Keywords
visceral artery pseudoaneurysm - percutaneous embolization - gastroduodenal artery
Introduction
Visceral artery aneurysms and pseudoaneurysms arise from the branches of celiac, superior
mesenteric, inferior mesenteric, and renal arteries.[1] Visceral artery aneurysms and pseudoaneurysm are rare causes of gastrointestinal
(GI) bleeding with a reported incidence of 0.01 to 0.2% on autopsy.[2] The etiology of visceral artery pseudoaneurysms includes infection, trauma, and
iatrogenic injuries. Atherosclerosis, fibromuscular dysplasia, or collagen vascular disorders are the causes
of true aneurysm. Most of the pseudoaneurysms are asymptomatic and are incidentally detected during
evaluation of other conditions. Pseudoaneurysms can rupture and present as upper or
lower GI bleed and hematuria.[1]
[2]
[3]
[4]
[5]
[6] The incidence of rupture of pseudoaneurysms can range from 2 to 80% depending on
the location with mortality rates as high as 100%.[4]
[5] While the true aneurysms smaller than 2 cm can be monitored, the pseudoaneurysms
need to be treated irrespective of their size.[1]
[4] Treatment options for pseudoaneurysms include endovascular management and surgery,
endovascular embolization, or stent graft placement are usually the first line of
therapies.[6]
Percutaneous embolization of pseudoaneurysms for difficult or inaccessible anatomical
sites has been mentioned in a few case reports and case series in the literature.
The largest case series included 21 patients.[7]
[8]
[9]
[10]
[11]
[12]
[13]
[14]
[15]
[16]
[17]
[18]
[19] Previous studies are limited by small size of the study population and lack of durable
follow-up. In this study, we would like to assess the safety and clinical effectiveness
of the percutaneous embolization in the management of visceral artery pseudoaneurysms
in a large group of patients.
Materials and Methods
A retrospective analysis of patients who had undergone percutaneous embolization for
visceral artery pseudoaneurysm at our tertiary care center between January 2012 and
May 2017 was performed. Ethical review was not required for this retrospective study.
The indications for direct percutaneous embolization included failure of selective
catheterization of the feeding artery supplying the pseudoaneurysm or prior embolization
of the proximal segment of the feeding artery.
A total of 26 patients (19 men and 7 women) with a mean age of 36 years (range 10–71
years) were included in this study. Patient demographics, details of percutaneous
embolization, technical success, procedure-related complications, and outcomes were
evaluated. Patients were followed for a mean 15 months (12–18 months) by ultrasound
to assess for recurrence of pseudoaneurysms.
Angiography and Endovascular Technique
Angiography and Endovascular Technique
Computed tomographic (CT) angiography of the abdomen and pelvis was obtained on a
six-slice multidetector CT scanner (Siemens Emotion, Siemens Healthcare GmbH). Angiography
was performed on single plane digital subtraction angiography (DSA) system (Axiom
Artis U, Siemens Healthcare GmbH). Angiography and endovascular interventions were
performed through a right transfemoral arterial access. Arterial access was obtained
through right common femoral artery and a 6F vascular sheath (Cordis) was placed.
Using a 5F Simmons angiography catheter (Cook Medical) and 0.035-inch hydrophilic
guidewire (Radifocus, Terumo), the celiac trunk, superior mesenteric artery, and inferior
mesenteric artery were catheterized. Superselective catheterization of the common
hepatic artery, gastroduodenal artery (GDA), left gastric artery (LGA), and splenic
artery were performed using a 4F Kumpe/Multipurpose catheter or a 3F microcatheter
(Cook Medical). DSA images were reviewed for presence of pseudoaneurysm. In 24 patients,
the feeding artery was not accessible for selective catheterization and two patients
had prior embolization of the feeding artery.
For percutaneous embolization, the pseudoaneurysms were accessed percutaneously using
an 18-G spinal needle (Meditop Corporation) under ultrasound guidance. Percutaneous
angiography was performed through needle, to assess the size of pseudoaneurysm, feeding
artery, and flow dynamics. Embolization was performed by using either N-butyl cyanoacrylate (NBCA) (Xion, Reevax Pharma) with lipiodol (Guerbet) or coil
(Cook Medical) or both under ultrasound and fluoroscopic guidance. The NBCA and lipiodol
were mixed in the ratio of 1:3, 1:2, and 1:1 depending upon the flow dynamics. A total
of 0.035 compatible coils of different sizes were used (3–30 mm, 3–40 mm, and 3–50
mm). The average fluoroscopy time was 3 minutes. A repeat catheter angiography was
performed to confirm the complete occlusion of the pseudoaneurysm.
Technical success was defined as complete occlusion of pseudoaneurysm at the end of
procedure confirmed by catheter angiography. All patients were followed for mean of
15 months by ultrasound to assess for recurrence of pseudoaneurysm. First ultrasound
was done after 1 month and subsequent follow-up was done at interval of 3 months.
Results
Patient demographics, etiology of pseudoaneurysms, and the arteries involved are summarized
in [Table 1]. The underlying cause of the pseudoaneurysm was pancreatitis in 20 patients, trauma
in two patient, iatrogenic injury in two patients, and incidentally detected pseudoaneurysm
in two patients. CT angiography was positive in all cases. Average size of pseudoaneurysm
was 15 mm (range 10–30 mm)
Table 1
Showing patients demographics, etiology, and arteries involved
|
Number of patients
|
26
|
|
M:F
|
19:7
|
|
Age, mean
|
36 (10–71 y)
|
|
Etiology
|
|
|
Pancreatitis
|
20
|
|
Trauma
|
2
|
|
Postoperative
|
2
|
|
Incidental
|
2
|
|
Arteries involved
|
|
|
Splenic artery
|
13
|
|
Gastroduodenal artery
|
4
|
|
Right hepatic artery
|
3
|
|
Left hepatic artery
|
1
|
|
Left gastric artery
|
2
|
|
Pancreaticoduodenal arteries
|
3
|
A failed selective catheterization was the indication for percutaneous embolization
in 24 patients ([Figs. 1]
[2]
[3]). A prior embolization of the feeding artery in two patients ([Fig. 4]) precluded endovascular embolization. Splenic artery was the most common vessel
involved (n = 13) followed by the GDA (n = 4), pancreaticoduodenal arteries (n = 3), right hepatic artery (n = 3), LGA (n = 2), and left hepatic artery (n = 1).
Fig. 1 A 38-year-old male patient with history of chronic pancreatitis and splenic artery
pseudoaneurysm. (A, B) Convention angiography shows a splenic artery pseudoaneurysm (black arrows). Transcatheter embolization failed due to difficult access. (C) Percutaneous puncture and angiography were performed to confirm the position of
the needle. (D) Embolization with NBCA and lipiodol was performed. (E) Postembolization angiography shows no filling of the pseudoaneurysm. (F) Postprocedure fluoroscopy image shows lipiodol cast in the pseudoaneurysm. NBCA,
N-butyl cyanoacrylate.
Fig. 2 A 38-year-old female patient with history of trauma and liver laceration. CT (A) axial and (B) oblique sagittal MIP images show a liver laceration (white arrows) and subcapsular hematoma (arrowhead) with a pseudoaneurysm (black arrows) arising from the right hepatic artery. Convention embolization failed due to difficult
arterial access. (C) Percutaneous puncture and angiography confirmed the pseudoaneurysm. (D) This was followed by embolization of the pseudoaneurysm with NBCA and lipiodol.
(E) Mild reflux of NBCA into right hepatic artery is noted. CT, computed tomography;
MIP, maximum intensity projection; NBCA, N-butyl cyanoacrylate.
Fig. 3 A 21-year-old male patient with chronic pancreatitis and prior embolization of right
hepatic artery. (a) Convention angiogram showed large pseudoaneurysm from proximal gastroduodenal artery
(black arrow). Transcatheter embolization failed due to difficult access. (b) Percutaneous puncture and angiography confirmed the position of the needle. (c) Embolization was performed with NBCA and lipiodol. NBCA, N-butyl cyanoacrylate.
Fig. 4 A 26-year-old female patient with history of cholecystectomy and pain abdomen. (A) Convention angiogram showing pseudoaneurysm arising from the right hepatic artery.
(B) Coil embolization of right hepatic artery was performed. (C) Postembolization angiography showed persistent filling of the pseudoaneurysm. Further
endovascular management was not feasible due to occlusion of proximal vessel by coils.
(D) Percutaneous puncture and angiography confirmed the position of the needle. (E) Embolization with NBCA and lipiodol was performed. (F) Postembolization angiography showed no filling of the pseudoaneurysm. NBCA, N-butyl cyanoacrylate.
Technical details, outcome, and complications are summarized in [Table 2]. Embolization was performed using NBCA and lipiodol combination in 23 patients (88.4%;
[Figs. 1]
[2]
[3]
[4]), coil in one patient (3.8%), and both coil and glue in two (7.7%) patients ([Fig. 5]). Embolization of the pseudoaneurysm was successful in all the cases. None of the
patients had immediate procedure-related complications. Minimal flow of NBCA into
nontarget vessel was seen in all the patients with NBCA embolization.
Table 2
Showing technical details, outcome, and complications
|
Abbreviation: NBCA, N-butyl cyanoacrylate.
|
|
Embolization
|
|
NBCA and lipiodol
|
23
|
|
Coil
|
1
|
|
Both
|
2
|
|
Outcome
|
|
Technical success
|
26 (100%)
|
|
Recurrence of pseudoaneurysm
|
Nil
|
|
Complications
|
|
Self-limiting abdominal pain
|
25
|
|
Splenic infarct
|
6
|
|
Liver abscess
|
1
|
Fig. 5 A 58-year-old male patient with chronic pancreatitis and splenic artery pseudoaneurysm.
(A, B) Convention angiogram shows splenic artery pseudoaneurysm. Transcatheter embolization
failed due to difficult access. (B) Percutaneous puncture and angiography confirmed the position of the needle. (C, D) Embolization was performed with coils. (E) Postembolization angiography showed persistent filling of pseudoaneurysm. (F) Further embolization was performed with NBCA and lipiodol. (G) Postembolization angiography showed no filling of the pseudoaneurysm.
Out of 13 patients with splenic artery pseudoaneurysms, six patients developed self-limiting
splenic infarcts. One patient with hepatic artery embolization developed infarction
and subsequent abscess requiring percutaneous drainage. Self-limiting abdominal pain
was seen in all the patients with NBCA embolization immediately after the procedure.
Follow-up ultrasound at a mean 15-month period showed no residual or recurrence of
pseudoaneurysm.
Discussion
Visceral artery aneurysms and pseudoaneurysms arise from the branches of celiac, superior
mesenteric, inferior mesenteric, or renal arteries.[1] Focal discontinuity in vessel wall by infection, inflammation, trauma, and iatrogenic
injury will lead to extravasation of blood contained by the vessel wall or by the
adjacent soft-tissue, leading to the formation of pseudoaneurysm.[2]
[3] The incidence of rupture of pseudoaneurysms varies from 2 to 80% with untreated
mortality rates reaching up to 100%.[5] Most of the pseudoaneurysms are asymptomatic and incidentally detected on imaging
during evaluation of other conditions.[3]
Noninvasive evaluation of visceral artery aneurysms and pseudoaneurysms is performed
with ultrasound, CT angiography, and magnetic resonance angiography. Pseudoaneurysm
is seen as an outpouching from the vessel wall with swirling motion and to-and-fro
waveform on the Doppler. In CT pseudoaneurysm is seen as low attenuation rounded focus
adjacent to artery in plain CT. High attenuation adjacent to the pseudoaneurysm indicates
pseudoaneurysm rupture. Contrast study shows enhancement similar to adjacent artery
in both arterial and venous phase. Convention angiography is the gold standard for
the diagnosis of pseudoaneurysms and has the advantage of real-time assessment of
vascular bed including identification of feeding vessel, collateral vessel, and concomitant
therapeutic intervention[3]
[5]
[6] In the present study, pseudoaneurysms were detected by CT angiography and were identifiable
on Doppler and conventional angiography in all cases.
Treatment options for visceral artery pseudoaneurysms include endovascular embolization,
stent graft placement, and surgery. Endovascular embolization and stent graft placement
are recommended as the first-line therapy. Surgical management of pseudoaneurysms
includes resection with a bypass, arterial ligation, and partial or complete organ
removal. Surgical treatment is invasive and is often associated with significantly
higher morbidity and mortality rates.[3]
[6]
[20] Success rate of transcatheter embolization for both aneurysms and pseudoaneurysms
is high (98%) with 30-day mortality rate of 3.4%.[9]
Direct percutaneous approach was first described by Cope and Zeit in 1986 for the
treatment of a femoral pseudoaneurysm.[21] Percutaneous embolization can be performed in pseudoaneurysms with difficult anatomy,
inaccessible sites, prior embolization of feeding vessel, and in whom feeding artery
cannot be identified on angiography.[9]
[10]
[11]
[12]
[13]
[14]
[15] Available literature on percutaneous embolization of visceral artery pseudoaneurysms
has been summarized in [Table 3]. Various embolic materials have been used for percutaneous embolization. Most commonly
used embolic material was thrombin, followed by NBCA and coils. In the present study,
a combination of NBCA and lipiodol was used in 23 patients (88.4%), coil in one patient
(3.8%), and both coil and NBCA in two patients (7.7%). Use of thrombin has been associated
with recurrence of pseudoaneurysm in significant number of cases requiring reintervention.
Coils and NBCA are not associated with recurrence or reformation of pseudoaneurysm.[7]
[8]
[9]
[16]
[17]
[18] None of the patients in our study had residual or recurrence of pseudoaneurysm.
Table 3
Summary of literature on percutaneous embolization of pseudoaneurysm
|
No. of cases
|
Embolic agent
|
Reintervention
|
Complications
|
|
Abbreviation: NBCA, N-butyl cyanoacrylate.
|
|
Gorsi et al[7]
|
21
|
NBCA
|
–
|
Pneumothorax
|
|
Fankhauser et al[9]
|
2
3
|
Coil
Thrombin
|
–
1
|
–
–
|
|
Vyas et al[8]
|
3
2
|
NBCA
Thrombin
|
–
1
|
–
–
|
|
Yadav et al[12]
|
4
|
NBCA
|
–
|
–
|
|
Nicholson et al[16]
|
4
|
Thrombin
|
4
|
–
|
|
Laganà et al[17]
|
2
|
Thrombin
|
2
|
–
|
|
Kuno et al[10]
|
1
|
Coil
|
–
|
–
|
|
Ghassemi et al[14]
|
1
|
Thrombin
|
–
|
–
|
|
Ward et al[13]
|
1
|
Thrombin
|
–
|
–
|
|
Santiagu et al[15]
|
1
|
Thrombin
|
–
|
–
|
|
Puri et al[18]
|
1
|
Thrombin
|
1
|
–
|
|
Barbiero et al[11]
|
1
|
Thrombin
|
–
|
–
|
|
Gulati et al[19]
|
1
|
NBCA
|
–
|
Jejunal stricture
|
Percutaneous embolization has been associated with a few complications. There are
no reports of procedural complications such as rupture of the pseudoaneurysm. Jejunal
stricture has been reported by Gulati et al due to reflux of NBCA into jejunal branches
[7]
[8]
[9]
[11]
[12]
[13]
[14]
[15]
[16]
[17]
[18]
[19] Most common complication in our study was self-limiting splenic infarct (six patients)
due to the reflux of NBCA into distal splenic branches. Self-limiting abdominal pain
was seen in all the patients with glue embolization. One patient with hepatic artery
embolization developed liver infarct and abscess which was managed with percutaneous
drainage.
Our study has many limitations. Despite being one of the largest studies, our study
included only 26 patients, and is retrospective in nature. The follow-up of our patients
was limited to ultrasound examination only. Additionally, the follow-up period was
limited to a mean of 15 months with no regular imaging follow-up after. A large prospective
study is required to validate the safety and efficacy of the procedure and the results
of our study.
Direct percutaneous embolization is safe and effective in the treatment of visceral
artery pseudoaneurysms. It can be considered as an alternative following a failed
endovascular approach or when an endovascular approach is not feasible.
