Keywords
embolization - N-BCA - Lipiodol
Introduction
Vascular embolization is a minimally invasive image-guided procedure to obtain a temporary
or permanent and partial or complete occlusion of blood vessels for a variety of clinical
indications, including active bleeding, varices, vascular malformations, and benign
and malignant tumors.[1]
[2]
[3]
[4] It is also used preoperatively for resectable tumors to reduce operative bleeding
or to prepare organs for resection.[5]
[6] Vascular embolization has the advantage of a lower rate of complications and superior
outcomes compared to open surgery.[7] Depending on the clinical scenario, different embolic materials could be used, including
mechanical occlusion devices (such as coils) for large vessel embolization, particulates
for small vessel embolization, and liquid/gel embolic materials (such as alcohol and
N-butyl-2-cyanoacrylate [N-BCA]) for both large and small vessel embolization.[8]
Ethyl esters of iodized fatty acids of poppy seed oil (Lipiodol Ultra Fluid, Guerbet,
Aulnay-sous-Bois, France) is a radiopaque contrast material extensively used in interventional
radiology. Lipiodol contains 480 mg/mL of iodine, has a viscosity of 34 to 70 mPa·s
at 20°C, and is slightly denser than water (1.28 g/cm3 at 20°C).[9] Lipiodol mixed with N-BCA has been used for vascular embolization since the 1980s.[10]
[11] Lipiodol increases the polymerization time of the mixture, hence limiting risks
of clogging of the catheters; allows viscosity adjustment of the mixture to the lesion
angioarchitecture and blood flow; and most of all allows the fluoroscopic visualization
of the embolic mixture due to iodine radiopacity. The ratio of Lipiodol to N-BCA is
empirically decided by the interventional radiologist depending on blood vessel diameter,
blood flow velocity, and the distance between the microcatheter tip and the target.[12] Despite its advantages, vascular embolization could be associated with some complications,
the most frequent being pain and postembolization syndrome (PES).[13]
[14] More rare but serious complications could occur, including pulmonary embolism and
hemorrhage.[15]
[16]
[17] This study aimed to assess the safety and efficacy of the combination of Lipiodol
and N-BCA for vascular embolization in Indian patients with various clinical conditions
and to meet the postmarketing obligation of the Drug Controller General of India (DCGI).
Materials and Methods
Study Design and Population
This is a prospective, multicenter, open-label, single-arm, postmarketing study conducted
on Indian patients requiring vascular embolization as part of therapeutic/palliative
strategy for their disease from November 2017 to May 2021. The study was approved
by the independent ethics committee of each site and authorized by the DCGI. This
study was performed in compliance with the Declaration of Helsinki and with the Indian
Good Clinical Practice guidelines. All patients provided informed consent. The study
was registered on ClinicalTrials.gov (NCT02625389).
All the patients included were 18 years old or above (age range: 18–69 years) and
presented with vascular lesions/anomalies, such as vascular malformations (arterial,
venous, arteriovenous, fistula) or tumors (benign such as angiomyolipoma, uterine
fibroid, hemangioma or malignant such as angiosarcoma) eligible for vascular embolization
using selective transarterial catheterization using the Lipiodol and N-BCA mixture,
and were not treated previously with this approach. Patients presenting with known
contraindications to vascular embolization (i.e., severe coagulation disorders, and
infectious syndrome) or on beta-blockers, metformin, interleukin II, or anticoagulant
therapy were excluded from the study. Those with acutely bleeding lesions were also
excluded from the study.
A first embolization procedure was performed on the same day or within 7 days after
enrolment, and a second optional embolization procedure was performed, if needed,
within 30 days after the first procedure. A safety follow-up was performed up to 30
days after the last procedure. The mixture was injected into the lesion after super-selective
cannulation of arterial feeders and at a ratio of 1:4 to 4:1 (Lipiodol to N-BCA) as
decided by the investigator. The total volume of Lipiodol used during embolization
depended on lesion size but did not exceed 15 mL.
Safety Evaluation
The primary outcome of this study was safety assessment of the Lipiodol and N-BCA
mixture assessed by calculating the rate of patients experiencing adverse drugs reactions
(ADRs) during the procedure after administration of the mixture (ADRs related to the
mixture). As a secondary safety outcome measure, the rate of adverse events (AEs),
regardless of a causal relationship, was also collected during the study from the
time of the first administration of the mixture to the end of the follow-up period.
Based on their severity, the AEs were categorized into mild (symptom awareness but
without significant interference in daily activities or clinical consequences), moderate
(disruption of daily activities or had clinical consequences), or severe (inability
to carry out daily activities and/or AE had definite clinical consequences) and further
translated as per the Society of Interventional Radiology (SIR) classification as
minor (A–B) or major (C–F) complications.[18] Additionally, the seriousness of these AEs was assessed.[19]
Efficacy Evaluation
Based upon the targeted volume of vascular occlusion (percentage of the lesion expected
to be embolized safely during the embolization procedure as decided by the interventional
radiologist based on the pre-embolization angiogram, the clinical situation, and local
practice), all the lesions planned for embolization were grouped into four groups
(group 1: target embolization <50% of the lesion; group 2: target embolization 50–75%
of the lesion; group 3: target embolization 75–99% of the lesion; and group 4: target
embolization 100% of the lesion).
Postembolization, based on the actual vascular occlusion achieved (percentage volume
of the lesion that was embolized), lesions were again regrouped into four groups (group
1: <50% of the lesion got embolized; group 2: 50–75% of the lesion got embolized;
group 3: 75–99% of the lesion got embolized; and group 4: 100% of the lesion got embolized).
To assess the efficacy of the mixture, consistency between the targeted and actual
vascular occlusion was evaluated by comparing the pre-embolization target volume and
actual vascular occlusion volumes in each group.
Statistical Methods
The sample size was determined based on the specified requirement of 125 participants
set by the DCGI. Only descriptive statistics were used to report all data collected
in the study. Quantitative variables were summarized in tables including sample sizes,
means, and standard deviations (SDs). Qualitative variables were described in terms
of percentages of the number of patients or lesions considered.
Results
Overall, 132 patients were included in 16 centers, and 8 patients were prematurely
discontinued from the study due to either withdrawal of consent (4 patients) or cancellation
of the procedure (4 patients). In all, 124 patients received the mixture for a first
embolization procedure, and 12 underwent a second embolization procedure. All 124
patients who received the mixture completed the study ([Fig. 1]). The mean (SD) age of the patients who received the mixture was 34.2 (±13.6) years,
and 55.6% of them were females ([Table 1]).
Fig. 1 Flowchart of patients undergoing vascular embolization.
Table 1
Demographic characteristics of patients
|
Parameter
|
Total (N = 124)
|
|
Age (y)
|
|
Mean (SD)
|
34.2 (13.6)
|
|
Median
|
31
|
|
Minimum; maximum
|
18; 69
|
|
Gender
|
|
Male
|
55 (44.4%)
|
|
Female
|
69 (55.6%)
|
|
Weight (kg)
|
|
Mean (SD)
|
56.1 (10.5)
|
|
Median
|
55
|
|
Minimum; maximum
|
34; 86
|
|
Height (cm)
|
|
Mean (SD)
|
159.9 (10.0)
|
|
Median
|
160
|
|
Minimum; maximum
|
131; 183
|
|
Body mass index (kg/m2)
|
|
Mean (SD)
|
21.93 (3.49)
|
|
Median
|
21.7
|
|
Minimum; maximum
|
15.1; 33.3
|
Abbreviation: SD, standard deviation.
Safety Results
Primary criterion: No AEs and ADRs were reported during the embolization procedure when the patient
was in the procedure room.
Secondary criteria: Overall, 23 patients (18.5%) reported a total of 42 AEs after leaving the procedure
room and up to 30 days from the last procedure. Sixteen patients (12.9%) had 26 AEs
related to the embolization procedure ([Table 2]). An AE-targeted medical therapy was needed for 17 patients (73.9%). Two patients
needed AE-targeted surgical therapy (cannula site changed in one patient, ventriculoperitoneal
shunt and external ventricular drain in the second patient). The most frequently reported
AEs were PES in six patients (4.8%), injection site pain in three patients (2.4%),
and pyrexia and skin necrosis in two patients each (1.6%). All except pyrexia were
considered related to the embolization procedure. All other AEs were reported in one
patient ([Table 2]).
Table 2
Adverse events related to the embolization procedure
|
Patients (N = 124)
|
|
At least one adverse event
|
16 (12.9%)
|
|
Postembolization syndrome
|
6 (4.8%)
|
|
Injection site pain
|
3 (2.4%)
|
|
Skin necrosis
|
2 (1.6%)
|
|
Catheter site swelling
|
1 (0.8%)
|
|
Injection site swelling
|
1 (0.8%)
|
|
Lower abdominal pain
|
1 (0.8%)
|
|
Dysphagia
|
1 (0.8%)
|
|
Burning sensation
|
1 (0.8%)
|
|
Cerebellar infarction
|
1 (0.8%)
|
|
Dyskinesia
|
1 (0.8%)
|
|
Hydrocephalus
|
1 (0.8%)
|
|
Blisters
|
1 (0.8%)
|
|
Back pain
|
1 (0.8%)
|
|
Mastication disorder
|
1 (0.8%)
|
|
Pain in extremity
|
1 (0.8%)
|
|
Hypertension
|
1 (0.8%)
|
|
Nontarget organ ischemia
|
1 (0.8%)
|
|
Vertigo
|
1 (0.8%)
|
As per the AE classification by the SIR, most AEs were graded as minor A complications
(10/42, 23.8%) or minor B complications (23/42, 54.8%). Six AEs (7.1%) were graded
as major C complications, and three AEs were reported as major D complications (3/42,
4.8%).
Only one patient (0.8%) with arteriovenous malformation (AVM) experienced two nonserious
ADRs: pregangrenous changes in the skin over embolized lesions with blister formation.
Both occurred 6 days after the procedure and were of mild intensity. They required
an ADR-targeted medication and resolved without sequelae (minor B complications).
In two patients, three nonfatal serious AEs (SAEs) were reported, considered related
to the embolization procedure and not related to the mixture. Out of three SAEs, two
life-threatening SAEs (hydrocephalus and a cerebellar infarction) leading to prolongation
of hospitalization were noted 2 days after the procedure in a 56-year-old patient
treated for a posterior fossa AVM (major D complications). The patient developed severe
and serious cerebellar infarction that was visible on magnetic resonance imaging,
and which happened due to the reflux of the mixture into the posterior cerebellar
artery. The hydrocephalus and the neurological and functional consequences of the
cerebellar infarction resolved within 62 days. The third SAE leading to prolongation
of hospitalization was noted in an 18-year-old patient being treated for an AVM of
the upper lip. While injecting the mixture into the superior labial artery, reflux
of the mixture occurred into an inferior labial artery, leading to local skin necrosis
in the nontarget area (major D complication). Signs and symptoms of this SAE were
noted 7 days after the procedure and resolved in 46 days.
Efficacy Results
One hundred and thirty-eight lesions in 124 patients were treated by embolization
on one occasion. One hundred and fourteen patients were treated for one lesion, six
patients for two lesions, and four patients for three lesions. Most of these lesions
were AVMs (66.7%) or other vascular neoplasms (9.4%; [Table 3]). Most of these lesions were localized in the skin and subcutaneous tissues (43.5%),
bone and other internal soft tissues (20.3%), urinary system (9.4%), liver (8.0%),
or reproductive system (7.2%; [Table 3]). Endocryl (n-butyl-2-cyanoacrylate, Samarth Life Sciences, Mumbai, India) was used
for 124 lesions (89.9%), while Histoacryl (n-butyl-2-cyanoacrylate, B. Braun, AG,
Melsungen, Germany) was used for embolization of 12 lesions (8.7%). The brand name
of the N-BCA glue was not specified for two lesions. In one patient, in addition to
the mixture, another embolic material (i.e., implant) was used during the first embolization
procedure. The most often used Lipiodol/N-BCA ratios for embolization were 4:1 (38.5%),
3:1 (14.6%), and 1:1 (14.2%). The mean (SD) duration of the procedure was 54 (±38)
minutes.
Table 3
Type of lesions to be treated
|
Patients (N = 124)
|
Lesions (N = 138)
|
|
Type of lesion
|
|
Arteriovenous malformation
|
84 (67.7%)
|
92 (66.7%)
|
|
Other vascular neoplasm
|
11 (8.9%)
|
13 (9.4%)
|
|
Arteriovenous fistula
|
4 (3.2%)
|
4 (2.9%)
|
|
Infantile hemangioma
|
3 (2.4%)
|
3 (2.2%)
|
|
Aneurysm
|
2 (1.6%)
|
2 (1.4%)
|
|
Noninvoluting congenital hemangioma
|
2 (1.6%)
|
2 (1.4%)
|
|
Angiosarcoma
|
1 (0.8%)
|
1 (0.7%)
|
|
Arterial malformation
|
1 (0.8%)
|
1 (0.7%)
|
|
Venous malformation
|
1 (0.8%)
|
1 (0.7%)
|
|
Other[a]
|
15 (12.1%)
|
19 (13.8%)
|
|
Location of lesion[b]
|
|
Skin and subcutaneous tissues
|
55 (44.4%)
|
60 (43.5%)
|
|
Bone and other internal soft tissues
|
27 (21.8%)
|
28 (20.3%)
|
|
Urinary system
|
11 (8.9%)
|
13 (9.4%)
|
|
Liver
|
8 (6.5%)
|
11 (8.0%)
|
|
Reproductive system
|
10 (8.1%)
|
10 (7.2%)
|
|
Lung
|
4 (3.2%)
|
5 (3.6%)
|
|
Intracranial intra-axial tissues
|
2 (1.6%)
|
2 (1.4%)
|
|
Spleen
|
2 (1.6%)
|
2 (1.4%)
|
|
Spinal cord
|
1 (0.8%)
|
1 (0.7%)
|
|
Stomach
|
1 (0.8%)
|
1 (0.7%)
|
|
Other[c]
|
5 (4.0%)
|
5 (3.6%)
|
a Including liver or neck hemangioma, uterine fibroid, and hypertrophied bronchial
arteries.
b Sum of locations is more than the number of patients as few patients were having
lesions at multiple locations.
c Including the left branchial and left axillary artery, occipital artery, right lateral
chest wall, gluteal region, scrotum, left groin, and pelvis.
The actual vascular occlusion percentage was available only for 135 of 138 lesions,
as for 3 lesions, investigators did not provide information about the actual percentage
of vascular occlusion achieved after embolization. The actual percentage of vascular
occlusion for lesions after the first embolization procedure met the pre-embolization
target in 100 out of 135 lesions (74.1%), exceeded the pre-embolization target in
19 lesions (14.1%), and was lower than the target score in 16 lesions (11.9%). Therefore,
the actual vascular occlusion achieved was equal to or higher than the pre-embolization
target in 119 lesions (88.1%). On further analysis, the actual vascular occlusion
percentage corresponded to the pre-embolization target in 10 of 12 lesions (83.3%)
for postembolization group 1 lesions, in 13 of 20 lesions (65.0%) for postembolization
group 2 lesions, in 35 of 44 lesions (79.5%) for postembolization group 3 lesions,
and in 42 of 59 lesions (71.2%) for postembolization group 4 lesions ([Table 4]).
Table 4
Actual and target vascular occlusion/lesion obliteration percentage for the first
embolization procedure
|
Actual vascular occlusion percentage
|
Target vascular occlusion percentage
|
|
<50%
|
≥50 and <75%
|
≥75 and <100%
|
100%
|
Total lesions (N = 138)
|
|
<50%
|
10
|
1
|
1
|
0
|
12
8.7%
|
|
83.3%
|
8.3%
|
8.3%
|
0.0%
|
|
≥50 and <75%
|
0
|
13
|
4
|
3
|
20
14.5%
|
|
0.0%
|
65.0%
|
20.0%
|
15.0%
|
|
≥75 and <100%
|
0
|
2
|
35
|
7
|
44
31.9%
|
|
0.0%
|
4.5%
|
79.5%
|
15.9%
|
|
100%
|
1
|
0
|
16
|
42
|
59
42.8%
|
|
1.7%
|
0.0%
|
27.1%
|
71.2%
|
|
Missing
|
0
|
1
|
2
|
0
|
3
|
|
Total
|
11
8.0%
|
17
12.3%
|
58
42.0%
|
52
37.7%
|
138
100.0%
|
A second embolization procedure was performed for 13 lesions in 12 patients (mostly
AVMs), exclusively with Endocryl: 1 lesion in 11 patients and 2 lesions in 1 patient.
The most often used Lipiodol/N-BCA ratio was 3:1 (36.6%). The mean (SD) duration of
the procedure was 55 (33) minutes.
The actual vascular occlusion after the second embolization procedure met the target
in 6 of 13 lesions (46.2%), exceeded the target in 5 lesions (38.5%), and was lower
than the target in 2 lesions (15.4%). The actual lesion obliteration/vascular occlusion
was equal to or higher than the target score in 11 lesions (84.6%). The actual lesion
obliteration percentage corresponded to the target score in 2 of 2 lesions with an
obliteration percentage of less than 50%, 1 of 2 lesions with an obliteration percentage
between ≥50 and less than 75%, 2 of 5 lesions with an obliteration percentage between
≥75 and less than 100%, and 1 of 4 lesions with an obliteration percentage of 100%.
Although vascular embolization was initially planned to be performed in this study
using a selective transarterial route only (as per study protocol), the embolization
of the portal vein in one patient with a liver neoplasm planned for liver resection
was performed via a percutaneous portal vein access. It was, therefore, reported as
a protocol deviation. Of note, this patient also had complete obliteration of the
portal vein branches in the desired part of the liver after the procedure, achieving
a 100% technical success rate.
Discussion
Our study has demonstrated a favorable safety profile of the lipiodol and N-BCA mixture,
with most reported AEs being mild to moderate (SIR minor A or B complications) and
mostly unrelated to the mixture. In addition, the study revealed a high success rate
in achieving the desired vascular occlusion, with 87.8% of treated lesions having
equal to or better than the targeted vascular occlusion.
In the present study, AVMs accounted for most treated lesions (i.e., 66.7%). In a
study performed on patients with craniofacial venous vascular malformations (low-flow
vascular malformation), preoperative embolization via direct puncture and injection
of N-BCA achieved complete embolization in 8 of 13 patients (62%) and partial embolization
in the remaining 5 patients (38%). All patients underwent a successful surgical resection.[20] Another randomized study reported 79.4% of cerebral AVM nidus reduction after embolization
with N-BCA/Lipiodol, comparable to that reported after embolization with polyvinyl
alcohol (PVA) microparticles (86.9%).[21] In the current study, the actual vascular occlusion/lesion obliteration percentage
was equal to or higher than the target in 88.1% of the treated lesions following the
first procedure and 84.6% following the second procedure. No information regarding
the need for surgical resection of the AVM after embolization was recorded. The high
lesion obliteration/vascular occlusion rate in the present study, which included patients
treated for various indications according to real-world clinical practice settings,
suggests equivalent efficacy to that reported in previous studies.[21]
[22]
[23]
In the present study, AEs considered related to the mixture were reported in only
one patient, and the incidence of AEs related to the procedure was 12.9%. Most of
the AEs related to the procedure were mild or moderate (96.2%). Other studies reported
a complication rate between 9 and 16% for AVM embolization.[24]
[25]
[26] PES is a postinflammatory clinical syndrome comprising symptoms such as pain, fever,
nausea, and vomiting. In the current study, the rate of PES was 4.8%. In the literature,
PES could be reported as such, and the incidence could vary depending on the indication
of the procedure, with higher incidences (18–86%) reported with the embolization of
bone metastases.[13]
[27]
[28] In a phase II study on vascular embolization for various indications (excluding
cerebral AVM), the rate of AEs of grade 3 (severe AEs) or above (based on common terminology
criteria for adverse events (CTCAE) version 4) that could be related to the procedure
was 17.2%, including many symptoms related to PES. The high rate of grade 3 AEs, as
compared to the present study, could be explained by the inclusion of patients with
traumatic or postsurgical bleeding.[29]
Embolotherapy for skin and subcutaneous AVM is usually associated with increased risk
of skin necrosis, as reported in several studies.[13]
[30] In the current study, two patients reported pregangrenous skin changes and mild
skin necrosis (1.6%), which resolved after treatment. The use of the N-BCA/Lipiodol
mixture for vascular embolization could be associated with a risk of possible uncontrolled
reflux during the injection with consequent ischemic complications.[31]
[32] Previous reports suggested that nontarget embolization such as cerebral or spinal
cord infarction could occur with vascular embolization.[24]
[33]
[34]
This study has a few limitations. No clinical success endpoint, as therapeutic or
palliative procedure, was defined in this study. The primary aim of this study was
to assess the safety of the vascular embolization procedure. Nevertheless, consistency
between target and actual lesion obliteration was assessed as technical measure of
efficacy of the Lipiodol/N-BCA mixture. Long-term follow-up was not performed. However,
complications encountered with vascular embolization occurred mostly within the first
weeks after the procedure.[29]
[35] The incidence of rebleeding was not specifically assessed, but the systematic reporting
of AEs did not identify such an event. The study primarily focused on AVM treatment,
while a limited number of other lesions such as arteriovenous fistulas, hemangiomas,
and aneurysms were included in the study. This limited representation of different
lesion types hampers the ability to generalize the study findings.
Conclusion
In conclusion, the Lipiodol and N-BCA mixture had a favorable safety profile for vascular
embolization, with no AEs reported during the procedure, and two mild AEs related
to the mixture reported after the procedure. Additionally, vascular embolization with
the Lipiodol and N-BCA mixture demonstrated a high success rate in achieving the desired
vascular occlusion, indicating its effectiveness in treating the targeted lesions.
