Anatomical parameters do not determine linear breaking strength or dimensions of the porcine biliary system

Abstract

Background

The porcine gall bladder and cystic duct gained attention in experimental research aimed at improving human clinical care. While the common bile duct has been investigated before, there is almost no data on the porcine cystic duct. Its relevance for research originates from its potential use in xenotransplantation.

Methods

We included 10 consecutive pigs (1♀, 9♂) that had been sacrificed after participation in another study. At necropsy, the distance from the biliary papilla to the gall bladder (gall bladder distance) was measured. The cystic duct was then subjected to linear traction. Ethical approval for the study was granted. Associations between force measurements and anatomical parameters were tested by correlation analysis.

Results

The pigs had a mean body weight of 21.9 kg (standard deviation 2 kg), a mean liver weight of 590 g (standard deviation 88 g), and a mean crown rump length of 67.3 cm (standard deviation 3 cm). The gall bladder distance was 8.2 cm (95% confidence interval: 7.6–8.7 cm). The cystic duct withstood mean linear traction forces of 4.8 N (95% confidence interval: 3.7–5.8 N) and could be elongated by a mean of 6 mm (95% confidence interval: 3.9–8 mm). Linear breaking strength was neither correlated to gall bladder distance (R = 0.3, 95% confidence interval: − 0.41 to 0.78, P = 0.406) or crown rump length (R = 0.42, 95% confidence interval: − 0.28 to 0.83, P = 0.222) nor liver weight (R = 0.02, 95% confidence interval: − 0.62 to 0.64, P = 0.954) or body weight (R = 0.36, 95% confidence interval: − 0.35 to 0.81, P = 0.304). This was also true for gall bladder distance.

Conclusions

Our study indicates that allometric parameters were not associated with the gall bladder distance or its resistance to traction forces. Consequently, substantial variation of these parameters can be expected in these surgically important parameters, which cannot be derived from easily accessible anatomical parameters.

Background

Pigs have been the preferred pre-clinical model for experimental surgery due to their unique anatomy and physiology (Swindle et al. 1994) and their similarity to humans (Baumgart et al. 2020; Oetzmann von Sochaczewski et al. 2020). The common bile duct has been evaluated for its biomechanical properties in pigs (Duch et al. 2004) and humans (Girard et al. 2019). Comparative studies were also conducted between the two species (Li et al. 2013), because of the importance of anatomical and biomechanical aspects in transplantation research (Girard et al. 2018). The biomechanical properties of the cystic duct have only been investigated once in dogs (Jian and Wang 1991), despite their relevance for pre-clinical interventional (Woo et al. 2016) and surgical procedures (Bos et al. 2017). In addition, the suitability of the cystic duct as a salvage option for complicated bile duct reconstruction in transplantation has first been described by Suh et al. (2004), but has been gaining ground recently, because several cohort studies reported its successful use without negative influence on graft- and overall survival (Hsu et al. 2022; Vijayashanker et al. 2021; Özçelik et al. 2021; Kubo et al. 2023). The cystic duct can be used to anastomose the right anterior intrahepatic duct to the recipient’s biliary system in right-lobe living-donor liver transplantation (Hsu et al. 2022). This approach allows a duct-to-duct reconstruction in the presence of an atypical biliary anatomy with two duct openings of the right liver lobe instead of using a bilio-digestive anastomosis with its associated risk of infection, but is only feasible for right-lobe grafts, whereas hepatico-jejunostomy remains the surgical approach of choice in left-lobe grafts (Vijayashanker et al. 2021). Although this is not a preferred approach, its long-term feasibility without a negative impact on graft- and overall survival has been demonstrated (Kubo et al. 2023), indicating a relevance of the cystic duct in transplant surgery. Knowledge of the anatomical peculiarities and biomechanical properties of the biliary system is a relevant aspect for reliable results that could withstand the test of transferability to humans (Lossi et al. 2016). We aimed to fill this knowledge gap by assessing anatomical and biomechanical properties of porcine cystic ducts.

Methods

We included 10 consecutive German Landrace pigs (Sus scrofa domestica), 9 males and 1 female, aged eight to twelve weeks, supplied in house by the Landwirtschaftliche Fakultät der Rheinischen Friedrich-Wilhelms-Universität Bonn, Lehr- und Forschungsstation Frankenforst (Königswinter-Vinxel, Germany) (Kratz et al. 2024). Feeding and husbandry practices prior to transfer to our facility followed the husbandry regulations and standard guidelines of the Zentralverband der Deutschen Schweineproduktion. Vaccinations against mycoplasma were administered on days 3 and 10 of life. The pigs were of conventional microbiological status and were given three days to acclimatise at our facility (Kratz et al. 2023a). Temperatures were regulated between 16 and 18 °C with a relative humidity of 50–60% and an air exchange of at least 8 times per hour. Pigs were housed alone in a box of 4–6 m² enriched with chains, balls, and additional materials to play with. An infrared heating lamp was always available for the pigs. Darkness cycles of twelve hours were used with artificial lighting between 7 and 19 o’clock (Baumgart et al. 2020; Kratz et al. 2023b). Drinking water was available ad libitum, and pigs were fed regular chow (Altromin 9023, Altromin Spezialfutter, Lage, Germany).

The animals were subject to preceding experiments that were approved by the Landesamt für Natur, Umwelt und Verbraucherschutz Nordrhein-Westfalen (permit-number 81-02.04.2020.A392) and conducted in accordance with national legislation (Lindner et al. 2020) and the directive 2010/63/EU (Kratz et al. 2024). Due to the lack of available data on porcine cystic ducts, a formal power analysis was impossible. Thus, we used a sample size of 10 animals, which were considered appropriate for exploratory research in porcine common bile ducts (Duch et al. 2004). However, these 10 animals would be sufficient to confirm the correlation between body weight and liver weight described with R = 0.89 (Soleimani et al. 2012). Using G*Power 3.1.9.2. (Faul et al. 2007), with a point biserial model, a sample size of 10 animals would be sufficient to confirm the result of Soleimani et al. (Soleimani et al. 2012) with α = 0.005 and β = 0.05, resulting in a statistical power of 95%. Only one animal was used per day, because the experimental surgery divided the animals into separate mini-experiments thereby increasing reproducibility (Kortzfleisch et al. 2020).

The crown rump length was measured from the top of the skull to the base of the tail (Kratz et al. 2023c). Experiments began with an immediate post-mortem necropsy as described in detail elsewhere (Kratz et al. 2023c) and were conducted within one hour after the pig’s demise to avoid structural changes due to cell death (Kratz et al. 2023c; Saxena et al. 2021). Pigs were sacrificed using an overdose of T61® [tetracaine/mebezonium/embutramide] of 0.5 ml/kg (Kratz et al. 2023b). In brief, the abdominal cavity was opened and the abdominal organs were explanted (Fig. 1A). The bile ducts and gall bladder were then identified and freed from surrounding tissue to identify the distal end of the cystic duct (Fig. 1B), which was then followed to the biliary papilla and transected. Our experimental setup has been described in detail before (Oetzmann von Sochaczewski et al. 2019a). The distance between the gall bladder and the biliary papilla was measured (Fig. 1C), and it was mounted on the motorised test stand (Sauter THM500N, Kern & Sohn, Balingen, Germany). The cystic ducts were subjected to linearly increasing traction forces by 10 mm per minute until the tissue ruptured (Fig. 1D). This was registered as a loss of force by the tensiometer (FL100, Kern & Sohn, Balingen, Germany), and the maximum value was recorded (Oetzmann von Sochaczewski et al. 2019b, 2021, 2019c).

Fig. 1

Photographic depiction of the procedures. A Explantation of the whole abdominal organ package. B Isolation of the gall bladder and the cystic duct from the liver. C Determination of the cystic duct length. D Measurement of linear breaking strength and lengthening of the cystic duct

Representative samples were taken from the site of the tissue rupture and one centimeter proximal to it. Specimens were immersed in 4% buffered formaldehyde solution (Sigma Aldrich, Darmstadt, Germany) and fixed for 24 h at room temperature. The specimens were then dehydrated in ascending concentrations of alcohol and embedded in Tissue-Tek III paraffin wax (Sakura Finetek, Alphen aan de Rhijn, The Netherlands). Sections of 3 µm were stained with Hematoxylin–Eosin and Elastica-van Gieson (Medite Tissue Stainer, Medite, Burgdorf, Germany) and Masson’s trichrome (Avantor, VWR, Darmstadt, Germany) (Kratz et al. 2023b).

Statistical analysis was performed using Graph Pad Prism 8. A Gaussian distribution was checked by the D’Agostino-Pearson test, supported by visual analysis of QQ plots. Correlations were calculated using Pearson’s R. Raw data for the numerical analysis are available via Zenodo (Ginghina et al. 2023).

Results

The pigs had a mean body weight of 21.9 kg (standard deviation 2 kg), a mean liver weight of 590 g (standard deviation 88 g), and a mean crown rump length of 67.3 cm (standard deviation 3 cm). The distance from the biliary papilla to the gall bladder had a mean length of 8.2 cm (95% confidence interval: 7.6–8.7 cm) and a tapered form towards the gall bladder (Fig. 1C). It had a mean diameter of 2.7 mm (95% confidence interval: 2.2–3.1 mm) at the junction with the common bile duct and was considerably smaller with 1.1 mm (95% confidence interval: 0.9–1.3) at the gallbladder neck. We did not observe a ramus lobi medialis dextri joining the cystic duct. The cystic duct, all specimens torn at the smallest diameter, withstood mean linear traction forces of 4.8 N (95% confidence interval: 3.7–5.8 N) and could be elongated by a mean of 6 mm (95% confidence interval: 3.9–8 mm).

Linear breaking strength was not correlated with the distance from the biliary papilla to the gall bladder (R = 0.3, 95% confidence interval: − 0.41 to 0.78, P = 0.406) or with crown rump length (R = 0.42, 95% confidence interval: − 0.28 to 0.83, P = 0.222). Similarly, liver weight (R = 0.02, 95% confidence interval: − 0.62 to 0.64, P = 0.954) or body weight (R = 0.36, 95% confidence interval: − 0.35 to 0.81, P = 0.304) were not correlated with linear breaking strength. The distance from the biliary papilla to the gall bladder was also not correlated with liver weight (R = − 0.22, 95% confidence interval: − 0.75 to 0.48, P = 0.544), body weight (R = − 0.07, 95% confidence interval: − 0.67 to 0.59, P = 0.852) or crown rump length (R = 0.43, 95% confidence interval: − 0.28 to 0.83, P = 0.221). The relationship between all investigated parameters is shown in a correlogram (Fig. 2).

Fig. 2

Correlogram of the investigated parameters. Correlation analysis of 10 pigs with Pearson’s R. All data had a normal distribution and their relationship to each other was linear

Cystic duct elongation was not included in the analysis, because it was not independent of linear breaking strength. Histologic analysis demonstrated the substantial destruction of the cystic duct by the traction forces (Fig. 3).

Fig. 3

Representative histology of the torn bile ducts depicting substantial destruction of the structures of the hepato-duodenal ligament. A Hematoxylin–Eosin staining with 1.25× magnification. B Elastica-van Gieson staining with 1.25× magnification. C Masson’s trichrome staining with 1.25× magnification. D Hematoxylin–Eosin staining with 10× magnification. E Hematoxylin–Eosin staining with 40× magnification

Discussion

Porcine models are preferred due to their unique physiology and anatomical similarity to humans (Swindle et al. 1994). The biomechanical properties of the porcine biliary system are not described beyond the common bile duct (Duch et al. 2004; Li et al. 2013), despite its relevance to pre-clinical research (Woo et al. 2016; Bos et al. 2017). The importance of such data for the common bile duct has been demonstrated by the improved outcomes due to stenting it after liver transplantation (Girard et al. 2018). For the cystic duct, knowledge of its biomechanical properties are also necessary for its trans-hepatic access (Ogawa et al. 2022) or endoscopic access (Woo et al. 2016) to the cystic duct to avoid surgery in high risk patients. Successful translational research requires such information, as a potentially promising treatment may end up as non-reproducible experimental research, if the similarities and differences between species are not taken into account. Previous reports on the simulation of endoscopic retrograde cholangio-pancreaticography have focused on fasting times and pig weight in relation to the length of the duodenoscopes used, but not on the biliary system (Noar 1995).

In xenotransplantation, determining the appropriate graft size is crucial. Previous research in German landrace pigs demonstrated an almost perfect correlation between porcine body weight and liver weight with a Pearson’s R of 0.89 (Soleimani et al. 2012), whereas such a relationship was not present in our results. Although our data could be considered somewhat compatible (Rafi and Greenland 2020) with the assumption that there would be a relationship between body and liver weight, but its magnitude is much smaller than described before (Soleimani et al. 2012). Such information is important, because decades of research have considered a pig weight around 30 kg to be ideal for xenotransplantation, from a surgical point of view (Chalstrey et al. 1971). However, the size and biomechanical properties of the liver vasculature were related to age of the pigs and were considered ideal at an age of approximately five months due to its similarity to the human vasculature at that age (Yifei et al. 2005). Therefore, these parameters need to be aligned between pigs and humans to come as close as possible to each other. To achieve this goal, it is necessary to identify the factors that can guide the selection of the appropriate animal size need to be identified. Additional research is needed to address this aspect of predicting organ weights from porcine body weight.

Another relevant issue regarding xenotransplantation is the pig breed: In an anatomical study from Egypt the ramus lobi medialis dextri was described to join the cystic duct in about one third of the specimens studied (Osman et al. 2008), whereas we did not encounter this variant. This is relevant because it would require an additional small-calibre anastomosis in the biliary system in xenotransplantation, which has been associated with biliary complications (Park et al. 2008). Consequently, avoiding such an anatomical variant by choosing the breed of pig could improve the outcome of xenotransplantation.

The only study of the biomechanical properties of the cystic duct beyond the common bile duct dates back to 1991 and was conducted in dogs (Jian and Wang 1991). This study showed that the elasticity of the cystic duct is greater than that of the common bile duct and that it can therefore withstand higher intraluminal pressures (Jian and Wang 1991). Based on the relatively low Young’s modulus of the cystic duct, compared to the other parts of the biliary system, one would have expected more stretch before tissue destruction. However, burst pressures have important drawbacks, because they do not allow comparison of absolute values between studies (Oetzmann von Sochaczewski et al. 2019a). In addition, previous studies have shown that stress testing in dogs yields much lower values in canine organs than in porcine organs (Stavropoulou et al. 2012), both in terms of burst pressures and linear breaking strength. In terms of absolute values, the linear breaking strength of porcine cystic ducts appears to be rather low, based on similar numerical results from oesophageal anastomoses in rats, whose body weight was 100 times less than that of our pigs (Tagkalos et al. 2020). However, in the absence of suitable comparators, absolute values are difficult to interpret.

A relevant limitation of our study is the use of a single breed, as our in house supplier could only provide German landrace pigs (Kratz et al. 2023c), but we considered this breed to be appropriate, because previous research on xenotransplantation has described an almost perfect correlation between porcine body weight and liver weight in the very same pig breed (Soleimani et al. 2012).

Conclusions

Our study demonstrated that the biliary ducts have a tapered shape with a doubled diameter at the junction of the cystic and the hepatic ducts. This finding complicated our initial approach to measure additional biomechanical parameters. This substantial difference in size is likely to influence these parameters in such a way that they would differ along the cystic duct, as it is the case in the oesophagus. In summary, we have demonstrated that surgically relevant parameters for xenotransplantation cannot be derived from easily accessible external anatomical parameters.