According to the
classification of Murad , the 5-year survival rate was 76%
in class II (115 patients) and 71% in class III (105 patients),
and did not differ significantly between the two groups
(PZ0.83). Among the causes of death there were recurrence
of BCS (despite anticoagulation) in four patients, one
patient died of leukaemia with a myelofibrosis 7 years after
LT, two deaths were unrelated to BCS (ovarian cancer at 42
months and sepsis due to recurrent cholangitis at 57 months)
and the cause of two sudden deaths at home could not be
ascertained. Among the nine patients dying after 1 year,
seven had a myeloproliferative disease.
Two hundred patients out of 235 (85%) received
anticoagulants whereas 10 patients (4%) did not. The
information was missing for 11%, the majority represented
by the 13 patients who died within the first postoperative
day. All remaining patients received life-long anticoagulant
treatment, except 10 patients in whom the cause of the BCS
was protein C or antithrombin III deficiency considered to
have been cured by LT. Of 197 patients living longer than 3
months, 162 patients had coumarins, 18 patients had other
treatments (heparin, aspirin, etc.), 10 patients had no
anticoagulant treatment, and for seven patients the
information was not available. Despite anticoagulants,
venous thrombosis (any site) recurred in 27 patients
(11%). Five patients had thrombosis at multiple sites
(Table 3). Among the six patients in whom BCS recurred,
one patient was successfully retransplanted (at 4 months),
one was successfully treated by TIPS, and four patients died
(24, 34, 56 and 104 months after LT). Of the 27 patients who
had venous thrombosis after LT, 11 died (mortality 41%).
Haemorrhage attributed to anticoagulants was observed
in 27 patients (11%). Two patients with intracranial
bleeding died and the mortality attributed to anticoagulants
was 1% (Table 4).
To assess the factors influencing survival, univariate
analysis of the variables reported in Table 5 was performed.
Creatinine, bilirubin, presence of SPSS/TIPS were found to
influence survival and impacted on early mortality. The
survival according to renal function is indicated in the
Multivariate analysis showed that only pre-LT renal
function and the presence of SPSS/TIPS had an independent
prognostic value (Table 6).
Data of the long-term renal function after LT were
available for 138 of the 151 1-year survivors (91%) at the
time of the study. A normal renal function (creatinine below
120 mmol/L) was documented in 75% of the patients, a mild
impairment (creatinine between 120 and 160 mmol/L) in
17% and a severe impairment (creatinine higher than
160 mmol/L) in 8%.
Patients at risk
171 148 129 111 948065 42 2816 248
Fig. 2. Survival of the 248 patients transplanted for BCS in the study
0-7 days 8 days
Fig. 1. Distribution of mortality after transplantation according to the
date of death (67/248 patients).
Recurrence of vein thrombosis after OLT (nZ27)
Portal vein thrombosis
Hepatic vein thrombosis
Vena cava thrombosis
Mortality after recurrence
Five patients had thrombosis at two or more sites.
Complications of anticoagulant therapy after OLT (nZ27)
Central nervous system bleeding
Mortality due to bleeding
aTwo patients with intracranial bleeding.
G. Mentha et al. / Journal of Hepatology 44 (2006) 520–528523
This study illustrates the status of LT for BCS in Europe
during 12 years. The results show that, while LT in BCS was
a demanding procedure, and early mortality was high, long-
term outcome was excellent, whatever the underlying
aetiology of the disease, with 5 and 10 years survival
figures similar to liver transplantation for more common
The population of the study, although limited to the 51
centres that answered the questionnaire and to 84% of the
patients transplanted for BCS during the period of the
investigation, was representative of general population
suffering from BCS, as shown by a similar distribution of
genders, underlying aetiologies and prevalence of cirrhosis
as in other studies reporting on BCS patients treated without
a liver transplant [8,11,12]. The relatively high mortality in
the series concerned mainly early deaths, reflecting the
complexity of LT for BCS, often performed on patients after
previous medical and surgical treatments. Univariate
analysis showed that a high bilirubin, a previous shunt
procedure, and renal failure were significantly associated
with increased mortality, and the two latter factors were
confirmed on multivariate analysis. Nearly, 40% of early
deaths occurred among patients who had a creatinine higher
than 160 mmol/L at the time of LT, while patients with a
normal renal function did well, with a survival of 77% at 10
years. Conversely, the need of being transplanted as an
emergency for the fulminant form of the disease was not a
significant risk factor. The results on early outcome in our
series expand on the results of previous investigations,
limited by the relative rarity of BCS. [13,14].
Creatinine>160 µmol/L or dialysis
Log rank P=0.0004
Patients at risk
>160 µmol/L or dialysis
Fig. 3. Survival of patients according to renal function at transplan-
tation: normal (serum creatinine %120 mMol/L), mildly impaired
(serum creatinine %160 mMol/L), severely impaired (serum creatinine
O160 mMol/L or dialysis).
Results of univariate analysis
Factor ValuesPatients 5-year survival (%)P
Renal functionNormal (creatinine %120 mMol/l)
Mildly impaired (creatinine
Severely impaired (creatinine
R160 mMol/L) or dialysis
O7 days and %500 days
Delay between diagnosis and OLT52
Underlying disease 45
Urgency status (ELTR)27
Emergency status: transplantation performed as an emergency according to ELTR data; TIPS, transjugular porto-systemic shunt.
G. Mentha et al. / Journal of Hepatology 44 (2006) 520–528524
Among the 248 patients of the study, late mortality was
low. Only nine patients died after 1 year. Seven of these had
a myeloproliferative disorder and five died of causes related
to it (four BCS recurrences and one acute leukaemia).
Although our study failed to show the impact of the
underlying disease on outcome, a longer follow-up may
have revealed a difference between patients with or without
a myeloproliferative syndrome. Myeloproliferative dis-
orders such as polycythaemia rubra vera are slow diseases
with a rate of leukaemic transformation of 10% at 15 years
and 25% at 25 years, and a nearly normal 10-year life
expectancy . Our study, however, shows that the slow
course of myeloproliferative syndromes is not appreciably
affected by transplantation, and that LT can be offered to
these patients in spite of the possible negative effect in the
long term. .
BCS recurring in a transplanted liver was described in
1983 (a 21-year-old woman who, at 1 year, discontinued
anticoagulants before a biopsy) , and there is a
consensus that anticoagulants are mandatory immediately
after LT, except for patients for whom the underlying
disease is cured by transplantation [18–20]. Among the 10
patients not submitted to anticoagulant treatment in the long
term, a 35 year-old patient with BCS thought to be related to
oral contraceptives had coumarin therapy stopped after
several months. The patient suffered a pulmonary embolism
1 year post OLT, and antiphospholipid antibodies were
found. Coumarin was re-instituted with an uneventful
course in the past 7 years. This case exemplifies that several
causes of BCS may coexist and if there is any doubt,
anticoagulant treatment should be maintained after OLT. In
our population, the thrombotic tendency manifested after
LT in several veins (mainly portal, hepatic and caval veins),
despite an apparently close monitoring of anticoagulant
treatment. The balance of the adverse events was strongly in
favour of anticoagulation, with a relatively low risk of death
from haemorrhagic complications (two deaths in 27
patients, 7%), as opposed to a high mortality when
thrombosis recurred (11 deaths in 27 patients, 41%).
One of the main issues in BCS is the indication and
timing of LT as opposed to medical, radiological or surgical
treatment, and how far these options should be pursued
[8,21–30]. Because all the patients were transplanted (i.e.
represented failures of more conservative treatments), our
study can offer only indirect arguments to the discussion.
Two points, however, came out strongly enough to be
considered in the balance. The first is that the overall
outcome of LT was good, especially in terms of late results,
and compared favourably to the results of conservative
treatments reported in other studies, in particular for patients
with severely impaired liver function. In this respect our
study validates Murad’s classification to determine when
LT should be undertaken: the outcome of the 105 class III
patients in the present series was better than the outcome of
the 55 patients in class III of Murad’s study treated without
transplantation (5-year survival 71% in our study vs 42% in
Murad’s series) . The periods of time considered in the
two studies are similar (1988–1999 in the present study and
1984–2001 in Murad’s study). The second point concerns
survival once transplantation is performed: renal function
was the most important prognostic factor of LT (while pre-
transplant Murad’s score was not a predictor of outcome
after transplantation), suggesting that alternatives to LT
should not be pursued up to the stage of renal impairment.
For patients with less severe disease—class II patients—
the outcome of the two series was similar (5-year survival
76% in our study vs 74% in Murad’s) showing that porto-
systemic shunts or other conservative treatments offer
results equivalent to LT and may be the therapies of choice
in selected patients.
For this reason, and because of the specificities of our
population (i.e. the shunt must have failed for our patients to
come to transplantation), we warn against interpreting the
increased mortality associated with a previous SPSS or
TIPS in our investigation as an argument to abandon shunts
as treatment options for BCS. In recent studies, TIPS was
very efficient in the short term, when successfully inserted
[27–29]. However, TIPS dysfunction and re-interventions
were frequent, including, ultimately, liver transplantation in
10–40% of the patients [27–29]. High revision probability
(47% at 1 year), transplantation and mortality (15%) figures
are reported even in the most experienced centres .
PTFE covered stents seem to have a higher patency rate
(67% at 1 year) and may be more successful in the long-term
. While a previous shunt procedure may complicate the
operation , is probably dangerous for the acute form of
the disease , and LT performed shortly after a shunt (i.e.
most likely rescue transplantation) was associated to a high
mortality in our study, porto-systemic shunts can offer
durable remission or cure in selected patients [21–23]. It has
been suggested that elective LT should be preferred to
shunts when patients with the BCS have a cirrhotic liver
[3,4], and our study might have shown that the results of a
subsequent transplant were particularly poor in patients who
had received shunt in the presence of cirrhosis. However,
this information is difficult to obtain even in a prospective
investigation, as the histological picture in BCS can vary
markedly within the same liver. The issue of the best
treatment in the individual patient should remain open.
In conclusion, this European series of 248 patients shows
that LT is an effective treatment for BCS—regardless of the
underlying disease—despite the complexity of the
Results of multivariate analysis (Cox model) nZ215
FactorPatients at riskP Relative risk
TIPS, transjugular porto-systemic shunt.
G. Mentha et al. / Journal of Hepatology 44 (2006) 520–528525
procedure, and its relatively high early mortality. On
multivariate analysis, renal failure before LT was the
strongest predictor of survival, and this result should
encourage considering liver transplantation early, when
more conservative options are not successful, in particular
when poor prognostic factors for survival without trans-
plantation can be identified.
All contributors to the project are acknowledged in
Appendix A. Fifty one centres have participated to the
ELTA project ‘LT FOR BUDD-CHIARI: EUROPEAN
† BARCELONA—Hospital Clı ´nic I Provincial de Barce-
lona (Prof. Luis GRANDE); C.S.U. Bellvitge (Prof. Joan
FIGUERAS, Dr Antonio RAFECAS).
† BELGRADE—KBC ‘Zvezdara ’(Prof. DAPCEVIC).
† BERGAMO—Ospedali Riuniti di Bergamo (Prof. Bruno
GRIDELLI, Dr Marco SPADA).
† BERLIN—Charite ´—Humboldt Univ.-Virchow-Klini-
kum (Prof. Peter NEUHAUS, Dr M. STOCKMANN).
† BERN—Inselspital—Universita ¨tspital Bern (Prof.
M. BU¨CHLER, Dr Ch. SEILER).
† BESANCON—C.H.U. Besanc ¸on (Prof. Georges MAN-
† BIRMINGHAM—The Queen Elizabeth Hospital (Prof.
Paul McMASTER, Miss Bridget GUNSON).
† BOLOGNA—S. Orsola Hospital (Prof. A. MAZ-
ZIOTTI, Dr Gian Luca GRAZI).
† BRUSSELS—Cliniques Universitaires Saint-Luc (Prof.
J. OTTE, Prof. Jan LERUT).
† BUDAPEST—Semmelweis Medical University (Prof.
Ferenc PERNER, Dr I. FEHERKASI).
† CAEN—C.H.U. Caen (Prof. Philippe SEGOL).
† CAMBRIDGE—Addenbrooke’s Hospital (Prof. Peter
FRIEND, Dr R.-K. PRASEEDOM).
† CLICHY—Ho ˆpital BEAUJON (Prof. Jacques BEL-
† COIMBRA—Hospitais da Uinersidade (Prof. A. LIN-
HARES FURTADO, Prof. Rui PERDIGOTO).
† COPENHAGEN—University Hospital Copenhagen
(Prof. Preben KIRKEGAARD, Dr Andre ` WETTERG-
† EDINBURGH—Royal Infirmary of Edinburgh (Prof.
James GARDEN, Ms Rosanne BATE).
† EL PALMAR (Murcia)—Hospital Universitario «
Virgen de la Arrixaca » (Prof. Francisco SA´NCHEZ
† ERLANGEN—Chirurgische Klinik der Univ. Erlangen-
Nu ¨rnberg (Prof. W. HOHENBERGER, Dr Rudolf OTT).
† GENEVE—University Hospital Geneva (Prof. Philippe.
MOREL, Dr Emiliano GIOSTRA).
† GO¨TEBORG—Sahlgrenska Univ. Hospital/University
of Go ¨teborg (Prof. Styrbjo ¨rn FRIMAN).
† GO¨TTINGEN—Georg-August-Universita ¨t Go ¨ttingen
(Prof. Burckhardt RINGE).
(Prof. Xavier ROGIERS, Dr D. BRO¨RING).
† HANNOVER—Medizinische Hochschule Hannover
(Prof. Ju ¨rgen KLEMPNAUER, Dr Thomas BECKER).
† HEIDELBERG—Universita ¨tsklinikum
(Prof. E. KLAR, Dr Gunther WEISS).
† HELSINKI—Univ. of Helsinki/Transplant. and Liver
Surgery (Prof. Krister HO¨CKERSTEDT).
† HUDDINGE—Huddinge University Hospital (Prof.
B. ERICZON, Mr Thomas JOHANSSON).
† INNSBRUCK—C.U.K Innsbruck/Universita ¨tsklinik fu ¨r
Chirurgie (Prof. Raimund MARGREITER, Dr Ruth
† LEIPZIG—Universita ¨tsklinikum
† LIEGE—CHU Lie `ge (Prof. M. MEURISSE, Dr
† LONDON—King’s College Hospital (Prof. John
O’GRADY, Prof Nigel HEATON); Royal Free Hospital
(Prof. A.K. BURROUGHS, Mr K. ROLLES).
† LYON—Ho ˆpital Edouard Herriot (Prof. Olivier BOIL-
LOT, Dr Je ´rome DUMORTIER).
† MADRID—Hospital Universitario « 12 de Octubre »
(Prof. E. MORENO GONZA´LEZ, Prof. C. LOINAZ-
SEGUROLA;Clinica Puerta de Hierro (Prof. J. ARDAIZ
SAN MARTIN, Dr C. BARRIOS PEINADO).
† MARSEILLE—Ho ˆpital de la Conception, Marseille
(Prof. Yves Patrice LE TREUT, Dr Xavier HANNA).
† MILANO—Ospedale Maggiore di Milano (Prof. Luigi
Rainero FASSATI, Dr Umberto MAGGI).
† NICE—C.H.U. Nice/Ho ˆpital de l’Archet II (Prof. Jean
† OSLO—The National Hospital, RIKSHOPITALE (Prof.
O. SOREIDE, Dr Kristian BJORO).
† PADOVA—Clinica Chirurgica III/University Hospital
of Padova (Prof. Giorgio Enrico GERUNDA, Dr
† PAMPLONA—Clinica Universitaria—Universidad de
Navarra (Prof. J. IGNACIO HERRERO).
† PARIS—Ho ˆpital COCHIN (Prof. Didier HOUSSIN,
Dr Jean-Marc THILLOIS); Ho ˆpital Saint-Antoine
(Prof. P. BALLADUR, Pr. O. CHAZOUILLe `RES);
Ho ˆpital Henri Mondor (Cre ´teil) (Prof. Daniel CHER-
† ROTTERDAM—University Hospital Rotterdam (Prof.
SW. SCHALM, S. de RAVE).
† SANTANDER—Hospital Universitario « Marques de
Valdecilla » (Prof. Luis Antonio HERRERA).
† STRASBOURG—Ho ˆpital de Hautepierre (Prof. Daniel
JAECK, Prof. Philippe WOLF).
G. Mentha et al. / Journal of Hepatology 44 (2006) 520–528526
† TOULOUSE—C.H.U. Toulouse (Rangueil) (Prof.
† VILLEJUIF—Ho ˆpital Paul Brousse (Prof. Henri
BISMUTH, Dr Vincent KARAM).
† WIEN—Universita ¨tsklinik
† WURZBURG—Chirurgische Univ. Wu ¨rzburg (Prof.
Appendix B. The questionnaire
The items of the questionnaire can be summarized as
follows: age, sex, dates of transplantation, of retransplanta-
tion, of death, and of the last outpatient visit, patient’s
current activities, cause of death or retransplantation.
Questions on the clinical presentation at onset of the
disease included: interval from clinical symptoms to
diagnosis of BCS, presence of ascites, wasting or
encephalopathy, history of bleeding, and history of
fulminant hepatitis-like presentation.
Questions on the histology before LT included a
description of the diagnosis, presence of cirrhosis and the
method by which histology was obtained: (percutaneous,
transjugular, or surgical liver biopsy during LT).
List of underlying disease included myeloproliferative
syndromes, anti-thrombin III-, protein C- or protein S-
deficiency, presence of a membranous web, treatment with
oral contraceptives, other or not determined causes.
Questions on pre-LT treatment concerned the use of
anticoagulants (anti-vitamin K, heparin), diuretics, surgical
porto-systemic shunts (SPSS), trans jugular porto systemic
shunts (TIPS), percutaneous angioplasty, or others. The
timing of the pre-LT treatment was also included.
Tests of the liver function (AST, ALT, albumin,
bilirubin, prothrombin time) were given pre-LT, in the
early period (!3 months), at the first year after LT and in
the long term. At the same time marks, renal function was
defined as normal (creatinine less than 120 mmol/L), mildly
impaired (creatinine 120–160 mmol/L), severely impaired
(creatinine more than 160 mmol/L), or patient on dialysis.
In case of a post-LT liver biopsy, the date and the results
Question on vein thrombosis included the site (hepatic
veins, portal vein, vena cava or others), and where and
when the thrombosis recurred after LT, during the early
period (!3 months), the first year or in the long term and
whether with or without anticoagulants, the type of
anticoagulants. The time and pattern of disease recurrence
were also asked.
The post-LT complications were defined as: variceal
hemorrhage, intra-abdominal bleeding, other bleeding,
infection, acute or chronic rejection, surgical complications
(artery, biliary tract, portal vein or others) and others.
 Janssen HL, Garcia-Pagan JC, Elias E, Mentha G, Hadengue A,
Valla DC. Budd–Chiari syndrome: a review by an expert panel.
J Hepatol 2003;38:364–371.
 Menon KVN, Shah V, Kamath PS. The Budd–Chiari syndrome. N
Engl J Med 2004;350:578–585.
 Putnam CW, Porter KA, Weil III R, Reid HA, Starzl TE. Liver
transplantation of Budd–Chiari syndrome. J Am Med Assoc 1976;
 Campbell DA, Rolles K, Jamieson N, O’Grady J, Wight D,
Williams R, et al. Hepatic transplantation with perioperative and
long term anticoagulation as treatment for Budd–Chiari syndrome.
Surg Ggynecol Obstet 1988;166:511–518.
 Halff G, Todo S, Tzakis AG, Gordon RD, Starzl T. Liver
transplantation for the Budd–Chiari syndrome. Ann Surg 1990;211:
 Ringe B, Lang H, Oldhafer KJ, Gebel M, Flemming P, Georgii A,
et al. Which is the best surgery for Budd–Chiari syndrome: venous
decompression or liver transplantation? A single-center experience
with 50 patients Hepatology 1995;21:1337–1344.
 Shaked A, Goldstein RM, Klintmalm GB, Drazan K, Husberg B,
Busuttil RW. Porto-systemic shunt versus orthotopic liver transplan-
tation for the Budd–Chiari syndrome. SGO 1992;174:453–459.
 Murad SD, Valla DC, de Groen PC, Zeitoun G, Hopmans JA,
Haagsma EB, et al. Determinants of survival and the effect of
portosystemic shunting in patients with Budd–Chiari syndrome.
 Christensen E. Multivariate survival analysis using Cox’s regression
model. Hepatology 1987;7:1346–1358.
 Adam R, Cailliez V, Majno P, Karam V, McMaster P, Caine RY, et al.
Normalised intrinsic mortality risk in liver transplantation: European
liver transplant registry study. Lancet 2000;356:621–627.
 Zeitoun G, Escolano S, Hadengue A, Azar N, El Younsi M, Mallet A,
et al. Outcome of Budd–Chiari syndrome: a multivariate analysis of
factors related to survival including surgical portosystemic shunting.
 Galati JS, Langnas AN, Donovan JP, Schafer DF, Sorrel MF,
Zetterman RK, et al. The Budd–Chiari syndrome: a therapeutic
quandary. Gastroenterology 1993;104:A904.
 Rao AR, Chui AK, Gurkhan A, Shi LW, Al-Harbi I, Waugh R, et al.
Orthotopic liver transplantation for treatment of patients with Budd–
Chiari syndrome: a single-center experience. Transplant Proc 2000;
 Ulrich F, Steinmuller T, Lang M, Settmacher U, Muller AR, Jonas S,
et al. Liver transplantation in patients with advanced Budd–Chiari
syndrome. Transplant Proc 2002;34:2278.
 Ganguli SC, Ramzan NN, McKusick MA, Andrews JC, Phyliky RL,
Kamath PS. Budd–Chiari syndrome in patients with haematological
disease: a therapeutic challenge. Hepatology 1998;27:1157–1161.
 Srinivasan P, Rela M, Prachalias A, Muiesan P, Portmann B,
Mufti GJ, et al. Liver transplantation for Budd–Chiari syndrome.
 Seltman HJ, Dekker A, Van Thiel DH, Boggs DR, Starzl TE. Budd–
Chiari syndrome recurring in a transplanted liver. Gastroenterology
 Goldstein R, Clark P, Klintmalm G, Husberg B, Gonwa T, Stone M.
Prevention of recurrent thrombosis following liver transplantation for
Budd-Chiari syndrome associated with myeloproliferative disorders:
treatment with hydroxyurea and aspirin. Transplant Proc 1991;23:
 Melear JM, Goldstein RM, Levy MF, Molmenti EP, Cooper B,
Netto GJ,et al. Hematologic aspects ofliver transplantation for Budd–
Chiari syndrome with special reference to myeloproliferative
disorders. Transplantation 2002;74:1090–1095.
G. Mentha et al. / Journal of Hepatology 44 (2006) 520–528527
 Schmid T, Sandbichler P, Pernthaler H, Voger L, Kunz F, Schmid K,
et al. Multiple venous thrombosis with recurrence of Budd–Chiari
syndrome after liver transplantation for paroxysmal nocturnal
hematuria. Clin Transplant 1989;3:194–197.
 Slakey DP, Klein AS, Venbrux AC, Cameron JL. Budd–Chiari
syndrome: current management options. Ann Surg 2001;233:
 Bismuth H, Sherlock DJ. Portasystemic shunting versus liver
transplantation for the Budd–Chiari syndrome. Ann Surg 1991;214:
 Orloff MJ, Daily PO, Orloff SL, Girard B, Orloff MS. A 27-year
experience with surgical treatment of Budd–Chiari syndrome. Ann
 Perello ´ A, Garcı ´a -Paga ´n JC, Gilabert R, Sua ´rez Y, Mojtinho,
Cervantes F, et al. TIPS is a useful long-term derivative therapy for
patients with the Budd–Chiari syndrome uncontrolled by medical
therapy. Hepatology 2002;35:132–139.
 Mancuso A, Fung K, Mela M, Tibbals J,
Burroughs AK, et al. TIPS for the Budd–Chiari syndrome: a single-
centre experience. J Hepatol 2003;38:751–754.
 Sharma S, Texeira A, Texeira P, Elias E, Wilde J, Olliff SP.
Pharmacological thrombolysis in Budd–Chiari syndrome: a single
centre experience and review of the literature. J Hepatol 2004;40:
 Hernandez-Guerra M, Turnes J, Rubinstein P, Olliff S, Elias E,
Bosch J, et al. PTFE-covered stents improve TIPS patency in Budd–
Chiari syndrome. Hepatology 2004;40:1197–1202.
 Attwell A, Ludkowski M, Nash R, Kugelmas M. Treatment of Budd–
Chiari syndrome in a liver transplant unit, the role of transjugular
intrahepatic porto-systemic shunt and liver transplantation. Aliment
Pharmacol Ther 2004;20:867–873.
 Kavanagh PM, Roberts J, Gibney R, Malone D, Hegarty J,
McCormick PA. Acute Budd–Chiari syndrome with liver failure:
the experience of a policy of initial interventional radiological
treatment using transjugular intrahepatic portosystemic shunt.
J Gastroenterol Hepatol 2004;19:1135–1139.
 Rossle M, Olschewski M, Siegerstetter V, Berger E, Kurz K,
Grandt D. The Budd–Chiari syndrome: outcome after treatment
with the transjugular intrahepatic portosystemic shunt. Surgery 2004;
G. Mentha et al. / Journal of Hepatology 44 (2006) 520–528528