Skip to main content
Journal of Clinical and Experimental Hepatology logoLink to Journal of Clinical and Experimental Hepatology
. 2024 Nov 28;15(2):102461. doi: 10.1016/j.jceh.2024.102461

Liver Explantation in Difficult Scenarios☆

Rajesh Dey 1,⁎, Shaleen Agarwal 1, Subash Gupta 1
PMCID: PMC11714708  PMID: 39802412

Abstract

Recipient hepatectomy is considered as the most difficult part of a liver transplant operation. This article describes techniques to deal with scenarios like massive caudate lobe, a recipient with a transjugular intrahepatic portosystemic shunt (TIPS) shunt in situ, a recipient with hepatocellular carcinoma, acute liver failure and a history of previous abdominal surgery.

Keywords: recipient hepatectomy, acute liver failure, hepatocellular cancer, caudate lobe


Recipient hepatectomy is a complex surgery and is often considered as the most difficult part of a liver transplant operation.1 End-stage liver disease is associated with many physiological disturbances such as coagulopathy, cardiorespiratory impairment, fluid and electrolyte imbalance and the inevitable blood loss during this extensive surgery further adds to the challenges.2 Coordinated effort between surgical and anaesthesia teams is required to minimise blood loss, maintain homeostasis and ensure optimal milieu for graft implantation.2 This article will discuss challenges in recipient hepatectomy.

Massive caudate lobe

Caudate lobe is a part of the liver located in front of IVC, behind the three major hepatic veins and cranial to the hilar plate. It is divided into three parts – the Spiegel lobe, para caval part and the caudate process. The anatomy of caudate lobe has been redefined over the years.3

The caudate lobe hypertrophies in cirrhosis of liver possibly from changes in portal blood flow, hepatic fibrosis and impaired venous outflow.4 This hypertrophy is much more pronounced when the main hepatic veins are occluded and drainage is via the small hepatic veins of the caudate lobe such as in hepatic venous outflow obstruction.5 The reason for caudate lobe hypertrophy in Budd Chiari syndrome (where hepatic veins are blocked) is intuitive, as the unimpeded separate venous drainage of the caudate lobe will assist in its compensatory enlargement. However, the reason for caudate lobe hypertrophy in other aetiologies of cirrhosis remains speculative. It is thought that in cirrhosis, there is an alteration of blood supply within liver, whereby most of the superior mesenteric venous blood, rich in hepatotropic factors, are preferentially directed to the left lobe whilst the right lobe receives more of the splenic venous blood. There is thus right lobe atrophy and left lobe hypertrophy. Caudate lobe receives a major part of blood supply from left portal vein and main portal vein and therefore hypertrophies along with the left lobe.

The inferior vena cava (IVC) must be freed for graft implantation in living donor liver transplant (LDLT) or during deceased donor transplantation when cavo cavostomy is planned. Multiple small hepatic veins need to be freed from the caudate lobe, which increases the risk of injury to the IVC with resultant bleeding and air embolism.6 Loss of tissue planes between the caudate lobe and IVC can make this dissection a daunting task and it certainly adds to the operating time.

Both in deceased donor liver transplantation (DDLT) and LDLT, it is of vital importance to identify and isolate the supra-hepatic and infra-hepatic IVC. The slings not only mark the location and direction of the IVC, it also allows the surgeon to quickly cross-clamp above and below in case bleeding is difficult to control.

Classical hepatectomy with IVC reconstruction has provided a relatively easier way for explant hepatectomy in situations where there is extensive caudate ‘wrapping’ or adhesion. In a situation where IVC reconstruction is feasible (as in DDLT/availability of grafts for IVC reconstruction), a rapid and relatively bloodless recipient hepatectomy is possible by this method.7 This situation may be encountered in cases where the retrohepatic cava is atretic and caval replacement with cryopreserved allografts is recommended in such situations. With the widespread adoption of ‘piggyback ‘technique in the DDLT scenario – the scope and practice of classical hepatectomy has reduced.

The surgeon needs to be more meticulous in scenarios where cross-clamp is not feasible. With recent advances in anaesthesia, a situation where ‘trial clamp’ fails – that is, the recipient becomes haemodynamically unstable during a trial clamp – is rare.However, should such a situation arise, a veno-venous bypass can be established. In trial clamping a cross-clamp is applied in the infra-hepatic and supra-hepatic IVC. The surgeon and the anaesthetist work in tandem and any resultant hemodynamic instability that is not amenable to correction by an increase in fluids or inotropes is noted and considered as a contraindication for proceeding with a cross-clamp without establishing a veno-venous bypass.

A ‘porta first’ approach – where the portal vein and hepatic hilum are divided before full mobilisation of the liver lobes – may allow for diminished blood loss. However, a prolonged anhepatic phase is also harmful as it may result in refractory acidosis and hyperlactatemia.8 A temporary porta caval shunt can be constructed to avoid bowel congestion and reduce reperfusion syndrome in such a scenario.9

Division of caudate wrap gives access to the right lateral surface of the IVC and aids in liver mobilisation (Figure 1, Figure 2).

Figure 1.

Figure 1

Division of caudate ‘wrap’.

C – caudate, IVC- Inferior Vena Cava.

Figure 2.

Figure 2

Extraction of a TIPS stent after cross-clamping.

The Black bold arrow marks the stent.

The assistant pulls the liver to the left and the operating surgeon further pushes up the right lobe of the liver so that the IVC becomes free. Occasionally, mobilisation from the left side by retracting the liver to the right may aid in caval mobilisation from the Caudate lobe.

In polycystic disease, the liver is often massively enlarged-access to the IVC is rendered difficult. Accessing the porta first and dividing the portal structures will help in reducing the vascularity and mobilisation of liver. A liberal incision is helpful.

Post-TIPS recipient hepatectomy

TIPS shunt had originally been developed as a bridge to liver transplantation. In recent years its use has increased and therefore it is not unusual to transplant patients with a working TIPS shunt. Interestingly, recipient hepatectomy is relatively easy to perform as there is no portal hypertension.10 However, these patients will not tolerate anhepatic phase well and serum lactate will increase rapidly.

Recipient hepatectomy is done in the usual manner with left lobe mobilisation first followed by Right lobe mobilisation and slinging of lower IVC. Cholecystectomy is done next followed by identification and ligation of the left hepatic artery distal to its branching. At this stage, the stent may be felt in the main portal vein branch which must be then dissected more proximally. Occasionally a portal venotomy may allow stent extraction. Usually, they can be retrieved easily, but may cause tear of the vein wall. Covered stents are easier to remove. Resecting the damaged part of the main portal vein and extending the portal vein with a cryopreserved allograft will be required in such situations.

Extra mobilisation up to the base of the right atrium may be needed in the supra-hepatic cava so that we can clamp easily for stent removal. A top caval dissection may result in injury to the supra-hepatic IVC. Mobilisation of infra-hepatic IVC to gain control over the bottom cava is a prerequisite which can allow a quick cross-clamp should there be an inadvertent IVC injury. Similarly, adequate mobilisation of supra-hepatic cava by ligating and dividing the phrenic veins can provide rapid access to the top cava for application of a cross-clamp in case of inadvertent caval injury. Of course, these difficulties may be recognised by preoperative imaging and the procedure planned accordingly.11

HCC

Liver transplantation is the definitive treatment for Hepatocellular carcinoma.12 Criteria for candidate selection are outside the scope of this chapter, hence will not be discussed here.

Tumour rupture and dissemination while mobilising the liver during recipient hepatectomy is possible, especially for exophytic lesions and those adherent to the diaphragm. Tumours that are adherent to surrounding structures such as the diaphragm or colon may require en bloc excision.

If the transplant is after a previous hepatic artery angiogram (previous TACE/TARE) hepatic arterial dissection may be present or occur whilst dissection is going on. Minimal direct handling of the arteries and maintaining a plane of dissection away from the arterial wall is essential. Ligation of hepatic artery should be beyond the distal branches after laying bare the branches to gain maximum length on the recipient side.13

The ‘no touch’ technique has been described to minimise the chances of tumour dissemination. In this technique, hilar dissection of hepatic arteries is done first followed by division of the hepatic artery and portal vein to devascularise the liver. A temporary portocaval shunt may be established to prevent oedema from splanchnic congestion. The left lobe is mobilised along with the division of hepatic veins. The right lobe is mobilised last. Total hepatectomy under total vascular exclusion and veno-venous bypass has also been described.

For tumours at the hilum, bile ducts should be divided distally and biliary reconstruction in case of multiple ducts can be done by Roux-en-Y Hepaticojejunostomy.

Previous upper abdominal surgery

Previous abdominal surgery in a liver transplant candidate poses major surgical challenges from vascularised adhesions resulting in difficult and bloody access to the abdomen and liver.14 Once access has been gained to the abdomen, attempt should be made to enter the area just around the liver in an extraperitoneal manner.

The liver is mobilised from top to bottom and the porta is approached from right to left rather than from anterior to posterior. The top-to-bottom approach means that the superior surface of the liver where the leaves of falciform ligament separate into the right and left coronary ligament is often free of intense adhesions. By accessing that part of the liver first, one can then release the adhesions towards the edge of the liver with ease.

The porta hepatis is often densely adherent with overlying omentum and bowel. The approach to the right border of the porta is more convenient in such scenarios. It helps by providing access to the Foramen of Winslow, which can be used to sling the porta hepatis.

One should stay away from Infra-colic adhesions and the bile duct should be preserved so that there is no need for hepatico jejunostomy which may be very difficult to fashion.

Occasionally the left lobe is hypertrophied and densely adherent to the spleen. In this situation, the adherent liver may be divided by an energy device. Attempts at separating from the spleen are liable to result in injury and splenectomy may add to complications such as pancreatic fistula and splenic and portal vein thrombosis. It is advisable to leave behind a part of liver parenchyma adherent to the spleen in such situations. At the end of surgery, this could be assessed and removed if feasible.

A sliver of liver tissue is often left behind on the adherent bowel wall, particularly while taking down a Roux-en-Y hepaticojejunostomy such as when operating after a Kasai procedure.

Collateralised adhesions should be divided between ligatures to avoid chances of delayed haemorrhage and use of diathermy minimised as late perforation may occur with thermal injury.

Acute liver failure

ALF and its consequent haemodynamic and metabolic imbalance lead to severe stress for the patient.15 Further, it is not uncommon for the surgery to take place during unscheduled hours. Preparation must start from the head end propped up all throughout so that cerebral oedema is not worsened.2

The liver is usually small and shrunken and surprisingly in spite of the raised prothrombin time, hypercoagulability is often seen. Due to the absence of portal hypertension explant hepatectomy is relatively easy in these patients; however, due to the absence of portal hypertension, clamping of portal vein can result in bowel ischemia and therefore the anhepatic time must be minimised. A temporary end-to-side portocaval shunt is helpful in such situations.

If the patient is unstable with hypotension and septic, an emergent hepatectomy followed by portacaval shunt will often stabilise the patient till the donor liver is ready to be implanted.

Biliary and infective complications are higher in ALF and therefore, if possible, graft selection should be careful. Single bile duct may minimise biliary complications.

Recipient explantation is one of the most complex and difficult parts of liver transplantation. Meticulous preoperative planning is essential to avoid intraoperative surprises. Different scenarios like TIPS, massive caudate lobe and acute liver failure need tailor-made approaches for optimum outcomes.

Funding

No funds were received in the preparation of this article.

Credit authorship contribution statement

The authors mentioned in the title page are the sole authors of this manuscript.

1. The reason for caudate lobe hypertrophy in Budd Chiari syndrome (where hepatic veins are blocked) is intuitive, as the unimpeded separate venous drainage of the caudate lobe will assist in its compensatory enlargement. However, the reason for caudate lobe hypertrophy in other aetiologies of cirrhosis remains speculative. it is thought that in cirrhosis, there is alteration of blood supply within liver, whereby most of the superior mesenteric venous blood, rich in hepatotropic factors, are preferentially directed to left lobe whilst the right lobe receives more of the splenic venous blood. There is thus right lobe atrophy and left lobe hypertrophy. Caudate lobe receives a major part of blood supply from left portal vein and main portal vein and therefore hypertrophies along with the left lobe.

2. “cadaveric” to “deceased donor” in “The Inferior vena cava (IVC) must be freed for graft implantation in Living donor liver transplant (LDLT) or during cadaveric transplantation”

3. The reviewer suggestion was “what is the way out in conditions where trial clamp fails ? The venovenous bypass option could have been discussed in more detail”.

4. Mention regarding difficulties in explanting a massive Polycystic liver.

Declaration of competing interest

The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.

Footnotes

☆

This article is part of a special issue entitled: Liver Transplant surgery published in Journal of Clinical and Experimental Hepatology.

References

  • 1.Makowka L., Stieber A.C., Sher L., et al. Surgical technique of orthotopic liver transplantation. Gastroenterol Clin N Am. 1988 Mar;17:33–51. [PMC free article] [PubMed] [Google Scholar]
  • 2.Kashimutt S., Kotzé A. Anaesthesia for liver transplantation. BJA Education. January 2017;17:35–40. [Google Scholar]
  • 3.Kumon Masamitsu. Anatomical study of the caudate lobe with special reference to portal venous and biliary branches using corrosion liver casts and clinical application. Liver Cancer. 14 February 2017;6:161–170. doi: 10.1159/000454682. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 4.Bolog N., Oancea I., Andreisek G., Mangrau A., Caruntu F. Diagnosis performance of different MR imaging signs of cirrhosis: the caudate to right lobe ratio, the posterior right hepatic notch, and the expanded gallbladder fossa. Curr Health Sci J. 2009 Jan;35:16–22. [PMC free article] [PubMed] [Google Scholar]
  • 5.Kikuchi Y., Yoshida H., Mamada Y., et al. Huge caudate lobe of the liver due to Budd-Chiari syndrome. J Nippon Med Sch. 2010 Dec;77:328–332. doi: 10.1272/jnms.77.328. [DOI] [PubMed] [Google Scholar]
  • 6.Smart K.J., Yaghoubian S. Uncommon occurrence of an air embolism during the preanhepatic phase of an orthotopic liver transplant. Case Rep Anesthesiol. 2021 Aug 9;2021 doi: 10.1155/2021/6628961. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 7.Chan T., DeGirolamo K., Chartier-Plante S., Buczkowski A.K. Comparison of three caval reconstruction techniques in orthotopic liver transplantation: a retrospective review. Am J Surg. 2017 May;213:943–949. doi: 10.1016/j.amjsurg.2017.03.045. [DOI] [PubMed] [Google Scholar]
  • 8.Ijtsma A.J., van der Hilst C.S., de Boer M.T., de Jong K.P., Peeters P.M., Porte R.J. The clinical relevance of the anhepatic phase during liver transplantation. Liver Transplant. 2009 Sep;15:1050–1055. doi: 10.1002/lt.21791. [DOI] [PubMed] [Google Scholar]
  • 9.Nacif L.S., Zanini L.Y., Sartori V.F., et al. Intraoperative surgical portosystemic shunt in liver transplantation: systematic review and meta-analysis. Ann Transplant. 2018 Oct 16;23:721–732. doi: 10.12659/AOT.911435. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 10.Salvalaggio P.R., Koffron A.J., Fryer J.P., Abecassis M.M. Liver transplantation with simultaneous removal of an intracardiac transjugular intrahepatic portosystemic shunt and a vena cava filter without the utilization of cardiopulmonary bypass. Liver Transplant. 2005 Feb;11:229–232. doi: 10.1002/lt.20292. [DOI] [PubMed] [Google Scholar]
  • 11.Gilbo N., Ceulemans L.J., Maleux G., et al. How to deal with a nonextractable transjugular intrahepatic portosystemic shunt complicating liver transplantation. Liver Transplant. 2018 Sep;24:1293–1297. doi: 10.1002/lt.25209. [DOI] [PubMed] [Google Scholar]
  • 12.Shimamura T., Goto R., Watanabe M., Kawamura N., Takada Y. Liver transplantation for hepatocellular carcinoma: how should we improve the thresholds? Cancers. 2022 Jan 14;14:419. doi: 10.3390/cancers14020419. PMID: 35053580; PMCID: PMC8773688. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 13.Agarwal S., Dey R., Pandey Y., Verma S., Gupta S. Managing recipient hepatic artery intimal dissection during living donor liver transplantation. Liver Transplant. 2020 Nov;26:1422–1429. doi: 10.1002/lt.25857. [DOI] [PubMed] [Google Scholar]
  • 14.Silva J.P., Berger N.G., Yin Z., et al. The effect of prior upper abdominal surgery on outcomes after liver transplantation for hepatocellular carcinoma: an analysis of the database of the organ procurement transplant network. Surgery. 2018 May;163:1028–1034. doi: 10.1016/j.surg.2017.10.072. [DOI] [PubMed] [Google Scholar]
  • 15.Mendizabal M., Silva M.O. Liver transplantation in acute liver failure: a challenging scenario. World J Gastroenterol. 2016 Jan 28;22:1523–1531. doi: 10.3748/wjg.v22.i4.1523. [DOI] [PMC free article] [PubMed] [Google Scholar]

Articles from Journal of Clinical and Experimental Hepatology are provided here courtesy of Elsevier

RESOURCES