Best Practice & Research Clinical Gastroenterology 31 (2017) 15e25

Contents lists available at ScienceDirect

Best Practice & Research Clinical Gastroenterology

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

Acute mesenteric ischemia (part I) e Incidence, etiologies, and how to

improve early diagnosis
rkka
Jussi M. Ka inen, MD, PhD Specialist in Vascular and Gastrointestinal Surgery a, *,
Stefan Acosta, MD, PhD Professor of Vascular Surgery b
a
Heart Center, Kuopio University Hospital, P.O. Box 100, 70029 Kuopio, Finland
b
Department of Clinical Sciences Malmo, Lund University, Sweden

a b s t r a c t
Keywords: Acute mesenteric ischemia (AMI) is generally thought to be a rare disease, but in fact, it is more common
Acute mesenteric ischemia cause of acute abdomen than appendicitis or ruptured abdominal aortic aneurysm in patients over 75
Acute on chronic mesenteric ischemia
years of age. In occlusive AMI, surgical treatment without revascularization is associated with as high as
Nonocclusive mesenteric ischemia
Mesenteric venous thrombosis
80% overall mortality. It has been shown that early diagnosis with contrast-enhanced computed to-
Etiology mography and revascularization can reduce the overall mortality in AMI by up to 50%. However, only a
Incidence minority of patients with AMI are being treated actively with revascularization in the United States, and
Diagnosis the situation is very likely similar in Europe as well. What can we do to improve diagnostic performance,
Computed tomography so that more patients get proper treatment? The diagnosis is a collaborative effort of emergency
department surgeons, gastrointestinal and vascular surgeons, and radiologists. The etiological categori-
zation of AMI should be practical and guide the therapy. Furthermore, the limitations of the diagnostic
examinations need to be understood with special emphasis on computed tomography ndings on pa-
tients with slowly progressing acute-on-chronic mesenteric ischemia.
2016 Elsevier Ltd. All rights reserved.

Introduction multi-slice contrast-enhanced computed tomography (CT), the

In 1926, A.J. Cokkinis wrote the diagnosis is impossible, the
prognosis hopeless and the treatment useless [1]. Ninety years later,
it is still a common presumption that acute mesenteric ischemia
(AMI) is a rare condition which inescapably leads to the death of the
patient. The reason for such dreary reputation is not so much based
on facts but because AMI is too often found too late when the treat-
ment outcome is inevitably poor. In the early 1990s in Finland, two-
thirds of patients with AMI were treated with mere surgical explo-
ration or comfort care resulting in certain death. With good luck, in
one-third of the cases, the patient could be treated with bowel
resection yielding 50% survival. Open surgical revascularization was
attempted in only 7% of the cases and the results were discouraging.
The overall mortality of patients with AMI was more than 80% [2].
What has changed in the past two decades? The most important
evolutionary step is that today, we have the ability to perform
* Corresponding author.
E-mail addresses: jkarkkai@gmail.com (J.M. K
arkk
ainen), stefan.acosta@med.lu.
se (S. Acosta).
most important diagnostic examination in AMI, to nearly all pa-
tients with acute abdominal pain at any given time. Second, our
endovascular capabilities have taken a leap from the conventional
time-consuming catheter-directed thrombolysis to mechanical
thrombectomy using dedicated aspiration catheters, and to utiliz-
ing stents that are designed especially for visceral arteries. We have
seen that with early diagnosis and treatment, more than half of
patients with AMI can be rescued [3]. In the United States, endo-
vascular treatment has become nearly as common therapeutic
approach as open revascularization in AMI according to studies
based on the Nationwide Inpatient Sample (NIS) by Schermerhorn,
Lo, and colleagues [4,5]. The in-hospital mortality of American
patients undergoing open or endovascular repair for AMI declined
from 51% in year 1995 to 26% in 2010 [5].
However, what does not seem to have changed, is that even
today, according to NIS, the overall revascularization rate in AMI
was no more than 6% in the year 2010 in the United States [5]. In
another study, Beaulieu and co-workers found a total of 23744
hospital admissions for AMI registered in the NIS database from
2005 through 2009. At that time, only 3% received an attempt at
open (n 514) or endovascular (n 165) revascularization, while

http://dx.doi.org/10.1016/j.bpg.2016.10.018
1521-6918/ 2016 Elsevier Ltd. All rights reserved.
16 rkka
J.M. Ka inen, S. Acosta / Best Practice & Research Clinical Gastroenterology 31 (2017) 15e25

17% were treated with bowel resection alone, and an alarming rate
of 80% received no intervention whatsoever [6]. There are no other
large population based data of the revascularization rate in AMI, but
we can only assume that not all patients in Europe, either, get
timely diagnosis and proper treatment.
To overcome the diagnostic-related challenges in AMI, we must
rst acknowledge that AMI is not a rare entity but actually a quite
common condition in elderly patients. Second, we must understand
the complex pathophysiology and diverse clinical presentation of
the disease. The CT signs at early stages of AMI are often subtle and
difcult to detect, and therefore, the key to diagnosis is clinical
suspicion. Third, the modern treatment of AMI requires a multi-
disciplinary team of gastrointestinal surgeons, vascular surgeons,
and interventional radiologists. We need a practical etiological
categorization of AMI that will guide the treatment and a simple
algorithm for the various treatment options in different situations.

Incidence

Between 1970 and 1982 in Malmo (Sweden), autopsies were

conducted on as many as 87% of the deceased in the population of Fig. 1. The age-related incidence rates of acute mesenteric ischemia (AMI), ruptured
abdominal aortic aneurysm (RAAA), acute pancreatitis, acute appendicitis and acute
approximately 250 000 inhabitants [7]. According to this data, the
cholecystitis in Kuopio/Finland between the years 2009 and 2013 [13].
annual incidence of AMI, diagnosed at autopsy or operation, was 12
per 100 000 inhabitans. The distribution of etiology in AMI was
In the United States, the National Health and Nutrition Examination
roughly 2/3 thromboembolic occlusive mesenteric ischemia, 1/6
Survey demonstrated that warfarin use has grown 2.5 times and
non-occlusive mesenteric ischemia (NOMI), and 1/6 mesenteric
statin use has almost tripled over a decade from 1999 to 2010 [5]. The
venous thrombosis (MVT). Thus, the most common cause of AMI
increased use of oral anticoagulants may have contributed to the
was acute occlusion of the superior mesenteric artery (SMA) with
decrease in the incidence of embolic events. Between 1970 and 1982
incidence rate of 8.6/100 000/year. The SMA occlusion was caused
in Malmo , SMA embolism was the most common cause of AMI with
by embolism in 70% and thrombosis in 30% of the cases [8]. The
embolism-to-thrombosis ratio of 1.4:1 [16]. From 1993 to 2000,
incidence rate of fatal NOMI was given as 2.0/100 000/year, and the
Endean et al. reported embolism-to-thrombosis ratio of 1:1 in 58
incidence rate of MVT with intestinal necrosis was estimated at 1.8/
cohort American patients with thromboembolic AMI [17] while Ryer et al.
100 000/year [9,10]. An interesting nding in the Malmo
reported 0.6:1 ratio in 78 patients between 1990 and 2010 [18].
was that the incidence of acute SMA occlusion was 1.5 times higher
Although there is no denitive proof of any change in the etiological
than the incidence of ruptured abdominal aortic aneurysm [7].
spectrum of AMI over time, it would seem that atherosclerotic
occlusive disease is currently the most common cause of AMI [19].
Recent data on the incidence of AMI

Due to current low autopsy rates, there is no recent population-
based data that would be comparable to the Malmo cohort. The
analysis of the NIS registry indicated that the incidence of AMI
declined from 8.4 to 6.7/100 000/year between the years 1995 and
2010 in the United States [5]. Similarly, in two contemporary
Swedish series of AMI, the reported incidence rates of acute oc-
clusion of the SMA were lower than during the Malmo autopsy
study; between 5.3 and 5.4/100 000/year [11,12]. From year
2009e2013 in Kuopio (Finland), practically all patients with acute
abdomen from a well-dened population of 250 000 inhabitants
were treated in one institution (Kuopio University Hospital); in
Kuopio hospital area, the incidence rate of AMI was 7.3/100 000/
year for all etiologies and 4.5/100 000/year for occlusive AMI [13].
The incidence of AMI increased exponentially with age (Fig. 1). In
patients aged over 75 years, AMI was more prevalent cause of acute
abdomen than appendicitis (Table 1).
Changes in cardiovascular risk factors over time
AMI is a disease of the elderly, and the population is aging in
Finland, Sweden, United States, and many other western countries.
Aging of the population means more burden from cardiovascular
diseases. However, interestingly, the incidence of AMI does not seem
to have increased. On the contrary, it has been shown that the inci-
dence of cardiovascular events have declined, at least in Finland, and
the prevalence of cardiovascular risk factors (smoking, serum total
cholesterol, and systolic blood pressure) have also decreased [14,15].
Prevalence of asymptomatic mesenteric artery stenosis
A patient with acute abdominal pain and chronic calcied oc-
clusion of the SMA represents a special challenge for the clinician.
Does the patient have AMI or is the SMA occlusion just an incidental
nding? The prevalence of asymptomatic mesenteric artery ste-
nosis has been reported as 6e29% depending on the study [20].
Unfortunately, the studies vary a great deal in terms of the popu-
lation (e.g. American, European, Korean), the denition of mesen-
teric artery stenosis (e.g. the numbers of celiac artery (CA) and SMA
stenoses are often merged and not given as separate values), the
grade of the stenosis included (ranging from 1 to 100%, 50e100% or
70e100%), and the methods of assessing the stenosis (ultrasonog-
raphy, angiography, CT, or autopsy). What we really would like to
know, is 1) the prevalence of SMA occlusion or hemodynamically
signicant (70%) SMA stenosis, especially in the aged population,
and 2) how many of those people have signicant concomitant CA
and inferior mesenteric artery (IMA) obstruction (i.e. 2- or 3-vessel
disease). In addition, 3) the natural outcome of chronic SMA oc-
clusion is of great interest.
The current data on the prevalence of asymptomatic mesenteric
artery stenosis is listed in Table 2. Based on the available data, the
following conclusions can be drawn:
 The prevalence of a hemodynamically signicant (>70%) SMA
stenosis is approximately 2% in elderly patients aged roughly 70
years or more [21,22].
 rkka
J.M. Ka inen, S. Acosta / Best Practice & Research Clinical Gastroenterology 31 (2017) 15e25 17

Table 1
Incidence rates (IR; n/100 000/year) of acute abdomen diagnoses for two age groups in Kuopio/Finland. This table contains unpublished data derived from previously published
studies [13,19].

Age group 75 years >75 years

Mean population 226290 inhabitants 21894 inhabitans

n/years IR 95% CI n/years IR 95% CI

AMI (all etiologies) 34/5y 3.0 2.0e4.0 56/5y 51.2 37.8e64.6

Occlusive AMI 15/5y 1.3 0.6e2.0 41/5y 37.5 26.0e48.9
Embolism 5/5y 0.4 0.1e0.8 16/5y 14.6 7.5e21.8
Atherosclerosis/thrombosis 10/5y 0.9 0.3e1.4 25/5y 22.8 13.9e31.8
NOMI 10/5y 0.9 0.3e1.4 15/5y 13.7 6.8e20.6
Acute MVT 9/5y 0.8 0.3e1.3 0/5y e
Other 3/5y 0.3 e 0/5y e
RAAA 25/5y 2.2 1.3e3.1 23/5y 21.0 12.4e29.6
Acute appendicitis 1042/5y 92.1 86.5e97.7 44/5y 40.2 28.3e52.1
Acute cholecystitis 169/2y 37.3 31.7e42.9 104/2y 238 192e283
Acute pancreatitis 193/2y 42.6 36.6e48.6 33/2y 75.4 49.6e101.1

CI, condence interval, AMI, acute mesenteric ischemia; NOMI, non-occlusive mesenteric ischemia; MVT, mesenteric venous thrombosis; RAAA, ruptured abdominal aortic
aneurysm.

Table 2
The current data on the prevalence of asymptomatic mesenteric artery stenosis (based on a Pubmed search of English literature including the past 20 years).

 In an autopsy cohort of 120 unselected Finnish subjects (mean age 62, range 18e93 years), four (3%) had >50% stenosis of the CA but none had SMA stenosis [24].
 In asymptomatic British patients undergoing abdominal ultrasonography (for unknown reasons), 18% of patients aged 65 years or more (n 97) had >70% stenosis of CA
or SMA (the rate of SMA stenosis was not reported). In patients under 65 years (n 87), only 3% had CA stenosis and none had SMA stenosis [25].
 CA stenosis rate and morphology was investigated in 400 Korean patients (mean age 57, range 17e85 years) referred for chemoembolization of hepatic tumors. More
than 50% CA stenosis with at least 10 mmHg gradient, based on angiography and CT, was found from 29 patients (7%). The cause of the stenosis was compression by
median arcuate ligament in 16, atherosclerosis in 3, and indeterminate in 10 patients [26].
 In 242 Dutch patients (median age 65 years) undergoing elective colorectal surgery, >50% CA stenosis was found in 21%, SMA stenosis in 3%, and IMA stenosis in 25%
based on preoperative CT [27].
 In 553 asymptomatic elderly American patients (mean age 77), 16% had CA stenosis and 2% had SMA stenosis dened as >70% by ultrasonography. Half of those with
SMA stenosis had also concomitant CA stenosis. None developed AMI during mean 6.5 years of follow-up [21,22].
 Out of 980 American patients with aortograms obtained for undisclosed reasons, 72 patients (mean age 68 years, range 31e95) with >50% mesenteric artery stenosis
were followed for 1e6 years. Four out of fteen patients with 3-vessel disease developed AMI, and 13 (86%) either developed AMI, had abdominal symptoms, or died
during the follow-up [23].

 In the elderly, 50e80% with SMA stenosis have concomitant
disease of the CA [21e23].
 Isolated CA stenosis is much more common than SMA steno-
sis, and the etiology of isolated CA stenosis is more often
external compression by median arcuate ligament than
atherosclerosis especially in young people [24e27]. Isolated
CA stenosis is usually asymptomatic with the rare exception of
median arcuate ligament compression syndrome (MALS). In
one study, only three of 21 patients with MALS had abdominal
symptoms [28].
 There are only little data on the natural outcome of asymptomatic
mesenteric artery stenosis. It seems that isolated CA stenosis is not
a signicant risk factor for AMI. However, concomitant obstruc-
tion of SMA, CA and IMA is a major risk factor for symptomatic
mesenteric ischemia. In one study, as many as 86% of patients
with 3-vessel disease could be dened as symptomatic [23].
Etiologies and clinical presentation
The classical etiological categorization of AMI
Conventionally, AMI is stratied into four groups by etiology;
arterial embolism, arterial thrombosis, venous thrombosis and
NOMI [29]. Chronic atherosclerotic obstruction of the mesenteric
arteries is traditionally associated with chronic mesenteric
ischemia (CMI). This classical categorization has not evolved since
the time when angiography was the gold standard diagnostic
investigation and bowel resection was the main therapeutic option.
Owing to modern CT technology, we now get more information on
the morphology of the arterial obstruction (Fig. 2AeD) and on the
degree of bowel injury in AMI [19]. A practical categorization
should guide the modern therapy. Therefore, the etiological cate-
gorization of AMI may need to be updated.
Controversies regarding current classication of AMI
The denition of arterial versus venous mesenteric ischemia is
clear-cut (Fig. 3). MVT is the usual cause of venous mesenteric
ischemia, and MVT is easy to detect in contrast-enhanced CT.
However, with early anticoagulation treatment, bowel ischemia is
often avoided in MVT. In other words, MVT does not automatically
result in venous mesenteric ischemia [19,29].
Arterial AMI is dened as either occlusive or non-occlusive (i.e.
NOMI). This categorization is rather controversial. By denition,
occlusive means 100% occlusion of the SMA whereas some might
categorize 99% stenosis as non-occlusive. In modern practice, the
cut-off between total and near-total occlusion is impractical. First,
the differentiation of 100% versus 99% obstruction is often
impossible based on CT (as in, for example, Fig. 2C); nevertheless,
the clinician has to choose the initial therapeutic approach based
on the CT ndings. Second, if the patient has AMI, revasculari-
zation is needed in both cases; total occlusion or severe stenosis.
It is generally accepted that 70% stenosis of the SMA is hemody-
namically signicant, and patients with concomitant severe
obstruction of the CA and IMA are at risk of developing AMI upon
chronic SMA stenosis; these cases should be categorized as
occlusive [19,30,31].
In most hospitals, it is the emergency department surgeon or
the on-duty gastrointestinal surgeon who has to make the deci-
sion whether to consult a vascular surgeon or an interventional
18 rkka
J.M. Ka inen, S. Acosta / Best Practice & Research Clinical Gastroenterology 31 (2017) 15e25
Fig. 2. The different morphologies of SMA obstruction in patients with AMI. A)
Embolic occlusion (arrowhead) of the SMA in a 94-year old patient. B) Thrombotic
occlusion (arrowhead) on superimposed calcied stenosis of the SMA. Incidentally, this
SMA thrombosis proved to be chronic occlusion based on an earlier CT scan. The pa-
tient had AMI with bowel necrosis nevertheless. C) 99% atherosclerotic stenosis
(arrowhead) of the SMA without thrombosis; a narrow patent lumen was demon-
strated in angiography. D) Chronic calcied (previously asymptomatic) occlusion
(arrowhead) of the SMA that could not be recanalized by endovascular means.
Fulminant bowel and liver ischemia manifested after elective nephrectomy and the
patient was successfully treated with surgical SMA bypass and bowel resection. Cases
AeC were treated by endovascular means, and all four patients survived.
radiologist, or whether the treatment should go to another di-
rection without revascularization. The assessment of mesenteric
artery stenosis and collateral circulation, based on CT, is
demanding even for vascular surgeons and radiologists, and may
be impossible for the gastrointestinal or the acute care surgeons
who are not accustomed to interpret vascular images on a daily
basis. Therefore, it is practical that patients who require revas-
cularization as the primary treatment in AMI are categorized as
occlusive mesenteric ischemia (Fig. 3). In NOMI, the primary
treatment is usually conservative (unless bowel necrosis has
already developed).
Embolism
Acute occlusion of the SMA is the most common cause of AMI. In
roughly half of the cases, the acute occlusion is caused by embolism
that is usually of cardiac origin [17,19]. Typically, SMA embolism is
characterized by acute onset of symptoms; the abdominal pain is
severe without any localization. Pain out of proportions to the
clinical signs is often used to describe the early clinical presenta-
tion in acute occlusion of previously healthy SMA. However, the
clinical presentation and severity of the disease varies a great deal
depending on individual anatomy and location of the occlusion in
the mesenteric arterial tree [32]. In an autopsy study of 122 patients
with embolic AMI, the blood clot was located near the aortic ostium
of the SMA in 16%, in the root of the SMA in 39%, and in peripheral
branches in 16% of the cases (the clot location was unclassiable in
29%) [16].
In sudden embolic occlusion of the SMA root, most part of the
small intestine and right-side of the colon suffers from severe
ischemia, and the patient cannot usually survive without revas-
cularization. In about 30% of the cases, the SMA embolus is
located distal to the middle colic artery, in which case, the patient
might survive even without revascularization but may still need
massive bowel resection [16]. In embolic occlusion of a small
SMA branch, removal of the embolus is usually not possible
or necessary; the treatment is simple bowel resection of the
ischemic segment.
The risk factors for SMA embolism are atrial brillation, recent
myocardial infarction, congestive heart failure, and prior embolic
events (e.g. stroke) [32]. In addition, a new threat of mesenteric
artery embolism has emerged in conjunction with the development
of endovascular aortic repair technique extending above the
infrarenal segment. An incidence rate of 5% for bowel ischemia was
reported in a recent series from London (U.K.) of 99 patients un-
dergoing fenestrated endovascular aortic repair [33]. This serious
complication with 80% mortality was associated with shaggy
aorta, i.e. aortic irregularity and increased thrombus volume in the
paravisceral segment, and was thought to be related to graft
manipulation and catheterisation of visceral vessels.
Mesenteric arterial occlusive disease
Thrombosis represents the other half of the cases of acute SMA
occlusion. Most often, acute thrombosis of the SMA is a complica-
tion of atherosclerotic occlusive disease. Thrombosis may also
develop upon isolated dissection of the SMA or aortic dissection
extending to the visceral arteries, but the natural history and the
treatment of these two conditions differ signicantly from the
atherosclerotic thrombosis, and therefore, both entities represent
their own distinct etiological category of AMI (not covered in this
review). Atherosclerotic thrombus usually forms in the proximal
SMA with superimposed high-grade stenosis or ruptured athero-
sclerotic plaque [17,19].
Unorthodoxly, we chose to rename the category of thrombosis
as mesenteric arterial occlusive disease (analogous to peripheral
arterial occlusive disease) (Fig. 3). Basically, this is a synonym for
atherosclerotic occlusive disease. In the conventional etiological
classication of AMI, the category of thrombosis is a bit contro-
versial, because the word thrombosis by itself includes the
assumption that the arterial occlusion in AMI is always acute.
However, it is not. A signicant number of AMI patients develop
acute intestinal ischemia upon chronic calcied occlusion (or se-
vere stenosis) of the mesenteric arteries; this is often referred to as
acute on chronic mesenteric ischemia in the literature [31]. In a series
of 37 patients with AMI caused by mesenteric arterial occlusive
disease, less than half presented with clearly visible thrombotic clot
in contrast-enhanced CT while the other half presented with
calcied obstruction of the SMA and other mesenteric arteries [19].
Thus, the atherosclerotic SMA obstruction in AMI can be either
acute or chronic, and thrombotic or non-thrombotic (i.e. calcied);
these are all manifestations of the atherosclerotic vascular disease
(Fig. 2BeD).
The clinical presentation of AMI caused by mesenteric arterial
occlusive disease is more varied than in embolic AMI depending on
the acuteness and extent of the arterial obstruction, and moreover,
on the compensatory blood ow from the collateral arteries. Pa-
tients with acute thrombotic occlusion of the SMA may present
with fulminant bowel ischemia, or the symptoms can be insidious
and obscure such as vague abdominal pain, diarrhea and vomiting.
 rkka
J.M. Ka inen, S. Acosta / Best Practice & Research Clinical Gastroenterology 31 (2017) 15e25 19

Fig. 3. A suggested practical etiological categorization of acute mesenteric ischemia. In occlusive mesenteric ischemia, the primary treatment approach is revascularization (red
color). In venous mesenteric ischemia and non-occlusive mesenteric ischemia, the initial treatment is usually conservative (blue color).

In slowly progressing mesenterial atherosclerosis, the collateral
arteries are often well developed and the bowel is accustomated to
withstanding ischemia, and therefore, it may take several days to
weeks before the disease evolves into irreversible bowel ischemia.
Among the 37 AMI patients with mesenteric arterial occlusive
disease, only 20% had symptoms for less than 24 hours while 40%
had more than three days of symptoms prior to the nal diagnosis
[19]. Even so, the prognosis was favorable after endovascular
revascularization especially in patients with slowly progressing
AMI. The mythical six-hour limit until irreversible bowel necrosis
occurs in AMI can go down to history. And even though time is
bowel, there is no general rule as to how long the bowel can stay
viable in atherosclerotic occlusive AMI.
Acute on chronic mesenteric ischemia
Prior history of CMI (postprandial abdominal pain, fear of eating,
and weight loss) is common in the acute-on-chronic presentation
pattern of AMI. These premonitory symptoms have been reported
prior to the nal admission in 25e84% of patients with AMI caused
by mesenteric arterial occlusive disease [3,31]. In a contemporary
series of 55 such patients, up to 80% were initially misdiagnosed and
inappropriately treated at hospital with medical therapy [31].
Some patients had even undergone exploratory laparotomy or
cholecystecomy for gastrointestinal issues without relief of symp-
toms before the disease culminated in AMI. Due to the variable na-
ture of mesenteric arterial occlusive disease, the diagnosis and the
presentation pattern of thrombotic AMI is much more complicated
than of embolic AMI. There are often other factors (besides the
obstruction of the mesenteric arteries) that contribute to the
development of AMI, such as dehydration, low cardiac output, ane-
mia, and hypercoagulable states. Sometimes, mere uid resuscita-
tion, correction of anemia, and administration of antibiotics may
reverse the acute ischemic process, but even so, the patient will
remain at risk of recurrent AMI unless treated with revascularization.
Acute-on-CMI should be suspected in patients with acute
abdominal pain, or slowly progressing abdominal symptoms, and
prior history of other manifestations of the atherosclerotic vascular
disease (i.e. coronary artery disease, peripheral arterial occlusive
disease, cerebrovascular disease). Patients with progressive symp-
toms of CMI are at high risk of developing AMI at any given day and
should be treated accordingly. The situation is comparable to acute
coronary syndrome. If a patient has chest pain and ischemic EKG
changes at rest, the risk of acute myocardial infarction is eminent.
Similarly, if abdominal pain persists after a meal, the patient may
already have gone into an irreversible state of acute intestinal
ischemia. Angiography and revascularization has to performed ur-
gently, regardless of whether the target is coronary or mesenteric
20 rkka
J.M. Ka inen, S. Acosta / Best Practice & Research Clinical Gastroenterology 31 (2017) 15e25
arteries. Inammatory markers (C-reactive protein) is often strik-
ingly elevated only after 24 hours of continuous symptoms in the
acute-on-chronic condition [34]. Metabolic acidosis indicates that
irreversible bowel necrosis may already have developed.
Non-occlusive mesenteric ischemia (NOMI)
In systemic circulatory failure, the blood ow is redistributed to
vital organs, and consequently, vasoconstriction of the mesenteric
arteries may cause severe intestinal hypoperfusion despite the
mesenteric arteries being patent. This is called NOMI. The potential
clinical scenarios in which NOMI can develop are severe heart failure,
hypovolemia, sepsis, use of vasoconstrictive medication (inotropes)
or intra-aortic balloon pump, hypotension caused by dialysis or
major surgery (especially cardiac or aortic surgery), and abdominal
compartment syndrome [35]. The diagnosis of NOMI can be
extremely challenging especially in patients that are intubated and
mechanically ventilated in the intensive care unit. In early NOMI, the
initial treatment is usually conservative and aims at restoring in-
testinal perfusion by treating the underlying condition. Worsening
metabolic acidosis and distended abdomen may indicate the need
for laparotomy. The role of imaging in NOMI is to rule out occlusive
mesenteric ischemia and to detect signs of bowel gangrene (Fig. 4). In
the Malmo autopsy cohort, there were 62 patients with fatal NOMI,
of whom 25 (40%) had stenosis of the SMA and 14 of those subjects
had concomitant stenosis of the CA [9]. Although there was no
indication whether or not those lesions were hemodynamically
signicant, in retrospect, it may be argued that those patients were
misclassied since NOMI essentially excludes signicant mesenteric
artery stenosis. Anyhow, endovascular stenting is an option to
consider whenever signicant SMA stenosis complicates NOMI.
Mesenteric venous thrombosis (MVT)
MVT is dened as acute thrombosis of the superior mesenteric
vein and its branches with or without extension of the thrombus to
the portal vein (Fig. 5A). Intestinal ischemia rarely develops unless
the peripheral veins of the mesentery (venous arcades and vasa
recta) are involved. Thus, isolated portal vein thrombosis seldom
causes bowel infarction. In acute MVT, mild intestinal oedema may
gradually develop into arterial spasm and transmural bowel
infarction within days to weeks. A subacute form of MVT de-
velops gradually within several weeks and the pain may be mild,
sometimes accompanied by (bloody) diarrhea. In chronic MVT,
Fig. 4. A young obese patient with severe respiratory failure due to unknown respiratory infection. CT showed dilated colon (white arrowheads), diminutive aorta (black
arrowhead), and narrow SMA (white open arrowhead) as a sign of severe NOMI, mesenteric vasoconstriction and septic shock. In laparotomy, the entire colon appeared to be
gangrenous and the colon was resected following damage control principles. Unfortunately, the patient died a month later due to many complications associated with prolonged
invasive ventilation in the intensive care unit.
complications such as portal hypertension and hypersplenism may
develop several months after the acute thrombotic event
(Fig. 5BeD) [36]. The treatment of acute MVT aims at ensuring
blood return from the intestine until sufcient venous collaterals
develop. The therapeutic options range from simple anti-
coagulation to invasive catheter-directed thrombolysis, and very
rarely, surgical venous thrombectomy may be needed. During the
course of treatment, the superior mesenteric vein either recanalizes
or obliterates.
Diagnostic examinations
Laboratory tests
Classically, patients with AMI have leukocytosis, metabolic
acidosis, elevated D-dimer, and elevated serum lactate. The chal-
lenge is that the conventional laboratory tests are very unspecic at
presentation such as leukocytosis or D-dimer. None of them can be
used to rule in the possibility of AMI. D-dimer may be used as an
exclusion test, where a negative test very likely excludes acute
thromboembolic occlusion of the SMA at presentation [37,38].
Arterial lactate is often negative in the early phase of AMI. In a study
of 50 AMI patients, half had negative lactate values at admission [3].
The arterial lactate level rises when the liver is unable to clear it
anymore, and when the test is positive, there may already be
advanced irreversible bowel ischemia without possibilities to save
the ischemic bowel with revascularization [39]. A biomarker
capable of detecting early hypoperfusion in AMI, when irreversible
bowel injury has not yet developed, would be of great clinical value.
Unfortunately, there are currently no standardized blood tests that
could be used widely in patients with acute abdominal pain to
screen for AMI in such way as the troponin test is used for screening
acute myocardial infarction in patients with acute chest pain [38].
Known pitfalls at admission are referring AMI patient with
elevated troponin test to a cardiologist or making incorrect diag-
nosis because of elevated pancreas amylase or abnormal liver test.
These abnormal laboratory tests are relatively common in AMI [40].
In Kuopio (Finland), nearly 30% of patients with AMI were rst
referred to an internist at the emergency department due to cardiac
comorbidities or obscure symptoms (e.g. diarrhea and vomiting)
[3]. A few patients were rst seen by a neurologist either because of
simultaneous embolic stroke and SMA embolism, or transient
ischemic attack and acute-on-CMI. Thus, AMI is not exclusively
surgeons' headache.
 rkka
J.M. Ka inen, S. Acosta / Best Practice & Research Clinical Gastroenterology 31 (2017) 15e25 21

Fig. 5. A) This previously healthy 40-year old patient with acute MVT (white arrowheads) and oedematous small bowel segment (white open arrowheads) was treated with
anticoagulation alone for six months. The patient was screened negative for hypercoagulable disorders, and the MVT was determined as idiopathic. Unfortunately, ve years later,
although the patient had patent MVT (B), the portal vein was occluded with cavernomatous transformation of the hepatoduodenal ligament (C; black arrowheads). The patient
developed severe portal hypertension and hypersplenism (B; asterisk) with pancytopenia and oesophageal varices (D). The liver function remained normal.

Imaging
CT has replaced conventional angiography as the gold standard
imaging modality in AMI. Digital subtraction angiography is an
invasive procedure that is currently used to perform therapeutic
interventions more commonly than just plain diagnostics. The
utility of duplex ultrasound is limited to screening for CMI. Ultra-
sound is not recommended in AMI (if CT is available). It is time
consuming and of limited diagnostic value; visualizing the
mesenteric vessels is often obscured by extensive bowel gas that is
associated with intestinal paralysis in AMI. Gadolinium enhanced
magnetic resonance imaging can be used to detect proximal
obstruction of the SMA and CA, but it has less value in the evalu-
ation of distal arterial occlusions and other intestinal ndings in
AMI [41]. Moreover, magnetic resonance imaging takes too much
time in a severely ill patient. Plain radiographs are not useful in
AMI; gas in the portal veins is a sign of extensive intestinal pneu-
matosis and implies that bowel necrosis has developed.
Accuracy of CT
When AMI is suspected, CT should be performed with contrast
enhancement in arterial and venous phases (the biphasic protocol)
[41]. The arterial phase (with 1 mm slice thickness) enables accu-
rate detection of vascular pathology and detailed reconstructions of
the main mesenteric arteries and even collaterals. The venous
phase is required for the assessment of bowel wall and solid organ
perfusion, and other pathology. The sensitivity and specicity of
biphasic CT in AMI has been estimated as 89e100% in two sys-
tematic reviews [42,43]. However, this may be an overestimate. A
closer look at the individual studies in the two reviews revealed
that the studies were performed exclusively in patients with clin-
ical suspicion of AMI, and roughly 70e100% of the study patients
had advanced bowel ischemia (i.e. AMI that was diagnosed at
laparotomy or autopsy, or withdrawal of active treatment that
resulted in rapid death) [44e50]. In practice, AMI is rarely sus-
pected prior to imaging. Furthermore, the goal is to detect AMI
before advanced bowel ischemia has developed. At early phase of
the disease, CT ndings are much more subtle and difcult to
detect. In two recent studies from Sweden and Finland with total of
147 AMI patients, the suspicion of AMI was mentioned in the CT
referral in only one-third of the cases and more than 70% of the CTs
were performed in venous phase alone [19,51]. The sensitivity of CT
in AMI was no more than 67e85% in these two studies that were
performed in the clinical routine in patients with unclear acute or
subacute abdomen; half of the patients with AMI had advanced
bowel ischemia (dened as above) whereas the other half had
reversible bowel ischemia.
Thus, without clinical suspicion of vascular disease, the routine
CT of the acute abdomen will be performed in venous phase alone
[52]. Nevertheless, most of the vascular and intestinal ndings can
be detected from the venous phase images. What seems to be more
important than the contrast enhancement protocol is clinical sus-
picion, which is a major factor in the correct interpretation of the CT
ndings in AMI [19,51]. A study of 97 AMI events from Finland
demonstrated that the initial CT report was correct in 97% of the
cases where the clinician had mentioned AMI suspicion in the CT
referral, and where not, the corresponding rate was only 81%
(p 0.04). Also, the need for bowel resection was signicantly
higher in the latter group [19]. Direct inquiry for mesenteric
22 rkka
J.M. Ka inen, S. Acosta / Best Practice & Research Clinical Gastroenterology 31 (2017) 15e25

ischemia encourages the radiologist to examine and report even
the faintest intestinal and vascular abnormalities. However, it
should be mandatory that the radiologists assess and report the
patency of the mesenteric vessels routinely in the abdominal CT
evaluation, not only when asked to.
Ischemia-specic CT ndings in AMI
CT ndings in AMI can be either specic or non-specic for
bowel ischemia (Table 3). Decreased or absent bowel wall enhance-
ment is probably the most specic CT sign of intestinal ischemia
with 96% specicity, but it has very low sensitivity (16e62%) [52].
Sometimes, the bowel wall may become abnormally hyper-
enhanced in AMI, for example, during reperfusion conditions
following occlusive AMI or NOMI. Impaired venous drainage of
contrast medium in MVT may also cause increased enhancement of
the bowel wall. Pneumatosis intestinalis have been reported in
6e28% of cases with AMI and portomesenteric venous gas in 3e14%
[52]. Although pneumatosis is very specic for bowel ischemia, it
has low sensitivity, and pneumatosis does not yet prove that the
bowel damage is irreversible.
Non-specic CT ndings in AMI
A very common but probably the least specic nding in AMI is
bowel wall thickening. In contrast, the bowel wall may become
dilated and paper thin in transmural infarction [53]. Mesenteric fat
stranding is caused by oedema of the mesentery, and a small
amount of ascites can be found in many patients with AMI. Another
very common but unspecic CT nding is luminal dilatation of the
small (or large) bowel which is caused by intestinal paralysis in
AMI.
The vascular CT ndings are not direct signs of bowel ischemia
but represent the etiology in AMI. An embolic or thrombotic clot of
the SMA usually suggests acute arterial occlusion, and therefore, is
considered a strong sign of AMI. Chronic occlusive mesenteric
atherosclerosis is not a specic nding and may be incidental.
However, if the SMA and the other mesenteric arteries are severely
obstructed, and there are associated intestinal ndings (specic or
unspecic), AMI is a likely cause of acute abdominal pain. Thus, as
discussed before, the chronic state may eventually become acute
without visible thrombosis in CT. Sometimes, the acute thrombotic
clot is so small that it cannot be visualized by imaging [19].
Intraluminal lling defect and engorgement of the superior
mesenteric vein together with mesenteric oedema are indications
of acute MVT in a patient with acute abdominal pain. In NOMI, the
CT ndings may be minimal; the signs to look for are those related
to bowel necrosis. Sometimes, irregular narrowing (i.e. vasospasm)
of the SMA and mesenteric arcades may be visible in NOMI, but it is
difcult to detect in CT (Fig. 4). Diminutive aorta and inferior vena
cava in CT are signs of severe systemic shock in NOMI [53,54].
Synchronous embolism (e.g. in arm, leg, brain, or abdominal
solid organs) was found in 68% of patients with SMA embolism in
the Malmo autopsy study [16]. Synchronous embolism within the
artery of spleen, liver or kidney may not be visible in CT, but
sometimes it can be detected as uneven perfusion pattern, i.e., solid
organ infarction. Solid organ infarction is found in CT in 15e36% of
patients with AMI, and not exclusively in patients with embolism
but also in patients with mesenteric arterial occlusive disease
[44,46,50].
Can patient have AMI even in absence of ischemia-specic CT
ndings?
In a study of 27 patients with AMI caused by mesenteric arterial
occlusive disease, contrast-enhanced CT and conventional digital
subtraction angiograms were obtained from all patient; twenty
patients with intermittent CMI were used as a control group, and all
images were blindly evaluated by three experienced radiologists
knowing that the patients were suspected of mesenteric ischemia
[34]. In this study, one third of patients with AMI presented without
any ischemia-specic CT signs (dened as thrombotic SMA clot,
absent or decreased bowel wall enhancement, or pneumatosis).
There were moderate to substantial interobserver variability in the
CT interpretation of these ndings. Another interesting observation
in that study was that the presence of ischemia-specic CT signs
was noted in 77% of patients who required bowel resection, but also
in 50% of patients who did not need bowel resection (after endo-
vascular revascularization). Hence, neither pneumatosis nor
decreased bowel wall enhancement are denitive signs of irre-
versible bowel necrosis. In contrast, the study showed that all pa-
tients with AMI had at least some level of abnormal intestinal
ndings in their CT scans, such as mesenteric fat stranding in 96%,
bowel lumen dilatation in 93%, and bowel wall thickening in 70%.
Only few of the control group patients with CMI had such ndings
(due to chronic ischemic colitis) [34].
In conclusion, the answer to the question whether a patient may
have AMI even in absence of ischemia-specic CT ndings is yes.
Nevertheless, unspecic intestinal CT ndings can usually be
detected, but these ndings are easily disregarded in the CT analysis
without the clinical suspicion of AMI. The same applies to patients
with NOMI as well [19]. In case of SMA embolism, the embolic clot

Table 3
Computed tomography (CT) ndings in acute mesenteric ischemia (AMI).

Specic
 SMA embolus (oval shaped clot in a previously unaffected artery)
 SMA thrombosis (blood clot with superimposed calcied lesion)
 Mesenteric venous thrombosis (with surrounding oedema of the mesentery)
 Unenhanced or poorly enhanced bowel wall segment(s)
 Intestinal pneumatosis
 Portal venous gas
Non-specic
 Chronic calcied occlusion or hemodynamically signicant (70%) stenosis of the SMA
 Concomitant severe obstruction of CA and/or IMA (2- or 3-vessel disease)
 Abnormal (increased) enhancement of the bowel wall
 Bowel wall thickening (oedema, hyperdense hemorrhage)
 Luminal dilatation of the intestine (paralysis, transmural bowel necrosis)
 Mesenteric fat stranding
 Ascites
 Free gas (bowel perforation)
 Solid organ infarction (synchronous embolism or organ hypoperfusion)
 rkka
J.M. Ka inen, S. Acosta / Best Practice & Research Clinical Gastroenterology 31 (2017) 15e25 23

is the most specic CT nding. In a study including 24 patients with
embolic AMI, the embolic clot was detected in CT in all expect two
patients with distal small branch embolism [19]. In ve of the 24
cases with SMA embolism, neither decreased bowel wall
enhancement nor pneumatosis could be detected by an experi-
enced radiologist.
Characteristics of AMI in elderly patients
As shown in Fig. 1, the spectrum of differential diagnoses in
acute abdomen is very different in the elderly than in the younger
patients. The incidence of AMI is roughly 10-fold in 80-year-old
people versus in 60-year-old population [8,13]. The proportion of
patients with atherosclerotic etiology of AMI is signicantly higher
in the elderly (Table 4). Based on the Kuopio cohort, embolic eti-
ology of AMI was as common as mesenteric arterial occlusive dis-
ease in patients aged 75 years or less. In patients over 75 years, the
atherosclerotic etiology was twice as common cause of AMI as
embolism. Another distinctive difference between these two age-
groups was the higher incidence of MVT in the younger popula-
tion. Patients with MVT were signicantly younger (mean age 53
years) than patients with arterial AMI (mean age 76e79 years) [19].
Unsurprisingly, the elderly patients had more cardiovascular
comorbidities and the in-hospital mortality was higher.
In the analysis of the two age groups with AMI (Table 4),
younger patients were more prone to undergo bowel resection
than the elderly patients (although the difference was not statis-
tically signicant). This may be due to the higher proportion of
acute-on-chronic presentations of AMI among the elderly. It was
shown that these patients with slowly progressing disease had
better outcome after emergency endovascular revascularization
than patients with fulminant AMI [3]. Perhaps the most important
message of this review is that there is only a small step from CMI to
AMI. This is relevant especially when interpreting CT ndings of an

Table 4
Characteristics of 95 patients with acute mesenteric ischemia (AMI) stratied by age. This table contains unpublished data derived from a previously published study [19].

Age group 75 years (n 36) >75 years (n 59) pa

Comorbidities
Age, mean standard deviation, years 63 10 83 5 <0.001
Sex, male 29 (81) 16 (27) <0.001
Obesity (body mass index 30 kg/m2) 12 (33) 9 (15) 0.050
Hypertension 16 (44) 44 (75) 0.004
Hyperlipidemia 9 (25) 24 (41) 0.182
Diabetes 12 (33) 14 (24) 0.484
Chronic renal insufciency 1 (3) 3 (5) 1.000
Atrial brillation 8 (22) 24 (41) 0.065
Coronary artery disease 10 (28) 31 (53) 0.020
Peripheral arterial disease 10 (28) 14 (24) 0.808
Chronic heart failure 3 (8) 12 (20) 0.153
Stroke or transient ischemic attack 6 (17) 20 (34) 0.068
Recent major surgery (30 days) 1 (3) 6 (10) 0.247
History of DVT or pulmonary embolism 1 (3) 2 (3) 1.000
Medication
Warfarin 5 (14) 11 (19) 0.778
Acetylic salicylic acid 14 (39) 37 (63) 0.034
Clopidogrel 2 (6) 4 (7) 1.000
Statin 15 (42) 33 (56) 0.208
AMI etiology
Embolism 8 (22) 16 (27) 0.594
Mesenteric arterial occlusive disease 9 (25) 28 (48) 0.029
Non-occlusive mesenteric ischemia 10 (28) 15 (25) 0.800
Mesenteric venous thrombosis 9 (25) 0 (0) <0.001
Clinical characteristics
Duration of symptoms
<24 hours 13 (36) 18 (31) 0.654
1e3 days 11 (31) 23 (39) 0.509
>3 days 12 (33) 18 (31) 0.822
Acute kidney injury 5 (14) 10 (17) 0.778
Acute myocardial infarction 1 (3) 4 (7) 0.647
Clinical AMI suspicion prior to imaging 8 (22) 21 (36) 0.251
Primarily admitted to surgical emergency unit 27 (75) 43 (73) 0.502
Diagnostic delay, median (IQR), hours 5.5 (2e29) 13.3 (2e24) 0.750
Laboratory ndings
WBC count, median (IQR), 109/L 12.9 (10e19) 13.0 (9e17) 0.378
CRP, median (IQR), mg/L 145 (31e207) 149 (30e213) 0.936
Plasma creatinine, median (IQR), mmol/L 76 (62e106) 78 (62e112) 0.840
GFR <30 ml/min/1.73m2 4 (11) 7 (12) 1.000
Metabolic acidosis 11 (31) 12 (20) 0.321
Therapy and outcome
Endovascular therapy 13 (36) 35 (59) 0.028
Surgical revascularization 0 (0) 2 (3) 0.524
Laparotomy 18 (50) 20 (34) 0.120
Bowel resection 15 (42) 15 (25) 0.098
Massive unresectable bowel necrosis 2 (6) 3 (5) 0.921
In-hospital death 7 (19) 23 (39) 0.047

Data are presented as n (%) unless otherwise stated.

DVT, deep vein thrombosis; IQR, interquartile range; WBC, white blood cell; CRP, C-reactive protein; GFR, glomerular ltration rate.
a
Fisher's exact test or the chi-squared test was used to compare nominal data and the ManneWhitney U test for nonparametric data.
24 rkka
J.M. Ka inen, S. Acosta / Best Practice & Research Clinical Gastroenterology 31 (2017) 15e25
elderly patient with prolonged abdominal symptoms and elevated
inammatory markers (C-reactive protein); if there are abnormal
intestinal ndings (listed in Table 3) and severe calcied obstruc-
tion of several mesenteric arteries (SMA being the most important),
the patient is likely suffering from acute episode of mesenteric
ischemia and needs emergent revascularization.
Do we need to fear contrast-induced renal failure?
Chronic renal insufciency and acute kidney injury are consid-
ered as risk factors for contrast-induced renal failure in patients
with CT performed for acute abdominal pain. On the other hand, CT
without contrast enhancement is associated with diagnostic delay
and increased mortality in AMI [51]. Among the 95 Finnish AMI
patients with CT obtained prior to treatment, only 5% of patients
with arterial AMI had chronic renal insufciency prior to admission
[19]. The prevalence of acute kidney injury at admission (prior to
imaging) was 13% and 40% in AMI patients with mesenteric arterial
occlusive disease and NOMI, respectively; none of the patients with
SMA embolism or MVT had acute kidney injury. In general, the risk
of contrast-induced renal failure in patients with AMI is negligible
especially if the patient has had normal kidney function before
falling ill [55]. The acute kidney injury in AMI is usually caused by
the disease itself, especially in NOMI. AMI patients are often
dehydrated because of bowel paralysis, vomiting and diarrhea.
Intravascular administration of isotonic crystalloids should be
started prior to imaging. At discretion, CT can be performed with
reduced amount of contrast agent in patients with impaired renal
function. However, contrast-enhanced CT should not be denied to a
severely ill patient because of the fear of contrast-induced renal
failure; as a consequence, the inevitable delay in the diagnosis of
AMI is a greater risk. In some situations, the rst image acquisition
can be performed without contrast enhancement, and if the diag-
nosis is not apparent in the unenhanced CT (e.g. cholecystitis,
appendicitis, diverticulitis, or pancreatitis), the examination should
be continued immediately with contrast-enhanced protocol.
Prognosis
An active endovascular-rst strategy in Kuopio with 76%
revascularization rate resulted 42% overall mortality rate in pa-
tients with occlusive AMI [3]. In comparison to the previous 82%
mortality for AMI in Finland, this is a signicant improvement [2].
Even so, it has to be individually considered, when the aggressive
treatment is futile and doomed to result in unnecessary suffering.
Due to the old age and comorbidities in AMI patients, the long-
term survival is poor regardless of the treatment. The most frag-
ile and severely ill patients should be treated with adequate
comfort care alone.
Summary
AMI appears to be relatively common in elderly patients e more
common than appendicitis in patients aged over 75 years (Table 1).
This is important to acknowledge, because clinical suspicion is
a major factor in the early diagnosis AMI and in the correct inter-
pretation of CT ndings. If AMI is suspected, contrast-enhanced CT
should be performed without fear of contrast-induced nephropa-
thy, preferably in arterial and venous phases. Clinicians should be
aware that the clinical presentation of AMI varies a great deal
depending on the etiology, and moreover, on the presentation
pattern of the arterial obstruction. The management and prognosis
of patients with acute occlusion of the SMA, acute-on-CMI, and
intermittent CMI are different, and these conditions should be
distinguished. CMI may insidiously evolve to AMI, and the
symptoms of acute-on-CMI are often obscure; the early CT ndings
are difcult to interpret and easily missed. The current role of
biomarkers in the early diagnosis of AMI is limited. Although
normal D-dimer may exclude acute thromboembolic occlusion of
the SMA and acute MVT, it is not known, how the D-dimer behaves
in acute-on-chronic presentation of AMI, or in NOMI.
Practice points
 Acute occlusion of the SMA should be suspected in pa-
tients with acute abdominal pain and atrial fibrillation,
acute or recent cardiac infarction, prosthetic heart valve,
or other synchronous embolic event (e.g. stroke, limb
embolism, spleen infarction).
 AMI should be suspected in patients with acute or sub-
acute abdominal pain (diarrhea/vomiting), and prior
manifestations of the atherosclerotic vascular disease
(e.g. coronary heart disease, peripheral vascular disease,
cerebrovascular disease), or prior symptoms of CMI.
 CT of the acute abdomen should be performed with
contrast enhancement, without fear of contrast-induced
nephropathy, if severe illness is suspected. The visceral
vessels and associated intestinal findings should be
assessed and reported routinely by radiologists.
 Acute occlusion of the SMA is the most common cause of
AMI. However, some patients develop AMI upon severe
chronic occlusive disease of the mesenteric arteries. In
these patients, even the unspecific CT findings together
with prolonged symptoms and elevated inflammatory
markers are suspicious of acute intestinal ischemia.
 CMI patients with progressive symptoms and severe
weight loss are at eminent risk of developing fulminant
AMI. These patients require rapid revascularization, with
similar urgency as patients with acute coronary
syndrome.
Research agenda
 The utility of the next generation dual-energy CT in
increasing visualization of the differences between
ischemic and perfused bowel segments should be
investigated in human patients with AMI.
 Research in the field of biomarkers should be targeted to
discover a blood test capable of detecting early intestinal
hypoperfusion (NOMI) in patients requiring treatment in
the intensive care unit for any underlying disease or
surgery.
Conict of interest
None.
Acknowledgements
The authors would like to acknowledge Interventional Radiol-
ogist of Finland society and Rattima
ki Cardiovascular Society for
funding the writing of this review. The funders did not have any
role in the making of this manuscript.
 rkka
J.M. Ka inen, S. Acosta / Best Practice & Research Clinical Gastroenterology 31 (2017) 15e25 25

References [29] Clair DG, Beach JM. Mesenteric ischemia. N Engl J Med 2016;374(10):959e68.
[30] Kolkman JJ, Bargeman M, Huisman AB, Geelkerken RH. Diagnosis and man-
re, Tindall & Cox; agement of splanchnic ischemia. World J Gastroenterol 2008;14(48):
[1] Cokkinis AJ. Mesenteric vascular occlusion. London: Baillie
7309e20.
1926. p. 1e93.
[31] Bjornsson S, Resch T, Acosta S. Symptomatic mesenteric atherosclerotic
rvinen O, Laurikka J, Salenius JP, Tarkka M. Acute intestinal ischemia e a
[2] Ja
disease-lessons learned from the diagnostic workup. J Gastrointest Surg
review of 214 cases. Ann Chir Gynecol 1994;83:22e5.
2013;17(5):973e80.
[3] Karkkainen JM, Lehtima ki TT, Saari P, Hartikainen J, Rantanen T, Paajanen H,
[32] Wyers MC. Acute mesenteric ischemia: diagnostic approach and surgical
et al. Endovascular therapy as a primary revascularization modality in acute
treatment. Semin Vasc Surg 2010;23(1):9e20.
mesenteric ischemia. Cardiovasc Interv Radiol 2015;38(5):1119e29.
[33] Patel SD, Constantinou J, Hamilton H, Davis M, Ivancev K. Editor's choice e a
[4] Schermerhorn ML, Giles KA, Hamdan AD, Wyers MC, Pomposelli FB. Mesen-
shaggy aorta is associated with mesenteric embolisation in patients under-
teric revascularization: management and outcomes in the United States,
going fenestrated endografts to treat paravisceral aortic aneurysms. Eur J Vasc
1988e2006. J Vasc Surg 2009;50(2):341e8.
Endovasc Surg 2014;47(4):374e9.
[5] Lo RC, Fokkema M, Curran T, Hamdan AD, Wyers MC, Schermerhorn ML. The
[34] Karkkainen JM, Saari P, Kettunen H-P, Lehtim aki TT, Vanninen R, Paajanen H,
decline of mesenteric ischemia-related mortality in the last decade. Abstract
et al. Interpretation of abdominal CT ndings in patients who develop acute
presented in 2014 Society for Clinical Vascular Surgery symposium. (article in
on chronic mesenteric ischemia. J Gastrointest Surg 2016;20(4):791e802.
preparation).
[35] Kolkman JJ, Mensink PB. Non-occlusive mesenteric ischaemia: a common
[6] Beaulieu RJ, Arnaoutakis KD, Abularrage CJ, Efron DT, Schneider E, Black JH.
disorder in gastroenterology and intensive care. Best Pract Res Clin Gastro-
Comparison of open and endovascular treatment of acute mesenteric
enterol 2003;17(3):457e73.
ischemia. J Vasc Surg 2014;59(1):159e64.
[36] Harnik IG, Brandt LJ. Mesenteric venous thrombosis. Vasc Med 2010;15(5):
[7] Acosta S. Epidemiology of mesenteric vascular disease: clinical implications.
407e18.
Semin Vasc Surg 2010;23(1):4e8.
[37] Acosta S, Nilsson TK, Bjo rck M. D-dimer testing in patients with suspected
[8] Acosta S, Ogren M, Sternby NH, Bergqvist D, Bjo rck M. Incidence of acute
acute thromboembolic occlusion of the superior mesenteric artery. Br J Surg
thrombo-embolic occlusion of the superior mesenteric arteryda population-
2004;91:991e4.
based study. Eur J Vasc Endovasc Surg 2004;27(2):145e50.
rck M. Fatal nonocclusive [38] Acosta S, Nilsson T. Current status on plasma biomarkers for acute mesenteric
[9] Acosta S, Ogren M, Sternby NH, Bergqvist D, Bjo
ischemia. J Thromb Thrombolysis 2012;33(4):355e61.
mesenteric ischaemia: population-based incidence and risk factors. J Intern
[39] van den Heijkant TC, Aerts BAC, Teijink JA, Buurman WA, Luyer MDP. Chal-
Med 2006;259(3):305e13.
rck M. Mesenteric venous lenges in diagnosing mesenteric ischemia. World J Gastroenterol 2013;19(9):
[10] Acosta S, Ogren M, Sternby NH, Bergqvist D, Bjo
1338e41.
thrombosis with transmural intestinal infarction: a population-based study.
rnsson S, Resch T, Bjo
[40] Acosta S, Block T, Bjo rck M, Nilsson T. Diagnostic pitfalls
J Vasc Surg 2005;41(1):59e63.
at admission in patients with acute superior mesenteric artery occlusion.
[11] Acosta S, Wadman M, Syk I, Elmsthl S, Ekberg O. Epidemiology and prog-
J Emerg Med 2012;42(6):635e41.
nostic factors in acute superior mesenteric artery occlusion. J Gastrointest
[41] Oliva IB, Davarpanah AH, Rybicki FJ, Desjardins B, Flamm SD, Francois CJ, et al.
Surg 2010;14(4):628e35.
rck M. Acute thrombo-embolic occlusion of the superior ACR appropriateness criteria imaging of mesenteric ischemia. Abdom Im-
[12] Acosta S, Bjo
aging 2013;38(4):714e9.
mesenteric artery: a prospective study in a well dened population. Eur J Vasc
[42] Cudnik MT, Darbha S, Jones J, Macedo J, Stockton S, Hiestand B. The diagnosis
Endovasc Surg 2003;26(2):179e83.
of acute mesenteric ischemia: a systematic review and meta-analysis. Ann
[13] Karkkainen JM, Lehtima ki TT, Manninen H, Paajanen H. Acute mesenteric
Emerg Med 2013;20(11):1087e100.
ischemia is a more common cause than expected of acute abdomen in the
[43] Menke J. Diagnostic accuracy of multidetector CT in acute mesenteric
elderly. J Gastrointest Surg 2015 Aug;19(8):1407e14.
rja
-Koskenkari P, Pietil ischemia: systematic review and meta-analysis. Radiology 2010;256(1):
[14] Salomaa V, Havulinna AS, Koukkunen H, Ka a A,
93e101.
Mustonen J, et al. Aging of the population may not lead to an increase in the
[44] Kirkpatrick IDC, Kroeker MA, Greenberg HM. Biphasic CT with mesenteric CT
numbers of acute coronary events: a community surveillance study and
angiography in the evaluation of acute mesenteric ischemia: initial experi-
modelled forecast of the future. Heart 2013;99(13):954e9.
ence. Radiology 2003;229(1):91e8.
[15] Borodulin K, Vartiainen E, Peltonen M, Jousilahti P, Juolevi A, Laatikainen T,
[45] Zandrino F, Musante F, Gallesio I, Benzi L. Assessment of patients with acute
et al. Forty-year trends in cardiovascular risk factors in Finland. Eur J Public
mesenteric ischemia: multislice computed tomography signs and clinical
Health 2015;25(3):539e46.
rck M. Clinical implications for performance in a group of patients with surgical correlation. Minerva Gas-
[16] Acosta S, Ogren M, Sternby NH, Bergqvist D, Bjo
troenterol Dietol 2006;52(3):317e25.
the management of acute thromboembolic occlusion of the superior mesen-
[46] Aschoff AJ, Stuber G, Becker BW, Hoffmann MHK, Schmitz BL, Schelzig H, et al.
teric artery: autopsy ndings in 213 patients. Ann Surg 2005;241(3):516e22.
Evaluation of acute mesenteric ischemia: accuracy of biphasic mesenteric
[17] Endean ED, Barnes SL, Kwolek CJ, Minion DJ, Schwarcz TH, Mentzer RM.
multi-detector CT angiography. Abdom Imaging 2009;34(3):345e57.
Surgical management of thrombotic acute intestinal ischemia. Ann Surg
[47] Ofer A, Abadi S, Nitecki S, Karram T, Kogan I, Leiderman M, et al. Multidetector
2001;233(6):801e8.
CT angiography in the evaluation of acute mesenteric ischemia. Eur Radiol
[18] Ryer EJ, Kalra M, Oderich GS, Duncan AA, Gloviczki P, Cha S, et al. Revascu-
2009;19(1):24e30.
larization for acute mesenteric ischemia. J Vasc Surg 2012;55(6):1682e9.
[48] Akyildiz H, Akcan A, Oztrk A, Sozuer E, Kucuk C, Karahan I. The correlation of
[19] Lehtim rkka
aki TT, Ka inen JM, Saari P, Manninen H, Paajanen H, Vanninen R.
the D-dimer test and biphasic computed tomography with mesenteric
Detecting acute mesenteric ischemia in CT of the acute abdomen is dependent
computed tomography angiography in the diagnosis of acute mesenteric
on clinical suspicion: review of 95 consecutive patients. Eur J Radiol
ischemia. Am J Surg 2009;197(4):429e33.
2015;84(12):2444e53.
[49] Barmase M, Kang M, Wig J, Kochhar R, Gupta R, Khandelwal N. Role of mul-
[20] Mensink PBF, Moons LMG, Kuipers EJ. Chronic gastrointestinal ischaemia:
tidetector CT angiography in the evaluation of suspected mesenteric ischemia.
shifting paradigms. Gut 2011;60(5):722e37.
Eur J Radiol 2011;80(3):e582e7.
[21] Hansen KJ, Wilson DB, Craven TE, Pearce JD, English WP, Edwards MS, et al.
[50] Yikilmaz A, Karahan OI, Senol S, Tuna IS, Akyildiz HY. Value of multislice
Mesenteric artery disease in the elderly. J Vasc Surg 2004;40(1):45e52.
computed tomography in the diagnosis of acute mesenteric ischemia. Eur J
[22] Wilson DB, Mostafavi K, Craven TE, Ayerdi J, Edwards MS, Hansen KJ. Clinical
Radiol 2011;80(2):297e302.
course of mesenteric artery stenosis in elderly Americans. Arch Intern Med
[51] Wadman M, Block T, Ekberg O, Syk I, Elmsthl S, Acosta S. Impact of MDCT
2006;166(19):2095e100.
with intravenous contrast on the survival in patients with acute superior
[23] Thomas JH, Blake K, Pierce GE, Hermreck AS, Seigel E. The clinical course of
mesenteric artery occlusion. Emerg Radiol 2010;17(3):171e8.
asymptomatic mesenteric arterial stenosis. J Vasc Surg 1998;27(5):840e4.
rvinen O, Laurikka J, Sisto T, Salenius JP, Tarkka MR. Atherosclerosis of the [52] Wiesner W, Khurana B, Ji H, Ros PR. CT of acute bowel ischemia. Radiology
[24] Ja
2003;226(3):635e50.
visceral arteries. Vasa 1995;24(1):9e14.
[53] Costa AF, Chidambaram V, Lee JJ. Multidetector computed tomography of
[25] Roobottom C, Dubbins P. Signicant disease of the celiac and superior
mesenteric ischaemia. Insights Imaging 2014;5(6):657e66.
mesenteric arteries in asymptomatic patients: predictive value of Doppler
[54] Mazzei MA, Volterrani L. Nonocclusive mesenteric ischaemia: think about it.
sonography. Am J Roentgenol 1993;161(5):985e8.
Radiol Med 2014;120(1):85e95.
[26] Park CM, Chung JW, Kim HB, Shin SJ, Park JH. Celiac axis stenosis: incidence
[55] Acosta S, Bjornsson S, Ekberg O, Resch T. CT angiography followed by endo-
and etiologies in asymptomatic individuals. Korean J Radiol 2001;2(1):8e13.
vascular intervention for acute superior mesenteric artery occlusion does not
[27] Kornmann VNN, Van Werkum MH, Bollen TL, Van Ramshorst B, Boerma D.
increase risk of contrast-induced renal failure. Eur J Vasc Endovasc Surg
Compromised visceral circulation does not affect the outcome of colorectal
2010;39(6):726e30.
surgery. Surg Today 2014;44(7):1220e6.
[28] Gms H, Gms M, Tekbas G, Onder H, Ekici F, etinakmak MG, et al.
Clinical and multidetector computed tomography ndings of patients with
median arcuate ligament syndrome. Clin Imaging 2012;36(5):522e5.
 Reproduced with permission of copyright owner.
Further reproduction prohibited without permission.