a gastrosplenic trunk (with absent hepatic a.) (3-6%), hepatosplenic trunk (with absent left gastric a.) (5-6%), and
hepatogastric trunk (with absent splenic a.) (1%) are formed [7, 16]. If two main branches are absent then the three main
branches originate independently, consequently there is no real celiac trunk (1 -2%) [7, 16 ]. A special form of absence of
the celiac trunk is the celiacomesenteric trunk where all three main branches of celiac trunk and SMA arise from a
common trunk (2%) [7]. Absent branches may originate from the aorta or one of the remaining two main arteries. In the
latter situation the common hepatic a. arising from the SMA is a well known example (hepatomesenteric trunk) [7, 16 ]. A
similar situation may occur for the left gastric a. in that it may also originate from the left hepatic a., as shown by
Nakamura et al. in 14% of the Japanese [11], although this variation is very rare in other raaces [16 ]. Finally, the splenic
a. may also show this pattern. A common splenomesenteric trunk, which gives off the splenic a. and SMA, has been
reported with an incidence of less than 1% [1, 7, 9, 14]. Some patients with such a variation have been presented as a
case report [5].
Tandler provided an embryological explanation for these variations in 1904 [15]. During development, three groups of
collateral aa. arise from the abdominal aorta as dorsal, lateral and ventral branches. The latter develop initially as paired
vessels, which then coalesce in the median line to form the four roots for the gut the four roots being connected by
ventral longitudinal anastomoses. In the majority of cases the first three roots coalesce to form the celiac trunk and
separate from the fourth root. The future SMA is developed from the fourth root, which migrates caudally with the
ventral migration of the gut. If this separation takes place at a higher level, one of the celiac branches arises from the
SMA. The splenic a. arising from SMA (splenomesenteric trunk), as in the present patient, is an example of this
exceptional condition.
The common hepatic a. is a branch of the celiac trunk. It arises from the SMA in 1.5 -7% of population [6, 7, 16]
exceptionally it may arise separately from the aorta (0.2-3%) [7, 11, 16]. The normal pattern of the common hepatic a. is
to form the gastroduodenal and proper hepatic a., which then divides distally into right and left hepatic branches. This
pattern is present in only 54 -75% of the population, the remainder having aberrant hepatic branches from the left gastric
a. or SMA [3, 6, 16 ]. According to van Damme et al. division of the hepatic a. into its right and left hepatic branches may
take place at any point between the proper hepatic a. and the aortic origin of the main hepatic trunk [16]. The right
hepatic a. may arise from the gastroduodenal a. (no proper hepatic a.), the common hepatic a. or the celiac trunk [16 ]. In
the latter situation, the right and left hepatic branches take their origin from the celiac trunk separately (2-4%) [7, 16].
Therefore, it has been called a duplicated hepatic a. [4]. Direct origin of the right hepatic a. from the aorta, as seen in the
present case, is one exception among other rare variations, such as the right anterior hepatic a. arising from the SMA [10]
and the right hepatic a. from the right renal a. [2].
The recognition of a variant splanchnic arterial supply has important diagnostic and therapeutic implications. Operational
strategies in cases of arterial aberrations interfering with major abdominal surgical techniques, such as liver
transplantation and resection, hepatic a. infusion chemotherapy, gastrectomy, biliary reconstruction, and
pancreaticoduodenectomy, have been well discussed [16]. Settembrini et al. reported two patients with aneurysms
involving the origin of the splenic a. that arose anomalously from the SMA just behind the pancreas [13]. The origin of
the splenic a. from the SMA might have been carried downwards by the caudal migration of the SMA behind the head of
the pancreas. These authors, therefore modified their surgical technique to reach the neck of the aneurysm and carried out
a successful repair.
Aberrant arterial branches may also interfere with interventional radiological procedures. In the case of the left gastric a.
arising from the left hepatic a., transarterial chemoembolization through the left hepatic a. may damage the gastric
mucosa. In our institution transarterial chemoembolization of hepatocellular carcinoma has been achieved through either
aberrant hepatic branches [12] or collateral vessels [8]. Present day catheter and guidewire technologies allow many
selective and superselective catheterizations in abdominal interventional procedures to be performed, as in other parts of
the body.
In conclusion, different anatomical variations involving the celiac trunk and SMA should be borne in mind during both
surgical and radiological evaluations. Knowledge of such variations would result in the accurate interpretation of disease
in diagnostic imaging, as well as the optimum elective procedure in surgical or interventional radiological management.
References
1. Adachi B (1928) Das Arteriensystem der Japaner. Kenkyusha Press, Tokyo
2. Braun MA, Collins MB, Wright P (1991) An aberrant right hepatic artery from the right renal artery anatomical
vignette. Cardiovasc Intervent Radiol 14 349-351
3. Chen CY, Lee RC, Tseng HS, Chiang JH, Hwang JI, Teng MM (1998) Normal and variant anatomy of hepatic arteries
angiographic experience. Chung Hua I Hsueh Tsa Chih (Taipei) 61 17-23