Keywords direct intrahepatic portosystemic shunt - Budd-Chiari syndrome - thrombolysis - catheter-directed
thrombolysis
Introduction
Budd-Chiari syndrome (BCS) occurs due to obstruction of the hepatic venous outflow
tract that can occur at the level of hepatic veins, ostial opening of hepatic vein
into inferior vena cava (IVC), or part of IVC above the level.[1 ] Direct intrahepatic portosystemic shunt (DIPS) is an endovascular side-to-side portocaval
shunt created from IVC to portal vein through liver parenchyma.[2 ] This ameliorates the portal hypertension associated with BCS. Although anatomic
recanalization of hepatic veins is always the initial aim in BCS, however, in absence
of recanalizable hepatic vein or failure of previously placed hepatic vein stent,
DIPS remains the next alternative therapy.[3 ]
[4 ] BCS is increasingly being diagnosed with high clinical suspicion leading to increased
frequency of patients undergoing DIPS creation.[5 ]
[6 ] The procoagulant state in BCS has been identified in 88% by European studies and
85% by Indian studies, posing a risk of stent thrombosis and requiring anticoagulation.
Primary and secondary patency rates of DIPS at 1 year range from 66 to 100% and occlusion
of DIPS stent is associated with reappearance of symptoms of portal hypertension.[7 ] Failure to recanalize these occluded stent leaves the patients with only curative
option of liver transplant. So, recanalizing occluded stent helps in averting an early
transplant.
Early identification and recanalization of occluded DIPS stent are of utmost importance
considering the risk life-threatening variceal hemorrhage. The purpose of this study
is to describe an innovative way of performing thrombolysis in thrombotically occluded
DIPS stent and compare it with the conventional method of continuous catheter-directed
thrombolysis (CDT).
Materials and Methods
Patients and Group
This retrospective study was performed after obtaining institutional review board
approval (IEC/2021/86/MA15) and the need for informed consent was waived (ClinicalTrials.gov
ID: NCT05117684).
Patients of BCS with DIPS occlusion were identified during their scheduled follow-up
(every 3 months in for the 1st year after DIPS and thereafter every 6 months) or their
visit to the hospital for new onset ascites and/or upper GI bleed. All consecutive
patients with thrombotic occlusion of DIPS stent underwent revision of DIPS stent
(from January 2015 to February 2021) if cannulation of DIPS stent was possible from
jugular approach using various hardware and technique. The data of all patients who
underwent DIPS revision was collected from the Hospital Information System (HIS).
Procedural details were retrieved from the electronic database and the patients were
divided into two groups, based on the technique of thrombolysis used for DIPS revision.
Patients were treated with conventional method of continuous CDT-group 1, or balloon
occluded thrombolysis (BOT)-group 2 within 1 to 2 weeks of diagnosis. BOT was conceptualized
and started in 2018, with first case being done in January and thereafter all cases
of DIPS occlusion underwent BOT for recanalization.
Laboratory values including liver function tests (LFT), kidney function tests (KFT),
prothrombin time (PT)/ international normalized ratio (INR), albumin-bilirubin score
(ALBI), and procoagulant workup at the time of patient presentation and at 1 and 6-month
follow-up were collected from the HIS. Flow velocities using Doppler ultrasound (mid-shunt
velocity, shunt velocity at IVC end, and main portal vein velocity) were recorded
at the time of DIPS creation and post-DIPS revision. Portosystemic gradient was measured
and recorded at the time of DIPS creation and during DIPS revision (pre- and postprocedure).
Stent patency was checked at 6-month follow-up using Doppler ultrasound and flow within
the stent and flow velocities were recorded. All the patients received oral anticoagulant
(Acitrom 2–5 mg), with individual dose titration in order to maintain the INR value
between 2.0 and 3.0.
Procedure was done under conscious sedation. All the procedures were performed by
an interventional radiologist (IR), having at least 10 years of experience. The thrombosed
stent was cannulated using 5F multipurpose or Cobra catheter (Cook Medical, Bloomington,
IN) via transjugular route. If cannulation failed despite using both catheters, then
angle stiffening cannula from RUPS set (Cook Medical) was used to direct the catheter
within the thrombosed stent. Heparin (2500-3000 IU) was administered via sheath at
the beginning of procedure. Urokinase was used as thrombolytic agent in either technique.
Conventional Catheter-Directed Thrombolysis
After cannulating the thrombosed stent, the 5F multipurpose or cobra catheter was
advanced to the portal end and a bolus dose of 100,000 IU of Urokinase was injected
all throughout the thrombi in the stent while gradually withdrawing the catheter.
After 20 minutes of injection of thrombolytic agent, balloon angioplasty was performed
(using 8 or 10 mm noncompliant balloon, Mustang, Boston Scientific) to macerate the
thrombi and later thrombectomy was performed using manual suction and aspiration of
thrombi using 10F long sheath. This was followed by sweeping a compliant balloon (Swan-Ganz
catheter, Edwards Life Sciences) across the DIPS stent to aid detachment of fragmented
thrombi adhered to the wall of the stent. These maneuvers though helped in debulking
the thrombus load and the clearance of residual thrombus was performed by continuous
thrombolytic infusion (using 40,000-60,000 IU/hour of Urokinase) using multi-sideport
infusion catheter (Cook Medical) for 12 to 48 hours. Venography was repeated every
12 hours to assess the response and stopped once complete luminal recanalization (wall
to wall patency within the stent) was achieved. Infusion of thrombolytic agent was
continued for maximum of 48 hours and stopped thereafter irrespective of the response.
Balloon-Occluded Thrombolysis
In BOT, after cannulating the thrombosed stent angioplasty ([Fig. 1 ]) was performed using an 8 or 10 mm noncompliant angioplasty balloon (Mustang, Boston
Scientific). Thereafter, a 10F long sheath was advanced up to the DIPS stent and a
Swan-Ganz catheter was placed inside the caval end of the DIPS stent and the balloon
was inflated to occlude the stent lumen at the caval end. Thereafter, 100,000 IU of
urokinase mixed with iodinated contrast media (to help visualization of urokinase)
was administered through the Swan-Ganz catheter to get inside the organized thrombi
in the stent. The aim was to prevent the flow of thrombolytic agent into the systemic
circulation and create a static column of thrombolytic agent within the stent, leading
to a longer contact between thrombolytic agent and thrombus that would help in achieving
a better thrombi clearance in a short duration with a limited dose of thrombolytic
agent. After 15 to 20 minutes, the thrombolytic solution within the stent was aspirated.
It was followed by a repeat angioplasty using an 8 or 10 mm noncompliant balloon,
manual thrombo-suction using 10F sheath and a balloon sweep across the stent using
Swan-Ganz catheter. In case of any residual thrombi, BOT and angioplasty were repeated
one to two times (till the total dose of urokinase did not exceed 200000 IU) to achieve
complete thrombus clearance.
Fig. 1 Schematic representation of balloon-occluded thrombolysis technique for thrombotically
occluded direct-intrahepatic portosystemic-shunt (DIPS) stent (A –D ). (A ) Thrombotically-occluded DIPS stent with thrombotic occlusion represented in gray
color. (B ) Balloon angioplasty with a noncompliant balloon (C ) Balloon-occluded thrombolysis performed after inflating a compliant balloon at the
inferior vena cava end of the DIPS stent and injecting thrombolytic agent to create
a static column (balloon-occluded thrombolysis and balloon angioplasty may be performed
couple of times alternatively to achieve complete recanalization) (D ) Recanalized DIPS stent with wall-to-wall flow.
Restenting within the existing DIPS stent was done in case of incomplete thrombi clearance
by either method of thrombolysis using 10 mm covered stent (Fluency BARD) in the liver
parenchymal track with or without bare stent in the portal vein (wall stent Boston).
The length of stent was either same or 2 cm longer than the previously placed covered
stent. End-point of procedure was to achieve complete luminal/wall-to-wall contrast
opacification of the stent, with no filling defects, no opacification of peripheral
portal vein branches, and/or varices on venogram.
Definition of Outcomes
Successful recanalization: Restoration of normal flow within the DIPS stent (postprocedure venogram and ultrasound
Doppler showing wall-to-wall contrast opacification with no filling defects and no
opacification of peripheral portal vein branches and/or varices) after thrombolysis
(CDT or BOT), without the need for restenting.
Besides technical success, the groups were compared for mean total dose of thrombolytic
agent administered, duration of hospital stay, and primary-assisted patency rate at
1 and 6-month follow-up. The data was analyzed for all complications associated with
the procedures as per the Society of Interventional Radiology (SIR) guidelines.
Statistical Analysis
All analyses were carried out on SPSS (version 23.0, SPSS Inc., Chicago, Illinois,
United States). Qualitative data were expressed as proportion or percentages (%).
Quantitative data were expressed as mean ± SD, median, and ranges depending on data
distribution. Continuous variables were compared parametrically using Student's t -test or nonparametrically using Mann–Whitney U test. p -Value less than 0.05 was taken as statistically significant in all cases. The Kaplan–Meier
survival analysis method was used for calculating overall and transplant-free survival.
Results
A total of 42 patients developed thrombotic occlusion of DIPS stent and 33 of them
were included for analysis. No color flow was seen on ultrasound Doppler in all patients
at the time of diagnosis. As per the treatment received, the patients were grouped
in CDT or BOT group ([Fig. 2 ]).
Fig. 2 Flowchart showing patient selection, inclusion, and exclusion criteria and group
stratification. BOT, balloon-occluded-thrombolysis; CDT, catheter-directed thrombolysis;
DIPS, direct intrahepatic portosystemic shunt.
Demographic, Clinical, and Laboratory Data
There were 12 patients in CDT group (11 males and 1 female) and 21 in BOT Group (13
males and 8 females). Most common clinical presentation was ascites and abdominal
pain followed by variceal bleed in either groups. The patient characteristics of both
groups at the time of presentation with DIPS occlusion are outlined in [Table 1 ]. The groups were found to be similar with respect to age, sex, PT/INR, LFT, and
KFT. The mean velocities on Doppler ultrasound at the time of DIPS creation did not
show any significant difference between the two groups. The mean portosystemic gradient
at the time of DIPS creation and DIPS occlusion were also compared with no significant
difference found between the two groups ([Supplementary Tables 1 ] and [2 ]). Recanalization of DIPS was performed within 1 to 2 weeks of diagnosis.
Table 1
Demographic, clinical, and laboratory findings of study patients at the time of presentation
with thrombotically-occluded DIPS stent
Characteristics
Values, n (%) or median (range)
Group 1 (n =12)
Group 2 (n =21)
p -Value
Age at diagnosis (years)
32.5 ± 16.8
34.7 ± 15.5
0.708
Sex (male/female)
12/1
14/8
0.065
Etiology
Polycythemia rubra vera
Myeloproliferative disorders
Factor V Leiden gene mutation
Methylene tetrahydrofolate reductase gene mutation
Prothrombin gene mutation
JAK
Not identified
Not evaluated
1
3
1
1
1
1
2
2
2
3
2
3
0
1
5
5
Symptoms at diagnosis
Ascites
Abdominal pain
Upper GI bleed
Nonsymptomatic
11 (92%)
6 (50%)
4 (33.3%)
1 (8.3%)
17 (81%)
10 (47.6%)
5 (23.8%)
3 (14.3%)
Biochemical parameters
Serum bilirubin (mg/dL)
Aspartate transaminase (IU/L)
Alanine transaminase (IU/L)
Serum albumin (g/dL)
PT-INR
Platelets (lacs/mm3 )
Serum creatinine (mg/dL)
ALBI score
2.4 ± 1.0
50 (IQR: 46–68)
40 (IQR: 31.5–55)
2.9 ± 0.4
2.2 ± 0.9
215 ± 114.4
0.8 ± 0.2
1.5 ± 0.4
2.1 ± 0.8
44 (IQR: 34.7–57.7)
38 (IQR: 25.5–46.7)
2.8 ± 0.6
2.7 ± 1.5
150 ± 86.7
0.8 ± 0.4
1.4 ± 0.5
0.370
0.408
0.400
0.712
0.533
0.232
0.796
0.903
Abbreviations: ALBI, albumin-bilirubin score; DIPS, direct intrahepatic portosystemic
shunt; GI, gastrointestinal; IQR, interquartile range; PT-INR, prothrombin time-international
normalized ratio.
Management and Procedure-Related Complications
CDT group (
n
= 12) : Eight patients (66.7%) showed complete recanalization of blocked DIPS stent with
continuous CDT, while remaining four patients (33.3%) required placement of another
10 mm covered stent within the DIPS due to incomplete recanalization.
The average dose of urokinase administered in patients of this group was 8.3 ± 2.9 × 105 IU. One patient had developed large puncture site hematoma in neck, which was managed
with manual compression (minor complication according to Society of Intervention Radiology
guidelines). The mean duration of hospital stay was 3.5 ± 1.0 days. At 1-month follow-up,
3 of 8 patients developed stent rethrombosis. However, the 6-month follow-up did not
reveal any additional case of DIPS block ([Table 2 ]).
Table 2
Comparison of outcome in both groups
Group I
Group II
p -Value
Length of hospital stay (days)
3.5 ± 1.0
1.8 ± 0.7
0.001
Total dose of urokinase (105 IU)
8.3 ± 2.9
2.2 ± 0.4
0.001
Complication rate
1/12
0/21
–
Technical success
66.7% (8/12)
81% (17/21)
0.420
Primary-assisted patency rate at 1 month
62.5% (5/8)
100% (17/17)
0.024
Primary-assisted patency rate at 6 months
100% (5/5)
100% (17/17)
–
BOT group (
n
= 21) : Seventeen patients (81%) had complete restoration of wall-to-wall flow after BOT
performed two to three times using a maximum of 200000 IU of urokinase ([Fig. 3 ]), while four patients (19%) required placement of another 10 mm covered stent within
the DIPS. The mean dose of urokinase administered in patients of this group was 2.2 ± 0.4 × 105 IU. No procedure-related major or minor adverse events were reported in this group.
Fig. 3 Thrombotic occlusion of direct intrahepatic portosystemic shunt (DIPS) stent, being
recanalized using balloon-occluded thrombolysis. . (A ) DIPS venogram shows nonopacification of the stent. (B ) Swan-Ganz balloon inflated at inferior vena cava end of stent with a column of contrast-urokinase
mixture within proximal part of stent. (C ) Swan-Ganz balloon inflated at mid-part of stent with a column of contrast-urokinase
mixture within distal part of stent. (D ) Postthrombolysis venogram normal wall to wall opacification of DIPS stent.
Unlike CDT group, where all the patients required 2 to 5 days of hospital stay, 7
patients (33%) in BOT group could be managed in the day-care and got discharged on
the same day of the procedure. The mean duration of hospital stay in BOT group was
significantly lower (1.8 ± 0.7 days). There was no recurrence of stent block in any
of the patients at 1- and 6-month follow-up ([Table 2 ]).
Comparison of Outcome between Two Groups
Success rates of 66.7 and 81% were achieved in CDT group and BOT group, respectively,
with no statistically significant difference (p -value = 0.420). However, a significant difference was noted in the mean urokinase
dose administered (p -value = 0.001) and length of hospital stay (p value = 0.001) between the two groups. The primary-assisted patency at 1-month follow-up
was also significantly higher in BOT group compared to CDT group (p -value = 0.024). The follow-up Doppler at 6 months did not differ between the two
groups ([Table 2 ]).
The mean Doppler velocities and portosystemic gradients were found to be within the
normal limits in either group after DIPS revision. No significant difference was found
between the groups in the postrevision Doppler velocities and portosystemic gradients
([Supplementary Tables 3 ] and [4 ]).
Similarly, time taken for resolution of ascites and abdominal pain in both the groups
after DIPS recanalization did not show significant statistical difference (p -values = 0.07 and 0.385, respectively; [Table 3 ]).
Table 3
Outcomes following DIPS recanalization
Symptom/Sign
Result
p -Value
Time for ascites resolution (weeks)
Group I
Group II
2.2 ± 0.8
2.7 ± 0.5
0.07
Time for abdominal pain resolution
Group I
Group II
1.75 ± 0.5
1.5 ± 0.5
0.385
Abbreviation: DIPS, direct intrahepatic portosystemic shunt.
Among the biochemical markers, serum albumin levels and ALBI score showed statistically
significant improvement at 1 month in both the groups, while serum bilirubin, platelet
counts, and serum creatinine showed significant improvement only in BOT group ([Table 4 ]).
Table 4
Comparison between pre- and postprocedure laboratory values in both groups
Variables
Pre-DIPS revision
Post-DIPS revision (at 1 month)
p -Value
PT-INR
Group I
Group II
2.2 ± 0.9
2.7 ± 1.5
2.5 ± 0.7
2.7 ± 0.6
0.327
0.429
Serum total bilirubin (mg/dL)
Group I
Group II
2.4 ± 1.0
2.0 ± 0.8
2.1 ± 0.9
1.5 ± 0.7
0.066
0.041
Aspartate aminotransferase (IU/L)
Group I
Group II
50 (IQR: 46–68)
44 (IQR: 34.7–57.7)
49.5 (IQR: 42–70)
47.5 (IQR: 38.5–53.8)
0.116
0.062
Alanine aminotransferase (IU/L)
Group I
Group II
40 (31.5–55)
38 (IQR: 25.5-46.7)
42.5 (IQR: 34.2–53)
43.5 (IQR: 33.2–51.5)
0.674
0.953
Serum albumin (g/dL)
Group I
Group II
2.9 ± 0.4
2.8 ± 0.6
3.5 ± 0.4
3.3 ± 0.8
0.011
0.003
Platelet count (lakhs/mm3 )
Group I
Group II
215 ± 114
150 + 86.7
231 ± 161
192.6 ± 99
0.799
0.001
Serum creatinine (mg/dL)
Group I
Group II
0.8 ± 0.2
0.8 + 0.4
0.7 ± 0.4
0.6 ± 0.3
0.575
0.026
ALBI score
Group I
Group II
1.5 ± 0.4
1.4 ± 0.5
2.1 ± 0.3
1.9 ± 0.7
0.036
0.007
Abbreviations: ALBI, albumin-bilirubin score; DIPS, direct intrahepatic portosystemic
shunt; IQR, interquartile range; PT-INR, prothrombin time-international normalized
ratio.
Discussion
The study population had a mean age of 32.5 years in group I and 34.7 years in group
II. The most common clinical presentation was in the form of ascites and abdominal
pain followed by variceal bleed in either groups, whereas 20% patients were asymptomatic.
Previous literature also showed that ascites is the most common symptom in BCS occurring
in 62 to 83% of patients, while 10% of BCS patients presented with variceal bleed
and another 10% patients were asymptomatic.[8 ] In the present study, three BCS patients had polycythemia rubra vera (1 in group
1 and 2 in group 2), six had myeloproliferative disorders (3 in group 1 and 3 in group
2), inherited thrombophilia (factor 5 mutation) was present in 3 patients (1 in group
1 and 2 in group 2), prothrombin mutation was present in 1 patients, and 14 patients
had undiagnosed prothrombotic state. Previous studies have shown incidence of myeloproliferative
disorders in 16 to 62% of patients with BCS and among various disorders, polycythemia
vera was the most common type. Pregnancy and puerperium were seen more commonly in
Asia, while oral contraceptive pills use was noted in 30% of patients with BCS in
Western studies. Antiphospholipid antibodies were found in 18 to 25% BCS patients.[9 ]
Although DIPS has become an integral part in BCS management, shunt thrombosis poses
a major concern. The patients with blocked DIPS stent present with reappearance of
ascites and/or varices on upper gastrointestinal endoscopy.[10 ] Revision of thrombosed stents can be easily accomplished in most cases using CDT.
However, prolonged administration and higher dose of thrombolytic agent due to the
reduced contact time between the thrombus and thrombolytic agent as the thrombolytic
agent quickly escape into the circulation after being released by the multi-side hole
catheter. Besides, the systemic dispersion of thrombolytic agent in high doses and
for longer duration may lead to hemorrhagic complications.[11 ] The use of BOT significantly increases the contact time between the thrombi and
the thrombolytic agent and reduces the overall dose of thrombolytic agent. In BOT,
static column of thrombolytic drug enables the administration of topical high-concentration
lytic agents thar is later aspirated, thus avoiding systemic effects of the thrombolytic
agent ([Figs. 1C ], [3B ], and [3C ]). CDT requires longer duration of thrombolytic infusion leading to high cumulative
doses of urokinase administered. This is evident from the significant difference in
average dose of urokinase administered in either groups.
The duration of hospital stays in CDT group ranged from 2 to 5 days. Patients treated
with CDT require ICU care during continuous infusion of thrombolytic agent. Periodic
digital subtraction angiography acquisitions are needed to look for the resolution
of thrombus that leads to increased patient discomfort and additional radiation dose
to the patient. In BOT hospital stay ranged from 1 day to a maximum of 3 days and
the procedure was completed in a single session. Seven patients (33.3%) were treated
in a day-care facility with postprocedure observation period of about 4 hours, and
were discharged on the same day. BOT provides similar results, with a lower dose of
thrombolytic agent and a short hospital stay. Thus, it has the potential to save time
and resource and may reduce the cost of the treatment. In this study, success rates
of 66.7 and 81% were achieved in CDT group and BOT group, respectively. Previous literature
with CDT by Li et al, in BCS with occluded shunt showed that only 70% patients were
able to achieve thrombolysis at 1-month follow-up with thrombus length and preopening
IVC proportion of thrombus independently favoring thrombolysis.[12 ] Also, studies show that patients of cirrhosis and coagulation disorder have high
risk of bleeding complication, thereby requiring lower dose of anticoagulation. As
BOT is not a frequently used therapy, there is no data available. But BOT therapy
gives advantage in patients with bleeding diathesis and Budd-Chiari-related portal
hypertension with less risk of complications and good favorable outcomes.
Bleeding complications associated with thrombolytic therapy can lead to extended hospital
stay, multiple transfusions, stroke, and death.[13 ]
[14 ] In this study, one such complication was documented in the form of expanding neck
hematoma in CDT group. However, no hemorrhagic complication was noted in BOT group.
BOT group showed a higher technical success when compared to CDT group, that is, 81
versus 66.7%. However, this did not turn out to be statistically significant. Moreover,
the two groups did not behave differently in terms of improvement in ultrasound Doppler
velocities, portosystemic gradient, and clinical resolution post-DIPS revision. Thus,
it can be inferred that although the technical success is similar in both groups,
yet the morbidity associated with higher thrombolytic dosage and longer hospital stay
in CDT group certainly places BOT group at an advantage.
Follow-up at 1 month revealed three cases of recurrence of stent thrombosis on sonography
in CDT group, whereas there were no such findings in BOT group. It was postulated
that this recurrence could be due to microthrombi persisting along the stent wall
(not visualized on angiograms) in CDT group. These remnant microthrombi may serve
as a nidus for recurrence intrastent thrombus.
The study had few limitations. The sample size is small that is due to rarity if the
disease. The available follow-up was only for 6 months. Patient factors prothrombotic
state may have some role to play in stent thrombosis/rethrombosis.
With the available results, it may be concluded that BOT is a quick and effective
treatment for thrombosed DIPS stent requiring short hospital stay and less dose of
thrombolytic agent as compared to conventional CDT. This novel technique (BOT) has
a potential to expedite thrombolysis and reduce thrombolysis-related bleeding complications
in thrombosed DIPS stent.
