Peripheral Vascular Disease

Comparison of Long-Term Outcomes of Endovascular

Management for Membranous and Segmental Inferior
Vena Cava Obstruction in Patients With Primary
Budd–Chiari Syndrome
Qianxin Huang, MD*; Bin Shen, MD*; Qingqiao Zhang, MD, PhD; Hao Xu, MD;
Maoheng Zu, MD; Yuming Gu, MD; Ning Wei, MD; Yanfeng Cui, MD; Rui Huang, MD

Background—Endovascular management is important for the treatment of primary Budd–Chiari syndrome, which is caused
by inferior vena cava (IVC) obstruction. The aims of this study were to compare long-term outcomes of endovascular
management for primary Budd–Chiari syndrome patients with membranous obstruction of IVC (MOVC) and segmental
obstruction of IVC (SOVC) and explore the optimal endovascular strategy for these conditions.
Methods and Results—Clinical data of 265 patients with Budd–Chiari syndrome who received endovascular management
(MOVC group, n=136; SOVC group, n = 129) were retrospectively reviewed. Cumulative IVC patency rates were
generated by the Kaplan–Meier method and compared by log-rank test. In total, 245 patients were followed up from 3
to 72 months after treatment. The difference of long-term outcomes of balloon dilation alone versus stent placement was
not significant in each group. The overall cumulative 1-, 3-, and 5-year primary IVC patency rates were 98.3%, 90.7%,
and 83.8% in the MOVC group and 88.3%, 79.1%, and 67.9% in the SOVC group (P=0.007), respectively. The long-term
IVC patency rates were lower in the SOVC group than in the MOVC group for patients who underwent balloon dilation
alone (P=0.001) and did not significantly differ for patients who underwent stent placement between both the groups
(P=0.687).
Conclusions—The long-term treatment outcome of endovascular management was better for primary Budd–Chiari
syndrome patients with MOVC than for those with SOVC. Balloon dilation alone could be the optimal treatment
for patients with MOVC. However, stent placement should be more strongly recommended for patients with SOVC.  
(Circ Cardiovasc Interv. 2016;9:e003104. DOI: 10.1161/CIRCINTERVENTIONS.115.003104.)
Key Words: Budd–Chiari syndrome ◼ endovascular procedures ◼ inferior vena cava ◼ retrospective studies
◼ stents

B udd–Chiari syndrome (BCS) is a vascular disorder char-

acterized by obstruction of hepatic venous outflow from
the hepatic venules to the entrance of the inferior vena cava
(IVC) into the right atrium.1,2 In contrast to Western coun-
tries, IVC obstruction is a common cause of BCS in China.
This disorder is classified into membranous obstruction of
IVC (MOVC; obstructed segment of IVC is ≤1.0 cm) and
segmental obstruction of IVC (SOVC; obstructed segment
of IVC is >1.0 cm).3–6 Treatments for this disorder include
medical management, surgical operation, and endovascular
intervention. Medical management alone has a limited abil-
ity to arrest progression of the disease, as reported in a study
of 237 patients, where 72% failed to show a significant sur-
vival benefit.7 Surgery has been associated with high mortality
rates (≤50%) and complicated by dysfunction rates as high as
32%.8,9 Endovascular intervention has proven more effective
than medical management and is associated with lower rates
of mortality than open surgical procedures.10–12 It has become
the primary treatment of choice for BCS management in
China because of its minimal invasiveness and good efficacy.
Although several studies reported the safety and efficacy
of endovascular intervention for the management of primary
BCS caused by IVC obstruction,13–16 few studies compared
long-term outcomes of endovascular management between
MOVC and SOVC patients. Moreover, there are still con-
troversies about the optimal endovascular strategy for these
conditions according to previous reports.5,13–16 Two previous
studies showed that balloon dilation alone was sufficient in
most BCS patients with MOVC and SOVC, and <3% had
recurrence in 3 to 8 years of follow-up.5,13 However, other

Received August 9, 2015; accepted February 8, 2016.

From the Department of Interventional Radiology, the Affiliated Hospital of Xuzhou Medical College, Xuzhou, China.
*Drs Q. Huang and Shen contributed equally to this work.
The Data Supplement is available at http://circinterventions.ahajournals.org/lookup/suppl/doi:10.1161/CIRCINTERVENTIONS.115.003104/-/DC1.
Correspondence to Qingqiao Zhang, MD, PhD, Department of Interventional Radiology, the Affiliated Hospital of Xuzhou Medical College, No.99,
Huaihai West Rd, Xuzhou 221006, China. E-mail 1427286069@qq.com
© 2016 American Heart Association, Inc.
Circ Cardiovasc Interv is available at http://circinterventions.ahajournals.org DOI: 10.1161/CIRCINTERVENTIONS.115.003104

1
Downloaded from http://circinterventions.ahajournals.org/ at University of Colorado on April 1, 2016
2   Huang et al   Endovascular Management of Budd–Chiari Syndrome
WHAT IS KNOWN
• Endovascular intervention comprising balloon dila-
tion and stent placement has become the treatment
of choice for primary Budd–Chiari syndrome (BCS)
with membranous and segmental inferior vena cava
(IVC) occlusion because of its minimal invasiveness
and satisfactory efficacy.
• To date, however, few studies have compared the
long-term outcomes of endovascular management
between membranous and segmental IVC occlusion
in patients with primary BCS.
WHAT THE STUDY ADDS
• The long-term outcome of endovascular treatment
was better for primary BCS patients with membra-
nous IVC occlusion than for those with segmental
IVC occlusion.
• For BCS patients with membranous IVC occlusion,
balloon dilation alone resulted in favorable outcomes,
suggesting stenting can be reserved for technical an-
gioplasty failure. For BCS patients with segmental
IVC occlusion, the efficacy of balloon dilation alone
was poor and routine stenting should be considered.
• Increased serum alkaline phosphatase and segmental
IVC occlusion were independent risk factors for re-
occlusion in patients with BCS.
scholars reported a high frequency (>70%) of reocclusion in
patients who underwent balloon dilation alone, and primary
stent placement was suggested in both MOVC and SOVC
patients.14–16
In this study, we compared long-term outcomes of endo-
vascular management, including balloon dilation alone and
adjunctive stent placement, in primary BCS patients with
MOVC and SOVC. We also determined the risk factors for
postoperative reocclusion and explored the optimal endo-
vascular strategy for these conditions (Table I in the Data
Supplement).
Methods
Patients
This retrospective cohort study was approved by the ethics committee
at the Affiliated Hospital of Xuzhou Medical College. All patients in
our department received detailed information about the endovascular
procedures. Patients provided written informed consent before treat-
ment. Inclusion criteria were as follows: (1) patients aged between 18
and 80 years, (2) primary BCS with MOVC or SOVC, (3) occlusion
of an IVC segment above the ostium of the renal veins, (4) at least 1
patent hepatic vein or accessory hepatic vein, and (5) no acute throm-
bosis in the IVC or hepatic veins. Exclusion criteria were as follows:
(1) patients with secondary BCS, (2) IVC patency or stenosis, (3)
IVC occlusion involving the ostium of the renal veins, (4) obstruction
of all 3 main hepatic veins and the accessory hepatic vein, (5) primary
BCS complicated by hepatocellular carcinoma, (6) contraindications
for anticoagulation, and (7) severe hepatic, renal, or heart failure.
From January 2009 to December 2014, 507 consecutive pa-
tients with primary BCS underwent endovascular management in
our department. Among them, IVC obstruction was identified in
Downloaded from http://circinterventions.ahajournals.org/ at University of Colorado on April 1, 2016
357 patients. On the basis of the inclusion and exclusion criteria,
92 patients were excluded and 265 patients (mean age, 49.1±10.3
years; range, 18–79 years) were included in this study. Patient diag-
noses were established on the results of abdominal ultrasonography
(265/265), magnetic resonance angiography (230/265), or computed
tomographic angiography (35/265) before treatment. Finally, IVC an-
giography was performed in all patients to identify the site, degree,
and extent of IVC obstruction. According to the length of the IVC
occlusion, patients were classified into the MOVC group (n=136) or
the SOVC group (n=129). The endovascular treatment strategy was
recorded, and the treatment steps are shown in Figure 1. Baseline
clinical characteristics of all 265 patients are summarized in Table 1.
Underlying pathogenic factors for BCS were identified in 31 patients.
Twenty patients had positive anticardiolipin IgG antibodies. Six pa-
tients had antithrombin deficiency. Hyperhomocysteinemia was pres-
ent in 5 patients. JAK2 V617F mutation and protein C and protein S
were not examined.
Endovascular Procedures
Under local anesthesia, a 6F catheter sheath was placed in the right
femoral vein, and a 5F pigtail catheter (Cook Medical, Bloomington,
IN) was advanced into the IVC below the distal part of the obstruction
via the sheath. Angiography was performed, and the pigtail catheter
was retained at the distal end of the IVC occlusion as a localizer. We
punctured the right jugular vein and inserted a 6F catheter sheath. A
5F angiographic catheter (Cordis Corp, Miami, FL) was advanced
to the proximal end of the IVC and angiography was performed.
Combination of the 2 venographic results delineated the extent of
IVC occlusion and collateral circulation.
A steel needle with a curved head17,18 was advanced through the
5F angiographic catheter via the right jugular approach to penetrate
the IVC occlusion from the proximal end under fluoroscopic guid-
ance. The weakest point of the IVC obstruction was carefully probed,
rather than using brute force to penetrate it. Once the steel needle
went through the IVC occlusion, the 5F angiographic catheter was
advanced along the steel needle track. The needle was withdrawn,
and an ultrasmooth exchange guidewire (Cook Medical; 260-cm
long, 0.035 inch in diameter) was advanced through the 5F angio-
graphic catheter to the distal end of the IVC. Then a balloon cath-
eter (Optimed, Ettlimgen, Germany; 40-mm long, 20, 24, 26, or 28
mm in diameter) was placed across the occlusion via the exchange
guidewire. The diameter of the balloon used was 20% more than the
diameter of the IVC below the occlusion. The balloon was manually
Figure 1. Endovascular treatment steps for Budd–Chiari syn-
drome with inferior vena cava occlusion. IVC indicates inferior
vena cava.
3   Huang et al   Endovascular Management of Budd–Chiari Syndrome

Table 1.  Baseline Characteristics of all 265 Patients inflated twice for 1 minute each time at the site of the IVC occlusion.
The purpose of the dilation was disappearance of the waist at the
MOVC Group SOVC Group obstruction site, which was confirmed by fluoroscopy. After dilation,
Variable (n=136) (n=129) P Value the balloon catheter was removed. Subsequently, IVC angiography
Sex, n
was performed through the right femoral vein to evaluate the extent
of IVC patency. The pressure gradient between the IVC and the right
 Male/female 70/66 61/68 0.496 atrium was recorded before and after treatment. Then, a 5F angio-
graphic catheter was inserted through the right femoral vein into the
 Age, y 18–79 18–78 0.084
hepatic vein or the accessory hepatic vein to perform angiography. If
(50.0±10.1) (47.8±10.9)
the IVC occlusion could not be penetrated through the right jugular
 Length of IVC ≤1.0 1.1–7.5 vein, penetration of the occluded IVC segment from the distal end
occlusion, cm (2.8±2.6) was performed through the right femoral vein.
If IVC recoil (>50%) was present or the IVC–right atrium pres-
Duration of symptom, n (%) sure gradient was higher than 15 cmH2O (1 cmH2O=0.098 kPa) after
 <1 mo 18 (13.2) 8 (6.2) 0.156 balloon dilation alone, a self-expandable stainless steel Z-type IVC
stent (Shenyang Yongtong Technology Co, Ltd, Shenyang, China;
 1–6 mo 22 (16.2) 22 (17.1) 70-, 80-, or 90-mm long, 28 or 30 mm in diameter) was deployed
 >6 mo 96 (70.6) 99 (76.7) to the site of the IVC obstruction as a bail-out, using a 12F delivery
sheath (Check-Flo Performer, Cook Incorporated, Bloomington, IN)
Symptoms and signs, n (%) through the right femoral vein. If the IVC diameter below the occlu-
 Abdominal fullness 41 (30.1) 34 (26.4) 0.494
sion was <26 mm, a 28-mm diameter stent was used. Otherwise, a
30-mm diameter stent was chosen. The length of the used stent was
 Abdominal pain 22 (16.2) 17 (13.2) 0.491 required to cover the entire IVC obstruction without covering the
opening of the renal veins.
 Thoracoabdominal 64 (47.1) 70 (54.3) 0.271
distention
 Edema of lower limbs 61 (44.9) 55 (42.6) 0.716
Post-Treatment Observations and Follow-Up
After endovascular treatment, all patients were prescribed subcutane-
 Ascites 48 (35.3) 70 (54.3) 0.002 ous low-molecular-weight heparin (Nadroparin Calcium, 5000 U/12
hours) for 3 to 4 days, followed by warfarin sodium for at least 1
 Gastrointestinal bleeding 15 (11.0) 10 (7.8) 0.362
year. The international normalized ratio was maintained at 2 to 3.
 Lower limbs pigmentation 63 (46.3) 54 (41.9) 0.465 Follow-up was performed at 1, 3, 6, and 12 months after treatment
and then annually or whenever symptoms recurred. Follow-up in-
 Leg ulcer 11 (8.1) 14 (10.9) 0.442
cluded clinical signs, imaging workups (abdominal color Doppler ul-
 Hepatosplenomegaly 108 (79.4) 100 (77.5) 0.708 trasound, magnetic resonance angiography, or computer tomographic
angiography), and laboratory test results (coagulation function, liver
Laboratory test and kidney function). Patients were followed up until death or the end
 AST, U/L 30.4±12.2 29.2±10.9 0.438 of the study period.
 ALT, U/L 22.1±10.2 21.2±10.4 0.531
Statistical Analysis
 ALP, U/L 108.8±41.8 103.1±59.4 0.407
Quantitative data are expressed as mean±SD, and they were com-
 Albumin, g/L 39.8±6.0 39.2±5.7 0.407 pared with independent sample t test. The changes in the IVC–right
atrium pressure gradient before and after treatment were assessed
 Creatinine, μmol/L 62.2±19.3 58.6±25.0 0.234 with a paired-samples t test. Categorical data are expressed as a ra-
 TBIL, μmol/L 34.7±20.5 29.5±17.0 0.047 tio (or percentage), and the χ2 test or Fisher exact test was used for
comparisons. The follow-up time was calculated as the time from
 BUN, mmol/L 5.8±1.6 5.4±2.0 0.124 the date of primary endovascular treatment to the date of patients’
 Prothrombin time, s 14.7±2.2 16.3±11.2 0.139 death or December 31, 2014. Patients who died during follow-up and
patients lost to follow-up were censored at the time of last known
 Fibrinogen, g/L 2.6±0.8 2.5±0.9 0.466 contact. Missing outcome data because of technical failure or loss to
 WBC count, 109/L 3.8±1.6 3.5±2.1 0.467 follow-up were not included in the analyses of long-term outcomes.
Cumulative patency rates were calculated with Kaplan–Meier curves
 Hemoglobin, g/dL 124.8±21.3 120.4±22.9 0.157 and compared using the log-rank test. A Cox proportional hazards re-
gression model was used to analyze risk factors for reocclusion. The
 Platelet count, 10 /L
9
97.1±43.8 105.1±68.1 0.311
proportional hazard assumption in the Cox model was assessed with
Child-Pugh Class, n the Schoenfeld residuals method,19 and all proportionality assump-
tions were appropriate. Baseline variables were first investigated by
 A/B/C 92/42/2 78/47/4 0.383
univariate Cox proportional hazards regression analysis. Candidate
 Child–Pugh score 6.3±1.2 6.5±1.3 0.127 variables with a P value of <0.10 in the univariate analyses were con-
sidered in the multivariate adjusted model. A difference with P<0.05
 Rotterdam score 0.5±0.5 0.8±0.6 0.002 was considered statistically significant. All statistical analyses were
 Clichy score 4.9±0.7 5.1±0.9 0.037 performed using SPSS, version 16.0 (SPSS Inc, Chicago, IL).
 New Clichy score 3.8±1.3 4.0±1.3 0.274
Results
 MELD score 10.9±2.8 10.8±3.0 0.864
ALP indicates alkaline phosphatase; ALT, alanine aminotransferase; AST, Immediate Outcomes
aspartate aminotransferase; BUN, blood urea nitrogen; IVC, inferior vena cava; Technical success was defined as eliminated IVC obstruc-
MELD, model of end-stage liver disease; MOVC, membranous obstruction of IVC; tion along with a decrease in the IVC–right atrium pressure
SOVC, segmental obstruction of IVC; TBIL, total bilirubin; and WBC, white blood cell. gradient. The overall technical success rate of endovascular
Downloaded from http://circinterventions.ahajournals.org/ at University of Colorado on April 1, 2016
4   Huang et al   Endovascular Management of Budd–Chiari Syndrome

Table 2.  Improvements in Symptoms and Signs After Treatment at Discharge

Membranous IVC Obstruction Group (n=136) Segmental IVC Obstruction Group (n=129)
Before After Treatment, n (%) Before After Treatment, n (%)
Treatment, Treatment,
Symptoms and Signs n (%) Disappeared Improved Stabilized n (%) Disappeared Improved Stabilized
Abdominal fullness 41 (30.1) 39 (28.6) 2 (1.5) 0 (0.0) 34 (26.4) 29 (22.5) 5 (3.9) 0 (0.0)
Abdominal pain 22 (16.2) 22 (16.2) 0 (0.0) 0 (0.0) 17 (13.2) 16 (12.4) 1 (0.8) 0 (0.0)
Thoracoabdominal 64 (47.1) 50 (36.8) 13 (9.6) 1 (0.7) 70 (54.2) 48 (37.2) 18 (13.9) 4 (3.1)
distention
Edema of lower 61 (44.9) 51 (37.5) 10 (7.4) 0 (0.0) 55 (42.6) 50 (38.7) 5 (3.9) 0 (0.0)
limbs
Ascites 48 (35.3) 38 (27.9) 8 (5.9) 2 (1.5) 70 (54.2) 39 (30.2) 26 (20.1) 5 (3.9)
Gastrointestinal 15 (11.0) 15 (11.0) 0 (0.0) 0 (0.0) 10 (7.8) 10 (7.8) 0 (0.0) 0 (0.0)
bleeding
Lower limbs 63 (46.3) 33 (24.3) 22 (16.2) 8 (5.8) 54 (41.8) 23 (17.8) 18 (13.9) 13 (10.1)
pigmentation
Leg ulcer 11 (8.1) 7 (5.1) 3 (2.2) 1 (0.7) 14 (10.8) 5 (3.9) 7 (5.4) 2 (1.5)
Hepatosplenomegaly 108 (79.4) 60 (44.1) 36 (26.5) 12 (8.8) 100 (77.5) 48 (37.2) 33 (25.6) 19 (14.7)

treatment was 99.2% (263/265) in this study. The mean
IVC–right atrium pressure gradient decreased from 25.7±8.1
cmH2O (range, 17–34) before treatment to 6.8±3.7 cmH2O
(range, 2–13) after treatment (t value=6.3, P<0.001). Patients
who underwent stent placement had a longer occlusion length
than those who underwent balloon dilation alone in the SOVC
group (3.7±1.6 versus 2.5±1.3 cm, respectively; P<0.001).
Regardless of the occlusion length, technical success was
obtained in 127 of 129 patients in the SOVC group. Of the
265 cases, 152 had 1 patent hepatic vein, 49 had 2 patent
hepatic veins, and 17 had 3 patent hepatic veins. At least 1
large and patent accessory hepatic vein was identified in the
other 47 patients. Angioplasty of the hepatic vein or the acces-
sory hepatic vein was not performed in any patient. The renal
veins were not covered by stents in any patient who under-
went stent placement. The main clinical symptoms and signs
significantly improved or stabilized after treatment (Table 2).
Technical failure occurred in 2 patients with SOVC. The
occluded IVC segment was rigid in both the patients. After
IVC penetration, the balloons used for dilation ruptured in
both the patients (20-mm balloon in 1 patient and 24-mm bal-
loon in the other patient). No further endovascular treatment
was performed in these patients.
Three patients (1.1%) had serious intraoperative com-
plications. Two patients with MOVC developed pericardial
tamponade after the puncture needle was inserted into the
pericardium by mistake. The 2 patients were treated suc-
cessfully using a 4F pigtail catheter for pericardial drainage.
One patient with SOVC sustained IVC rupture and hemor-
rhage caused by dilation of a 24-mm balloon catheter. A
26-mm balloon was used for temporary plugging, followed
by a covered stent (70-mm long and 28 mm in diameter;
Shanghai MicroPort Co, Ltd, Shanghai, China) placed at the
ruptured portion through the right jugular vein. Successful
IVC repair was achieved in this patient. There were no pro-
cedure-related deaths. No stents were dislocated into the
right atrium.
Long-Term Outcomes
Patients in this study were followed up for 3 to 72 (mean,
43.2±20.0) months after treatment. Eighteen patients were lost
to follow-up. Of the remaining 245 patients, 36 experienced
reocclusion during follow-up (Table 3). Twenty-four patients
with reocclusion underwent another balloon dilation or stent
placement, and IVC patency was again achieved. The other 12
patients refused further intervention. Of the 52 stented cases,
46 cases of mild intimal hyperplasia were found by ultrasound
examination. There were other 6 cases of reocclusion at the
end of the study follow-up. No significant difference in the
occlusion length of IVC was detected between patients with

Table 3.  Comparison of Reocclusion Incidence Between MOVC and SOVC Groups During
Long-Term Follow-Up
Overall Balloon Dilation Stent Placement
Reocclusion,
Group Patients, n Reocclusion, n (%) Patients, n Reocclusion, n (%) Patients, n n (%)
MOVC 128 12 (9.3) 115 11 (9.5) 13 1 (7.6)
SOVC 117 24 (20.5) 78 19 (24.3) 39 5 (12.8)
P value 0.014 0.005 0.616
MOVC indicates membranous inferior vena cava obstruction; and SOVC, segmental inferior vena cava obstruction.

Downloaded from http://circinterventions.ahajournals.org/ at University of Colorado on April 1, 2016

5   Huang et al   Endovascular Management of Budd–Chiari Syndrome

and without reocclusion in the SOVC group (3.2±1.6 versus
2.8±1.5 cm, respectively; P=0.286).
The cumulative 1-, 3-, and 5-year primary IVC patency
rates of the 245 cases were 93.5%, 85.2%, and 76.1%, respec-
tively. The cumulative 1-, 3-, and 5-year secondary IVC
patency rates were 98.7%, 94.9%, and 89.9%, respectively.
The overall cumulative 1-, 3-, and 5-year primary IVC patency
rates were 98.3%, 90.7%, and 83.8% in the MOVC group
and 88.3%, 79.1%, and 67.9% in the SOVC group, respec-
tively (P=0.007). For patients who underwent balloon dilation
alone, the cumulative 1-, 3-, and 5-year primary IVC patency
rates were 98.2%, 91.0%, and 83.2% in the MOVC group and
83.7%, 76.9%, and 62.9% in the SOVC group, respectively
(P=0.001). For patients who underwent bail-out stenting, the
cumulative 1-, 3-, and 5-year primary IVC patency rates were
100.0%, 88.9%, and 88.9% in the MOVC group and 97.2%,
85.0%, and 77.3% in the SOVC group, respectively (P=0.687;
Figure 2). There were no significant differences in the cumu-
lative 1-, 3-, and 5-year primary IVC patency rates between
balloon dilation alone and bail-out stenting for patients in the
MOVC group (P=0.866) or SOVC group (P=0.065).
Two patients in the SOVC group experienced stent frac-
ture. These were identified by computed tomographic angiog-
raphy during clinical follow-up, one at 26 months and another
one at 28 months after treatment. The IVC lumen was still
patent in both the patients. Three patients (2 in the SOVC
group and 1 in the MOVC group) had right hepatic vein occlu-
sion, and 1 patient in the SOVC group had accessory hepatic
vein occlusion caused by stent implantation. All 4 cases of
hepatic vein occlusions were successfully treated by balloon
angioplasty. None of the patients presented raised creatinine
or renal vein thrombosis during follow-up. Five patients (3
in the MOVC group and 2 in the SOVC group) died during
follow-up. Three of them died of liver failure, 1 each at 5, 13,
and 14 months. One patient died of hepatocellular carcinoma
21 months after stent placement. One patient died of causes
unrelated to BCS 12 months after treatment.
Risk Factors for Reocclusion
According to univariate analysis, the risk factors for reocclu-
sion were sex (male), segmental occlusion, increased level
of serum alkaline phosphatase (ALP), and decreased levels

Figure 2. Kaplan–Meier curves for comparisons of long-term inferior vena cava (IVC) patency states between the membranous obstruc-
tion of IVC (MOVC) group and segmental obstruction of IVC (SOVC) group. A, The overall comparison. B, Comparison of patients who
underwent balloon dilation alone. C, Comparison of patients who underwent bail-out stenting.

Downloaded from http://circinterventions.ahajournals.org/ at University of Colorado on April 1, 2016

6   Huang et al   Endovascular Management of Budd–Chiari Syndrome
Table 4.  Multivariable Analysis for Postoperative
Reocclusion
Multivariable Cox Model
Variables HR (95% CI)
Sex (male) 1.650 (0.783–3.477)
IVC obstruction type (SO) 3.496 (1.623–7.352)
ALP (per unit increase) 1.010 (1.006–1.014)
Albumin (per unit decrease) 1.008 (0.920–1.103)
Hemoglobin (per unit decrease) 0.984 (0.964–1.005)
ALP indicates alkaline phosphatase; CI, confidence interval; HR, hazard ratio;
IVC, inferior vena cava; and SO, segmental occlusion.
of serum albumin and hemoglobin. Multivariate analysis
confirmed that an increased ALP concentration and seg-
mental occlusion were independent risk factors for reocclu-
sion (Table 4). According to receiver–operator characteristic
analysis to determine the prognostic value of ALP for reoc-
clusion risk, the area under the receiver–operator character-
istic curve was 0.693 (95% confidence interval, 0.597–0.789;
P<0.001). For segmental occlusion, the area under the
receiver–operator characteristic curve was 0.614 (95% confi-
dence interval, 0.512–0.716; P=0.035; Figure 3). The duration
of symptoms, prognostic index, and treatment with balloon
dilation alone were not correlated with post-treatment reoc-
clusion in this study.
Discussion
In this retrospective study, balloon dilation was used as a pri-
mary method of treatment, and adjunctive stent placement
was performed whenever there was significant residual steno-
sis after balloon dilation. Angioplasty of the occluded hepatic
Figure 3. Receiver–operator characteristic (ROC) curve for
the alkaline phosphatase value prognostic for reocclusion risk
showed an adequate discrimination capacity (area under the
curve [AUC], 0.693) similar to that for the segmental form of infe-
rior vena cava (IVC) obstruction (AUC, 0.614). SOVC indicates
segmental obstruction of IVC.
Downloaded from http://circinterventions.ahajournals.org/ at University of Colorado on April 1, 2016
veins was not attempted in this study because of the presence
of at least 1 patent hepatic or accessory hepatic vein, which
was sufficient to maintain venous return to the IVC.5,20 In this
study, endovascular treatment was technically successful in
P Value
263 of 265 BCS patients with IVC obstruction. Long-term
0.188 follow-up of the 245 patients was carried out, and satisfactory
0.002 therapeutic outcomes were achieved.
Specific exclusion criteria were established in this study
0.000
to reduce the influence of confounding factors on the com-
0.870 parison of long-term outcomes of endovascular intervention
0.127 between patients with MOVC and patients with SOVC. For
instance, if stenting was used in SOVC patients with IVC
occlusion involving the ostium of the renal veins, the renal
veins would be covered by the stent, possibly resulting in
renal vein obstruction. However, for patients with MOVC, the
occlusion site was far away from the renal veins, and the use of
a stent did not affect the renal veins. Therefore, patients with
IVC occlusion involving the ostium of the renal veins were
excluded to make the 2 groups more comparable. Patients
with BCS complicated by hepatocellular carcinoma were also
excluded because these patients often had poor prognosis, and
this would affect the assessment of long-term outcomes of
endovascular management. At least 1 main hepatic or acces-
sory hepatic vein was needed to achieve recanalization in
patients with occlusion of all 3 main hepatic veins or acces-
sory hepatic vein. Thus, these patients were excluded to allow
better comparisons of the endovascular procedures between
the 2 groups. In our routine clinical practice, the prevalence
of pediatric BCS is not high, but endovascular treatment with
balloon dilation and stent placement is also performed in these
patients. However, the IVC diameter and length in pediatric
patients is still in the process of growth and development, and
this might affect the results of long-term follow-up. Thus, we
excluded the pediatric population in this study. In addition, we
did not include protein C and protein S in the preinterventional
workup because previous research indicated that deficiencies
of protein C and protein S were rare in patients with BCS.21
Furthermore, the examination of protein C and protein S lev-
els is not part of the routine laboratory workup at our center.
In this study, the IVC occlusion in all patients with BCS was
complete, and it was usually tenacious and rigid. It was difficult
to cross this type of occlusion with a guidewire. Therefore, in the
majority of the patients, a pigtail catheter was inserted first via
the femoral approach and retained at the distal end of the IVC
occlusion to act as a localizer. Then, a steel needle with a curved
head was inserted to penetrate the IVC occlusion from the top
to the bottom end via the right jugular approach.17,18 This dual
approach improved the technical success rate and decreased
the occurrence of complications, such as IVC perforation and
hemopericardium. In most cases of BCS, there was a sufficient
distance between the proximal end of the IVC occlusion and the
right atrium. Thus, the steel needle could pass through into the
right atrium safely and reach the top end of the occlusion. This
was helpful to avoid piercing the pericardium in the following
penetration. In this study, 2 patients (0.7%) developed pericar-
dial tamponade because the site of the IVC occlusion was close
to the right atrium. The incidence of pericardial tamponade of
this dual approach was acceptable, when compared with that in
a previous report (1.9%).22
7   Huang et al   Endovascular Management of Budd–Chiari Syndrome
For patients with BCS caused by MOVC, endovascu-
lar management of balloon dilation alone can often achieve
satisfactory long-term outcomes. Yang et al13 showed that 38
BCS patients with MOVC underwent balloon angioplasty suc-
cessfully without stent placement. Only 1 patient experienced
recurrence during a follow-up period of ≤8 years. Ciesek
et al23 reported 5 white patients with MOVC, 3 of whom
received endovascular treatment with balloon dilation alone.
All procedures were successful, and no recurrence occurred
during follow-up (mean, 46±11 months). Ding et al24 reported
on 93 patients with BCS who had hepatic venous obstruction
(91 had membranous obstruction) with or without IVC obstruc-
tion, of which 91 (97.9%) were treated with balloon dilation
alone. In that study, long-term outcomes were satisfactory
with a low reocclusion rate of 8.9% (8/90). In our study, bal-
loon dilation alone was performed in most cases in the MOVC
group, and the recurrence rate was 9.6% (11/115), which was
similar to that reported by Ding et al.24 In contrast, Zhang
et al25 used IVC stent placement as the first-choice treatment
for MOVC, and long-term outcomes were also excellent. In
the MOVC group of our study, there was no significant dif-
ference (P=0.866) in long-term IVC patency rates between
balloon dilation alone and bail-out stenting. These results
show that both balloon dilation and bail-out stenting can
result in good long-term IVC patency for the management of
MOVC in patients with primary BCS. From the viewpoint of
patient costs, balloon dilation alone minimizes the operative
cost and procedure complexity, making it a more appropriate
first-choice endovascular treatment for patients with MOVC.
Moreover, the complications of long-term indwelling stents,
such as stent fracture, migration to the heart, and occlusion of
the ostium of the hepatic or accessory hepatic veins,26,27 serve
as a reminder that caution should be used when choosing and
placing stents in BCS patients with MOVC.
For BCS patients with SOVC, the usage of an IVC stent
remained a controversial issue. Srinivas et al28 successfully
treated 5 BCS patients with SOVC by endovascular interven-
tion and 4 of them (80%) underwent initial stent implanta-
tion; whereas Wu et al5 maintained that balloon angioplasty
alone should be the optimal treatment for this condition. In
our study, 84 of 129 patients underwent balloon dilation alone
because there was no significant IVC recoil after balloon dila-
tion. Stent placement was not used unless the residual lesions
pushed back into the lumen of the IVC once the balloon cath-
eter was removed. According to our clinical experience, in the
majority of primary BCS patients with IVC obstruction (mem-
branous or segmental), there is a weak point in the occlusion.
Thus, we used the technique of using a steel needle to find
this weak point instead of using brute force for penetration.
Sufficient dilation with a balloon catheter along the weak area
of IVC obstruction could be helpful to improve the efficacy
of balloon dilation alone. During the 72 months of follow-
up, the reocclusion incidence for patients in the SOVC group
who underwent balloon dilation alone was 24.3% (19/78).
This was a better success rate than that described in previ-
ous reports on balloon dilation alone.16,25 In addition, this was
not significantly higher than that for patients in our study who
underwent bail-out stenting (12.8%, 5/39; P=0.145). In our
opinion, the advantages of our technique played an important
Downloaded from http://circinterventions.ahajournals.org/ at University of Colorado on April 1, 2016
role in reducing the need for stent implantation in patients
with SOVC in this study.
Although no significant difference was detected for cumula-
tive 1-, 3-, and 5-year primary IVC patency rates between bal-
loon dilation alone and bail-out stenting for patients with SOVC
(P=0.065), the results demonstrated that the cumulative 1-, 3-,
and 5-year primary IVC patency rates for patients who under-
went balloon dilation alone were lower in the SOVC group
than those in the MOVC group (P=0.001). Thus, the long-term
outcomes of balloon dilation alone were worse in BCS patients
with SOVC than in those with MOVC, resulting in lower long-
term treatment efficacy of endovascular management for BCS
patients with SOVC (P=0.007). Indeed, stent implantation was
used more frequently for patients with SOVC in several previous
reports.16,25,28 Zhang et al25 reported on 115 patients with BCS, in
whom stents were placed whether they had MOVC or SOVC.
At a mean follow-up of 45 months, stent patency reached 90%.
Our study showed that the primary IVC patency rate for patients
who underwent stent placement was 87.2% (34/39) in the SOVC
group, which was consistent with the previous study.
Patients who presented IVC reocclusion during follow-up
were suggested to undergo another balloon angioplasty, and
secondary stent placement was recommended for both SOVC
and MOVC patients if IVC recoiled >50% intraoperatively
after balloon dilation. In addition, multivariate analysis in
our study showed that SOVC was an independent risk factor
for reocclusion in patients with BCS receiving endovascular
management. Therefore, to decrease the rate of reocclusion,
it is reasonable to recommend stent placement more strongly
for patients with SOVC. Furthermore, we found that increased
ALP was an independent risk factor for reocclusion. Previous
reports demonstrated that elevation of ALP in patient with
BCS was caused by an intrahepatic cholestatic process sec-
ondary to hepatic venous outflow obstruction.29 Our results
further indicated that elevated ALP levels positively corre-
lated with reocclusion in patients with BCS after endovascular
treatment. This is a finding that has not been demonstrated in
other research, indicating that higher ALP levels may play a
role in the occurrence of reocclusion.
Conclusions
The long-term treatment outcome of endovascular manage-
ment was better for primary BCS patients with MOVC than
for those with SOVC. For BCS patients with MOVC, bal-
loon dilation alone could be the optimal treatment and stent
placement should be used as a bail-out only when there is a
technical angioplasty failure. For BCS patients with SOVC,
the efficacy of balloon dilation alone was worse than that for
MOVC patients and stenting should be more strongly recom-
mended for these patients.
Disclosures
None.
References
1. Menon KV, Shah V, Kamath PS. The Budd-Chiari syndrome. N Engl J
Med. 2004;350:578–585. doi: 10.1056/NEJMra020282.
2. Seijo S, Plessier A, Hoekstra J, Dell’era A, Mandair D, Rifai K, Trebicka J,
Morard I, Lasser L, Abraldes JG, Darwish Murad S, Heller J, Hadengue A,
8   Huang et al   Endovascular Management of Budd–Chiari Syndrome
Primignani M, Elias E, Janssen HL, Valla DC, Garcia-Pagan JC; European
Network for Vascular Disorders of the Liver. Good long-term outcome
of Budd-Chiari syndrome with a step-wise management. Hepatology.
2013;57:1962–1968. doi: 10.1002/hep.26306.
3. Qi X, Guo X, Fan D. Difference in Budd-Chiari syndrome between the
West and China. Hepatology. 2015;62:656. doi: 10.1002/hep.27628.
4. Xu K, He FX, Zhang HG, Zhang XT, Han MJ, Wang CR, Kaneko M,
Takahashi M, Okawada T. Budd-Chiari syndrome caused by obstruction
of the hepatic inferior vena cava: immediate and 2-year treatment results
of transluminal angioplasty and metallic stent placement. Cardiovasc
Intervent Radiol. 1996;19:32–36.
5. Wu T, Wang L, Xiao Q, Wang B, Li S, Li X, Zhang J. Percutaneous bal-
loon angioplasty of inferior vena cava in Budd-Chiari syndrome-R1. Int J
Cardiol. 2002;83:175–178.
6. Okuda K. Membranous obstruction of the inferior vena cava (obliterative
hepatocavopathy, Okuda). J Gastroenterol Hepatol. 2001;16:1179–1183.
7. Darwish Murad S, Valla DC, de Groen PC, Zeitoun G, Hopmans JA,
Haagsma EB, van Hoek B, Hansen BE, Rosendaal FR, Janssen HL.
Determinants of survival and the effect of portosystemic shunting in pa-
tients with Budd-Chiari syndrome. Hepatology. 2004;39:500–508. doi:
10.1002/hep.20064.
8. Copelan A, Remer EM, Sands M, Nghiem H, Kapoor B. Diagnosis and
management of Budd Chiari syndrome: an update. Cardiovasc Intervent
Radiol. 2015;38:1–12. doi: 10.1007/s00270-014-0919-9.
9. Langlet P, Valla D. Is surgical portosystemic shunt the treatment of choice
in Budd-Chiari syndrome? Acta Gastroenterol Belg. 2002;65:155–160.
10. Xu PQ, Dang XW. Treatment of membranous Budd-Chiari syndrome:
analysis of 480 cases. Hepatobiliary Pancreat Dis Int. 2004;3:73–76.
11. Lee BB, Villavicencio L, Kim YW, Do YS, Koh KC, Lim HK, Lim JH,
Ahn KW. Primary Budd-Chiari syndrome: outcome of endovascular man-
agement for suprahepatic venous obstruction. J Vasc Surg. 2006;43:101–
108. doi: 10.1016/j.jvs.2005.09.003.
12. Meng QY, Sun NF, Wang JX, Wang RH, Liu ZX. Endovascular treatment
of Budd-Chiari syndrome. Chin Med J (Engl). 2011;124:3289–3292.
13. Yang XL, Cheng TO, Chen CR. Successful treatment by percutaneous
balloon angioplasty of Budd-Chiari syndrome caused by membranous ob-
struction of inferior vena cava: 8-year follow-up study. J Am Coll Cardiol.
1996;28:1720–1724. doi: 10.1016/S0735-1097(96)00385-3.
14. Zhang C, Fu L, Zhang G, Xu L, Shun H, Wang Z, Zhu J. Ultrasonically
guided inferior vena cava stent placement: experience in 83 cases. J Vasc
Interv Radiol. 1999;10:85–91.
15. Pelage JP, Denys A, Valla D, Sibert A, Sauvanet A, Belghiti J, Menu Y.
Budd-Chiari syndrome due to prothrombotic disorder: mid-term patency
and efficacy of endovascular stents. Eur Radiol. 2003;13:286–293. doi:
10.1007/s00330-002-1475-2.
16. Qiao T, Liu CJ, Liu C, Chen K, Zhang XB, Zu MH. Interventional endo-
vascular treatment for Budd-Chiari syndrome with long-term follow-up.
Swiss Med Wkly. 2005;135:318–326. doi: 2005/21/smw-10947.
Downloaded from http://circinterventions.ahajournals.org/ at University of Colorado on April 1, 2016
17. Zhang Q, Huang Q, Shen B, Sun J, Wang X, Liu H. Efficacy and safety of
endovascular intervention for the management of primary entire-inferior
vena cava occlusion. Cardiovasc Intervent Radiol. 2015;38:665–671. doi:
10.1007/s00270-014-0980-4.
18. Zhang QQ, Xu H, Zu MH, Gu YM, Shen B, Wei N, Xu W, Liu HT,
Wang WL, Gao ZK. Strategy and long-term outcomes of endovascular
treatment for Budd-Chiari syndrome complicated by inferior vena caval
thrombosis. Eur J Vasc Endovasc Surg. 2014;47:550–557. doi: 10.1016/j.
ejvs.2014.01.014.
19. Schoenfeld D. Partial residuals for the proportional hazards regression
model. Biometrika. 1982;69:239–241.
20. Buhe S, Miyaki T, Saito T, Sawuti A, Terayama H, Naito M, Yi SQ, Itoh
M. A study of the accessory hepatic vein to segments VI and VII with
a morphological reconsideration of the human liver. Surg Radiol Anat.
2008;30:201–207. doi: 10.1007/s00276-008-0315-8.
21. Qi X, De Stefano V, Wang J, Bai M, Yang Z, Han G, Fan D. Prevalence
of inherited antithrombin, protein C, and protein S deficiencies in portal
vein system thrombosis and Budd-Chiari syndrome: a systematic review
and meta-analysis of observational studies. J Gastroenterol Hepatol.
2013;28:432–442. doi: 10.1111/jgh.12085.
22. Xue H, Li YC, Shakya P, Palikhe M, Jha RK. The role of intravascular
intervention in the management of Budd-Chiari syndrome. Dig Dis Sci.
2010;55:2659–2663. doi: 10.1007/s10620-009-1087-7.
23. Ciesek S, Rifai K, Bahr MJ, Boozari B, Steinmann E, Helfritz FA,
Wedemeyer H, Manns MP, Lotz J, Gebel M. Membranous Budd-Chiari
syndrome in Caucasians. Scand J Gastroenterol. 2010;45:226–234. doi:
10.3109/00365520903406719.
24. Ding PX, Zhang SJ, Li Z, Fu MT, Hua ZH, Zhang WG. Long-term safety
and outcome of percutaneous transhepatic venous balloon angioplasty for
Budd-Chiari syndrome. J Gastroenterol Hepatol. 2016;31:222–228. doi:
10.1111/jgh.13025.
25. Zhang CQ, Fu LN, Xu L, Zhang GQ, Jia T, Liu JY, Qin CY, Zhu JR. Long-
term effect of stent placement in 115 patients with Budd-Chiari syndrome.
World J Gastroenterol. 2003;9:2587–2591.
26. Lin X, Fang L, Wang Y. Multiple heart injuries caused by fracture and mi-
gration of the inferior vena stent. Eur Heart J. 2013;34:625. doi: 10.1093/
eurheartj/ehs119.
27. MacNicholas R, Olliff S, Elias E, Tripathi D. An update on the diagno-
sis and management of Budd-Chiari syndrome. Expert Rev Gastroenterol
Hepatol. 2012;6:731–744. doi: 10.1586/egh.12.56.
28. Srinivas BC, Dattatreya PV, Srinivasa KH, Prabhavathi, Manjunath
CN. Inferior vena cava obstruction: long-term results of endovascu-
lar management. Indian Heart J. 2012;64:162–169. doi: 10.1016/
S0019-4832(12)60054-6.
29. Esfeh JM, Hanouneh IA, Zein NN. A 25-year-old man with very high
alkaline phosphatase. Cleve Clin J Med. 2011;78:793–800. doi: 10.3949/
ccjm.78a.10166.
Comparison of Long-Term Outcomes of Endovascular Management for Membranous and
Segmental Inferior Vena Cava Obstruction in Patients With Primary Budd −Chiari
Syndrome
Qianxin Huang, Bin Shen, Qingqiao Zhang, Hao Xu, Maoheng Zu, Yuming Gu, Ning Wei,
Yanfeng Cui and Rui Huang

Circ Cardiovasc Interv. 2016;9:

doi: 10.1161/CIRCINTERVENTIONS.115.003104
Circulation: Cardiovascular Interventions is published by the American Heart Association, 7272 Greenville
Avenue, Dallas, TX 75231
Copyright © 2016 American Heart Association, Inc. All rights reserved.
Print ISSN: 1941-7640. Online ISSN: 1941-7632

The online version of this article, along with updated information and services, is located on the
World Wide Web at:
http://circinterventions.ahajournals.org/content/9/3/e003104

Data Supplement (unedited) at:

http://circinterventions.ahajournals.org/content/suppl/2016/02/23/CIRCINTERVENTIONS.115.003104.DC1.html

Permissions: Requests for permissions to reproduce figures, tables, or portions of articles originally published
in Circulation: Cardiovascular Interventions can be obtained via RightsLink, a service of the Copyright
Clearance Center, not the Editorial Office. Once the online version of the published article for which
permission is being requested is located, click Request Permissions in the middle column of the Web page
under Services. Further information about this process is available in the Permissions and Rights Question and
Answer document.

Reprints: Information about reprints can be found online at:

http://www.lww.com/reprints

Subscriptions: Information about subscribing to Circulation: Cardiovascular Interventions is online at:

http://circinterventions.ahajournals.org//subscriptions/

Downloaded from http://circinterventions.ahajournals.org/ at University of Colorado on April 1, 2016

 SUPPLEMENTAL MATERIAL

Supplemental Table 1. Univariable and multivariable analysis for postoperative

reocclusion

Univariable Cox model Multivariable Cox Model

Variables at baseline HR 95%CI P HR 95%CI P

value value

*
Sex (male) 2.315 1.110-4.826 0.025 1.650 0.783-3.477 0.188

Age at diagnosis (per 1 years

0.990 0.957-1.025 0.585
increase)

*
IVC obstruction type (SO) 2.494 1.246-4.992 0.010 3.496 1.623-7.352 0.002

Treatment method (balloon

1.607 0.562-4.597 0.376
dilation alone)

Duration of symptom (per 1

1.003 0.999-1.007 0.153
months increase)

Abdominal fullness (present vs

1.141 0.519-2.512 0.742
absent)

Abdominal pain (present vs

0.039 0.001-3.620 0.161
absent)

Thoracoabdominal distention
0.569 0.268-1.207 0.142
(present vs absent)

Edema of lower limbs (present vs 1.229 0.591-2.555 0.582

absent)

Ascites (present vs absent) 1.196 0.570-2.512 0.636

Gastrointestinal bleeding (present

1.536 0.668-1.316 0.212
vs absent)

Lower limbs pigmentation

0.048 0.001-4.544 0.603
(present vs absent)

Leg ulcer (present vs absent) 0.681 0.162-2.865 0.600

Hepatosplenomegaly (present vs
0.774 0.312-1.918 0.580
absent)

AST (present vs absent) 1.016 0.988-1.044 0.275

ALT (per unit increase) 1.007 0.976-1.039 0.661

*
ALP (per unit increase) 1.009 1.005-1.014 0.000 1.010 1.006-1.014 0.000

*
Albumin (per unit decrease) 0.942 0.894-0.993 0.025 1.008 0.920-1.103 0.870

TBIL (per unit increase) 0.994 0.973-1.015 0.570

Creatinine (per unit increase) 0.992 0.973-1.012 0.438

BUN (per unit increase) 1.015 0.838-1.229 0.879

Prothrombin time (per unit

1.004 0.971-1.039 0.802
increase)

INR (per unit increase) 0.958 0.652-1.597 0.124

Fibrinogen (per unit increase) 1.262 0.867-1.837 0.224

WBC count (per unit increase) 1.015 0.854-1.207 0.863

*
Hemoglobin (per unit decrease) 0.985 0.971-0.999 0.042 0.984 0.964-1.005 0.127
Platelet count (per unit
1.003 0.999-1.007 0.114
increase)

Child-Pugh score (per unit

0.950 0.694-1.301 0.749
increase)

Rotterdam score (per unit

1.224 0.632-2.373 0.549
increase)

Clichy score (per unit increase) 0.943 0.581-1.533 0.814

New Clichy score (per unit

0.857 0.604-1.215 0.385
increase)

MELD score (per unit increase) 1.039 0.923-1.169 0.529

*Variables included in the multivariable analysis.

HR: hazard ratio; CI: confidence interval; IVC: inferior vena cava; SO: segmental

occlusion; ALT: alanine aminotransferase; AST: aspartate aminotransferase; ALP:

alkaline phosphatase; TBIL: total bilirubin; BUN: blood urea nitrogen; WBC: white

blood cell; MELD: Model for End-Stage Liver Disease