The hepatopancreatic ampulla (HPA) has a complex anatomy related in particular to the different modes of convergence of the choledoc duct (CD) and the main pancreatic duct (MPD) [1,2]. Thus, there are multiple anatomical variants essential to know to prevent certain risks of pathologies of varying severity, ranging from biliary cysts to biliary tract cancers [3].

Indeed, the relationship between the terminal portions of CD and MPD and its variants has long been important in the sense that the reflux of bile, duodenal or pancreatic contents in this ductal system may or may not be possible in certain anatomical circumstances [4-6]. The objective of this study was to describe the configuration of the hepato-pancreatic junction by the radioscopy injection method and to review the literature.

This is a fundamental observational study for descriptive purposes. It concerned 30 pieces of fresh pancreas taken from anatomical subjects during forensic autopsy (numbered from P1 to P36). The average age was 39.43 years (Standard deviation = 13.9). The sex ratio was 3.3. The average size of the subjects was 175.10 cm [extreme 163cm-187cm]. The average length of the pancreas was 22.48 cm +/- 2.4.

In the laboratory, the transverse length of the pancreas was measured with a tape measure, from the right edge of the head from the middle of the descending portion of the duodenum to the end of the tail. Then the choledoc duct was spotted on the posterior face of the specimen and then catheterized with an intracath of adapted caliber. Subsequently, the pancreas was sectioned sagittally over its entire height using a scalpel at the body-tail junction. The main pancreatic duct was spotted in the section and catheterized.

Then the specimen was sent to the imaging department to take X-ray shots using a versatile digital radio-fluoroscopy device with a tilting table. First of all, a first shot without injection was taken. Then, the contrast agent was injected at constant pressure, under radioscopic control, into the main pancreatic duct (1st step), then into the choledoch (2nd step) and finally both at the same time (3rd step). The intra-pancreatic ducts were spotted until they were dislodged at the duodenal mucosa. When the injection was deemed sufficient, front incidence shots were taken for the different steps. For each preparation, photographs were taken. The assessment of the permeability of the duodenal papillae was judged on the injection clichés.

The cephalo-isthmic configuration of pancreatic ducts was studied. The method of constitution of the hepatopancreatic ampulla was evaluated on the basis of the Flati classification (Figure 1) [1]. The actual dimensions (length and diameter of the AHP) were taken by the ruler at the PACS (Picture Archiving and Communication System). Finally, the anatomical pieces were secondarily sent to the Anatomy and Organogenesis Laboratory. The parameters studied were morphological and morphometric. The data were analyzed by SPSS software version 25.0.

The different modalities of junction between the choledoc duct and the main pancreatic duct were mainly at the level of a hepatopancreatic ampoule but in a variable way. This hepatopancreatic bulb always opened at the level of a single major duodenal papilla.

In 18 cases (60%), a 'Ya' type was noted with an average length of 5.4mm of the common portion (extremes 1.9mm-9.9mm) and an average diameter of 1.6mm (extremes 1mm-3.2mm) (Figure 2).

In 1 case (3.3%), a 'Yb' type was observed with a length of 12mm of the common portion and a diameter of 1.9mm (Figure 3).

In 2 cases (6.7%), the hepatopancreatic bulb was shaped like a 'V' with an average length of 0.95mm of the common portion and an average diameter of 0.7mm (Figure 4).

In 8 cases (26.7%), the hepatopancreatic ampoule had the shape of a 'U' with the main choledoc and pancreatic ducts that opened at the level of the same duodenal papilla without having a common portion (Figure 5).

In 1 case (3.3%), the hepatopancreatic ampoule had the shape of an 'II' where there was no common portion between the choledocal duct and the main pancreatic duct and a distinct duodenal papilla for each duct (Figure 6).

The anterograde injection-radioscopy method is a means of seeing almost all the ramifications of the pancreatic ducts. This method has allowed many authors to provide new data on pancreatic ducts. It is an easily feasible exploration method, already proven for years [7-10]. Indeed, dissection makes it easy to follow the duct that drains the tail and body of the pancreas during its relatively straight path to the head of the pancreas. But it becomes much more difficult to follow the duct, to identify its tributaries and its communications once it has entered the head of the gland.

The rise of endoscopic retrograde pancreatography, as a clinical procedure, has prompted some authors after catheterization of the main choledocal and pancreatic ducts to obtain pancreatograms in post-mortem anatomical subjects. In 1911, Baldwin was already providing convincing results by an injection-radioscopy method rather than the pancreatic duct dissection method [11]. In addition, anterograde injection, giving similar results, is a much more physiological test providing details of the anatomy of pancreatic ducts. Over the years, several classifications of pancreatic ducts have been proposed (Millbourn, Sigfusson, Dawson, Skandalakis, and Flati) [1,10,12,23]. We used the Flati classification because it is recent and applied in comparable series like Wilasrusmee [1,14].

In most cases, the main pancreatic duct joins the choledoc duct to form a common portion before discharging through the same orifice in the second portion of the duodenum, with a clear predominance of the 'Ya' type (60%), consistent with literature. 'Yb' type (common portion > 10 mm) occurred in 3.3% of cases (Flati reported 4%). Long common channels (> 15 mm) are considered pathological and predisposing to congenital choledochal cysts [15-17] and gallbladder carcinoma due to pancreatic juice reflux [18-21]. Type 'U' (separate openings in common papilla) accounted for 26.7% (8 cases), and Type 'II' (separate papillae) accounted for 3.3% (1 case) [22-24].

The anatomical study of the hepatopancreatic ampoule in the population is interesting in the sense that it makes it possible to identify or estimate risk groups for the occurrence of certain pathologies of the bilio-pancreatic pathways. These anatomical data have important clinical applications and open up new surgical perspectives, including the prevention of the occurrence of cysts or biliary tract cancer based on early detection of anatomical variants.