Surg Radiol Anat
DOI 10.1007/s00276-015-1515-7
ANATOMIC VARIATIONS
A mixed-type intermesenteric trunk as a major contributor
to the ascending, transverse, and descending colons: a case report
Ilke Ali Gurses1 Ozcan Gayretli1 Sule Canberk2 Aysin Kale1
Received: 9 December 2014 / Accepted: 29 June 2015
Springer-Verlag France 2015
Abstract
Purpose We present a case in which a mixed-type
intermesenteric trunk was the major arterial supply for the
ascending, transverse, and descending colons.
Methods We dissected a cadaver of a man aged 74 years
that was used for a routine abdominal dissection course of
2nd year medical students.
Results We observed that a mixed-type intermesenteric
trunk supplied the majority of the colon, originating from the
inferior mesenteric artery. The vessel was non-tortuous and
had a counter clockwise course. It gave branches that supply
the marginal artery at the splenic and hepatic flexures and at
the transverse colon and finally it anastomosed with the
ileocolic artery at the ileocecal junction. Through the inter-
mesenteric trunk, the inferior mesenteric artery supplied the
descending, transverse, and ascending colons with contri-
butions from the sigmoidal and ileocolic arteries.
Conclusions The intermesenteric trunk is an important
central connection between the superior and inferior
mesenteric arteries. It probably is an embryologic remnant
that constituted a longitudinal anastomosis between both
mesenteric arteries.
Keywords Colon
Blood supply
Intermesenteric trunk

Meandering mesenteric artery
Superior mesenteric
artery
Inferior mesenteric artery
& Ilke Ali Gurses
iagurses@gmail.com
1
Department of Anatomy, Istanbul Faculty of Medicine,
Istanbul University, Millet Caddesi, Fatih, 34093 Istanbul,
Turkey
2 During a routine abdominal dissection for 2nd year medical
Department of Pathology, Haydarpasa Numune Education
and Research Hospital, Tibbiye Caddesi, No: 40,
Uskudar, 34668 Istanbul, Turkey
Introduction
The coeliomesenteric axis has a wide range of variation
and collateral blood flow between its branches is common
[3, 5, 7, 9]. Arterial connections that constitute a collateral
arterial supply between the superior mesenteric artery
(SMA) and inferior mesenteric artery (IMA) consist of two
main contributors [3, 6, 9]. The first, which is named the
marginal artery (of Drummond), is a peripheral arterial
arch that lies close to the colonic wall [1, 3, 6, 9]. It is
formed by branches of the right, middle and left colic
arteries and runs from the ascending colon to the sigmoid
colon and in some cases to the upper segments of the
rectum [1, 3]. The second arterial connection lies centrally
along the colonic mesentery and communicates the SMA
and IMA or their respective branches, the middle and left
colic arteries [1, 3, 6]. Although this connection has many
synonyms including Riolans arch, Artery of Moskowitz,
the intermesenteric trunk and the meandering mesenteric
artery, its definition remains controversial [1, 3, 6]. For
example, the intermesenteric trunk has a straight course
parallel to the vertebral column. On the other hand, the
meandering mesenteric artery has a significant tortuous
course probably due to underlying occlusive disease [1, 6].
In this report, we present a case in which a variant artery
that originated from the IMA was the only arterial source
for the majority of the territories of the right, middle, and
left colic arteries.
Case report
students, we encountered mesenteric arterial variations in a
male cadaver, aged 74 years. The SMA did not give the
123

right and middle colic branches. As we dissected proxi-
mally to the SMA, we observed a horizontal artery within
the proximal mesocolon that crossed the superior mesen-
teric vessels anteriorly. Further dissection revealed that the
left colic artery was also absent. The variant artery origi-
nated from the IMA, had a straight cranial course to the
splenic flexure close to the mesocolon, and gave an angular
branch at the flexure. As it traversed the transverse colon, it
gave a branch to the transverse colon that corresponded to
the middle colic artery. The middle colic branch bifurcated
into its right and left branches. As the variant artery
reached the hepatic flexure, it gave a dominant colic branch
to the flexure. Between the hepatic flexure and cecum, it
coursed caudally and gave numerous colic branches and
finally anastomosed with the ileocolic artery at the ileo-
cecal junction. There were no further variations regarding
the SMA and IMA in the cadaver. The SMA gave 4 jejunal
and 3 ileal arteries before giving the ileocolic artery. We
also observed 2 sigmoidal arteries and the superior rectal
artery originated from the IMA. We observed the marginal
artery (MA) (of Drummond) along the descending and
transverse colons. The angular, middle, and dominant colic
branches of the variant artery and the first sigmoidal artery
contributed to the MA at this region. We believe that the
first sigmoidal artery also contributed to the arterial supply
of the distal descending colon. The MA was absent
between the hepatic flexure and ileocecal junction, where
the colon was supplied with the vasa recta, which origi-
nated from the terminal portion of the variant artery.
Similar to the distal descending colon, the proximal
ascending colon was probably supplied by the terminal
portion of the variant artery and the ileocolic artery through
their anastomoses.
Bertelli et al. [1] named the central connection between
the SMA and IMA as the intermesenteric trunk (IMT) and
classified it into three types. In the direct type, the IMT
connects the SMA and the IMA; in the mixed type, the
IMT connects one of the mesenteric arteries with a main
branch of the other mesenteric artery; and in the indirect
type, the IMT connects two main branches of the two
mesenteric arteries [1]. We believe that the variant artery in
our case was a mixed-type IMT because it connected the
IMA with the ileocolic artery (Fig. 1a).
On manual examination, there were no signs of occlusion
regarding the SMA or IMA. To evaluate the existence of
arterial occlusion, we collected tissue samples from the SMA
and IMA for microscopic examination. We fixed the samples
in 10 % buffered formalin, embedded them in paraffin wax,
and stained them with hematoxylin and eosin. We observed
atherosclerotic plaques in the SMA and IMA (Fig. 1b, c) and
calcification in the IMA (Fig. 1d). Nevertheless, we
observed no occlusions in the SMA or IMA (Fig. 1b, c).
123
Surg Radiol Anat
Discussion
We presented a case in which the IMT supplied the arterial
territories of the right, middle, and left colic arteries with
contributions from the ileocolic and sigmoidal arteries. We
have avoided naming this artery as Riolans arch because
this eponym is a misnomer [6]. We also did not consider
this vessel as the meandering mesenteric artery. The
meandering mesenteric artery has a long tortuous course
and is usually associated with severe arterial stenotic dis-
ease [3]. The vessel in our case had a straight cranial course
and the SMA and IMA showed only minimal atheroscle-
rosis. Lange et al. [6] considered the ascending branch of
the left colic artery (ALCA) as another SMAIMA con-
nection. We believe that the vessel in our case was not an
ALCA. The left colic artery was absent in our case and
even if it were a variant ALCA we observed no arteries or
branches that corresponded to the descending branch of the
left colic artery. The marginal artery that formed between
the first sigmoidal artery and the angular colic artery sup-
plied the territory of the descending branch (Fig. 1a). We
did not consider the variant vessel to be the MA. The MA
in our case was present close to the colonic wall, separate
to the variant artery. We decided the variant artery in our
case was consistent with the mixed-type IMT described by
Bertelli et al. [1]. The SMAleft colic artery and middle
colic arteryIMA connections were the most common
connection types that Bertelli et al. [1] reported with a
mixed-type IMT. The IMT in our case connected the IMA
to the more proximal ileocolic artery of the SMA.
There is a possibility that the variant vessel in our case
may have been a middle colic artery originating from the
IMA or its branches. Although there is no detailed
description of the course and branching pattern of a middle
colic artery originating from the IMA, illustrations in
previous studies clearly show that when the middle colic
artery originates from the IMA its right and left branches
are absent [1, 5, 7]. In our case, the IMT gave a colic
branch to the transverse colon that bifurcated into the right
and left branches, as the middle colic artery would. We
believe that this colic branch corresponded to the middle
colic artery itself.
We also observed that the MA was absent between the
ileocecal junction and right hepatic flexure. The splenic
flexure is considered as a watershed area between the
midgut and hindgut arterial trees [3, 4]. This area is also
referred to as Griffiths critical point due to increased
ischemic colitis after colonic resection [3, 4]. Nevertheless,
the right half of the colon, with its rich arterial support
from the SMA, is less prone to postoperative ischemia [4].
In our case, the splenic flexure was supplied by the MA that
formed between the branches of the IMT and the first
Surg Radiol Anat
Fig. 1 The abdominal anatomy (a) and microscopic images of SMA
(b) and IMA (c, d). In microscopic images, arterial occlusion is not
present. White arrows show atherosclerotic plaques in SMA (b) and
IMA (c) and the black arrow shows the calcification in IMA (d). Ao
abdominal aorta, L liver, D2 duodenum, descending part, P head of
pancreas, TI terminal ileum, C caecum, AC ascending colon, TC
transverse colon, DC descending colon, SC sigmoid colon, 1 ileocolic
artery, 2 superior mesenteric vein, 3 superior mesenteric artery, 4
Fig. 2 Schematic diagrams of the development patterns of the ventral
segmental arteries. The illustration is based on the studies of Tandler
[10] and Kitamura et al. [8]. DA dorsal aorta, Ao abdominal aorta, LA
longitudinal anastomosis, G left gastric artery, L splenic artery,
H common hepatic artery, CT coeliac trunk, J jejunal arteries, I ileal
inferior mesenteric artery, 5 intermesenteric trunk, 6 sigmoidal
arteries, 7 superior rectal artery, 8 angular colic branch of the IMT to
the splenic flexure, 9 middle colic branch of the IMT, 10 dominant
colic branch of the IMT to the hepatic flexure, 11 terminal portion of
the IMT that supplies the ascending colon, black arrowheads
marginal artery (of Drummond), white arrowhead anastomosis
between the ileocolic artery and the MMA
arteries, IC ileocolic artery, RC right colic artery, MC middle colic
artery, SMA superior mesenteric artery, LC left colic artery,
S sigmoidal arteries, SR superior rectal artery, IMA inferior mesen-
teric artery
123

sigmoidal artery; therefore, it was protected as long as the
IMA or IMT was preserved. The ascending colon consti-
tuted the watershed area with the absence of the MA
because it was supplied by the vasa recta, which originated
from the anastomosis between the terminal portion of the
IMT and the ileocolic artery.
In a clinical setting, the SMA is more important than the
IMA for colonic perfusion up to the descending colon [6].
On the other hand, the IMA could be ligated at its origin in
left colon or rectum resections and aortic surgery [3, 6].
Although intraoperative arterial pressure and colic oxy-
genation decreases during distal colic resections, surgery
would not result in ischemia of the proximal loop in most
cases [3, 4, 6]. Nevertheless, a high tie of the IMA in distal
colonic surgery in our case would have devastating results.
The ventral segmental arteries of the aorta supply the
primitive gut after the fusion of the paired dorsal aortae [2].
Between 5 and 8 mm stages of the embryo, the 7th and 8th
segmental arteries form the coeliac trunk (CT), the 9th
13th (predominantly 13th) segmental arteries form the
SMA, and the 20th ventral segmental artery forms the
IMA. In addition, a longitudinal anastomosis exists
between the CT and SMA, thus forming an arterial plexus
[2, 10]. Retention of the connections between both arteries
provides a possible explanation for the variations between
the CT and SMA, including the coeliomesenteric trunk [2,
10]. We were unable to find another article showing this
anastomotic plexus between the SMA and IMA. Kitamura
et al. [8] reported a case whose IMA originated from the
SMA. They applied the concept of longitudinal anasto-
mosis to all ventral segmental branches and explained how
IMA may have originated from the SMA. They also sug-
gested that the longitudinal anastomosis may be responsi-
ble for intermesenteric arterial communications [8]. This
explanation seems reasonable because when the middle
colic artery is absent the left colic artery usually replaces
its distribution [1, 5]. This replacement suggests that the
longitudinal anastomosis may reach more caudally than the
level of the SMA. Considering the central location of the
variations and communications between TC, SMA and
IMA, we concur with Kitamura et al. [8] and believe that
the IMT, which is centrally located within the mesocolon,
may represent the distal remnant of the longitudinal anas-
tomosis between SMA and IMA. Similarly, Bertelli et al.
[1] considered the IMT as an embryologic remnant
between the SMA and IMA. It seems that in our case, the
longitudinal anastomosis between IMA and SMA remained
123
Surg Radiol Anat
intact distal to the right colic artery. In addition, the ventral
segmental branches of right, middle, and left colic arteries
lost their proximal portions along with the longitudinal
anastomosis between the right colic and ileocolic arteries
(Fig. 2). In our case, the territories of the most distal
branches of the SMA (right and middle colic arteries) along
with the left colic artery were supplied by the IMT.
In conclusion, the IMT is probably a remnant of the
longitudinal anastomosis and is important for the arterial
supply of the colon in surgical practice.
Acknowledgments We thank our donor-cadaver and his immediate
family for their invaluable gift and for making this research possible.
We also thank David F. Chapman BSc. for editing the revised version
of our manuscript.
Compliance with ethical standards
Conflict of interest The authors declare no conflict of interests.
References
1. Bertelli L, Lorenzini L, Bertelli E (1996) The arterial vascular-
ization of the large intestine. Anatomical and radiological study.
Surg Radiol Anat 18(Suppl 1):S1S59
2. Evans HM (1910) Development of the vascular system. In:
Kiebel F, Mall FP (eds) Manual of human embryology, vol 2. JB
Lippincott Co, Philadelphia and London, pp 642652
3. Gourley EJ, Gering SA (2005) The meandering mesenteric
artery: a historic review and surgical implications. Dis Colon
Rectum 48(5):9961000
4. Griffiths JD (1956) Surgical anatomy of the blood supply of the
distal colon. Ann R Coll Surg Engl 19(4):241256
5. Hollinshead HW (1956) Anatomy for surgeons: vol 2 the thorax,
abdomen, and pelvis. HoeberHarper, New York, pp 502529
6. Lange JF, Komen N, Akkerman G, Nout E, Horstmanshoff H,
Schlesinger F, Bonjer J, Kleinrensink GJ (2007) Riolans arch:
confusing, misnomer, and obsolete. A literature survey of the
connection(s) between the superior and inferior mesenteric
arteries. Am J Surg 193(6):742748
7. Lippert H, Pabst R (1985) Arterial variations in man. Classifi-
cation and frequency. J. F. Bergmann Verlag, Munchen,
pp 4853
8. Kitamura S, Nishiguchi T, Sakai A, Kumamoto K (1987) Rare
case of the inferior mesenteric artery arising from the superior
mesenteric artery. Anat Rec 217(1):99102
9. Rosenblum JD, Boyle CM, Schwartz LB (1997) The mesenteric
circulation. Anatomy and physiology. Surg Clin North Am
77(2):289306
10. Tandler J (1904) Uber die varietaten der arteria coeliaca und
deren entwickelung. Anat Embryol 25(2):473500