Case report

Pharmacological thrombolysis in Budd Chiari syndrome: a singlecentre experience and review of the literature

S. Sharma1, A. Texeira2, P. Texeira2, E. Elias2,*, J. Wilde3, S.P. Olliff1

1Department of Radiology, Queen Elizabeth Hospital, University Hospital Birmingham NHS Trust, The Liver Unit, Birmingham B15 2TH, UK2The Liver Unit, Queen Elizabeth Hospital, University Hospital Birmingham NHS Trust, Birmingham B15 2TH, UK

3Department Haematology, The Liver Unit, Queen Elizabeth Hospital, University Hospital Birmingham NHS Trust, Birmingham B15 2TH, UK

Background/Aims: To review our experience of thrombolytic therapy in patients with acute Budd Chiari syndrome

(BCS).Methods: Records of 10 patients with BCS, treated by thrombolysis over a 12-year period were retrospectively

analysed for demographics, clinical presentation/duration, primary disease, thrombolytic regimen, and follow-up. The

same characteristics were also studied in previously reported patients. The agent used was recombinant tissue

plasminogen activator (tPA) in all patients.

Results: Thrombolysis was used 12 times in 10 patients. Infusion was made systemically in three patients, into the

hepatic artery in one patient, locally into a hepatic vein and/or IVC in four patients and locally within TIPS/portal vein

in two patients. Only one infusion made systemically was partially successful. Adjunctive balloon angioplasty and/or

stent insertion was undertaken for all eight procedures (in six patients) where local infusion was into the hepatic vein orTIPS. Six of these were ultimately successful (in five patients) and two were unsuccessful. Thrombolysis was more likely

to be successful in the presence of a short history of thrombosis, when the thrombolytic agent was locally infused and

when it was combined with a successful radiological procedure. Mean follow-up was 4.5 years (range 1–10 years). No

serious bleeding complication occurred.

Conclusions: We observed no benefit from thrombolysis when delivered systemically or arterially except in one case.

Thrombolysis was useful in adjunctive management of BCS when the drug was infused locally into recently thrombosed

veins that had appreciable flow following partial recanalisation. Thrombolysis was clearly of benefit in the

repermeation of occluded/partially occluded hepatic veins/TIPS when early detection of new thrombus followedinterventional procedures such as balloon angioplasty or stenting of hepatic veins.

q 2003 Published by Elsevier B.V. on behalf of European Association for the Study of the Liver.

Keywords: Thrombolysis; Budd Chiari syndrome; TIPS; Hepatic venous outflow obstruction

1. Introduction

Budd Chiari syndrome (BCS) is a rare condition caused

by hepatic venous outflow obstruction, most commonly due

to thrombotic occlusion of the hepatic veins (HV) and/or the

intra- or supra-hepatic inferior vena cava (IVC) [1]. We

have developed an algorithm for management of BCS

involving a primary preference for dilating/recanalising

hepatic veins to restore venous outflow whenever

possible[2,3]. Bypass by TIPS or surgical shunt is

reserved for those symptomatic patients in whom

restoration of hepatic vein outflow is unsuccessful or

impossible [4]. Numerous case reports describe treatment

by thrombolysis for which there is a theoretical basis in

the BCS [5–18]. However there are no published

guidelines for thrombolysis in this condition and our

own usage has been on a case by case basis.

Warren et al. [5] first reported the successful use of

streptokinase in a patient with BCS. Guerin et al. [6]

suggested early thrombolysis to be an alternative to surgery

in acute BCS. To be effective, it was suggested that the

treatment should consist of early intensive thrombolysis for

up to 1 week, followed without interruption by APTT-

controlled heparin infusion [19]. More recently Slakey et al.

Journal of Hepatology 40 (2004) 172–180

www.elsevier.com/locate/jhep

0168-8278/$30.00 q 2003 Published by Elsevier B.V. on behalf of European Association for the Study of the Liver.

doi:10.1016/j.jhep.2003.09.028

Received 5 March 2003; received in revised form 23 September 2003;

accepted 29 September 2003* Corresponding author.

E-mail address: elwyn.elias@uhb.nhs.uk (E. Elias).

[18] reported successful thrombolytic treatment in five

patients with underlying haematological disease.

Nevertheless thrombolysis is of uncertain therapeutic

utility in this condition and exposes the patient to the risk of

bleeding and pulmonary embolism [20,21]. Contraindica-

tions to thrombolytic agents include—a history of cerebro-

vascular accident, recent invasive or surgical procedure,

recent prolonged cardio-pulmonary resuscitation, marked

hypertension and active peptic ulcer disease [22].

We report 12 years experience of thrombolytic treatment

of BCS both during initial treatment and after a primary

radiological intervention such as TIPS or hepatic vein stent

insertion. We discuss several issues that may determine the

success of thrombolytic treatment, the potential risks

involved and complications encountered.

2. Materials and methods

Ten patients (2M/8F; age range 16–58; mean age 33 years) with acuteand subacute BCS, who received thrombolysis as primary or adjunctivemanagement between September 1990 and January 2002 formed the studygroup. General characteristics and laboratory profile of these patients atpresentation are shown in Table 1A and B. We retrospectively analyseddemographics, clinical presentation, primary disease, duration of symp-toms, details of thrombolytic treatment (agent, regimen and route), results,complications and follow-up (Tables 2 and 3). The same characteristicswere studied in a review of previously reported patients (Table 4).

The extent of thrombosis in the major veins was graded as mild (þ ;involving HV or IVC), moderate (þþ ; involving any two of HV, IVC orportal vein), and marked (þþþ ; involving HV, IVC and portal/splenic/superior mesenteric vein). The underlying occlusion or obstruction of thehepatic veins and/or IVC was also graded as short (less than 2 cm) or long(more than 2 cm). The amount of thrombus within IVC and/or hepatic veinswas graded as minor (i.e. significant remaining patent vein lumen) or major(i.e. extensive or complete occlusion by thrombus).

Written informed consent was obtained in all cases prior tothrombolytic treatment. Thrombolytic therapy was arbitrarily consideredas early when it was instituted within 4 weeks of the start of symptoms,and late when it was started at 4 weeks or later. Treatment was by local orsystemic thrombolysis, usually in association with other recanalisationprocedure(s) viz. hepatic vein/IVC balloon dilatation or stenting.Adjunctive balloon angioplasty was undertaken for all procedures wherelocal infusion into hepatic veins or IVC was used, irrespective of theinitial length of stenosis or thrombotic occlusion. End points wereresolution of the thrombosis or a bleeding complication. Treatment wasabandoned if there was no sign of improvement on the check venogramperformed 1 or 2 days after starting therapy. APTT-controlled heparininfusion was continued during thrombolysis and extended until oralanticoagulation with Warfarin, and definitive haematological treatmentwas instituted.

The procedure was considered successful when there was sustaineddilatation of the vein caliber, improved blood flow, reduction in pressuregradients and improvement was maintained on subsequent review,accompanied by alleviation of symptoms and near normalization of liverfunction tests. It was considered partially successful when flow in the HVand/or IVC could be partially restored; or flow in the IVC could be restoredbut not in the HV; accompanied by only suboptimal recovery in laboratoryparameters and symptoms. The procedure failed when the flow in affectedveins could not be restored. Mesocaval shunting or TIPS were used asrescue procedures. Two patients (nos. 9 and 10) already had TIPS insertedfor BCS at the time of thrombolysis.

The initial clinical diagnosis was confirmed using ultrasound withDoppler. Ultrasound revealed the extent of venous thrombosis/obstruction,ascites and evidence of portal hypertension. Hepatic venography andinferior vena cavography were performed via the transjugular routeinitially. Ultrasound guided percutaneous transhepatic puncture of hepaticT

ab

le1

Gen

era

lch

ara

cter

isti

csa

nd

lab

ora

tory

pro

file

,a

tp

rese

nta

tio

n,

of

pa

tien

tso

fB

CS

(A),

an

dB

CS

wit

hT

IPS

thro

mb

osi

s(B

)

Pat

ien

t,ag

e/se

xP

rim

ary

dis

ease

Ch

ild

pu

gh

sco

reA

scit

esg

rad

eS

eru

mb

ilir

ub

in

(mm

ol/

l)

Ser

um

alb

um

in

(g/l

)

AS

T

(U/l

)

Alk

alin

ep

ho

sph

atas

e

(U/l

)

Ure

a

(mm

ol/

l)

Cre

atin

ine

(mm

ol/

l)

(A)

(1)

58

/MP

NH

C(1

1)

31

29

26

55

23

51

7.6

18

0

(2)

45

/FM

yel

op

roli

fera

tiv

ed

iso

rder

(3)

23

/MP

oly

cyth

aem

iaru

bra

ver

aB

(8)

31

83

81

26

16

04

.58

7

(4)

28

/FE

ssen

tial

thro

mbo

cyth

aem

iaB

(9)

34

23

04

88

19

3.3

81

(5)

34

/FE

ssen

tial

thro

mbo

cyth

aem

ia3

(6)

16

/FM

yel

op

roli

fera

tiv

ed

iso

rder

B(8

)1

10

48

38

19

13

.47

1

(7)

30

/FP

rote

inC

defi

cien

cyC

(12

)2

13

52

97

81

57

(8)

40

/FP

rote

inC

defi

cien

cyB

(9)

31

42

41

15

26

92

.06

2

(B)

(9)

29

/FE

ssen

tial

thro

mbo

cyth

aem

iaC

(11

)3

27

31

38

52

43

87

5

(10

)2

7/F

Pro

tein

Cd

efici

ency

A0

30

43

39

36

52

.78

0

Mis

sin

gd

ata

are

un

kno

wn

.A

scit

esar

eg

rad

edas

mil

d(1

),m

od

erat

e(2

)o

rse

ver

e(3

).

S. Sharma et al. / Journal of Hepatology 40 (2004) 172–180 173

Table 2

Characteristics of eight patients with Budd Chiari syndrome treated by tPA thrombolysis

Patient Length

of history

# tPA regimen and time

elapsed in starting therapy

Site of infusion Extent of thrombosis

and occlusion

Result Complication related

to thrombolysis

Rescue procedure Follow-up time

(1) Not known Not known Systemic HV (mild); þ ; major Unsuccessful None TIPS day after Died after 1 year

(2) Few months (Delayed)

Bolus of 10 mg

Systemic HV, IVC, PV, SMV (marked);

þþþ ; major

Unsuccessful None – Died day after

40 mg/h

50 mg/2 h

(3) 3 weeks (Early) Local (hepatic vein) HV (mild); þ ; minor/short Successful None – 9 years

0.5 mg/h for 24 h

(4) 3 weeks (Early)

Bolus of 5 mg

Hepatic artery HV, IVC (moderate);

þþ ; major

Unsuccessful None TIPS 7 days after 4 years

1 mg/h for 48 h

(5) Several weeks (Delayed)

Bolus of 10 mg

Systemic HV, IVC (moderate);

þþ ; major

Partially

successful

Bruise over sternal

aspiration site

Meso-caval shunt

10 days after

10 years

2 mg/h for 72 h

(6) Few weeks (Early)

Bolus of 5 mg

Local (hepatic

vein and IVC)

HV (mild); þ ; short/minor Successful None – 6 years

1 mg/h for 48 h

(7) 3 Weeks (Early)

1 mg/h for 96 h

Local (hepatic vein) HV (mild); þ ; major Unsuccessful None Mesocaval shunt

2 day after

7 year

(8) 4 weeks First time: (delayed)

Bolus of 5 mg

First time: local

(hepatic vein and IVC)

HV, IVC (moderate);

þþ ; major/short

Initially failed None – 8 years

1 mg/h for 48 h

3 weeks Second time: (early)

0.5 mg/h for 96 h

Second time: local;

(hepatic vein and IVC)

HV, IVC (moderate);

þþ ; major/short

Then successful None

S.

Sh

arm

aet

al.

/Jo

urn

al

of

Hep

ato

log

y4

0(2

00

4)

17

2–

18

01

74

vein segments was used when the hepatic veins could not be entered fromthe IVC but were visibly present on ultrasound [23]. The femoral approachto the IVC was also used in selected cases. Percutaneous liver biopsy wasnot generally done as patients had ascites and were candidates forthrombolysis and anticoagulation. After fluoroscopic placement of 5fmultiple side hole infusion catheter in the target vein, the infusion wasstarted in the angio suite and later continued in liver ITU. Patients weremonitored for vital signs and bleeding complications. For systemicinfusion a peripheral IV line in the arm was used. Hepatic artery infusionwas performed by a 5f catheter placed in the common hepatic artery.Resolution of the thrombus was assessed by injecting contrast through thecatheter used for local thrombolysis. Once a satisfactory venogram wasobtained and balloon angioplasty/stenting completed, the catheter wasremoved.

Early in our series thrombolysis was used as a sole therapy, but morerecently it has been used in combination with other recanalisationprocedures. Thrombolysis was indicated when venography suggestedsignificant intraluminal thrombus complicating or preventing adequaterestoration of venous flow. If the vein rethrombosed or failed to display awide and smooth lumen, thrombolysis was continued or repeated. Balloonangioplasty and/or stent insertion were also performed and repeated asrequired.

Patients were discharged from hospital 1–3 weeks after treatment andfollowed up at 3 month intervals or sooner if indicated. At the follow-upvisits, physical examination and liver function tests and ultrasound weredone. Contrast venography and pressure measurements were repeated ifthere was any clinical or ultrasound suggestion of rethrombosis ornarrowing, reversed or poor venous flow. Patients with TIPS were followedup by ultrasound with routine venography at 12 months or earlier if therewere clinical or ultrasound signs of stenosis or occlusion.

The underlying thrombotic disorder was sought in all patients andappropriately trated.

3. Results

Eight surviving patients have been followed up for 1–9

years (mean 4.5 years).

3.1. Thrombolysis

3.1.1. Non-TIPS patients

Eight patients (nos. 1–8, Tables 1A and 2) had

thrombolysis either before or after their initial interven-

tional radiology investigation and treatment. Three patients

had successful radiological dilatation/recanalisation pro-

cedures of HV and/or IVC (nos. 3, 6 and 8). One procedure

(no. 5) was partially successful in reducing IVC thrombus

such that surgical mesocaval shunt could be performed.

One patient (no. 2) died the day after thrombolysis

commenced. TIPS (patients 1 and 4) and mesocaval

shunt (patient no. 7) were definitive treatments after

angioplasty and thrombolysis proved unsuccessful in the

other three patients.

3.1.2. Following TIPS for Budd Chiari syndrome

Patients 9 and 10 had TIPS performed for treatment of

Budd Chiari and later developed thrombosis of TIPS and

portal vein. They were successfully thrombolysed and

recanalised (Tables 1B and 3). A large thrombus occluding

the TIPS was discovered during routine follow-up ultra-

sound in one patient (no. 10) while clinically asymptomatic.Tab

le3

Ch

ara

cter

isti

cso

ftw

oB

CS

pa

tien

tsw

ith

thro

mb

ose

dT

IPS

,tr

eate

db

yth

rom

bo

lysi

s

Pat

ien

t,

age/

sex

Pri

mar

yd

isea

seL

eng

tho

f

his

tory

tPA

regim

enan

dti

me

elap

sed

sin

cest

arti

ng

ther

apy

Sit

eo

fin

fusi

on

Res

ult

Com

pli

cati

on

rela

ted

toth

rom

bo

lysi

s

Res

cue

pro

ced

ure

Fo

llow

-up

tim

e

(9)

29

/FE

ssen

tial

thro

mbo

cyth

aem

ia

7w

eek

sF

irst

tim

e:(l

ate)

Bolu

s5

mg

Lo

cal

(TIP

S)

Init

ial

succ

ess

tem

po

rary

No

ne

Nil

1y

ear

1m

g/h

for

6d

ays

8w

eek

sS

eco

nd

tim

e:(l

ate)

Bolu

s5

mg

Lo

cal

(SM

V)

Lat

erco

mp

lete

ly

succ

essf

ul

No

ne

0.5

mg

/hfo

r4

8h

(10

)2

7/F

Pro

tein

Cd

efici

ency

**

Bolu

s5

mg

(no

tk

no

wn)

1m

g/h

for

48

h

Loca

l(T

IPS

)S

ucc

essf

ul

Ble

edfr

om

nec

ksh

eath

Nil

1y

ear

**

Thro

mb

osi

sd

isco

ver

edd

uri

ng

rou

tin

efo

llo

w-u

pin

asy

mp

tom

atic

pat

ien

t.

S. Sharma et al. / Journal of Hepatology 40 (2004) 172–180 175

Table 4

Characteristics of reported cases of thrombolytic therapy in BCS

Author Number ofcases, age/sex

Primaryaetiology

Lengthof history

Agent, regimen and timeelapsed in starting therapy

Site of infusion Extent ofthrombosis

Result Complicationrelated toprocedure

Follow-upduration

Assistedrecanalisationprocedure

Greenwoodet al. [7]

1, 41/M ? ? Urokinase:(not known)Bolus 4400 U/kg4400 U/kg/hfor 55 h

Local (IVC) Mild; þ Successful Intraperitonealhaemorrhage

Well at 1year, died18 monthsafter

Nil

Casselet al. [6]

1, 46/F Polycy-thaemiavera

3 months Streptokinase:(delayed)Bolus 600,000 U

Systemic Mild; þ Successful Mild pyrexia 12 months Nil

100,000 U/h for96 h

Guerinet al. [10]

1, 34/M Recentpregnancyand oralcontraceptives

? Streptokinase:(not known)Bolus of 250,000 U150,000 U/h/48 h;

Systemic Mild; þ Successful None ? Nil

Lys-plasminogen 100 mg/2 h160,000/h/48 h

Warrenet al. [5]

1, 22/F Oral contra-ceptives

? Streptokinase:(not known)250,000 U/h

Systemic Moderate; þþ Successful Recurrentpulmonaryemboli

4 months Nil

100,000 U/h/72 h

Sholar [8] 2, 33/F PNH Case 1:5 days

Case 1 (first time)Streptokinase: (early)

Case 1 First time:local (hepatic vein)

Marked; þþþ Successful None Case 1:2 years

Nil

7500 U/h, then5000 U/h(total 72 h)(Second time, 2 monthsafter)

Second time:systemic

Moderate; þþ Successful

Urokinase: (delayed)250,000 U/40 min250,000 U/2 h

33/M Case 2:7–10 days

Case 2Streptokinase: (early)

Case 2: systemic Moderate; þþ Successful None Case 2:5 years

Nil

250,000 U/30 min100,000 U/h/48 h

McMullinet al. [15]

2, 33/F PNH Case 1:3 weeks

Case 1: tPA30 mg/24 h (early)

Case 1: systemic Moderate; þþ Successful None Case 1:6 years

Nil

22/F PNH Case 2:severalweeks

Case 2: tPAFirst time15 mg/3 h (delayed)

Case 2: Systemic Mild; þ Unsuccessful None Case 2:2 years

Nil

Second time (after 9 days) Hepatic artery Mild; þ Unsuccessful24 mg/48 h (delayed)Third time (after 2 days) Systemic Mild; þ Successful25 mg/3 h (delayed)Then, 50 mg/24 h

Raju et al. [16] 1, 59/M ? Few hours Urokinase: (early) Local (IVC) Mild; þ Successful None 12 months Nil300,000 U bolus300,000 U/h for 72 h

S.

Sh

arm

aet

al.

/Jo

urn

al

of

Hep

ato

log

y4

0(2

00

4)

17

2–

18

01

76

3.1.3. Outcome versus assisted recanalisation procedure

An appropriate adjunctive procedure (balloon dilatation

or stenting of HV/IVC) was used to ensure improved blood

flow before (nos. 6 and 7) or after (nos. 3 and 8)

thrombolysis in all four patients in whom local infusion

was made. Three patients had a successful outcome. In one

patient (no. 7) the patency of the HV could not be

maintained during thrombolysis even after HV dilatation.

3.1.4. Outcome versus route of infusion

None of the three systemic infusions were completely

successful. An infusion into the hepatic artery failed to

restore flow in the hepatic veins in one patient. Systemic

infusion in one patient (no. 5) in whom the patency of

hepatic veins could not be restored, cleared the IVC of

thrombus and enabled mesocaval shunt surgery with an

excellent long-term outcome. Of five local infusions in four

patients, three patients were ultimately successful treated by

thrombolysis and dilatation/stenting.

Local infusion into thrombosed TIPS/portal vein was

successful in two patients.

3.1.5. Outcome versus length of history (age of thrombus)

No success could be achieved in a patient (no. 1) in

whom the length of history was unknown but presumed

long. No success was achieved in a patient (no. 2) with a

history of several months. All other patients had symptoms

ranging from a few to several weeks duration. Four (nos. 3,

6, 8 and 9) of these were treated successfully, one (no. 5)

had partial success and two (nos. 4 and 7) were unsuccess-

ful. One patient (no. 8) with a few weeks history did not

respond until the hepatic veins were recanalised by

transhepatic balloon dilatation.

3.1.6. Outcome versus length of venous stenosis

Two patients (nos. 3 and 8) had short focal stenoses due to

‘webs’ in the hepatic vein and IVC, respectively. The short

length of occlusion was only appreciated after the diffusely

thrombosed lumen was satisfactorily cleared. Hepatic vein

balloon angioplasty and thrombolysis was very effective in

one patient (no. 3). Extensive IVC thrombosis below a web

(patient no. 8) was treated by serial thrombolysis and balloon

dilatation. The central hepatic vein in this patient remained

occluded even after IVC clearance and was recanalised 8 days

later by a combined transhepatic/transjugular dilatation

procedure. Patient 6 also had a short length occlusion of

the hepatic vein confluence, which was traversed, dilated and

stented prior to an acute thrombosis, which then responded to

thrombolysis. These three patients have had excellent long-

term clinical benefit. In contrast, patient 7 had very extensive

thrombosis of hepatic veins and failed to respond to

thrombolysis and angioplasty.

3.1.7. Outcome versus extent of thrombosis

Eight procedures were employed for thrombosis that

was graded as mild or moderate in extent, i.e. either onlyDe

Ste

fano

[12]

1,

29/F

PN

HW

ith

pre

gnan

cy?

tPA

firs

tti

me:

(not

know

n)

50

mg/5

hS

ame

repea

ted

afte

r48

h

Syst

emic

Moder

ate;

þþ

Succ

essf

ul

tem

pora

ryS

ever

ehae

mat

uri

a?

(Cae

sare

andel

iver

yat

33

wee

ks)

Nil

Sec

ond

tim

e:(n

ot

know

n)

50

mg/5

hS

yst

emic

Mil

d;þ

Succ

essf

ul

Wel

lat

2w

eeks

Sam

ere

pea

ted

afte

r48

h

Ishig

uch

iet

al.

[9]

1,

42/F

PN

HC

hro

nic

*(s

ever

alm

onth

s)

Uro

kin

ase:

(del

ayed

)10,0

00

U/m

info

r35

min

Then

240,0

00

U/d

ayfo

r7

day

s

Loca

l(I

VC

)M

oder

ate;

þþ

Succ

essf

ul

None

14

month

sIV

Can

gio

pla

sty

and

sten

ting

McK

eeet

al.

[9]

1,

26/F

Post

-par

tum

3w

eeks

Str

epto

kin

ase:

(ear

ly)

Thre

ein

ject

ions:

10

mg,

10

mg,

5m

g(t

ota

l25

mg)

Syst

emic

Mil

d;þ

Succ

essf

ul

Mil

dw

hee

ze9

month

sN

il

Kw

anet

al.

[14]

1,

47/F

PN

H2

wee

ks

tPA

:(e

arly

)100

mg

over

3h

Syst

emic

Mil

d;þ

Par

tial

lysu

cces

sful

None

Die

d10

day

saf

ter

Nil

Ilan

etal

.[1

1]

1,

29/F

Post

-par

tum

Few

day

sS

trep

tokin

ase:

(ear

ly)

for

72

h(d

ose

?)L

oca

l(I

VC

)M

oder

ate;

þþ

Par

tial

lysu

cces

sful

None

Die

d9

month

saf

ter

IVC

and

HV

Angio

pla

sty

?,n

ot

kn

ow

n;

*,

IVC

mar

ked

lyn

arro

wan

dca

lcifi

ed.

S. Sharma et al. / Journal of Hepatology 40 (2004) 172–180 177

hepatic veins involved or hepatic veins and IVC were

thrombosed. Those with a smaller amount of thrombus in

the hepatic veins (patients 3 and 6) did well. Patient 8

initially had very extensive IVC thrombus and short

length occlusion of hepatic veins, which were ultimately

successfully treated by thrombolysis and dilatation. One

patient (no. 2) who had extensive thrombotic occlusion of

all hepatic veins, spleno-portal-mesenteric axis and had

extensive gastro-esophageal varices succumbed despite

aggressive thrombolytic therapy.

3.1.8. Outcome versus treatment of underlying

hematological condition

All patients were treated for the underlying cause of

thrombosis. In one patient (no. 9), local infusion made into

the TIPS and portal vein resulted in a slow recanalisation

that persisted only for the next 8 weeks. Essential

thrombocythaemia was only diagnosed at this stage. Repeat

local thrombolysis with balloon dilatation along with the

specific anti-platelet treatment, resulted in a more prolonged

patency.

3.1.9. Complications

Only two patients developed minor bleeding compli-

cations (Tables 2 and 3). None required transfusion. In one

patient (no. 10) the treatment was stopped after ooze around

the neck sheath, but thrombolysis was complete by this

time.

4. Discussion

We describe the successful use of thrombolysis in five

patients when deployed in association with repermeation of

thrombus within recanalised hepatic veins, inferior vena

cava or TIPS. Success may be attributable to more than one

factor. Firstly, the thrombus was recent in origin and of

known age particularly when it followed diagnostic or

therapeutic interventional radiology. Secondly, thromboly-

sis was instituted immediately after mechanically re-

establishing venous patency. Maintenance of flow through

the thrombosed vein appears critical to the success of the

technique. When hepatic venous outflow obstruction is

complete, blood is likely to bypass the severely congested

liver via retrograde flow in the portal vein and/or collaterals

and there will be poor contact of systemically administered

thrombolytic drug to the thrombosed hepatic veins. Local

thrombolysis also requires some local blood flow to

maintain and improve the patent channel. Thirdly, local

infusion into the vein immediately proximal to or within the

thrombus achieves high concentrations of the thrombolytic

agent around the thrombus. Reduction of portal pressure by

recanalisation procedure immediately before thrombolysis

also reduces the risk of a complicating variceal

haemorrhage.

Thrombolysis can be repeated after previously successful

or unsuccessful treatment. TIPS or shunt surgery can be

done after thrombolysis fails. Indeed three of our patients

who had failed thrombolysis were successfully salvaged by

TIPS and shunt surgery. In one of the patients, whom the

HVs remained occluded after thrombolysis, a surgical

mesocaval shunt could be created because of repermeation

of the previously occluded vena cava. Vena caval occlu-

sions have been treated with combinations of local

thrombolysis, balloon angioplasty and metallic stent(s)

[13]. Local thrombolysis via the transjugular and transhe-

patic approach has been used successfully in early post-liver

transplant portal vein thrombosis [24,25]. Blum et al. [26]

recently reported the successful placement of TIPS followed

immediately by local low dose thrombolysis in seven

cirrhotic patients with portal vein thrombosis, without

complication. Before the administration of the thrombolytic

therapy, balloon angioplasty was performed in the throm-

bosed main portal vein to restore some patency. Portal vein

blood flow could not be established with mechanical means

alone. Mechanical thrombectomy using an Amplatz

Thrombectomy Device (ATD) is recently reported as a

useful complementary technique in patients with thrombosis

of large native vessels, grafts or TIPS shunts where

thrombolysis alone has either failed or is contraindicated

[27,28]. While hepatic decompression is immediate after

TIPS or surgical shunt, thrombolysis produces a slow

improvement.

TIPS is now established as an effective treatment for

variceal bleeding and ascites and also for selected cases of

BCS [29]. TIPS is generally accepted to have a significant

risk of stenosis or occlusion but Budd Chiari patients may

have a potentially greater risk especially if there is an

aggressive underlying thrombotic tendency. Our two

patients show that thrombolysis can have a beneficial role

in combination with balloon dilatation and/or restenting in

completely occluded thrombosed TIPS in Budd Chiari

patients.

Both local and systemic administration of thrombolytic

drugs have been used in the treatment of BCS. There are no

studies directly comparing the efficacy of local versus

systemic infusion. Catheter directed infusion is of proven

efficacy in arterial and venous thrombosis [30]. We used the

hepatic artery to deliver tPA for 2 days in one of our patients

with blocked HVs with no success. tPA infusion into the

hepatic artery was used by Mc Mullin et al. [15] to no

advantage in their patient after initially failed systemic

infusion. Success with systemic infusion of thrombolytic

agents has been described [5,6,8–10,12,15], but we had

partial success only in one of our patients. Surgical

thrombectomy is usually not technically possible in BCS

patients to restore flow [31].

No data is available to direct the choice of thrombolytic

agent in the setting of BCS. tPA is a natural human enzyme

with no antigenicity. It has a very short half-life of 5 min

and may be discontinued if bleeding occurs. tPA was

S. Sharma et al. / Journal of Hepatology 40 (2004) 172–180178

reported to be most effective and produced earlier

reperfusion following pulmonary embolism [32]. In the

treatment of acute myocardial infarction, the accelerated

regimen using tPA offers a small advantage over

streptokinase (SK) in lysing clots from the coronary

arteries but with a marginally higher incidence of stroke

[33]. Urokinase or tPA should be preferred in a patient who

has high levels of circulating antistreptococcal antibodies

due to previous SK administration. tPA is more expensive

than SK per therapeutic dose.

Thrombolysis in advanced age is associated with

increased risk of bleeding complications [34]. BCS

patients tend to be younger and therefore more likely

to withstand thrombolytic therapy without complication

as compared to those treated for myocardial infarction or

stroke. The thrombus ‘load’ may have a bearing on the

eventual outcome of thrombolysis [13]. It is known that

the patients with massive thrombus have a poorer

prognosis [35]. In our experience the finding of a short

underlying occlusion or stenosis increased the chance of

eventual clinical success for thrombolysis/recanalisation

even when it was associated with a large amount of

thrombus occluding or partially occluding the veins.

This is a retrospective, observational study in a small

number of patients. The patients treated are hetero-

geneous in terms of their clinical presentation, extent of

thrombosis, chronicity, underlying etiologies, risk factors,

and thrombolytic dosage regimens used. In most

published reports the thrombolytic treatment was pri-

mary, whereas in our cases it was often adjunctive.

We propose that thrombolysis has an important role in

the management of acute and subacute forms of BCS in

selected patients in combination with other interventional

radiological techniques. Local infusion into partially

recanalised veins with some appreciable flow is best. It

is potentially repeatable and does not preclude other

more invasive treatments. As an adjunct, thrombolysis is

best employed in a patient who presents early, has a

thrombus that is limited in extent and not completely

occlusive. Close clinical, imaging and laboratory follow-

up along with appropriate treatment of underlying

haematological disorders can ensure improved long-term

venous patency and hence patient survival. Thrombolysis

alone especially when systemically administered appears

of limited value, contrary to the earlier literature. Pooling

and analysis of data from other centres is needed before

more extensive conclusions and recommendations can

be made.

References

[1] Janssen HLA, Garcia-Pagan J-C, Elias E, Mentha G, Hadengue A,

Valla D-C. Budd Chiari syndrome: a review by an expert panel.

J Hepatol 2003;364–371.

[2] Griffith JF, Mahmoud AE, Cooper S, Elias E, West RJ, Olliff S.

Radiological intervention in Budd Chiari syndrome: techniques and

outcome in 18 patients. Clin Radiol 1996;51:775–784.

[3] Fisher NC, McCafferty I, Dolapci M, Wali M, Buckels JAC, Olliff SP,

et al. Managing Budd Chiari syndrome: a retrospective review of

percutaneous hepatic vein angioplasty and surgical shunting. Gut

1999;44:568–574.

[4] Olliff S. Radiological treatment of Budd Chiari syndrome. Imaging

1998;10:81–88.

[5] Warren RL, Schlant RC, Wenger NK, Galambos JT. Treatment of

Budd Chiari syndrome with streptokinase (abstr). Gastroenterology

1972;62:200.

[6] Cassel GA, Morley JE. Hepatic vein thrombosis treated with

streptokinase. S A Med J 1974;16:2319–2320.

[7] Greenwood LH, Yrizarry JM, Hallett Jr JW, Scoville Jr GS. Urokinase

treatment of Budd Chiari syndrome. Am J Radiol 1983;141:

1057–1059.

[8] Sholar PW, Bell WR. Thrombolytic therapy for inferior vana cava

thrombosis in paroxysmal nocturnal haemoglobinuria. Ann Internl

Med 1985;103:539–541.

[9] McKee CM, Mayne EE, Crothers JG, Callender ME. Budd Chiari

syndrome treated with acylated streptokinase–plasminogen complex.

J Roy Soc Med 1985;78:768–769.

[10] Guerin JM, Meyer P. Need for early thrombolysis in Budd Chiari

syndrome. Gastroenterology 1988;94:1109.

[11] Ilan Y, Oren R, Shouval D. Postpartum Budd Chiari syndrome with

prolonged hypercoagulability state. Am J Obstet Gynecol 1990;162:

1164–1165.

[12] De Stefano V, Leone G, Teofili L, Ricerca BM, Pearelli L,

Menichella G, Caruso A, et al. Hepatic and portal vein thrombosis

in a pregnant women with paroxysmal nocturnal haemoglobinuria

(PNH) and severe thrombocytopenia: Efficacy of tissue plasmino-

gen activator in the treatment of Budd Chiari syndrome and

successful outcome of pregnancy (abstract). Thrombo Haemost

1991;65:1188.

[13] Ishiguchi T, Fukatso H, Itoh S, Shimamoto K, Sakuma S. Budd Chiari

syndrome with long segmental inferior vana cava obstruction:

treatment with Thrombolysis, angioplasty and intravascular stents.

J Vasc Interven Radiol 1992;3:421–425.

[14] Kwan T, Hansard P. Recombinant tissue-plasminogen activator for

acute Budd Chiari syndrome secondary to paroxysmal nocturnal

haemoglobinuria. NY State J Med 1992;92:109–110.

[15] McMullin MF, Hillmen P, Jackson J, Ganly P, Luzzatto L. Tisue

plasminogen activator for hepatic vein thrombosis in paroxysmal

nocturnal haemoglobinuria. J Internal Med 1994;235:85–89.

[16] Raju GS, Felver M, Olin JW, Satti SD. Thrombolysis for acute Budd

Chiari syndrome: Case report and literature review. Am J Gastro-

enterol 1996;91:1262–1263.

[17] Leebeek PWG, Lameris JS, van Buren HR, Gomez E, Madretsma

S, Sonneveld P. Budd Chiari syndrome, portal vein and mesenteric

vein thrombosis in a patient homozygous for factor V Leiden

mutation treated by TIPS and thrombolysis. Br J Haematol 1998;

102:929–931.

[18] Slakey DP, Klein AS, Venbrux AC, Cameron JL. Budd Chiari

syndrome: current management options. Ann Surg 2001;233:

522–527.

[19] Loliger EA. Fibrinolytic and anticoagulation therapy in Budd Chiari

syndrome. Review and recommendations (in German). Wien Klin

Wochenschr 1988;100:43–46.

[20] Sherlock S, DooleyJ J. The hepatic artery and hepatic veins: The liver

in circulatory failure. In: Sherlock S, Dooley J, editors. Diseases of the

liver and biliary system, 11th ed. Oxford: Blackwell Science; 2001. p.

192–197.

[21] Elliot MS, Immelman EJ, Jeffery P, et al. A comparative randomised

trial of heparin versus streptokinase in treatment of acute proximal

venous thrombosis: an interim report of prospective trial. Br J Surg

1979;66:838–843.

S. Sharma et al. / Journal of Hepatology 40 (2004) 172–180 179

[22] Antman EM, Braunwald E. Acute myocardial infarction. In: Antman

EM, Braunwald E, editors. Harrison’s principles of medicine, 15th ed.

Singapore: McGraw-Hill; 2001. p. 1386–1398.

[23] Cooper S, Olliff S, Elias E. Recanalisation of hepatic veins by a

combined transhepatic, transjugular approach in three cases of Budd

Chiari syndrome. J Interven Radiol 1996;11:9–13.

[24] Ciccarelli O, Goffette P, Lattere PF, Danse E, Wittenbolle X, Lerut J.

Transjugular intrahepatic portosystemic shunt approach and local

thrombolysis for treatment of early post transplant portal vein

thrombosis. Transplantation 2001;72:159–161.

[25] Bhattacharya T, Olliff SP, Bhattcharya S, Mirza DF, Mc Master P.

Percutaneous portal vein thrombolysis and endovascular stent for

management of post transplant portal vein conduit thrombosis.

Transplantation 2000;27:2195–2198.

[26] Blum U, Haag K, Rossle M, Ochs A, Gableman A, Boos S, Langer M.

Noncavernomatous portal vein thrombosis in hepatic cirrhosis:

treatment with transhepatic intrahepatic portosystemic shunt and

local thrombolysis. Radiology 1995;195:153–157.

[27] Muller HS, Hopfner M, Hilbert C, Krammer HH, Heller M.

Mechanical thrombectomy of acute thrombosis in transjugular

intrahepatic portosystemic shunt. Invest Radiol 2000;35:385–391.

[28] Uflacker R. Mechanical thrombectomy in acute and subacute

thrombosis with use of Amplatz device: arterial and venous

applications. J Vasc Interv Radiol 1997;8:923–932.

[29] Blum U, Rossle M, Haag K, Ochs A, Blum HE, Hauenstein KH, et al.

Budd-Chiari syndrome: technical, hemodynamic, and clinical results

of treatment with transjugular intrahepatic portosystemic shunt.

Radiology 1995;197:805–811.

[30] van Breda A, Katzen BT, Deutsch AS. Urokinase versus streptokinase

in local thrombolysis. Radiology 1987;165:109–111.

[31] Vogt PR, Anderson LC, Jenni R, et al. Dorsocranial liver resection

and direct hepato-atrial liver anastomosis for hepatic venous outflow

obstruction: long term outcome and functional results. Am J

Gastroenterol 1996;91:539–544.

[32] Bozzoli C, Renzi C, Puddu GM, Muscari A, Puddu P. Treatment of

pulmonary emobolism: Current status and future prospects (in

Italian). Clin Ter 1993;142:497–505.

[33] Ganz P, Barunwald E, editors. Acute myocardial ischaemia, 5th ed.

Heart disease: a text book of cardiovascular medicine, vol. 2.

Singapore: WB Saunders; 1997. p. 1184–1288.

[34] Braithwaite BD, Davies B, Birch PA, Heather BP, Earnshaw JJ.

Managemant of acute leg ischaemia in the elderly. Br J Surgery 1988;

85:217–220.

[35] Mahmoud AEA, Helmy AS, Billingham L, Elias E. Poor prognosis

and limited therapeutic options in patients with Budd Chiari syndrome

and portal vein system thrombosis. Eur J Gastroenterol Hepatol 1997;

9:485–489.

S. Sharma et al. / Journal of Hepatology 40 (2004) 172–180180