Downloaded from www.ajronline.org by 186.151.63.12 on 07/09/14 from IP address 186.151.63.12. Copyright ARRS.

For personal use only; all rights reserved

Symptomatic Intrahepatic
Portosystemic Venous Shunt:
Angiographic Findings and Transcatheter
Embolization with an Alternative
Approach
Shuichi Tanoue1
Hiro Kiyosue 1
Eiji Komatsu 2
Yuzo Hori 3
Tohru Maeda 2
Hiromu Mori 1

OBJECTIVE. Intrahepatic portosystemic venous shunt is relatively rare and not well recognized. Awareness of intrahepatic communications is important because they can cause encephalopathy, and most of these shunts can be completely cured by transcatheter embolization. In this
study, we describe the angiographic findings and transcatheter embolization techniques using
several approaches for the treatment of intrahepatic portosystemic venous shunt.
MATERIALS AND METHODS. Between 1989 and 2001, we treated 10 patients with
symptomatic intrahepatic portosystemic venous shunt by performing transcatheter embolization
with Gianturco coils, fibered platinum coils, detachable balloons, and detachable microcoils using one of three approaches to access the portal venous system: transileocolic obliteration (n = 2),
percutaneous transhepatic obliteration (n = 4), or retrograde transcaval obliteration (n = 4).
RESULTS. In all patients, complete obliteration or nearly complete obliteration was confirmed angiographically, and symptoms related to portalsystemic encephalopathy improved
after treatment. Complications were observed in three patients: adhesive ileus in a patient
treated by transileocolic obliteration and thrombosis of intrahepatic portal branches in two patients treated by percutaneous transhepatic obliteration.
CONCLUSION. On angiography, two types of intrahepatic portosystemic venous shunt
were seen: intrahepatic portal venoushepatic venous communication and intrahepatic portal
venousperihepatic venous communication. Transcatheter embolization is effective for treatment of intrahepatic portosystemic venous shunt. Retrograde transcaval obliteration is the
least invasive technique and is recommended as the first choice for treatment of portosystemic
venous shunt except in patients with multiple shunts.

Received June 7, 2002; accepted after revision

December 6, 2002.
1

Department of Radiology, Oita Medical University, 1-1,

Idaigaoka, Hasama-machi, Oita-gun, Oita, 879-5593, Japan.
Address correspondence to H. Mori.

2
Department of Radiology, Oita Prefectural Hospital,
476, Bunyo, Oita-shi, Oita, 870-8511, Japan.
3
Department of Radiology, Nagatomi Neurosurgical
Hospital, Omichi-Machi, Oita-shi, Oita, 870-0822, Japan.

AJR 2003;181:7178
0361803X/03/181171
American Roentgen Ray Society

AJR:181, July 2003

ntrahepatic portosystemic venous

shunt is a rare condition defined as
communication between the intrahepatic portal vein and systemic veins, including
the hepatic vein and perihepatic vein, via an
anomalous intrahepatic venous channel. After the
introduction of CT, MR imaging, and sonography, intrahepatic portosystemic venous shunt has
been encountered more frequently [13]. Clinical
manifestations of intrahepatic portosystemic
venous shunt depend on the shunt flow; a highflow shunt might cause hepatic encephalopathy
and hypoglycemia.
Conservative therapy (restriction of protein,
ingestion of lactulose, oral administration of
nonabsorbable antibiotics) [4], surgery (portal
vein ligation or hepatic lobectomy), and transcatheter embolization have been used for the
treatment of intrahepatic portosystemic venous
shunt. Transcatheter embolization is a well-established, useful, and less invasive treatment

for several vascular diseases. In this report, we

describe methods of embolization with three
approaches in respective types of intrahepatic
portosystemic venous shunt, and we refer to
the pathogenesis of the condition based on angiographic findings.
Materials and Methods
Patients
Between 1989 and 2001, 10 patients (two men
and eight women; age range, 3377 years; mean
age, 58.3 years) with symptomatic intrahepatic portosystemic venous shunt were treated by transcatheter embolization at multiple related facilities. The
patients had experienced various symptoms, including disturbance of consciousness (n = 6), tremors (n =
4), disorientation (n = 2), and somnolence (n = l),
that were thought to be caused by portalsystemic
encephalopathy. The diagnoses of intrahepatic portosystemic venous shunt were based on CT and
sonography performed before embolization in all

71

Downloaded from www.ajronline.org by 186.151.63.12 on 07/09/14 from IP address 186.151.63.12. Copyright ARRS. For personal use only; all rights reserved

Tanoue et al.
patients, and conservative therapies were tried initially. In two patients, intrahepatic portosystemic
venous shunt was associated with liver cirrhosis.
Laboratory test results and transcatheter portal
venous pressure measurements were as follows: serum ammonium ratio, 1.505.00 (mean, 2.65); Fischers ratio, 0.941.65 (mean, 1.31); and portal
venous pressure, 5.018.0 cm H2O (mean, 11.0 cm
H2O). Serum ammonium ratios are defined as ratios
of serum ammonium levels relative to normal values
because the units and normal values differed among
the institutions where patients were treated.

Transcatheter Embolization
Transcatheter embolization was performed via
one of the following three access routes: transileocolic obliteration, percutaneous transhepatic obliteration, and retrograde transcaval obliteration.
These access techniques are described and are illustrated in Figure 1.
Transileocolic obliteration.After exposure of
the distal ileum under a small abdominal incision,
a catheter was advanced into the portal venous
system via the ileocolic vein.
Percutaneous transhepatic obliteration.After
percutaneous puncture of the intrahepatic portal
branch under sonographic guidance, a catheter
was advanced into the portal venous system.
Retrograde transcaval obliteration.Two catheters were advanced in a retrograde manner into the
portal venous system via bilateral transfemoral

venous access. One flexible and straight catheter

(Tracker 38, BSJ, Tokyo, Japan) was advanced into
the main portal vein through the shunt to obtain a
portogram and measure portal venous pressure before and during embolization. Another catheter was
used for placement of embolic materials.
Retrograde transcaval obliteration and percutaneous transhepatic obliteration were performed
with the patient under local anesthesia, and transileocolic obliteration were performed with the patient under epidural anesthesia.
These approaches were selected as follows: If
the patient had one large shunt or a few shunts, retrograde transcaval obliteration was selected as the
access route for the initial treatment. If we failed
with this approach, another approach was attempted. If the patient had multiple shunts in a
unilateral lobe, percutaneous transhepatic obliteration was selected. Transileocolic obliteration was
selected only if multiple shunts were present in the
bilateral lobe or if treatments with other approaches had failed.
Embolization was performed by transileocolic
obliteration in two patients, percutaneous transhepatic obliteration in four patients, and retrograde
transcaval obliteration in four patients. Several embolic materials including Gianturco coils (William
Cook Europe, Bjaeverskov, Denmark), detachable
balloons (BSJ), and microcoils were used. Four patients were treated using Gianturco coils only, one
was treated with Gianturco coils and detachable balloons, and five were treated with Gianturco coils and

microcoils (fibered platinum coils, Detach Coil System [William Cook Europe], or both). Three to 30
coils that ranged from 5 to 15 mm in diameter were
used in each patient. When a 5-French catheter
could not be advanced into the shunt vessels because
of their extreme tortuosity, we used microcatheters
and microcoils. As a follow-up examination within
312 months after the procedure, CT, sonography, or
both were performed. The clinical follow-up period
ranged from 24 to 156 months.

Evaluation
All data including clinical data, radiologic findings, and clinical outcomes were collected retrospectively. The angiographic findings evaluated by
three radiologists were the type of drainage vein,
multiplicity, and associated intrahepatic venous abnormalities. The intrahepatic portosystemic venous
shunt was classified by the type of drainage vein and
multiplicity. Possible approaches for several types of
shunt, their technical success rates, and complications were investigated. We evaluated the effects of
these procedures on clinical symptoms, laboratory
test results, and portal venous pressures. Clinical
symptoms, except for consciousness level, were
evaluated without any scale by degree of patients
complaint. Consciousness level was evaluated using
the Glasgow Coma Scale. Laboratory tests included
serum ammonium ratios and Fischers ratios. The
changes in these data were analyzed using Wilcoxons signed rank test.

Fig. 1.Drawings illustrate three approaches to access intrahepatic portosystemic venous shunts.
A, For transileocolic obliteration, catheter (open arrow) is advanced into portal venous system via ileocolic vein (solid arrow) through small abdominal incision.
B, For percutaneous transhepatic obliteration, catheter (arrow) is advanced into portal venous system after percutaneous puncture of intrahepatic portal branch.
C, For retrograde transcaval obliteration, two catheters are retrogradely advanced into portal venous system through shunt vessel (arrowhead) via bilateral transfemoral
venous access. One catheter (open arrow), which is advanced into main portal vein through shunt, is straight catheter used for portography and to measure portal venous
pressure during procedure. Other catheter (solid arrow) is used to place embolic materials.

72

AJR:181, July 2003

Angiography and Embolization of Portosystemic Venous Shunts

Downloaded from www.ajronline.org by 186.151.63.12 on 07/09/14 from IP address 186.151.63.12. Copyright ARRS. For personal use only; all rights reserved

Fig. 2.Transcaval retrograde portogram shows intrahepatic portosystemic venous shunt with
aneurysmal dilatation (arrow) in 48year-old man. Arrowhead indicates
a catheter advanced into main portal vein via shunt.

Results

Transvenous portography was performed in

all patients, and eight of 10 patients underwent
transarterial portography before embolization.
Transarterial or transvenous portography showed
multiple shunts in five patients and single shunts
in the other five patients.
Intrahepatic portosystemic venous shunts
were divided into two types according to the
drainage vein. One was a communication between the intrahepatic portal vein and the hepatic
vein, whereas the other was a communication
between the intrahepatic portal vein and the in-

ferior vena cava via the perihepatic veins (adrenal vein or inferior phrenic veins). The former
type of shunt was identified in eight patients including two patients with multiple shunts.
None of the cases of intrahepatic portal
venoushepatic venous shunt were associated
with liver cirrhosis. Portograms or hepatic
venograms showed the eight cases were associated with intrahepatic vein anomalies including five portal vein aneurysms (Fig. 2),
one portal vein anastomosis (Fig. 3), and two
hepatic vein anastomoses (Fig. 4). The latter
type of shunt, the portal venousperihepatic

Fig. 3.Transcaval retrograde hepatic venogram shows intrahepatic portosystemic venous

shunt between left hepatic vein and left portal vein in 53-year-old woman. Note portal venous
anastomosis of medial branches (arrowheads).

AJR:181, July 2003

venous shunt, was observed in two patients

with liver cirrhosis.
Three of the five patients with multiple intrahepatic portosystemic venous shunts underwent percutaneous transhepatic obliteration. In
the remaining two patients, retrograde transcaval obliteration was tried but failed initially
because of technical difficulty in approaching
the shunts. These two patients were subsequently treated by transileocolic obliteration
(Fig. 5). One of the five patients with a single
shunt underwent percutaneous transhepatic
obliteration. Because the shunt was not completely obliterated after the initial procedure,
reembolization was subsequently performed
and the shunt was completely occluded. The
remaining four patients were treated by retrograde transcaval obliteration (Fig. 6). Reembolization was required in one of these patients
because of residual shunt flow. In nine of the
10 patients, intrahepatic portosystemic venous
shunts were shown to be completely obliterated on angiograms obtained after the obliteration procedures. Nearly complete obliteration
was achieved in the remaining patient who had
multiple shunts.
Procedure-related complications were observed in three patients: adhesive ileus was
seen in a patient treated by transileocolic
obliteration a few days after the procedure,
occlusion of the left portal vein due to coil
migration was found in a patient treated by
percutaneous transhepatic obliteration, and
thrombosis of the left portal venous branch
related to the puncture procedure was ob-

Fig. 4.Transcaval retrograde hepatic venogram shows hepatic

venous anastomosis between right hepatic vein and accessory hepatic
vein (arrow) in 64-year-old woman.

73

Downloaded from www.ajronline.org by 186.151.63.12 on 07/09/14 from IP address 186.151.63.12. Copyright ARRS. For personal use only; all rights reserved

Tanoue et al.

Fig. 5.Multiple intrahepatic portosystemic venous shunts in 62-year-old woman who did not have cirrhosis and who presented with memory disturbance and trembling.
Blood examination revealed hyperammonemia and low Fischers ratio. Patient was treated by transileocolic obliteration.
A, Transileocolic portogram revealed multiple intrahepatic portosystemic venous shunts in left lobe (arrowheads). Gianturco coils (William Cook Europe, Bjaeverskov, Denmark) and fibered microcoils were placed into shunt vessels.
B, Transileocolic portogram obtained after embolization shows complete obliteration of intrahepatic portosystemic venous shunts.

served in a patient treated by percutaneous

transhepatic obliteration. The latter two complications were seen on follow-up CT after
the procedure and did not cause any clinical
symptoms. No other complications, including peritoneal complications, were encountered after the treatments.
Symptoms related to portalsystemic encephalopathy completely disappeared in

eight patients and improved in two patients

after treatment. Serum ammonium ratios significantly decreased and Fischers ratios increased after embolization (Figs. 7A and
7B). The portal venous pressure tended to increase to levels above those recorded before
treatment (Fig. 7C). In the two patients with
liver cirrhosis, the pressures after treatment
were above the reference value.

Discussion

Portalsystemic encephalopathy usually

occurs in patients with portal hypertension
and is mainly induced by liver cirrhosis, although it can sometimes occur in patients
who do not have liver cirrhosis. Recent imaging studies have identified intrahepatic
portosystemic venous shunt in the latter
group of patients [122].

Fig. 6.Single intrahepatic portosystemic venous shunt in 72-year-old woman who did not have cirrhosis and who presented in coma. Blood examination revealed hyperammonemia and low Fischers ratio. Patient was treated by retrograde transcaval obliteration.
A, Retrograde transcaval portography was performed with catheter advanced into portal vein via shunt vessel (arrowhead). Portogram shows intrahepatic portosystemic
venous shunt between right portal vein and accessory hepatic vein (open arrow) with portal vein aneurysm (solid arrow). Gianturco coil (William Cook Europe, Bjaeverskov, Denmark), detachable microcoils, and fibered platinum microcoils were positioned in shunt just before aneurysmal dilatation.
B, Portogram obtained after procedure shows complete obliteration of intrahepatic portosystemic venous shunt.

74

AJR:181, July 2003

Portal Venous Pressure (cm H2O)

Angiography and Embolization of Portosystemic Venous Shunts

3
Fischers Ratio

Serum Ammonium Ratio

Downloaded from www.ajronline.org by 186.151.63.12 on 07/09/14 from IP address 186.151.63.12. Copyright ARRS. For personal use only; all rights reserved

Before
Embolization

After
Embolization

Before
Embolization

After
Embolization

20

10

Before
Embolization

After
Embolization

Fig. 7.Graphs show changes in laboratory data and portal venous pressures after treatment in study group.
A, Serum ammonium levels decreased significantly (p < 0.01) after treatment. Data are expressed as ratios relative to normal values because units and normal values differed among institutions.
B, Fischers ratios increased significantly (p = 0.028) after treatment. This value was not measured in four patients.
C, Portal venous pressure increased significantly (p = 0.018) after treatment. Dotted lines represent range of normal values. In two patients, portal venous pressure was
higher than normal values both before and after treatment; both patients had associated liver cirrhosis. Pressure levels were within normal range in other patients. Portal
venous pressure was not measured in three patients.

Although surgical occlusion had been used

for treatment of patients with this condition in
the past [20, 23], transcatheter embolization
and its usefulness have been reported in recent
years [2428]. In our patients, clinical symptoms and laboratory data improved immediately after occlusion treatment. Although the
portal venous pressures recorded after treatment tended to be higher than the levels before
treatment, this relative increase in pressure was
not associated with overt clinical problems.
Transcatheter embolization was performed
using one of three routes to access the intrahepatic portosystemic venous shunts: transileocolic obliteration, percutaneous transhepatic
obliteration, or retrograde transcaval obliteration. Of the patients in our study group, two
with multiple shunts underwent transileocolic
obliteration. This procedure offers the easiest
control of catheters through its anterograde access route. On the other hand, transileocolic
obliteration is the most invasive among the
three techniques used by our team because it
requires an abdominal incision with the patient
under general anesthesia or epidural tubing.
Furthermore, this procedure carries the risk of
adhesion because it requires an abdominal incision. One of our patients experienced adhesive ileus after the procedure. Therefore,
transileocolic obliteration should be limited to
patients with multiple shunts located in both
lobes of the liver.

AJR:181, July 2003

Percutaneous transhepatic obliteration is

useful for patients with contralateral distribution of shunts because it offers good catheter
control of the portal vein contralateral to the
punctured side. In our patients, three with multiple shunts in identical segments and one with
a single shunt underwent percutaneous transhepatic obliteration. The procedure resulted in
successful and complete obliteration as confirmed angiographically. Although percutaneous transhepatic obliteration is a relatively
invasive technique, Ohta et al. [29] reported
that 16.5% of their patients treated using percutaneous transhepatic catheterization developed procedure-related complications.
Retrograde transcaval obliteration is the
least invasive technique, but it has some applicable limitations with regard to the type, number, and location of intrahepatic portal
systemic venous shunts. The presence of a
large shunt close to the inferior vena cava allows easy catheterization of the main portal
vein and retrograde portography to confirm the
shunt and embolization. Because the inferior
vena cava and the proximal portion of the hepatic vein have relatively large diameters, supporting catheters during placement of embolic
material is difficult. Therefore, the catheter
type and shape and the embolic material
should be selected carefully. We recommend
the use of a preshaped catheter adjusted to the
hepatic vein, and detachable coils or a combi-

nation of microcatheters and microcoils. In our

group, four patients with a single shunt and
one patient with two shunts in the adjacent hepatic vein were successfully treated by retrograde transcaval obliteration. We believe that
this technique should be applied to patients
with a small number of shunts.
The cause of intrahepatic portosystemic
venous shunt is not completely understood; however, two major theories have been proposed. The
first is the congenital origin theory, which suggests the persistence of the communication between the portal and hepatic venous systems that
occurs during embryonal development. The second is the acquired theory, which suggests the
shunt results from portal hypertension, trauma, or
rupture of a portal vein aneurysm. Intrahepatic
portosystemic venous shunt in patients who do
not have liver cirrhosis or a history of trauma are
thought to be congenital in origin. Embryologically, the intra- and extrahepatic portal venous
systems develop by the selective persistence of
vitelline and umbilical systems between the
fourth week and third month of fetal life (Fig. 8).
In the congenital origin theory, intrahepatic portosystemic venous shunt is thought to represent
persistent communication between cranial and
caudal hepatic sinusoids formed by vitelline
veins and umbilical vein.
Macroscopically evident intrahepatic portosystemic venous shunts have been classified
according to their morphology [l, 46]. In the

75

Downloaded from www.ajronline.org by 186.151.63.12 on 07/09/14 from IP address 186.151.63.12. Copyright ARRS. For personal use only; all rights reserved

Tanoue et al.
our study, we divided the intrahepatic portosystemic venous shunts into two types on the
basis of pathogenic mechanisms. One is a
shunt that consists of an intrahepatic portal
venoushepatic venous pathway, whereas the
other is a shunt that consists of an intrahepatic
portal venousperihepatic venous pathway
that includes the inferior phrenic veins, adrenal
vein, and paraumbilical veins.
Intrahepatic Portal VenousHepatic Venous Pathway

This type of intrahepatic portosystemic

venous shunt is depicted as tubular or aneurys-

mal communication (single or multiple) between intrahepatic portal veins and hepatic
veins. To our knowledge, 42 cases have been reported in the English-language literature [l, 2,
421, 24, 3039]. These cases include 31 simple types and 11 multiple types. Portal vein aneurysms were reported in 29 (69%) of these 42
cases. Most of these cases (76%) were not associated with liver cirrhosis. Eight of our patients
had this type of shunt. Six (75%) of these patients had neither cirrhosis nor any other hepatic
disease. Associated anomalies of the hepatic
vessels, which included portal vein aneurysm,

C
76

hepatic venous anastomosis, and portal vein

anastomosis, were observed (Figs. 24). Chagnon et al. [2] indicated that portal vein aneurysms including intrahepatic portosystemic
venous shunts are congenital. The latter two
anomalies have not, to our knowledge, been reported previously. These anomalies are thought
to be formed in the hepatic sinusoid during fetal
development. Because of the low rate of coexisting liver cirrhosis and high rate of coexisting
anomalies, intrahepatic portal venoushepatic
venous shunts are likely to be of the congenital
origin type.

Fig. 8.Schematic drawings of normal

development of intrahepatic portal and
hepatic venous systems.
A, Drawing shows embryo at 5 weeks
gestation. Vitelline venous plexus is
surrounded by liver cords to form hepatic sinusoids. Bilateral umbilical
B veins (UV) form sinusoids. CV = cardinal
vein, HS = hepatic sinusoid, D = duodenum, VV = vitelline vein.
B, Drawing shows embryo at 8 weeks
gestation. Sinusoids start to develop,
forming portal and hepatic venous systems. Right umbilical vein and cranial
portion of left umbilical vein are regressed. Dorsal communication between caudal vitelline veins persists as
part of main portal vein. Note tubular
structure between left umbilical vein
and inferior vena cava, which is called
ductus venosus. DV = ductus venosus,
LVV = left vitelline vein.
C, Drawing shows fetus at 12 weeks
gestation. Note advanced differential
growth of portal and hepatic venous
systems. Presence of residual communication between hepatic venous system and portal venous system at this
stage corresponds with intrahepatic
portosystemic venous shunt after
birth. DV = ductus venosus, SMV = superior mesenteric vein, IVC = inferior
vena cava.
D, Drawing shows fetus with normally
developed portohepatic venous system
before birth. HV = hepatic vein, DV =
ductus venosus, PV = portal vein.

D
AJR:181, July 2003

Angiography and Embolization of Portosystemic Venous Shunts

Downloaded from www.ajronline.org by 186.151.63.12 on 07/09/14 from IP address 186.151.63.12. Copyright ARRS. For personal use only; all rights reserved

Intrahepatic Portal VenousPerihepatic Venous

Pathway

Radiologically, this type of intrahepatic portosystemic venous shunt shows some communications between the intrahepatic portal vein
and perihepatic veins, and it drains into the inferior vena cava. Intrahepatic portosystemic
venous shunts through persistent paraumbilical veins are occasionally encountered in patients with cirrhosis. Paraumbilical veins (also
called veins of Sappey) are known as potential
communications between the intrahepatic portal vein and veins of the abdominal wall [40].
Other perihepatic veins, such as the inferior
phrenic veins and capsular vein, might constitute the communicating venous system and
form portosystemic venous shunt in this type.
Intrahepatic portosystemic venous shunt between the right portal vein and inferior vena
cava via venous structures around the right
lobe are less common. To our knowledge, 22
cases of this type have been reported in the English-language literature [1, 3, 21, 22, 31, 32,
4148], and these shunts were described as being located in the bare area and posteroinferior
aspect of the right lobe. A high incidence of
liver cirrhosis (71.4%) is reported in patients
with this type of shunt. Our two patients with
this type of shunt had liver cirrhosis. Because
of the high rate of coexisting liver cirrhosis,
the acquired theory could explain this type of
shunt, and the perihepatic veins are thought
to develop in association with portal hypertension as intra- and extrahepatic collateral
pathways from existing venous structures including paraumbilical veins, inferior phrenic
veins, and adrenal vein.
Intrahepatic portosystemic venous shunts are
divided into two main types: intrahepatic portal
venoushepatic venous communication and intrahepatic portal venousperihepatic venous
communication. On the basis of the angiographic findings and clinical manifestations, the
former type is considered to be of congenital origin, whereas the latter is thought to be an acquired condition associated with portal
hypertension. Selection of the most suitable access route based on the morphology of shunt is
an important aspect of treatment of intrahepatic
portosystemic venous shunt by transcatheter embolization. In selected cases, retrograde transcaval obliteration is a useful, safe, and less
invasive technique than the other options.

References
1. Mori H, Hayashi K, Fukuda T, Matsunaga N, Futagawa W, Nagasaki M. Intrahepatic portosys-

AJR:181, July 2003

2.

3.

4.

5.
6.

7.

8.

9.

10.

11.

12.

13.

14.

15.

16.

17.

18.

temic venous shunt: occurrence in patients with

and without liver cirrhosis. AJR 1987;149:711
714
Chagnon SF, Vallee CA, Barge J, Chevalier LJ,
Le Gal J, Blery MV. Aneurysmal portahepatic
venous fistula: report of two cases. Radiology
1986;159:693695
Itai Y, Kurosaki Y, Saida Y, Niitsu M, Kuramoto
K. CT and MRI in detection of intrahepatic portosystemic shunts in patients with liver cirrhosis. J
Comput Assist Tomogr 1994;18:768773
Watanabe A. Portal-systemic encephalopathy in
non-cirrhotic patients: classification of clinical
types, diagnosis and treatment. J Gastroenterol
Hepatol 2000;15:969979
Park JH, Cha SH, Han JK, Han MC. Intrahepatic
portosystemic venous shunt. AJR 1990;155:527528
Chevallier P, Oddo F, Souci J, Diaine B, Padovani
B. Macroscopic intrahepatic portosystemic venous
shunt: review of the literature and reclassification. J
Radiol 2001;81:597604
Araki T, Ohtomo K, Kachi K, et al. Magnetic resonance imaging of macroscopic intrahepatic portal-hepatic venous shunts. Gastrointest Radiol
1991;16:221224
Kawakatsu M, Vilgrain V, Belghiti J, Flejou JF,
Nahum H. Association of multiple liver cell adenomas with spontaneous intrahepatic portohepatic shunt. Abdom Imaging 1994;19:438440
Gheorghiu D, Leibowits O, Bloom RA. Case report: asymptomatic aneurysmal intrahepatic
porto-hepatic venous shuntdiagnosis by ultrasound. Clin Radiol 1994:49:6465
Horiguchi Y, Kitano T, Imai H, Ohsuki M,
Yamauchi M, Itoh M. Intrahepatic portal-systemic shunt: its etiology and diagnosis. Gastroenterol Jpn 1987;22:496502
Tanaka S, Kitamura T, Fujita M, et al. Intrahepatic
venous and portal venous aneurysms examined by
color Doppler flow imaging. J Clin Ultrasound
1992;20:8998
Jabra AA, Taylor GA. Ultrasound diagnosis of
congenital intrahepatic portosystemic venous
shunt. Pediatr Radiol 1991;21:529530
Tarazov PG. Spontaneous aneurysmal intrahepatic portosystemic venous shunt. Cardiovasc Intervent Radiol 1994;17:4445
Wittich G, Jantsch H, Tscholakoff D. Congenital
portosystemic shunt diagnosed by combined realtime and Doppler sonography. J Ultrasound Med
1985;4:315318
Okuda K, Kanda Y, Fukuyama Y, Musha H, Sumikoshi T. Hepatic vascular anomalies in nonparasitic cysts of the liver. Acta Hepatogastroenterol
(Stuttg) 1976;23:110113
Kozuka S, Sassa R, Kakumu S. An enormous intrahepatic shunt between portal vein and hepatic
one. Angiology 1975;26:365371
Charnsangavej C, Soo CS, Bernardino ME,
Chuang VP. Portal-hepatic venous malformation:
ultrasound, computed tomographic, and angiographic findings. Cardiovasc Intervent Radiol
1983;6:109111
Takayasu K, Moriyama N, Shima Y, et al. Spontaneous portal-hepatic venous shunt via an intrahepatic portal vein aneurysm. Gastroenterology
1984;86:945948

19. Hondo T, Teragawa H, Munemori M, et al. Portalhepatic venous shunt through a portal aneurysm
complicated by hepatic encephalopathy and pulmonary hypertension. Intern Med 1997;36:790793
20. Miyauchi A, Okada S, Hashimoto T, et al. Surgical treatment of an enormous aneurysmal portahepatic venous fistula: report of a case. Surg
Today 1995;25:855858
21. Santamaria G, Pruna X, Serres X, Inaraja L, Zuasnabar A, Castellote A. Congenital intrahepatic portosystemic venous shunt: sonographic and magnetic
resonance imaging. Eur Radiol 1996;6:7678
22. Florio F, Nardella M, Balzano S, Giacobbe A, Perri
F. Congenital intrahepatic portosystemic shunt.
Cardiovasc Intervent Radiol 1998;21:421424
23. Jeng LB, Chen ME. Intrahepatic portohepatic
venous shunt: a case report of successful surgical
resection. Arch Surg 1993;128:349352
24. Ohmoto K, Furui S, Saito M, Kokubo T, Itai Y, Ito
M. Enormous intrahepatic communication between the portal vein and hepatic vein. Clin Radiol 1986;37:513514
25. Okada Y, Endo T, Kusano S, Yoshida M. Multiple
intrahepatic portosystemic venous shunts: treatment with steel-coil embolization. AJR 1991;157:
971973
26. Maeda T, Mori H, Aikawa H, Komatsu E. Therapeutic embolization of intrahepatic portosystemic
shunts by retrograde transcaval catheterization.
Cardiovasc Intervent Radiol 1993;16:245247
27. Yoshimitsu K, Andoh H, Kudo S, Matsuo Y, Shimoda Y. Multiple intrahepatic portosystemic
venous shunts: treatment by portal vein embolization. Cardiovasc Intervent Radiol 1993;16:
4951
28. Kim IO, Cheon JE, Kim WS, et al. Congenital intrahepatic venous shunt: treatment with coil embolisation. Pediatr Radiol 2000;30:336338
29. Ohta M, Hashizume M, Kawanaka H, et al. Complications of percutaneous transhepatic catheterization of the portal venous system in patients
with portal hypertension. J Gastroenterol Hepatol
1996;11:630634
30. Naganuma H, Ishida H, Niizawa M, Masamune O.
Macroscopic intrahepatic portal-hepatic venous
shunt not associated with chronic hepatic disease.
Abdom Imaging 1994;19:153156
31. Kudo M, Tomita S, Tochio H, Minowa K, Todo
A. Intrahepatic portosystemic venous shunt: diagnosis by color Doppler imaging. Am J Gastroenterol 1993;88:723729
32. Mori H, Hayashi K, Amamoto Y. Membranous
obstruction of the inferior vena cava associated
with intrahepatic portosystemic shunt. Cardiovasc Intervent Radiol 1986;19:209213
33. Hanatate F, Matsuoka H, Mizuno Y, Kitamura T.
Porto-hepatic venous shunt via portal vein aneurysm with splenomegaly. J Gastroenterol 1995;30:
786789
34. Kumano H, Kinoshita H, Hirohashi K. Aneurysm of
intrahepatic portal vein shown by percutaneous transhepatic portography. (letter) AJR 1994;163:1000
35. Yamaguchi T, Kubota Y, Seki T, et al. Acquired
intrahepatic portal vein aneurysm. Dig Dis Sci
1992;37:17691771
36. Stroz P, Asch M, Salem S, Sniderman K, Gallinger S.
Transhepatic embolization of a portohepatic venous

77

Downloaded from www.ajronline.org by 186.151.63.12 on 07/09/14 from IP address 186.151.63.12. Copyright ARRS. For personal use only; all rights reserved

Tanoue et al.
fistula. J Vasc Interv Radiol 1996;7:705707
37. Sakai T, Katoh T, Asada K, Yagi K, Yoshida R. Portal hepatic venous shunt via an intrahepatic portal
vein aneurysm. Intern Med 1992;31:899903
38. Ito Y, Tarao K, Tamai S, et al. Portal vein aneurysm
in the liver associated with multiple vascular malformations. J Gastroenterol 1994;29:776781
39. Bezzi M, Mitchell DG, Needleman L, Goldberg
BB. Iatrogenic aneurysmal portal-hepatic venous
fistula: diagnosis by color Doppler imaging. J Ultrasound Med 1998;7:457461
40. Martin BF, Tudor RG. The umbilical and paraumbilical vein of man. J Anat 1980;130:305322
41. Park JH, Cha SH, Han JK, Han MC. Intrahepatic

portosystemic venous shunt. AJR 1990;155:527528

42. Lalonde L, VanBeers B, Trigaux JP, Delos M, Pringot J. Focal nodular hyperplasia in association with
spontaneous intrahepatic portosystemic venous
shunt. Gastrointest Radiol 1992;17:154156
43. Kadoya M, Takashima T, Matsui O, Kitagawa K,
Chohtoh S. Unusual portosystemic collateral penetrating the liver parenchyma. Gastrointest Radiol 1985;10:115118
44. Kudo M. Intrahepatic portosystemic venous shunt
in liver cirrhosis: is it congenital or acquired? AJR
1993;160:421422
45. Ohnishi K, Hatano H, Nakayama T, Kohno K,
Okuda K. An unusual portal-systemic shunt, most

likely through a patent ductus venosus: a case report. Gastroenterology 1983;85:962965

46. Nagafuchi Y, Watanabe H, Kitahara K, et al. Intrahepatic portosystemic venous shunt: report of
two cases assessed by helical computed tomography. Scand J Gastroenterol 1996;31:11321135
47. Ardic S, Deniz E, Arbatli M, Ardic FA, Aysun A.
A rare intrahepatic portacaval tubular shunt in a
patient with spastic paraparesis. Australas Radiol
1997;41:181184
48. Kanematsu M, Hoshi H, Imaeda T, Yokoyama R.
Three-dimensional CT demonstration of intrahepatic portosystemic venous shunt draining into the
inferior vena cava. Br J Radiol 1997;70:418420

The ARRS foundation, The Roentgen Fund, sponsors worthy educational, scientific, and research initiatives that contribute
to excellence in radiology and medicine. A gift to The Roentgen Fund means progress in radiology. For more information
call 800-438-2777.

78

AJR:181, July 2003