Best Practice & Research Clinical Gastroenterology 31 (2017) 97e104

Contents lists available at ScienceDirect

Best Practice & Research Clinical Gastroenterology

journal homepage: https://ees.elsevier.com/ybega/default.asp

10

Visceral aneurysms: Old paradigms, new insights?

M.J.E. van Rijn, MD, PhD *, S. ten Raa, MD, PhD,
J.M. Hendriks, MD, PhD Professor, Head of the Department of Surgery,
H.J.M. Verhagen, MD, PhD Professor, Head of the Department of Vascular Surgery
Department of Vascular and Endovascular Surgery, Erasmus Medical Center, Rotterdam, The Netherlands

a b s t r a c t
Keywords: True visceral artery aneurysms (VAAs) are a rare entity with an incidence of 0.01e2%. The risk of rupture
Visceral artery aneurysm varies amongst the different types of VAAs and is higher for pseudo aneurysms compared with true
Visceral artery pseudo aneurysm
aneurysms. Size, growth, symptoms, underlying disease, pregnancy and liver transplantation have all
Endovascular treatment
Open repair
been associated with increased risk of rupture. Mortality rates after rupture are around 25%. The splenic
Rupture artery is most commonly affected and the etiology is predominantly atherosclerosis. Open repair can be
done by simple ligation or reconstruction of the artery, while endovascular options include embolization
or using a stent graft. Location, collateral circulation and medical condition of the patient should all be
taken into account when an intervention is planned. We compared types of treatment and searched for
risk factors for rupture but unfortunately, the level of evidence found in the literature is low. Therefore,
deciding when and how to treat a patient with a VAA based on the current literature, remains chal-
lenging for clinicians.
© 2016 Elsevier Ltd. All rights reserved.

Introduction
VAAs are defined as aneurysms of the celiac (CA), superior
(SMA) or inferior mesenteric (IMA) arteries and their branches.
They are a rare entity with a reported incidence of 0.01%e2% [1].
The incidence of 0.78% in nearly 3600 abdominal arteriographic
studies may better reflect their true frequency in the general
population, although postmortem studies have found an incidence
of splenic artery aneurysms (SAA) of 10.4% [2,3]. The first descrip-
tion of a VAA was done by Beaussiers in 1770 when he found a
splenic artery aneurysm on autopsy. Both this case, and a second
case reported by Parker in 1844, were omitted from the literature
for many years and mistakenly given to Crisp. In 1871 Quincke first
described the “classic” triad of jaundice, biliary colic and upper
gastrointestinal hemorrhage caused by a hepatic artery aneurysm
rupture. Kehr performed the first successful surgical procedure in
1903 when he ligated a hepatic artery aneurysm.
Depending on the location of the aneurysm, different symptoms
can be expected. However, clinical symptoms and signs are
frequently unspecific and since VAAs are rare, they are not often
suspected, leading to a delay in diagnosis. The mortality rate of a
ruptured VAA is around 25%, although various rates have been re-
ported in the literature and differ between location [4e6]. Risk of
rupture is also related to size, growth rate and underlying disease
[4,7]. With the increasing use of diagnostic tools like ultrasonog-
raphy, computed tomography angiography (CTA) and magnetic
resonance imaging (MRI), the incidence incidental finding of
asymptomatic VAAs has increased.
Treatment of VAAs can be done by either open or endovascular
repair. Discriminating between VAAs that can be monitored and
those that require an intervention remains a challenge as no ran-
domized controlled trials (RCTs) have been performed in this area.
This chapter provides an overview of the currently available liter-
ature on VAAs. We will separately describe them by anatomic
location and distinguish true VAAs from visceral artery pseudo
aneurysms (VAPAs).
Epidemiology, etiology and natural behavior

* Corresponding author. Department of Vascular and Endovascular Surgery,
Erasmus University Medical Center, 's-Gravendijkwal 230, 3015 CE, Rotterdam, The
Netherlands. Fax: þ31 10 7032396.
E-mail address: m.vanrijn@erasmusmc.nl (M.J.E. van Rijn).
In VAAs, all three layers of the arterial wall are intact, while
VAPAs are actually contained ruptures lined by the adventitia or
perivascular tissues. VAAs are a focal dilatation of the artery with a
diameter more than 1.5 times the normal diameter of the vessel.

http://dx.doi.org/10.1016/j.bpg.2016.10.017
1521-6918/© 2016 Elsevier Ltd. All rights reserved.
98 M.J.E. van Rijn et al. / Best Practice & Research Clinical Gastroenterology 31 (2017) 97e104
They are located in the CA, SMA, IMA or their branches. Renal artery
aneurysms are usually not considered VAAs as they have a slightly
different etiology. The splenic artery is most commonly affected
(60%). Second is the hepatic artery (20%), followed by the SMA (5%)
and the CA (4%, see Fig. 1) [5,6,8,9]. Other possible locations are the
IMA, gastric, gastroepiploic, intestinal, pancreatic, gastroduodenal
(GD) and pancreaticoduodenal (PD) arteries which together ac-
count for 11% of the locations. Table 1 summarizes the most com-
mon location per artery [10,11]. VAPAs mostly occur in the hepatic
artery (39%), the CA or its branches (39%) [12].
VAAs are rare with an incidence of 0.01e2% [1]. With the
increasing use of diagnostic tools for complaints unrelated to the
VAA, the finding of asymptomatic VAAs has increased and the
growing number of interventions in the arterial bed and the biliary
tract has increased the number of VAPAs. Multiple etiologies ac-
count for the development of VAAs [5,8]. The most common
pathway is through atherosclerosis (32%), followed by medial
degeneration (24%) and abdominal trauma (22%). Hyperflow con-
ditions (e.g., pregnancy, portal hypertension), connective tissue
disorders (e.g., Marfan, EhlerseDanlos, fibromuscular dysplasia),
vasculitis (e.g., polyarteritis nodosa, Takayasu, Kawasaki), neurofi-
bromatosis and the antiphospholipid syndrome have all been
associated with the formation of VAAs (see Table 2). VAPAs are the
result of iatrogenic injury, infection or abdominal trauma.
Recently a paper by Corey et al. was published describing the
natural behavior of VAAs while under surveillance [7]. During a
study period of 20 years, 176 VAAs with a mean size of 16.28 mm
(8e41 mm), were monitored instead of immediately treated. Of
these, 91.3% remained stable over time (mean of 36.1 months,
ranging from 2 to 155 months) without any change in size. There
were no ruptures in this group of patients and only 5.8% required an
intervention during follow-up. None of the hepatic, jejunal or IMA
aneurysms grew in time. The only aneurysms showing some
growth were located in the CA, SA, PDA, GDA and SMA.
Comparing intact and ruptured VAAs
Shukla et al. [6] studied 261 patients of which 181 were
repaired. In 77 patients, the VAA had ruptured (rVAA), in 104 it was
intact (iVAA). The percentage of men was 63.2% in de rVAA group
compared with 42.0% in the iVAA group (p ¼ 0.005). Of the patients
Fig. 1. Distribution of visceral artery aneurysms by arterial bed. VAA ¼ visceral artery aneurysm, SMA ¼ superior mesenteric artery, CA ¼ celiac artery.
Table 1
Location per type of aneurysm.
Type of VAA Most common location per artery
Splenic artery Distal third of the artery
Bifurcation distal to short gastrics
Splenic hilum
Hepatic artery Extrahepatic (80%):
- Common hepatic artery 60%
- Right hepatic artery 30%
- Left hepatic artery 10%
Intrahepatic (20%)
SMA Proximal 5 cm
CA Distal to site of chronic vascular compression
VAA ¼ visceral artery aneurysm, SMA ¼ superior mesenteric artery, CA ¼ celiac
artery.
with a rVAA, 29.7% were in hemodynamic shock. The most com-
mon cause of iVAA was degenerative disease (31.7%), for rVAA this
was inflammatory/pancreatitis (33.8%). VAPAs were more common
in rVAAs (81.8%) compared with iVAAs (35.3%, p < 0.001). The
perioperative complication rate was higher for rVAAs (13.7% vs 1%,
p ¼ 0.003). Mortality for rVAAs at 30 days was 13%, at 1 year 32.5%
and at 3 years 36.4%, all significantly higher compared with iVAAs
(0%, 4.1% and 8.3% respectively). The 30-day mortality for rVAAs
was highest in splenic artery aneurysms (27.7%) and the compli-
cation rate in rVAAs was highest for SMA (62.5%) and hepatic an-
eurysms (41.1%).
Pitton et al. [5] described 233 patients with 253 VAAs. Fifteen
percent of the patients presented with a rupture, all of them had
symptoms. Of the rVAAs, 76.3% were VAPAs. Only 3.1% of the VAAs
ruptured. There was no significant difference between the diameter
of rVAAs and iVAAs (14.8 vs 16.3 mm). The greatest diameters were
found in splenic artery aneurysms. After treatment, the 30-day
mortality was 6.7% in rVAAs compared with no mortality in iVAAs.
A univariate regression model examining risk factors associated
with rupturing was presented in the study by Corey et al. [7] An
odds ratio of 1.1 (p ¼ 0.004) was found for aneurysm size, of 11.2
(p ¼ 0.0002) for PDA and GDA aneurysms and of 32.5 (p ¼ 0.01) for
EhlerseDanlos. There was no difference in 5-year survival between
patients that underwent surveillance compared with those that
underwent early repair.
 M.J.E. van Rijn et al. / Best Practice & Research Clinical Gastroenterology 31 (2017) 97e104
Table 2
Etiology.
Etiology of VAA
Atherosclerosis
Medial degeneration
Abdominal trauma
Infection/inflammation
Other:
Connective tissue disorder,
hyperflow conditions,
vasculitis, neurofibromatosis,
antiphospholipid syndrome
VAA ¼ visceral artery aneurysms.
In general, elective repair is recommended for all VAPAs
regardless of size, and for VAAs with a diameter >2.0 cm. Corey
et al. suggest that this cut-off can be increased to 2.5 mm [7].
Because there are no RCTs performed, this recommendation is not
supported by a high level of evidence. Other recommendations on
treatment are shown in Table 3 [9,13]. Treatment can be done either
by open or endovascular repair, with the latter being first choice of
treatment in many centers. Laparoscopic ligation of the aneurysm
has also been described [14]. If an endovascular approach is used,
the most common way of excluding the aneurysm is by emboliza-
tion (using coils, glue, plugs or Onyx). In VAPAs the efferent and
afferent artery on either side of the aneurysm has to be closed,
while VAAs may sometimes be treated by coil packing of the
aneurysm only. Covered stent placement may be another treatment
option for both types of aneurysms depending on location and
tortuosity of the vessel as well as diameters of the artery. A
multilayer stent graft has been available since 2009. This is a flow-
diverting bare metal stent that improves laminar flow while
decreasing flow within the aneurysm. Like covered stent grafts,
they have the benefit of keeping the vessel patent. They are no RCTs
comparing the different types of endovascular techniques.
In open repair, ligation is the simplest solution, but there is al-
ways the risk of distal ischemia. Reconstructing the artery after
resection of the aneurysm (either primarily or using a prosthetic or
vein graft) or an aneurysmorrhaphy (suturing the aneurysm sac to
restore its normal lumen dimension) preserves blood flow to the
organs, but may be challenging especially in small or very distal
aneurysms [8].
No RCT has compared open with endovascular repair for the
treatment of VAAs or VAPAs. Sachdev et al. retrospectively studied
59 patients with 61 VAAs comparing open (24 patients) and
endovascular (35 patients) techniques [15]. VAPAs were more likely
to be treated endovascular (86% vs 14%, p < 0.1). Length of stay was
shorter for elective cases treated endovascular compared with open
repair (2.4 days compared with 6.6 days, p < 0.001). There was no
difference in 30-day mortality (3 and 4%), complications (26 and
33%) or reinterventions (20 and 17%) between the two groups.
Table 3
Treatment guidelines.
Treatment of visceral aneurysms
VAA
Size >2.0 cm
Growth during follow-up
Symptoms related to the aneurysm
Women who wish to become pregnant
Patient requiring a liver transplant
Non-atherosclerotic etiology (i.e. connective tissue disorder)
Multiple hepatic VAAs
VAPA
Always
VAA ¼ visceral artery aneurysms, VAPA ¼ visceral artery pseudo aneurysm.
99
Batagini et al. also compared both techniques retrospectively [16].
Of the 113 patients, 57 were treated endovascular and 56 open.
Percentage % Operative time was shorter and blood loss less for endovascular
32
repair compared with open repair. Complications and reinterven-
24 tions did not differ between the groups. Length of stay was shorter
22 for the endovascular group (1 vs 6 days, p < 0.001). The clinical
10 success was 91.2% for the endovascular and 92.9% for the open
12
group (p ¼ 0.74). The overall survival (94.7% and 96.4%) was not
significantly different. Other success rates reported are in the range
of 73e100% [12,17e19]. After endovascular occlusion, recanaliza-
tion and persistent perfusion create a risk for the patient as the
aneurysm is not excluded and risk of rupture is still present. This
can occur in up to 25% of patients after embolization [20]. Follow-
up after treatment is therefore necessary. However, the material
used to treat the VAA is usually radiopaque which causes artifacts
on imaging making it hard to interpret whether or not the VAA is
excluded.
Splenic artery aneurysms (SAA)
Risk factors and etiology
SAAs are the third most common true aneurysms in the
abdomen after aortic and iliac artery aneurysms (Illustration 1). In a
series of 250 necropsies in 108 males and 142 females (age ranging
from 34 to 100 years) the spleen and pancreas were removed en
bloc and the splenic artery was completely dissected [2]. The inci-
dence of SAA found in this study was 10.4%. The largest study
reviewing patients with a SAA was published by Abbas et al. [4]. The
authors retrospectively reviewed 217 patients diagnosed with a
SAA in the Mayo Clinic. The female to male ratio was 4:1, consistent
with a cohort study by Lakin et al. performed in the Cleveland Clinic
[21]. Concomitant VAAs were present in 3.3%, mostly extrahepatic
aneurysms. Concomitant nonvisceral aneurysms were present in
14.3%, mostly renal artery and aortic aneurysms. SAAs are more
common in women after multiple pregnancies [22,23]. This is
thought to be due to degenerative changes in the wall of the artery
as a result of hormonal shifts, as well as to the higher levels of the
hormone relaxin (22). The mean number of pregnancies in the
study from the Mayo Clinic was 3.5. Other known risk factors are
hypertension, present in 50 % of the patients in both studies, and
portal hypertension described in 7.8% [22e24]. Alpha-1 antitrypsin
deficiency and medial fibrodysplasia have also been associated
with SAA [22,23].
Rupture risk
The rate of rupture was 5% in both studies [4,21]. The classic
presentation is the ‘double rupture phenomenon’. After a period of
severe abdominal pain and hemodynamic instability follows a
period of normalization. This occurs due to containment of the
rupture within the lesser sac and therefore tamponade of the
rupture. If the aneurysm then ruptures into the peritoneal cavity,
recurrent hemodynamic instability occurs. Males have a higher
risk of rupture compared with women. None of the ruptures
occurred in women of childbearing age, nor in women using es-
trogen replacement therapy in the Mayo Clinic study. Other
studies did show that pregnancy increases the risk of rupture [25].
Calcification was present in 90% of ruptured and 84.5% of intact
SAAs and thus did not seem to protect against rupture in the Mayo
Clinic cohort. In the Cleveland Clinic cohort increased calcification
was associated with decreased size of the SAA (p ¼ 0.013). The
mean size of rSAAs in males was 3.2 cm compared with 2.6 in
iSAAs. For females this was 2.3 cm compared with 2.1 cm. The
smallest ruptured SAA was 2.3 cm. From this, a diameter of 2.0 cm
100 M.J.E. van Rijn et al. / Best Practice & Research Clinical Gastroenterology 31 (2017) 97e104
Fig. 2. CTA showing a splenic artery aneurysm. This was an accidental finding in a
patient that was screened for a cardiac transplantation.
is advised to use as a cut-off for elective repair, but since this was
not an RCT, the evidence to choose this cut-off is not very strong. In
the non operative group, only 10.1% showed growth during follow-
up (1e371 months). The most rapid growth was 1 cm in 63
months, the slowest 0.2 cm in 194 months. In patients with a SAA,
the risk of rupture increases after liver transplantation [23].
Pseudo aneurysms in SAAs are far less common than true SAAs.
They mostly occur in patients with severe pancreatitis and pseu-
docysts and occur more often in men. The release of pancreatic
enzymes causes a necrotizing arteritis leading to weakening of the
wall of the artery and rupture [22,26].
Treating SAA: open repair
Through a median laparotomy, the SAA can be completely
resected, after which a reconstruction can be performed with an
end-to-end anastomosis. If the aneurysm is in the distal part of the
artery and a reconstruction is impossible, sometimes a splenectomy
is necessary. Another option is an aneurysmorrhaphy. The aneu-
rysm sac is then partly excised and the wall is sutured to restore the
normal lumen dimensions. Most common complications after open
repair are hemorrhage, wound infection and pancreatitis. After a
splenectomy, an overwhelming post-splenectomy infection (OPSI)
can occur. Although this is a rare entity, it has a high mortality rate
when there is a delayed or inadequate treatment [27]. Patients
should therefore be vaccinated after a splenectomy against
meningococci, pneumococci and Haemophilus influenza.
Treating SAA: endovascular repair
The spleen has a rich blood supply with collateral flow from the
short gastric arteries and the gastroduodenal artery via the gas-
troepiploic arteries. Therefore, occlusion of the splenic artery does
not inevitably lead to ischemia of the spleen and embolization of
the artery is usually a good option [9,13]. This can be done using
coils, plugs, glue, thrombin or Onyx. Often there is tortuosity in the
artery, which causes difficulty in placement of a covered stent graft,
although in more straight anatomy (usually proximal in the artery),
this can be done and has the advantage of keeping the vessel pat-
ent. In the more distal part of the artery, super selective emboli-
zation is usually the preferred treatment. General complications
that can occur are access site complications leading to bleeding or
ischemic problems of the leg or arm, coil or glue migration,
recanalization of the artery and contrast induced complications
(anaphylactic shock or nephropathy). Specific for SAAs are infarc-
tion of the spleen, abscess formation, pancreatitis and splenic vein
thrombosis. Asplenia is very uncommon, but vaccination is offered
if extensive distal embolization is performed [13,28,29].
Comparing treatments
A systematic review comparing open and endovascular repair
was performed by Hogendoorn et al. [30] Data from 1321 patients
with true SAAs was used from 47 articles. There were significantly
more women in the conservatively treated group (p ¼ 0.001). In the
open group, the number of ruptured SAAs was 18.4% compared
with 8.8% in the endovascular group (p < 0.001). The 30-day
mortality rate was significantly higher in the open group (5.1%)
compared with the endovascular group (0.6%, p < 0.001). Endo-
vascular treatment required more reinterventions (3.2% vs 0.5%,
p ¼ 0.04). The late mortality rate was significantly higher in pa-
tients treated conservatively (4.9%) compared with open (2.1%) and
endovascular repair (1.4%). Both types of treatment had success
rates of over 96%. To investigate the cost-effectiveness of the three
treatment strategies, the same authors performed a decision
analysis using a Markov model [31]. They found that the endo-
vascular strategy is superior to open repair, being both cost-saving
and more effective in all age groups. Elderly patients (>80 years)
should be considered for conservative treatment as they only have
a small gain in health compared to the high costs.
Hepatic artery aneurysms (HAA)
Etiology and rupture risk
HAA is the second most common VAA and with increasing
biliary procedures, liver transplantations, and conservative man-
agement of trauma patients, it has become the most common VAPA
[9,13,23,26,29]. Most of the HAAs are solitary, but concomitant
VAAs are present in up to 31% of patients. They are mostly extra-
hepatic in the common hepatic artery (Table 1). The male to female
ratio is 3:2. Again, the Mayo Clinic has studied the largest group of
true HAAs and found that hypertension was present in 72% of pa-
tients and that most HAAs were the result of atherosclerosis. In this
series a group of patients with significant comorbidity and a HAA of
>3.0 cm was observed and none of them ruptured. Again, a cut-off
of 2.0 cm is generally used for the elective treatment of HAAs,
without much evidence to support this. The main risk factors for
rupture are the presence of multiple HAAs and non-atherosclerotic
etiology (e.g. fibromuscular dysplasia, polyarteritis nodosa or
mycotic) [32]. Rupture into the biliary tree is more common than
into the intraperitoneal cavity. This can lead to Quincky's triad;
 M.J.E. van Rijn et al. / Best Practice & Research Clinical Gastroenterology 31 (2017) 97e104
jaundice, biliary colic and upper gastrointestinal hemorrhage.
Mortality rates of up to 21% have been described after rupture.
Treating HAA: open repair
This depends on the location, the medical condition of the pa-
tient and the presence of collateral flow. Options are ligation of the
artery with or without reconstruction using a bypass or interposi-
tion graft (venous or prosthetic), partial hepatectomy or in very rare
instances a liver transplantation [23]. HAAs distal to the origin of
the gastroduodenal artery should be treated with reconstruction of
the artery whereas HAAs proximal to this point can be ligated
because the hepatic perfusion is than secured through collateral
flow from the gastroduodenal and right gastric arteries. Ligation
can only be performed if the portal vein is patent. Complications
include graft thrombosis, dissection of the reconstructed artery,
bile leak, liver failure, abscess, cholecystitis, hemorrhage and
wound infection.
Treating HAA: endovascular repair
For intra-hepatic aneurysms embolization is most commonly
used as open repair is difficult. For this, micro catheters must be
used. Aneurysms in the proper hepatic artery can be treated with a
covered stent graft with or without embolization of the gastrodu-
odenal artery, if there is a risk of recanalization of the aneurysm
through this artery. There must be adequate sealing zones for the
stent graft in order for this technique to successfully exclude the
HAA. Also, pre-intervention, the anatomy should be studied with
special attention to the risk of kinking of the stent graft. This would
lead to an occlusion of the stent and increases the risk of ischemia
of the liver. The common hepatic artery can be embolized if the
portal vein is patent. Pseudo aneurysms are predominantly treated
by endovascular repair as they often result from previous abdom-
inal interventions making open repair more complicated [15].
Complications include dissection, liver failure, abscess, cholecys-
titis, access and contrast induced complications as well as coil
migration, stent occlusion or recanalization of the vessel.
Superior mesenteric artery aneurysms (SMAA)
Etiology and rupture risk
These aneurysms are rare and commonly located in the prox-
imal 5 cm of the SMA. Etiologies are atherosclerosis, cystic medial
disease, polyarthritis nodosa, collagen vascular disorders and
infection (e.g. septic emboli from endocarditis). They are more
common in men [33]. A dilated SMA can also occur after extension
of an aortic dissection into the SMA or after a dissection solely
located in the SMA. Pseudo aneurysms may develop after the
arterial wall is destructed by pancreatic enzymes in patients with
pancreatitis or after trauma or abdominal surgery. Thrombus in the
aneurysm beyond the collateral circulation from the IMA and CA
may result in bowel ischemia [13,23,26].
A series from the Mayo Clinic described 21 patients with SMAA
[33]. Of the patients 14 (67%) were male. Eight patients (38%)
presented with a rupture, of which 88% were male. At time of
presentation, 48% were asymptomatic. Two of the eight patients
with rupture (25%) had concomitant symptoms of bowel ischemia
at initial presentation and bowel resection was necessary in three
patients. Thus, elective treatment of SMAAs may lead to bowel
preservation. Thirteen patients (62%) had calcified aneurysms, but
all ruptures were seen in noncalcified aneurysms. Laminated
thrombus was identified in aneurysms of three patients (14%).
101
Treating SMAA: open repair
Ligating the efferent and afferent arteries of the aneurysm is a
commonly used treatment depending on the location with respect
to the collateral circulation. Intraoperative test occlusion with
assessment of bowel viability can be performed first before pro-
ceeding with the ligation. Other options are aneurysmorrhaphy or
resection and reconstruction. Ligation in combination with a
bypass (using either an autologous vein or prosthetic graft) is also
possible. In case of a mycotic aneurysm or the presence of bowel
ischemia, a vein graft is preferred. Possible complications are bowel
ischemia, bypass occlusion, dissection, hemorrhage and wound
infection. We recently treated a patient in our center that presented
with bowel ischemia and an occluded SMAA. Bypass surgery was
performed from the common iliac artery to the SMA distally of the
aneurysm and he recovered uneventfully. However, he presented a
few weeks later with abdominal pain. On CTA it was shown that the
aneurysm had recanalized and ruptured. We excluded the ruptured
SMAA successfully with coiling, glue and Amplatzer plugs
(Illustration 2). With this case in mind we would recommend that
even in case of an occluded SMAA the artery should be ligated at
time of bypass surgery.
Treating SMAA: endovascular repair
Embolization is only possible if there is sufficient collateral cir-
culation and usually not possible for more distal aneurysms. Bowel
viability cannot be checked as with open surgery, so close moni-
toring of the patient after embolization is manditory. The use of
stent grafts (covered or multilayer) has also been described in the
literature, although series are small (Illustration 3) [13,23]. Com-
plications are bowel ischemia, dissection, coil migration, stent oc-
clusion and access and contrast related complications.
Celiac artery aneurysms (CAA)
Etiology and rupture risk
These are rare aneurysms, commonly located in the distal part of
the artery. The most common cause is atherosclerosis, although it
can also be the result of post stenotic dilatation when the artery is
compressed by the median arcuate ligament [9,23]. Concomitant
aneurysms (aortic, renal, femoral and popliteal) are present in
about 67% of patients [34]. Because of the location of the artery, a
symptomatic CAA may mimic pancreatitis. However, most CAAs are
asymptomatic (72%). The risk of rupture has been reported at
10e20% with high mortality rates (up to 100%). In a series of 18
patients from the Mayo Clinic, only one presented with a rupture
(6%). In this study, eight patients were observed. Only one patient
ruptured during follow-up, the other seven did not show any
growth over a mean period of 91 months [34].
Treating CAA: open repair
To gain access to the artery, a medial visceral rotation may be
necessary or even a thoracoeabdominal approach. Simple ligation
with careful observation for ischemic complications is possible,
especially in case of emergency surgery for rupture. However,
revascularization by performing bypass surgery (with vein or
prosthetic graft) or resection with reconstructing the artery, is
usually chosen. Possible complications may include ischemic
gastric ulcer, gangrenous cholecystitis, liver abscess, bypass occlu-
sion, hemorrhage and wound infection.
102 M.J.E. van Rijn et al. / Best Practice & Research Clinical Gastroenterology 31 (2017) 97e104

Fig. 3. A: Occluded SMA (arrow) on CTA. B: CTA image after surgery showing the bypass (indicated with the arrow and the white short lines) from the left common iliac artery to the
SMA distally of the aneurysm. C: Recanalized SMA and rupture.

Fig. 4. A: Aneurysm of a distal branch of the SMA (arrow). B: Exclusion of the SMAA by a covered stent graft (arrows pointing out the start and end of the stent graft).
 M.J.E. van Rijn et al. / Best Practice & Research Clinical Gastroenterology 31 (2017) 97e104
Fig. 5. A: This CTA shows a large abscess (arrow) formation after ischemia of the spleen following coiling of a CAA. B: The abscess was drained (arrows) percutaneously over a period
of several weeks. The patient was also treated with antibiotics.
Treating CAA: endovascular repair
Embolization of the in- and outflow of the aneurysm may be
performed in patients without liver dysfunction and intact collateral
circulation from the SMA and gastroduodenal artery. Several studies
have investigated the risk of CA coverage during thoracic endovas-
cular aneurysm repair (TEVAR) [35,36]. These studies found
ischemic complications in 6e11%. All of these patients had adequate
collateral flow on preoperative imaging and in case of stenosis in the
SMA, a stent was placed prior to TEVAR. Thus, careful monitoring for
ischemic events remains mandatory after embolizing a CAA. Other
endovascular options are covered or multilayer stent grafts. Com-
plications are foregut ischemia, gangrenous cholecystitis, liver and
spleen abscess (Illustration 4), dissection, coil migration, stent oc-
clusion and access and contrast related complications.
Other visceral artery aneurysms
True pancreaticoduodenal (PDAA), gastroduodenal (GDAA) or
inferior mesenteric artery aneurysms (IMAA) are extremely rare.
Small series describing these patients have shown rupture rates of
20e80% for GDAAs and 100% for PDAAs [9,23]. Mean size at time of
rupture has found to be as small as 9 mm (4e12 mm range)
[20].Therefore, it is advised that all of these aneurysms should be
treated regardless of size. Risk factors for GDAA are trauma, hyper-
tension and atherosclerosis. The PDA is the main collateral pathway
between the CA and the SMA. In case of a CA stenosis, blood flow is
increased in the PDA and this may cause a PDAA. The same theory
suggests that occlusion or stenosis of the SMA or CA could lead to the
formation of a GDAA [37]. Pseudo aneurysms in these arteries are
usually the result of pancreatitis. A review of 74 patients with GDAAs
showed that 52% of patients presented with a gastrointestinal
hemorrhage secondary to rupture. Only 7.5% of GDAAs remained
asymptomatic [37]. Reconstruction is usually technically difficult
and unnecessary if adequate collateral flow is present. Embolization
is then the first choice. IMAAs most commonly occur in the proximal
part of the artery, occur more often in men and are mainly due to
atherosclerosis. Rupture rates are unknown.
Summary
VAAs are rare and therefore hard to study. Their clinical signs
and symptoms, if any, are diverse and unspecific. The most com-
mon pathway is atherosclerosis. VAPAs usually result from iatro-
genic injury, infection or abdominal trauma. No RCTs have studied
rupture risk or compared treatment strategies and only one sys-
tematic review compared open and endovascular repair for SAAs.
103
Endovascular treatment seemed to be associated with lower mor-
tality, however, there were more iSAAs in this group compared with
the open group and all studies included in the review were small,
retrospective studies. All other evidence comes from postmortem,
arteriographic and retrospective studies, so no strong recommen-
dations can be made. Some studies have found an increased risk of
rupture for SAA in pregnant women and in patients post liver
transplantation. Elective repair is recommended in these patients
as well as if the diameter exceeds 2.0 cm, the aneurysm grows or is
symptomatic. The threshold of 2.0 cm is based on retrospective
series and the evidence for this is therefore low. Non-
atherosclerotic or multiple HAAs are recommended to be treated
as well as all PDAAs and GDAAs, regardless of size because of their
high risk of rupture. Open repair includes ligation, aneurysmor-
rhaphy or reconstruction with a graft, while endovascular treat-
ment includes embolization or using a stent graft and similar
success rates (up to 96%) have been reported. What treatment
strategy is chosen depends on the anatomical location and the
presence of collateral circulation as well as the medical condition of
the patient. An true evidence-based approach towards treating
VAAs remains a challenge for the clinician.
Financial disclosure
None.
Conflicts of interest
None.
Practice points
 VAAs have an incidence of 0.01e2% and the most com-
mon VAA is SAA (60%)
 Atherosclerosis is the most common pathway for VAA
 Risk of rupture may be related to gender, size, growth,
underlying disease, pregnancy and liver transplantation,
although the level of evidence is low
 Generally, VAAs >2 cm, that are symptomatic or growing
are treated, although there is little evidence supporting
this
 Multiple HAAs, PDAAs and GDAAs should be treated
regardless of size because of their high risk of rupture
 Treatment can be through open or endovascular repair
 VAPA is most common in the HA. VAPAs should always
be repaired
104 M.J.E. van Rijn et al. / Best Practice & Research Clinical Gastroenterology 31 (2017) 97e104
Research agenda
 RCTs comparing different types of treatment for VAAs
should be performed
 Detailed studies are necessary to define the behavior of
VAAs, especially the rate of growth and risk of rupture
 Long term follow-up studies are necessary to investigate
the effect of endovascular treatment of VAAs, especially
with respect to recanalization.
References
[1] Pulli R, Dorigo W, Troisi N, Pratesi G, Innocenti AA, Pratesi C. Surgical treat-
ment of visceral artery aneurysms: a 25-year experience. J Vasc Surg 2008;48:
334e42.
[2] Bedford PD, Lodge B. Aneurysm of the splenic artery. Gut 1960;1:312e20.
[3] Stanley JC, Wakefield TW, Graham LM, Whitehouse WM, Zelenock GB,
Lindenauer SM. Clinical importance and management of splanchnic artery
aneurysms. J Vasc Surg 1986;3:836e40.
[4] Abbas MA, Stone WM, Fowl RJ, Gloviczki P, Oldenburg WA, Pairolero PC, et al.
Splenic artery aneurysms: two decades experience at Mayo clinic. Ann Vasc
Surg 2002;16:442e9.
[5] Pitton MB, Dappa E, Jungmann F, Kloeckner R, Schotten S, Wirth GM, et al.
Visceral artery aneurysms: incidence, management, and outcome analysis in a
tertiary care center over one decade. Eur Radiol 2015;25:2004e14.
[6] Shukla AJ, Eid R, Fish L, Avgerinos E, Marone L, Makaroun M, et al. Contem-
porary outcomes of intact and ruptured visceral artery aneurysms. J Vasc Surg
2015;61:1442e7.
[7] Corey MR, Ergul EA, Cambria RP, English SJ, Patel VI, Lancaster RT, et al. The
natural history of splanchnic artery aneurysms and outcomes after operative
intervention. J Vasc Surg 2016;63:949e57.
[8] Gehlen JMLG, Heeren PAM, Verhagen PF, Peppelenbosch AG. Visceral artery
aneurysms. Vasc Endovasc Surg 2011;45:681e7.
[9] Hemp JH, Sabri SS. Endovascular management of visceral arterial aneurysms.
Tech Vasc Interv Radiol 2015;18:14e23.
[10] Dave SP, Reis ED, Hossain A, Taub PJ, Kerstein MD, Hollier LH. Splenic artery
aneurysm in the 1990s. Ann Vasc Surg 2000;14:223e9.
[11] Pilleul F, Beuf O. Diagnosis of splanchnic artery aneurysms and pseudoa-
neurysms, with special reference to contrast enhanced 3D magnetic reso-
nance angiography: a review. Acta Radiol 2004;45:702e8.
[12] Tulsyan N, Kashyap VS, Greenberg RK, Sarac TP, Clair DG, Pierce G, et al. The
endovascular management of visceral artery aneurysms and pseudoaneur-
ysms. J Vasc Surg 2007;45:276e83. discussion 83.
[13] Belli AM, Markose G, Morgan R. The role of interventional radiology in the
management of abdominal visceral artery aneurysms. Cardiovasc Interv
Radiol 2012;35:234e43.
[14] Sessa C, Tinelli G, Porcu P, Aubert A, Thony F, Magne JL. Treatment of visceral
artery aneurysms: description of a retrospective series of 42 aneurysms in 34
patients. Ann Vasc Surg 2004;18:695e703.
[15] Sachdev U, Baril DT, Ellozy SH, Lookstein RA, Silverberg D, Jacobs TS, et al.
Management of aneurysms involving branches of the celiac and superior
mesenteric arteries: a comparison of surgical and endovascular therapy. J Vasc
Surg 2006;44:718e24.
[16] Batagini NC, El-Arousy H, Clair DG, Kirksey L. Open versus endovascular
treatment of visceral artery aneurysms and pseudoaneurysms. Ann Vasc Surg
2016 Aug;35:1e8.
[17] Ikeda O, Tamura Y, Nakasone Y, Iryou Y, Yamashita Y. Nonoperative man-
agement of unruptured visceral artery aneurysms: treatment by transcatheter
coil embolization. J Vasc Surg 2008;47:1212e9.
[18] Lagana D, Carrafiello G, Mangini M, Dionigi G, Caronno R, Castelli P, et al.
Multimodal approach to endovascular treatment of visceral artery aneurysms
and pseudoaneurysms. Eur J Radiol 2006;59:104e11.
[19] Spiliopoulos S, Sabharwal T, Karnabatidis D, Brountzos E, Katsanos K,
Krokidis M, et al. Endovascular treatment of visceral aneurysms and pseu-
doaneurysms: long-term outcomes from a multicenter European study. Car-
diovasc Interv Radiol 2012;35:1315e25.
[20] Tetreau R, Beji H, Henry L, Valette PJ, Pilleul F. Arterial splanchnic aneurysms:
presentation, treatment and outcome in 112 patients. Diagn Interv Imaging
2016;97:81e90.
[21] Lakin RO, Bena JF, Sarac TP, Shah S, Krajewski LP, Srivastava SD, et al. The
contemporary management of splenic artery aneurysms. J Vasc Surg 2011;53:
958e65.
[22] Berceli SA. Hepatic and splenic artery aneurysms. Semin Vasc Surg 2005;18:
196e201.
[23] Sachdev-Ost U. Visceral artery aneurysms: review of current management
options. Mt Sinai J Med 2010;77:296e303.
[24] Ohta M, Hashizume M, Ueno K, Tanoue K, Sugimachi K, Hasuo K. Hemody-
namic study of splenic artery aneurysm in portal hypertension. Hepatogas-
troenterology 1994;41:181e4.
[25] Sadat U, Dar O, Walsh S, Varty K. Splenic artery aneurysms in pregnancyea
systematic review. Int J Surg 2008;6:261e5.
[26] Lu M, Weiss C, Fishman EK, Johnson PT, Verde F. Review of visceral aneurysms
and pseudoaneurysms. J Comput Assist Tomogr 2015;39:1e6.
[27] Sinwar PD. Overwhelming post splenectomy infection syndrome e review
study. Int J Surg 2014;12:1314e6.
[28] Chiaradia M, Novelli L, Deux JF, Tacher V, Mayer J, You K, et al. Ruptured
visceral artery aneurysms. Diagn Interv Imaging 2015;96:797e806.
[29] Jana M, Gamanagatti S, Mukund A, Paul S, Gupta P, Garg P, et al. Endovascular
management in abdominal visceral arterial aneurysms: a pictorial essay.
World J Radiol 2011;3:182e7.
[30] Hogendoorn W, Lavida A, Hunink MGM, Moll FL, Geroulakos G, Muhs BE, et al.
Open repair, endovascular repair, and conservative management of true
splenic artery aneurysms. J Vasc Surg 2014;60:1667. U653.
[31] Hogendoorn W, Lavida A, Hunink MGM, Moll FL, Geroulakos G, Muhs BE, et al.
Cost-effectiveness of endovascular repair, open repair, and conservative
management of splenic artery aneurysms. J Vasc Surg 2015;61:1432e40.
[32] Abbas MA, Fowl RJ, Stone WM, Panneton JM, Oldenburg WA, Bower TC, et al.
Hepatic artery aneurysm: factors that predict complications. J Vasc Surg
2003;38:41e5.
[33] Stone WM, Abbas M, Cherry KJ, Fowl RJ, Gloviczki P. Superior mesenteric
artery aneurysms: is presence an indication for intervention? J Vasc Surg
2002;36:234e7. discussion 7.
[34] Stone WM, Abbas MA, Gloviczki P, Fowl RJ, Cherry KJ. Celiac arterial aneu-
rysms: a critical reappraisal of a rare entity. Arch Surg 2002;137:670e4.
[35] Mehta M, Darling 3rd RC, Taggert JB, Roddy SP, Sternbach Y, Ozsvath KJ, et al.
Outcomes of planned celiac artery coverage during TEVAR. J Vasc Surg
2010;52:1153e8.
[36] Rose MK, Pearce BJ, Matthews TC, Patterson MA, Passman MA, Jordan WD.
Outcomes after celiac artery coverage during thoracic endovascular aortic
aneurysm repair. J Vasc Surg 2015;62:36e42.
[37] Habib N, Hassan S, Abdou R, Torbey E, Alkaied H, Maniatis T, et al. Gastro-
duodenal artery aneurysm, diagnosis, clinical presentation and management:
a concise review. Ann Surg Innov Res 2013;7:4.