DOI:10.2214/AJR.07.3356
AJR 2008; 191:555-559
© American Roentgen Ray Society
Gastric Varices with Gastrorenal Shunt: Combined Therapy Using Transjugular Retrograde Obliteration and Partial Splenic Embolization
Fumio Chikamori1,
Nobutoshi Kuniyoshi2,
Takahiko Kawashima3 and
Yasuhiro Takase3
1 Department of Surgery, Kuniyoshi Hospital, 1-3-4 Kamimachi, Kochi, Japan
780-0901.
2 Department of Internal Medicine, Kuniyoshi Hospital, Kochi, Japan.
3 Department of Surgery, Tsukuba Soai Hospital, Ibaraki, Japan.
Received October 26, 2007;
accepted after revision February 23, 2008.
Address correspondence to F. Chikamori
(chikamo1{at}i-kochi.or.jp).
Abstract
OBJECTIVE. This study was prospectively conducted to evaluate the
effectiveness of the combination of transjugular retrograde obliteration and
partial splenic embolization in the treatment of gastric varices with
gastrorenal shunt.
SUBJECTS AND METHODS. Between November 2002 and December 2006, 14
patients with gastric varices and gastrorenal shunt were treated by combining
transjugular retrograde obliteration and partial splenic embolization (group
1). These patients were compared with 19 patients with gastric varices and
gastrorenal shunt treated by only transjugular retrograde obliteration (group
2) for the disappearance rate of gastric varices, the cumulative survival
rate, and the occurrence rate of esophageal varices after transjugular
retrograde obliteration. Partial splenic embolization was performed 7–14
days before transjugular retrograde obliteration. No significant differences
were seen between the two groups in terms of demographic data, including age,
sex, and Child-Pugh classification.
RESULTS. The disappearance rate of gastric varices after
transjugular retrograde obliteration was 100% in both groups. The 3-year
cumulative survival rate after transjugular retrograde obliteration was 92% in
group 1 and 95% in group 2. The 3-year cumulative occurrence rate of
esophageal varices after transjugular retrograde obliteration was 9% in group
1 and 45% in group 2, a significant difference (p < 0.05).
CONCLUSION. The findings of this study indicate that partial splenic
embolization contributed to preventing portal congestion after transjugular
retrograde obliteration. We conclude that the combination of transjugular
retrograde obliteration and partial splenic embolization for gastric varices
is more effective than transjugular retrograde obliteration only in the
long-term prevention of esophageal varices after transjugular retrograde
obliteration.
Keywords: gastric varices • gastrorenal shunt • partial splenic embolization • transjugular retrograde obliteration
Introduction
Transvenous retrograde oblit eration has recently become the treatment of
choice for gastric varices with a gastrorenal shunt in many institutions in
Japan [1,
2]. The method of transvenous
retrograde obliteration has two approaches: the transjugular and the trans
femoral. The former is called transjugular retrograde obliteration
[1] and the latter,
balloon-occluded retrograde transvenous obliteration
[2]. As we reported in a
previous article [1],
transjugular retrograde obliteration is less invasive and is effective in
eradicating gastric varices. However, the high occurrence rate of esophageal
varices after transjugular retrograde obliteration is still a problem
[3]. Newly developed esophageal
varices after transjugular retrograde obliteration can be easily treated with
endoscopic injection sclerotherapy. However, the development of esophageal
varices after trans jugular retrograde obliteration means a continuous
pathologic state of portal circulation. Our previous study
[4] showed that transjugular
retrograde obliteration obliterates the gastrorenal shunt, which has an
abundant blood flow, and increases portal venous pressure. Partial splenic
embolization has the effect of decreasing splenic blood flow and portal venous
pressure [5]. So we considered
that partial splenic embolization may prevent the side effect of transjugular
retrograde obliteration. This study was undertaken to evaluate the efficacy
and safety of the combined treatment of transjugular retrograde obliteration
and partial splenic embolization compared with transjugular retrograde
obliteration only.
Subjects and Methods
Patients
Thirty-three consecutive patients with portal hypertension, seen between
November 2002 and December 2006, who had gastric varices and a gastrorenal
shunt were included in this study if the size of the varices was more than F2
(mediumsized and nodular). Fourteen patients with liver cirrhosis were admit
ted to our institution and treated by a combin ation of transjugular
retrograde obliteration and partial splenic embolization (group 1). Nineteen
patients with liver cirrhosis were admitted to an affiliated hospital and were
treated by only transjugular retrograde obliteration (group 2). We analyzed
the disappearance rate of gastric varices, the rate of complications, the
cumulative survival rate, and the occurrence rate of esophageal varices after
transjugular retrograde obliteration. Demographic data, including age, sex,
and Child-Pugh classi fication for the two groups, are listed in
Table 1. Our institutional
review board approved this study, and informed consent was obtained from every
patient.
No significant differences were seen between the two groups. In group 1,
three patients had a history of gastric variceal bleeding and were treated
electively, and 11 had no history of bleeding and were treated
prophylactically. In group 2, three patients were treated electively and 16
patients prophylactically. Two patients in group 1 and two patients in group 2
had hepato cellular carcinoma smaller than 3 cm in diameter. We encountered
complicated large gastroesophageal varices with a gastrorenal shunt in three
and four patients in groups 1 and 2, respectively. In these cases, endoscopic
injection sclerotherapy was nec essary before transjugular retrograde
obliteration.
View larger version (165K):
[in this window]
[in a new window]
[as a PowerPoint slide]
|
Fig. 1A —40-year-old man with alcoholic cirrhosis was admitted for
treatment of large gastric varices. Laboratory data on admission were as
follows: WBC, 4,900/µL; hemoglobin, 13.3 g/dL; platelets, 8.6 x
104/µL; albumin, 2.6 g/dL; total bilirubin, 2.4 mg/dL; glutamic
oxaloacetic transaminase, 59 U/L; glutamic pyruvic transaminase, 34 U/L;
prothrombin time, 59.0%; hepaplastin test, 48%; total bile acid, 81.1
µmol/L; arterial ketone body ratio (AKBR), 0.3. Plasma ammonia level was
145 µg/dL, and indocyanine green retention rate at 15 min
(ICG15) was 33%. Child-Pugh score was grade B. Antibodies to
hepatitis B and C were negative. Patient was treated with partial splenic
embolization and transjugular retrograde obliteration. Superior mesenteric
arterial portography shows that gastric varices are supplied by left gastric
vein and drained into gastrorenal shunt (arrow) and inferior phrenic
vein.
|
|
View larger version (145K):
[in this window]
[in a new window]
[as a PowerPoint slide]
|
Fig. 1B —40-year-old man with alcoholic cirrhosis was admitted for
treatment of large gastric varices. Laboratory data on admission were as
follows: WBC, 4,900/µL; hemoglobin, 13.3 g/dL; platelets, 8.6 x
104/µL; albumin, 2.6 g/dL; total bilirubin, 2.4 mg/dL; glutamic
oxaloacetic transaminase, 59 U/L; glutamic pyruvic transaminase, 34 U/L;
prothrombin time, 59.0%; hepaplastin test, 48%; total bile acid, 81.1
µmol/L; arterial ketone body ratio (AKBR), 0.3. Plasma ammonia level was
145 µg/dL, and indocyanine green retention rate at 15 min
(ICG15) was 33%. Child-Pugh score was grade B. Antibodies to
hepatitis B and C were negative. Patient was treated with partial splenic
embolization and transjugular retrograde obliteration. After microcoil
obliteration of inferior phrenic vein, 10 mL of 5% ethanolamine oleate with
iopamidol was injected into gastric varices (arrow), as shown on
retrograde shunt venography.
|
|
View larger version (148K):
[in this window]
[in a new window]
[as a PowerPoint slide]
|
Fig. 1C —40-year-old man with alcoholic cirrhosis was admitted for
treatment of large gastric varices. Laboratory data on admission were as
follows: WBC, 4,900/µL; hemoglobin, 13.3 g/dL; platelets, 8.6 x
104/µL; albumin, 2.6 g/dL; total bilirubin, 2.4 mg/dL; glutamic
oxaloacetic transaminase, 59 U/L; glutamic pyruvic transaminase, 34 U/L;
prothrombin time, 59.0%; hepaplastin test, 48%; total bile acid, 81.1
µmol/L; arterial ketone body ratio (AKBR), 0.3. Plasma ammonia level was
145 µg/dL, and indocyanine green retention rate at 15 min
(ICG15) was 33%. Child-Pugh score was grade B. Antibodies to
hepatitis B and C were negative. Patient was treated with partial splenic
embolization and transjugular retrograde obliteration. Retrograde shunt
venography on day 1 after transjugular retrograde obliteration shows that
thrombi have formed in gastric varices (arrow).
|
|
Methods
Gastrorenal shunt was diagnosed on CT and superior mesenteric arterial
portography (Fig. 1A). Partial
splenic embolization was performed 7–14 days before transjugular
retrograde obliteration. Through the right femoral artery, a 5-French catheter
was selectively advanced to the splenic artery. Through this catheter, a
3-French microcatheter was positioned in the peripheral splenic artery distal
to the great pancreatic artery. After selective splenic arteriography, more
than 70% splenic arterial embolization was performed using platinum microcoils
or stainless steel coils under fluoroscopic guidance. The splenic infarction
area was confirmed on CT performed 1 week after partial splenic embolization.
Systemic antibiotic medication was started on the day of partial splenic
embolization and continued for 1 week.
The transjugular retrograde obliteration technique and the skill level of
both interventional radiologists were the same at our hospital and the
affiliated hospital. The technical details of transjugular retrograde
obliteration have prev iously been reported
[1]. Through the right internal
jugular vein, we inserted an 8-French long cobrashaped sheath into the left
renal vein. We then inserted a 5- or 6-French angiographic catheter with an
occlusive balloon 11 or 20 mm in diameter into the gastrorenal shunt through a
previously inserted sheath. The balloon was inflated with 0.7–4.0 mL of
diluted contrast medium to stop the blood flow in the shunt. The communicating
routes of the gastrorenal shunt, such as the inferior phrenic and
retroperitoneal veins, were obliterated with a microcoil when a 3-French
microcatheter could be inserted into these veins or with absolute ethanol when
the microcatheter could not be inserted. After the procedure, gastric variceal
blood flow was completely controlled, and we injected 5–20 mL of 5%
ethanolamine oleate with iopamidol (EOI) into the gastric varices under
fluoroscopic guidance (Fig.
1B). After confirming the presence of thrombi in the gastric
varices on retrograde shunt venography the next day, we removed the catheter
(Fig. 1C). If retrograde shunt
venography revealed no thrombi in the varices, we repeated the procedure. The
catheter was left in the vein for 1 or more days, depending on how rapidly
thrombi formed in the gastric varices
[3].
The success of obliteration of the gastric varices was confirmed on
contrast-enhanced CT 1 week, 1 month, and 3 months after transjugular
retrograde obliteration. Eradication of the gastric varices was diagnosed by
endoscopic examination 1 week, 1 month, and 3 months after transjugular
retrograde obliteration. After eradication of the gastric varices, follow-up
by endoscopy and CT was performed every 6 months. Form of gastric or
esophageal varices was classified into three types—F1, small and
tortuous; F2, medium-sized and nodular; and F3, large and tumorous—and
evaluated. Esophageal varices with the "red color" sign
[6] that occurred after
transjugular retrograde obliteration were treated with endo scopic injection
sclerotherapy.
Statistical analysis was performed using Stat-View statistical software
(Abacus Concepts, Inc.). The clinical data were tested using the Student's
t test and the chi-square test. Survival curves were obtained by the
Kaplan-Meier method. Analysis of survival curves for significance was
performed using the log-rank test. Data were considered significant when the
p value was less than 0.05. The results are given as mean ±
SD.
Results
The success rate of gastric variceal eradication by transjugular retrograde
oblit eration was 100% in both groups. Eighty-six percent (12/14) of gastric
varices completely thrombosed on day 1 after transjugular retrograde
obliteration in group 1 and 79% (15/19), in group 2. In the remaining two
patients in group 1, the gastric varices were completely thrombosed on day 2.
In group 2, the varices were completely thrombosed on day 2 in three patients
and on day 3 in one. Major and minor complications for the two groups are
listed in Table 2. Mild to
moderate abdominal pain necessitating analgesic drugs was found in 100%
(14/14) of the patients in group 1 and in 63% (12/19) of those in group 2,
with a significant difference (p = 0.011). In group 2, one patient
had worsening ascites and one had worsening portal hypertensive gastropathy
necessitating medication. No significant differences were seen between the two
groups in minor complications, including fever of more than 38°C,
hematuria, and a small amount of pleural effusion. Form changes in gastric
varices in both groups after transjugular retrograde obliteration are listed
in Table 3. One hundred percent
(14/14) of gastric varices in group 1 and 95% (18/19) in group 2 were markedly
reduced in size 1 month after transjugular retrograde obliteration. The
varices dis appeared 3 months after transjugular retrograde obliteration in
both groups.
View larger version (7K):
[in this window]
[in a new window]
[as a PowerPoint slide]
|
Fig. 2 —Graph shows cumulative occurrence rate of esophageal varices
after transjugular retrograde obliteration in groups 1 and 2. Lower line
represents group 1, partial splenic embolization and transjugular retrograde
obliteration; upper line, group 2, trans jugular retrograde obliteration only.
Difference between the two groups was statistically significant (p
< 0.05).
|
|
The 3-year cumulative survival rate after transjugular retrograde
obliteration was 92% in group 1 and 95% in group 2. The 3-year recurrence rate
of gastric varices after transjugular retrograde obliteration was 0% in both
groups. The cumulative occurrence rates of esophageal varices with the red
color sign after transjugular retrograde obliteration in group 1 at 6 months
and 1, 2, and 3 years were 0%, 0%, 9%, and 9%, respectively, as opposed to
16%, 27%, 45%, and 45% in group 2 (Fig.
2). The differences between the two groups were statistically
significant (p = 0.045). Endoscopic findings of these varices are
listed in Table 4. Varices were
treated with endoscopic injection sclerotherapy. Gastrointestinal bleeding was
not observed after transjugular retrograde obliteration in either group. One
patient in group 1 and one patient in group 2 died of liver failure 5 and 6
months, respectively, after transjugular retrograde obliteration.
View this table:
[in this window]
[in a new window]
|
TABLE 4: Endoscopic Findings of Esophageal Varices Occurring After Transjugular
Retrograde Obliteration in the Two Groups
|
|
Discussion
There is no standard treatment for gastric varices
[7–10].
In patients with preserved liver function, periesophagogastric devascular
ization and splenectomy—the Hassab procedure
[11]—may be indicated.
However, in patients with poor liver function this surgical approach is
usually less well tolerated so that a minimally invasive alternative is
warranted. A transjugular intrahepatic portosystemic shunt (TIPS)
[12] can lower the portal
venous pressure and may be indicated in patients with bleeding esophageal
varices. However, the portal venous pressure in patients with gastric varices
is usually lower than in patients with esophageal varices
[13,
14].
We have been treating gastric varices using transjugular retrograde
obliteration since September 1991
[1]. With our method, the
sclerosant volume can be kept at a safe dose, unlike other methods
[2]. There was no recurrence
and no bleeding of gastric varices in any patient after transjugular
retrograde obliteration. The overall cumulative survival rates after
transjugular retrograde obliteration at 1, 3, and 5 years were 92%, 76%, and
61%, respectively [3]. Ninoi et
al. [15] reported that the
cumulative gastric variceal bleeding rate at 1 year was 20% in a TIPS group
and 2% in a transvenous obliteration group (p < 0.01). The
cumulative survival rates in the TIPS group at 1, 3, and 5 years were 81%,
64%, and 40%, respectively, as opposed to 96%, 83%, and 76% in the transvenous
obliteration group, respectively (p < 0.01)
[15]. On the basis of these
results, TIPS cannot be recommended for the treatment of gastric varices.
Although transjugular retrograde obliteration has a dramatic effect in
eradicating large gastric varices, it is weak for increasing portal venous
pressure [4] and for
accelerating the occurrence of esophageal varices in Child-Pugh classification
B and C patients in the long term
[3]. Flow in the paraesophageal
vein and the trunk of the left gastric vein is not controlled by transjugular
retrograde obliteration, and these vessels cannot be obliterated. Therefore,
it is not surprising that esophageal varices occur through the left gastric
vein or the cardiac venous plexus after transjugular retrograde obliteration.
Even if esophageal varices after transjugular retrograde obliteration can be
easily controlled by endoscopic injection sclero therapy before they bleed,
the occurrence of esophageal varices after obliteration is a disadvantage to
patients. Because the portal venous pressure in patients with gastric varices
and a gastrorenal shunt is lower than in patients with esophageal varices,
gastric varices may play a protective role in the development of esophageal
varices. The issue is how to prevent the development of esophageal varices
after the treatment of gastric varices by transjugular retrograde
obliteration.
Many reports have revealed that partial splenic embolization effectively
improves hypersplenism [16],
portal hypertensive gastropathy
[17], liver function
[18], and the prognosis of
patients who have undergone endoscopic injection sclerotherapy for esophageal
varices [19]. Our results
confirm that partial splenic embolization can reduce the hyperdynamic cycle of
portal hypertension and, in combination with transjugular retrograde
obliteration, decrease the occurrence rate of esophageal varices. We believe
that partial splenic embolization contributed to preventing the increase of
portal venous pres sure after transjugular retrograde obliteration. We further
believe that partial splenic embolization has the potential to compensate for
the drawbacks of transjugular retrograde obliteration; we will be using this
combination therapy for gastric varices in future patients. In this study, we
performed partial splenic embolization and transjugular retrograde
obliteration on separate days because the access routes of the two procedures
are different. The procedures can also be performed on the same day, thereby
saving the patient a trip to interventional radiology.
On the other hand, some patients have complicated gastroesophageal varices
with a gastrorenal shunt. Such cases are less common than isolated gastric
varices and are thought to be in a serious stage of portal hypertension
[20]. In these cases,
endoscopic injection sclerotherapy is necessary before transjugular retrograde
obliteration. Because endoscopic injection sclerotherapy and transjugular
retrograde obliteration bring further congestion in the portal venous system,
partial splenic embolization is strongly recommended. Randomized controlled
trials are also needed to evaluate whether the combined treatment of
transjugular retrograde obliteration and partial splenic embolization is more
effective than the devascularization proposed by Hassab
[11]. We conclude that the
combination of transjugular retrograde obliteration and partial splenic
embolization for gastric varices is more effective than transjugular
retrograde obliteration alone in the long-term prevention of esophageal
varices after transjugular retrograde obliteration.
References
- Chikamori F, Shibuya S, Takase Y, Ozaki A, Fukao K. Transjugular
retrograde obliteration for gastric varices. Abdom
Imaging 1996; 21:299
-303[CrossRef][Medline]
- Kanagawa H, Mima S, Kouyama H, Gotoh K, Uchida T, Okuda K.
Treatment of gastric fundal varices by balloon-occluded retrograde
obliteration. Gastroenterol Hepatol 1996;11
: 51-58[CrossRef]
- Chikamori F, Kuniyoshi N, Shibuya S, Takase Y. Eight years of
experience with transjugular retrograde obliteration for gastric varices with
gastrorenal shunts. Surgery 2001;129
: 414-420[Medline]
- Chikamori F, Kuniyoshi N, Shibuya S, Takase Y. Short-term portal
hemodynamic effects of transjugular retrograde obliteration of gastric varices
with gastrorenal shunts. Dig Surg 2000;17
: 332-336[CrossRef][Medline]
- Chikamori F, Kuniyoshi N, Shibuya S, Takase Y. Short-term portal
hemodynamic effects of partial splenic embolization for hypersplenism.
Hepatogastroenterology 2007;54
: 1847-1849[Medline]
- Beppu K, Iniokuchi K, Koyanagi N, et al. Prediction of variceal
hemorrhage by esophageal endoscopy. Gastrointest
Endosc 1981; 27:213
-218[Medline]
- Greig JD, Garden OJ, Anderson JR, Carter DC. Management of gastric
variceal hemorrhage. Br J Surg 1990;77
: 297-299[Medline]
- Ramond MJ, Valla D, Mosnier JF, et al. Successful endoscopic
obturation of gastric varices with butyl cyanoacrylate.
Hepatology 1989;10
: 488-493[CrossRef][Medline]
- Sarin SK, Sachdev G, Nanda R, Misra SP, Broor SL. Endoscopic
sclerotherapy in the treatment of gastric varices. Br J
Surg 1988; 75:747
-750[Medline]
- Trudeau W, Prindiville T. Endoscopic injection sclerosis in
bleeding gastric varices. Gastrointest Endosc1986; 32:264
-268[Medline]
- Hassab MA. Gastroesophageal decongestion and splenectomy in the
treatment of esophageal varices in bilharzial cirrhosis: further studies with
a report of 355 operations. Surgery 1967;61
: 169-17.[Medline]
- LaBerge JM, Ring EJ, Gordon RL, et al. Creation of transjugular
intrahepatic portosystemic shunts with the Wallstent endoprosthesis: results
in 100 patients. Radiology 1993;187
: 413-420[Abstract/Free Full Text]
- Chikamori F, Kuniyoshi N, Shibuya S, Takase Y. Correlation between
endoscopic and angiographic findings in patients with esophageal and isolated
gastric varices. Dig Surg 2001;18
: 176-181[CrossRef][Medline]
- Watanabe K, Kimura K, Matsutani S, Ohto M, Okuda K. Portal
hemodynamics in patients with gastric varices: a study in 230 patients with
esophageal and/or gastric varices using portal vein catheterization.
Gastroenterology 1988;95
: 434-440[Medline]
- Ninoi T, Nakamura K, Kaminou T, et al. TIPS versus transcatheter
sclerotherapy for gastric varices. AJR2004; 183:369
-376[Abstract/Free Full Text]
- Owman T, Lunderquist A, Alwmark A, Borjesson B. Embolization of the
spleen for treatment of splenomegaly and hypersplenism in patients with portal
hypertension. Invest Radiol 1979;14
: 457-464[Medline]
- Shimizu T, Onda M, Tajiri T, et al. Bleeding portal-hypertensive
gastropathy managed successfully by partial splenic embolization.
Hepatogastroenterology 2002;49
: 947-949[Medline]
- Tajiri T, Onda M, Yoshida H, Mamada Y, Taniai N, Kumazaki T.
Long-term hematological and biochemical effects of partial splenic
embolization in hepatic cirrhosis.
Hepatogastroenterology 2002;49
: 1445-1448[Medline]
- Tajiri T, Onda M, Taniai N, Yoshida H, Mamada Y. A comparison of
combination endoscopic therapy and interventional radiology with esophageal
transection for the treatment of esophageal varices.
Hepatogastroenterology 2002;49
: 1552-1554[Medline]
- Chikamori F, Kuniyoshi N, Kawashima T, Shibuya S, Takase Y.
Combination treatment of partial splenic embolization, endoscopic embolization
and transjugular retrograde obliteration for complicated gastroesophageal
varices. Hepatogastroenterology 2004;51
: 1506-1509[Medline]
CiteULike 
Complore 
Connotea 
Del.icio.us 
Digg 
Reddit 
Technorati What's this?
Top
Abstract
Introduction
