The goal of this study is to modify the rat orthotopic liver transplant model to better represent human liver transplantation and improve recipient survival. The presented method reestablishes hepatic arterial inflow by connecting the donor liver's common hepatic artery to the recipient liver's proper hepatic artery.
The rat orthotopic liver transplantation (OLT) model is a powerful tool to study acute and chronic rejection. However, it is not a complete representation of human liver transplantation due to the absence of arterial reconnection. Described here is a modified transplantation procedure that includes the incorporation of hepatic artery (HA) reconnection, leading to a marked improvement in transplant outcomes. With a mean anhepatic time of 12 min and 14 s, HA reconnection results in improved perfusion of the transplanted liver and an increase in long-term recipient survival from 37.5% to 88.2%. This protocol includes the use of 3D-printed cuffs and holders to connect the portal vein and infrahepatic inferior vena cava. It can be implemented for studying multiple aspects of liver transplantation, from immune response and infection to technical aspects of the procedure. By incorporating a simple and practical method for arterial reconnection using a microvascular technique, this modified rat OLT protocol closely mimics aspects of human liver transplantation and will serve as a valuable and clinically relevant research model.
The global burden of liver disease continues to increase, with a 30% rise in liver disease-related deaths from 2005 to 20131,2. Liver transplantation is often the only recourse for patients with end-stage liver disease. The liver is the second most frequently transplanted solid organ, and the number of liver transplants performed globally increased by 7.25% from 2015 to 20161,2. Despite its prevalence, post-transplantation survival rates have become stagnated3,4,5. The 15 year patient survival rate is reported to be 53%, and the 20 year patient survival rate may be as low as 21%3,5. While there are exciting new immunobiology initiatives that may lead to new treatments and improved clinical outcomes, there is not yet a reliable small animal model in which to test them.
The rat OLT model has been widely used in the investigation of liver transplantation, including rejection6,7,8,9,10, immune tolerance11, transplant ischemia-reperfusion injury12, immunosuppression13, and biliary tree injury14,15,16,17. However, a disadvantage of the model in its current form is its high post-operative morbidity and mortality18,19. This is a serious drawback that is at odds with the human operation, and it compromises the ability to draw clinically relevant conclusions from the model20.
Additionally, a large proportion of this morbidity can be attributed to an absent or imperfect hepatic artery (HA) reconnection18. Although a critical step in human liver transplantation, technical difficulties tend to compromise HA reconnection in the rat OLT model. As a result, bile duct (BD) anastomosis is tenuous and results in high rates of bile leakage and BD necrosis21. Beyond the high incidence of biliary complications22, an absence of arterial inflow alters the physiology of the graft liver post-transplantation23, with hypoxia in the donor liver graft24 and liver damage observed in inflamed lobes19,25,26. Rat OLT without arterial reconnection also tends to promote fibrosis27. The rat OLT protocol described below addresses these issues by incorporating a simple HA reconstruction step with a previously published rat OLT method28, resulting in preservation of liver parenchyma and improved survival rates.
Liver transplantation has three phases: (1) extraction of the liver graft from the donor, (2) preparation of the donor liver graft, and (3) replacement of the recipient liver with the liver graft. The procedure involves the manipulation of five anatomical structures: the suprahepatic inferior vena cava (SHVC), portal vein (PV), infrahepatic inferior vena cava (IHVC), hepatic artery (HA), and bile duct (BD).
OLT in the rat was first introduced by Lee et al. using microsuture anastomosis of the SHVC, PV and IHVC, and a pull-through technique for the BD29. This model was later improved through use of the two-cuff technique in 197930. Since then, several alternative techniques have been proposed, with the majority focusing on venous anastomosis and using a two-cuff technique with a few modifications31. Although HA anastomosis has been described previously in the rat OLT model using techniques such as microsuture, cuff, and intraluminal sleeves26,31,32,33,34, these techniques often require highly trained microsurgical skills, significantly alter rat physiology, and are hampered by thrombosis and/or biliary complications27,35.
Furthermore, the choice of surgical procedure can also influence the anhepatic time (time from PV clamping to reperfusion of graft through the reconstituted PV), which is critical for the success of rat liver transplantation. Specifically, high survival rates are observed with anhepatic times of 15-20 min36, and 30 min is the upper limit for success37,38. Therefore, the goal of this method is to implement a less invasive and more easily adoptable surgical rat OLT model that is able to reconnect the hepatic artery, promote efficient perfusion of the transplanted liver, maintain flow to the recipient bile duct, and preserve the recipient's physiological condition.
Detailed here are all the steps of this revised protocol, including the manipulation of the celiac trunk of the donor liver as well as the use of 1) a 1.5 mm stent to perform an extraluminal sleeve connection with the recipient proper HA, 2) a running suture for the SHVC reconstruction, 3) two 3D-printed plastic cuffs for PV and IHVC reconstruction39,40, 4) a microvascular sleeve reconnection for the HA18,27,41 and 5) a previously described BD stenting technique28. Two additional steps are also included: a cold flush via the PV, and an antibiotic regimen that is based on previous findings17. This optimized OLT protocol minimizes perioperative complications and morbidity and more closely models the surgical operating procedure employed in human liver transplantation.
The study was performed according to the guidelines of rodent handling and surgery, and the study protocol was approved by the University Health Network Animal Care Committee (UHN AUP #: 5840.3) and follows the guidelines of the Canadian Council of Animal Care. The study uses male Lewis rats (strain LEW/SsNHsd), 12-14 weeks old, weighing between 250-300 g.
1. Equipment set-up
2. Donor operation
3. Donor rat liver preparation ("back bench")
4. Recipient operation
5. Post-operative care
While establishing a non-HA anastomosis rat OLT model using a previously described protocol28, our team observed 50% and 37.5% survival rates at 21 days and 60 days post-operation, respectively. Although high rates of long-term survival without HA anastomosis have been reported by some groups28, these early results highlight the drawbacks of not having arterial inflow. By contrast, the optimized HA reconnection procedure significantly increased long-term survival from 37.5% to 88.2% (p = 0.015) (Figure 6).
Histological analysis of a representative subset of transplanted animals without HA reconnection (at days 6 and 13 post-operation) showed signs of hypoxic liver injury with centrilobular necrosis (Figure 7). Extensive liver necrosis was associated with tremendously elevated levels of alanine aminotransferase (ALT) and aspartate aminotransferase (AST) in these animals (Figure 7). In contrast, transplanted rats with HA reconnection showed no signs of liver injury, and histological analysis revealed a normal liver parenchyma structure with organized acini, lobules (e.g., central vein and portal triads with hepatic vein), arteries, and bile duct (Figure 7).
Although the mean anhepatic time over the course of 23 separate operations was acceptable (12 min and 14 s [± 78 s]), it is still possible that survival in the non-HA reconnection model can eventually be improved with increased practice. However, it is worth noting that three of the four animals transplanted without HA reconnection (which were being followed for long-term survival) were euthanized due to distress on days 56, 96, and 111 post-operation. Additionally, histological analysis of the livers revealed reactive changes after hypoxic liver injury including marked bile duct proliferation, periportal fibrosis and inflammation, and distorted liver parenchyma (Supplementary Figure 2). The presence of morphological features of hypoxic liver injury corroborate the findings that HA reconnection is important for efficient liver perfusion and normal function.
Figure 1: Schematic representation of the 3D-printed cuff design for portal vein and infrahepatic inferior vena cava. The first tie is tightened in the groove (ii) closest to the handle (iii), and the second tie is tightened in the groove (i) furthest away from the handle. The outside diameters are (iv) 2.38 mm for the portal vein (PV) and 2.15 mm for the infrahepatic inferior vena cava (IHVC). The inside diameters are (v) 1.74 mm for the PV and 1.38 mm for the IHVC. The lengths are (vi) 2.60 mm for the PV and 2.15 mm for the IHVC (exact specifications for all 3D-printed materials can be found in Supplementary Materials). Please click here to view a larger version of this figure.
Figure 2: Hepatic artery stent insertion in graft. (A) The opening end of the celiac trunk (i) is widened by cutting the splenic artery to the left gastric artery, which exposes the bifurcation of the common HA. (ii) The BD stent is tied prior to extraction of the donor rat liver. (iii) The PV cuff and (iv) IHVC cuff are inserted and tied by folding the ends of the vessels over the cuff. (B) (i) To insert the HA stent, the exposed common HA is stretched multiple times with forceps. (C) (i) The HA stent is placed securely in the common HA and tied with 8-0 prolene. (D) (i) The HA stent is flushed with (ii) Ringer's lactate solution (BD = bile duct, IHVC = infrahepatic inferior vena cava, HA = hepatic artery). Please click here to view a larger version of this figure.
Figure 3: Infrahepatic inferior vena cava connection using 3D-printed holder. (A) The (i) PV is connected using the same technique as the IHVC connection. The graft is (ii) clamped above the (iii) IHVC cuff. The recipient IHVC opening is (iv) sutured at the sides of the opening to a 3D-printed holder to keep it stretched open. A loose (v) 7-0 silk is tied around the recipient IHVC. (B) The cuff of the graft IHVC is (i) inserted within the recipient IHVC. The loose tie is now tightened. (C) The clamp is removed, and (i) the 3D printed holder is detached with scissors. (D) An additional (i) 7-0 silk is tied around the connection if not secure, but typically one tie is sufficient (PV = portal vein, IHVC = infrahepatic inferior vena cava). Please click here to view a larger version of this figure.
Figure 4: Microvascular sleeve connection of the hepatic artery. (A) (i) The BD stent is not connected to the recipient. (ii) The HA stent is placed in the graft, which is linked to the (iv) recipient's proper HA. (iii) The PV is connected. (B) 10-0 ethilon with a (i) curved needle is drawn through the HA stent to the sides of the recipient HA opening end. (C) The 10-0 ethilon is drawn back through the HA stent; therefore, the recipient's proper HA is pulled through the stent like a sleeve. (D) (i) A tie with 10-0 ethilon is made once the recipient's proper HA is pulled into the stent to the portion that first runs through the HA stent. (E) Shown here is a schematic of the HA anastomosis described in (B), (C), and (D) (BD = bile duct, HA = hepatic artery, PV = portal vein). *The opening end of the celiac trunk is widened by cutting the splenic artery to the left gastric artery. Please click here to view a larger version of this figure.
Figure 5: Bile duct connection using two stents. (A) (i) Graft BD stent is inserted into the recipient BD with the aid of the (ii) stent loosely tied at the opening of the recipient's BD. (iii) The PV is linked before the BD connection, which is located behind the BD. (B) The stent at the end of the recipient's BD is removed and used as a widened opening to (i) insert the BD stent tied to the graft. (C) The tie that is loosely securing the recipient stent is now used to tie the connection, and (i) another 7-0 silk is used to firmly keep the stent in place to avoid slipping or twisting of the stent. Please click here to view a larger version of this figure.
Figure 6: Transplant percent survival. Orthotopic rat liver transplantation without HA reconnection (n = 8) and with HA reconnection (n = 17). Animals are closely followed post-transplantation for signs of liver failure and/or infection for at least 60 days. Rats that showed no complications after surgery were considered survivors (*p = 0.015, as calculated by Kaplan-Meier estimation [long rank test]). Please click here to view a larger version of this figure.
Figure 7: Liver histopathological assessment. Representative hematoxylin and eosin-stained sections in animals (A) without and (B) with hepatic artery (HA) reconnection at days 6 and 13 after liver transplant (LTx). (C) Normal liver parenchyma showing portal triad (portal vein, artery, and bile duct), lobules including central vein, and acini. Hepatocytes next to the portal triad are zone 1 hepatocytes; hepatocytes next to the central vein within lobules are zone 3 hepatocytes; and hepatocytes between zones 1 and 3 are zone 2 hepatocytes (ALT = alanine aminotransferase, AST = aspartate aminotransferase, CV = central vein). Please click here to view a larger version of this figure.
Supplementary Figure 1: Stent and cuff dimensions. Please click here to view a larger version of this figure.
Supplementary Figure 2: Liver histopathological assessment showing disruption of liver parenchyma. Representative hematoxylin and eosin-stained sections in animals without HA reconnection at days 54, 96, and 111 after LTx. Please click here to view a larger version of this figure.
Supplementary Material 1: Porta cuff 200g – support 2.0. Please click here to view this file (Right click to download).
Supplementary Material 2: Cava cuff 200g – support 2.0. Please click here to view this file (Right click to download).
Supplementary Material 3: Liver retractor 200g. Please click here to view this file (Right click to download).
Supplementary Material 4: Back holder – 1.2. Please click here to view this file (Right click to download).
Supplementary Material 5: Cava 150g – 2.1. Please click here to view this file (Right click to download).
Supplementary Material 6: Porta 1.4.1. Please click here to view this file (Right click to download).
Supplementary Material 7: Holder arm McGil. Please click here to view this file (Right click to download).
Supplementary Material 8: Holder mini arm LAB. Please click here to view this file (Right click to download).
Supplementary Material 9: Holder and arm soft part 1.3. Please click here to view this file (Right click to download).
Supplementary Material 10: Holder base – 3.1. Please click here to view this file (Right click to download).
Small animal liver transplantation models are important for understanding transplant immunity and identifying novel therapeutic strategies32. The ideal small animal liver transplantation model replicates all steps of the human procedure, including arterial anastomosis. It can be challenging to interpret results from the rat OLT model, as most versions do not incorporate an HA anastomosis step, which leads to higher rates of complications and morbidity42. Some reconstruction procedures have used the renal artery, which requires removal of the kidney27. This protocol avoids organ removal, as it is beyond what occurs in the human procedure.
Arterial reconstructions can also be performed by manipulating the rat aorta31. However, these methods require extensive dissection and clamping of the aorta. If the clamp time is prolonged, then the recipient rat will have poor outcomes related to distal ischemia43. In humans, a LT surgical technique involves the ligation and division of the recipient gastroduodenal artery (GDA). However, the physiological and anatomical features of rodents make transplantation using this technique more physiologically challenging and may lead to complications (i.e., necrosis of the pancreas and bile duct35 and bile leakage44). The arterial reconnection in this protocol is intended to circumvent this challenge, maintain duct blood flow, and improve recipient outcome.
The use of a sleeve and stenting technique for reconstruction of the rat HA has been described previously27. In this technique, a stent is used as a guide, and the artery is reconstructed from the donor celiac trunk to the recipient common HA. The recipient common HA is then dissected out, and the recipient GDA is tied off27. As a result, the blood supply to the lower part of the recipient BD and head of the pancreas may become compromised. It is believed that the collateral circulation to this area often provides inadequate blood flow to the bile duct. For example, this protocol test-clamps the recipient GDA first with a microvascular clip, then divides the recipient BD. With the GDA clamped, the divided BD does not bleed. After removing the GDA clamp, brisk bleeding from the BD is observed. This protocol, which maintains good flow to the divided recipient bile duct, protects the physiology of the recipient liver tissue by providing adequate liver blood perfusion and preventing post-OLT hypoxic liver injury.
On the donor side, the HA stent is inserted into the celiac trunk during the graft preparation with ease by creating a patch from the celiac trunk, left gastric artery, and splenic artery. The stent can be inserted through the broad opening, which is less difficult than trying to insert the stent into the celiac trunk alone. It has been found that 24 G is an ideal size to use for the HA stent. The length of the stent should be 1.0-1.5 mm long, because it acts as an open gate to allow the recipient's proper HA to be readily pulled into the donor's common HA. With careful attention to where the 10-0 ethilon suture is placed, blood flowing through this connection will never touch the stent directly, and the recipient's proper HA will shield it from the inside, reducing the risk of complications. Importantly, the donor's HA is never clamped in order to avoid vasospasm. The success of arterial reconstruction is evaluated by leaving the donor GDA open. Successful anastomosis results in good blood flow from the donor GDA once reconstruction is completed.
In this protocol, similar to others, SHVC reconnection is the slowest step and ultimately dictates the duration of the anhepatic phase. As the duration of the anhepatic time increases, the risk ischemic injury and liver dysfunction increases45. Another critical component of OLT rat models is the sizes of the graft, stents, and cuffs. If the graft is too small, the graft may twist or flip, obstructing the vascular connections. The size of the stents and cuffs may require adjustments according to the age, sex, weight, and strain of the rat. The size of the cuffs used here was chosen as previously described28, and one cuff size that controlled for rat size was used. There were no signs of distress or complications (i.e., liver congestion, edema, ascites, or splenomegaly) during the follow-up period (to date: median = 133 days post-operation, minimum = 115 days post-op, maximum = 161 days post-op). Further studies to determine the suitable size of PV and IHVC for various rat strains accounting for both age and sex are warranted.
This modified rat OLT protocol uses 3D-printed cuffs for the PV and IHVC, as described previously39,40. Existing methods for connecting the PV and IHVC include a microsuture technique32, cuff technique46, and microsuture-temporary splint technique47. The 3D-printed cuff technique was chosen, since it allows the size of the cuff to be standardized according to rat strain and is easy to prepare and use. Large quantities of cuffs with the same dimensions can be printed at once. The outer surface of the cuff has two grooves to aid with securing ties and prevent slipping. A tail is also incorporated into the cuff design to allow for easier manipulation of the cuff. Overall, it is believed that incorporating 3D-printed cuffs leads to high success rates and reproducibility of the OLT procedure by shortening anhepatic time. It is determined that this technique also shortens the surgical learning curve.
In conclusion, the described protocol established a model that is more similar to human liver transplantation by incorporating an arterial reconnection step. This protocol can be adapted to study many immunologic and surgical aspects of liver transplantation and can serve as a model to test novel therapeutic interventions relevant to transplantation.
The authors have nothing to disclose.
This research was funded through funds from the Multi-Organ transplant program at the UHN and support from the Toronto General and Toronto Western Foundation.
10-0 Ethilon | Ethicon | 2830G | 10-0 Ethilon Black 1X5" BV100-4 Taper |
10mL Syringe | BD | B302995 | Luer-Lok Tip, Sterile, Disposable |
1mL Syringe | BD | B309628 | Luer-Lok Tip, Sterile, Disposable |
20mL Syringe | BD | B301031 | Luer-Lok Tip, Sterile, Disposable |
3D Printed Cuff for IHVC | Custom | ||
3D Printed Cuff for PV | Custom | ||
3D Printed Holder for IHVC | Custom | ||
3D Printed Holder for PV | Custom | ||
3mL Syringe | BD | B309657 | Luer-Lok Tip, Sterile, Disposable |
4-0 Sofsilk | Coviden | GS-835 | Wx coded braided silk, 30", Suture 1-Needle 26 mm Length 1/2 Circle Taper Point Needle |
5-0 Monocryl | Ethicon | Y433H | Undyed Monofilament 1X27" TF |
5mL Syringe | BD | B309646 | Luer-Lok Tip, Sterile, Disposable |
7-0 Silk | Teleflex Medical | 103-S | Black |
8-0 Prolene | Ethicon | 2775G | 8-0 Prolene Blue 1X24" BV130-5 EVP Double Armed |
Barraquer Micro Needle Holder Without Catch | Aesculap Surgical Instruments | FD231R | Curved 120 mm, 4 3/4″ |
Barraquer Needle Holder, Extra Fine Jaws 8.0mm, Curved With Out Lock | Rumex International Co. | 8-025T | Small Size, Titanium |
Barraquer Needle Holder, Fine Jaws 12.0mm, Curved With Out Lock | Rumex International Co. | 8-021T | Small Size, Titanium |
BD Insyte Autoguard BC 22 GA x 1.00 IN | BD Angiocath / Autoguard | 382523 | 22 G x 1.00" (0.9 mm x 25 mm) Wingless catheter, 37 mL/min |
BDPrecisionGlide Single-use Needles: Regular Bevel – Regular Wall. | BD | B305106 | PrecisionGlide stainless-steel needles with translucent, color-coded, polypropylene hubs. 22 G |
BD Precisionglide Syringe Needle 21G | BD | 305167 | Gauge 21, length 1.5 inch, hypodermic needle |
BD Precisionglide Syringe Needle 30G | BD | 305128 | Gauge 30, length 1 inch, hypodermic needle |
Betadine Solution by Purdue Products LP | Purdue Products Lp | 67618-150-17 | 10% povidone–iodine topical solution USP |
Bupivacaine Injection BP 0.5% | SteriMax Inc. | DIN:02443694 | 0.5% (100mg/20mL) |
Curved Tying Forceps | Duckworth & Kent | 2-501E | 6mm tying platforms, straight shafts, flat handle, length 88mm |
DC Temperature Controller | FHC Inc. | 40-90-8D | |
DK Iris Scissors (Curved) | Duckworth & Kent | 1-211B | Blunt tips, cut length 4mm, tip to pivot length 11mm, round handle, length 107mm |
Ethanol, 200 proof (100%), USP, Decon Labs | Decon Labs, Inc. | 2716 | Dilute to 70% with d2H2O |
Fine Adjustable Wire Retractor | Fine Science Tools | 17004-05 | Maximum spread: 3.5cm, Depth 5cm |
Harvard Apparatus Isoflurane Funnel-Fill Vaporizer | Harvard Appartus Limited | 34-1040SV | |
Heparin LEO(heparin sodium) | LEO Pharma Inc. | DIN:00453811 | 10,000 i.u./10 mL |
Ice-Pak | Cryopak | FIP88016 | 4.00 in. x 7.00 in., thickness 1.50 inch |
Isoflurane United States Pharmacopeia (USP) 99.9% | Piramal Healthcare Limited | DIN: 02231929 | 250 mL, Inhalation Anesthetic, NDC 66794-017-25 |
Khaw Transconjunctival Adjustable Suture Control Forceps | Duckworth & Kent | 2-502N | 5mm highly polished tying platforms, straight shafts, flat handle, length 84mm |
Lactate Ringer's Injected USP, 1000mL | Baxter Co. | DIN: 00061085 | JB2324 |
McPherson Tying Forceps | Duckworth & Kent | 2-500E | 6mm tying platforms, straight shafts, flat handle, length 90mm |
Metzenbaum Scissors – 14.5 cm | Fine Science Tools | 14024-14 | Straight Sharp/Blunt |
Micro Kitzmiller Clamp | Scanlan | 3003-630 | Jaw length 23mm, Length 11cm |
Microscope-Leica M525 F20 | Leica Microsystems | No catalog number | |
Non-woven Gauze Sponges | Fisherbrand | 22-028-556 | |
Olsen-Hegar with Suture Cutter | Fine Science Tools | 12002-14 | 15 mm cutting edge, 2mm jaw surface – 14cm |
OptixCare Eye Lube, 25gm | OptixCare | ES-KE8O-69U1 | Formerly Optixcare Surgical Eye Lubricant |
Piperacillin sodium salt | Sigma-Aldrich | P8396 | Penicillin analog |
Puritan 3" Standard Cotton Swab w/Wooden Handle | Puritan Medical Products Company LLC | 803-WC | Regular Cotton Tipped Applicator with Wooden Handle |
Round Handled Needle Holder Straight w/ Lock | Fine Science Tools | 12075-12 | Round handles allow easy fingertip adjustments – 12.5cm |
Shea Scissors Curved Blunt | Fine Science Tools | 14105-12 | Transplant scissors with light and delicate pattern – 12cm |
Stainless Steel Micro Serrefines Curved – 4mm | Fine Science Tools | 18055-06 | Jaw length 4mm, Jaw width 0.75mm, Total length 16mm, Jaw pressure 125g |
Stainless Steel Micro Serrefines Curved – 6mm | Fine Science Tools | 18055-05 | Jaw length 6mm, Jaw width 1mm, Total length 17mm, Jaw pressure 100g |
Stainless Steel Micro Serrefines Straight – 6mm | Fine Science Tools | 18055-03 | Jaw length 6mm, Jaw width 1mm, Total length 15mm, Jaw pressure 100g |
Surgical Platform | Custom, magnetic | ||
SurgiVet Vaporstick Anesthesia Machine | General Anesthetic Services, Inc | V7015 | |
T/Pump Localized Therapy | Stryker | TP700 Series | |
Vacuum-Pressure Pump | Barnant Co. | 400-1901 | |
Vannas Scissors with Microserrations Straight | Fine Science Tools | 15070-08 | Cutting edge: 5mm, Tip diameter: 0.1mm – 8.5cm |
Vetergesic Buprenorphine | Ceva Animal Health Ltd | NAC No.:12380352 | 0.324 mg/ml buprenorphine hydochloride Solution for Injection for Dogs and Cats |
Vetroson V-10 Bipolar Electrosurgical Unit | Summit Hill Laboratories | No catalog number |