CON: Portal Vein Thrombosis Does Not Impact Liver Transplantation Outcomes

Watch a video presentation of this article

Watch an interview with the author

Abbreviations

CT

computed tomography

DPV

donor portal vein

GV

gastric varix

IVC

inferior vena cava

Kid

kidney

LRV

left renal vein

LT

liver transplant

MRI

magnetic resonance imaging

ns

not significant

Panc

pancreas7

PV(S)

portal vein, stapled

PVT

portal vein thrombosis

SMV

superior mesenteric vein

Spl

spleen

SPS

spontaneous portosystemic shunt

SPV

splenic vein

SRS

spontaneous splenorenal shunt

Stom

stomach

TIPS

transjugular intrahepatic portosystemic shunt

US

ultrasound

VJG

venous jump graft

Key Points

• Portal vein thrombosis (PVT) is increasingly prevalent among candidates being evaluated for liver transplant (LT) and candidates on the wait list.

• Early detection, detailed surveillance, and medical management of PVT is imperative for wait‐listed LT candidates.

• Planned surgical techniques to reduce hemodynamic instability and optimize portal perfusion at the conclusion of transplant surgery are crucial.

• High‐volume and extensive operative experience with PVT is essential to optimize outcomes in LT candidates with PVT.

• Center experience with living donor and pediatric LT can augment techniques.

• Because most patients with PVT can achieve post‐LT outcomes comparable with those without PVT, lifesaving LT should not be withheld from patients with PVT.

PVT is increasingly prevalent in patients with cirrhosis who are being evaluated for LT. At the time of LT evaluation or during LT surgery, the prevalence rate of PVT ranged from 2% to 26%, ¹ , ² and up to 50% of these cases were unrecognized prior to LT. ¹ Although PVT was historically considered a contraindication, this notion is being challenged as the result of improvements in preoperative imaging, medical management and interventional radiological interventions, intraoperative surgical techniques, and postoperative therapies. As LT centers become more experienced in the care of patients with PVT, outcomes can be optimized. As such, patients with PVT should not be excluded from lifesaving LT.

In early studies, PVT has been reported to negatively impact post‐LT outcomes, but these studies analyzed all PVT as a single group. ³ Post‐LT mortality was found to be significantly higher in patients with PVT at 30 days. Nevertheless, there was no significant difference in survival at or beyond 1 year after LT. ³ A review of the Organ Procurement and Transplantation Network database demonstrated the presence of PVT to be an independent risk factor for 90‐day mortality and graft failure after LT, but PVT had no significant impact on long‐term outcomes in patients surviving longer than 180 days after LT. ⁴ The likely causative factor of early mortality was greater complexity during LT surgery with more blood loss and possible graft portal hypoperfusion. Several large single‐center studies have demonstrated that LT recipients with PVT have similar early postoperative mortality, 1‐ and 5‐year patient survival rates, ⁵ as well as 10‐year patient and graft survival. ⁶ Yerdel classified PVT between grades 1 and 4 based on the completeness and extensiveness of thrombosis. ⁷ The grade of thrombosis dictates the surgical technique, which is required to restore appropriate blood flow. Outcomes of grade 1 thrombosis were shown to be equivalent to controls. Grade 2 to 4 thrombosis has resulted in inferior outcomes compared with controls in earlier experiences. Multiple studies have since shown that with increasing center experience, outcomes of all grades of thrombosis have improved. When comparing early versus later era outcomes in patients with PVT undergoing LT, Yerdel et al. ⁷ demonstrated that patients transplanted in the later era experienced less bleeding, shorter operative times, and reduced in‐hospital mortality with resultant improved short‐ and long‐term survival. Ravaioli et al. ⁵ demonstrated that the number of patients with PVT increased over time (7% to 13%) with a higher rate of complete PVT. When moving from the first to the second era, the hospital mortality rate decreased for all PVT grades, and graft loss decreased significantly. During the first era, 1‐ and 5‐year survival rates were higher for partial PVT compared with complete PVT (90% and 75% versus 57% and 57%, P < 0.05, respectively), but this difference was no longer present in the second era. Table 1 provides a summary of outcomes from selected studies.

Table 1.

Summary of Post‐LT Outcomes From Selected Studies of Patients With PVT

Author	PVT/Total (%)	PVT Grade (Yerdel or Partial, Complete)	Post‐LT Outcomes
			Postoperative Hospital Mortality (%)	Primary Nonfunction (%)	1‐Year Patient Survival (%)	5‐Year Patient Survival (%)	Overall Mortality (%)
Yerdel et al. 7	63/779 (8%)	G1: 24 (38.1%)	No PVT: 12.4	No PVT: 1.4	No PVT: 84.2	No PVT: 76.7
		G2: 23 (36.5%)	PVT: 30 (P < 0.001)	G1: 0	PVT: 65.2	PVT: 65.6 (P = 0.04)
		G3: 6 (9.5%)	Significant reduction from era 1 to era 2 for PVT G1‐4	G2: 14.2
		G4: 10 (15.9%)		G3: 16.6
				G4: 0
				G1‐4: 6.6 (P = 0.02)
Gimeno et al. 6	83/864 (10%)	Confirmed intraoperatively, no further classification provided			No PVT: 78.8	No PVT: 67.8	No significant difference in 10‐year patient survival
					PVT: 76.7 (P = 0.48)	PVT: 63.2 (P = 0.48)	No significant difference in overall graft survival
Ravaioli et al. 5	91/889 (10%)	Diagnosed intraoperatively	6.6 versus 5.8 (P = ns)		No PVT: 86	No PVT: 73
		Partial: 51 (56%)			PVT: 85 (P = ns)	PVT: 68 (P = ns)
		Complete: 40 (44%)
John et al. 10	70/290 (24.1%)	Occlusive or partially occlusive thrombosis of main portal vein					No difference in 60‐day or 6‐month post‐LT mortality

Given the critical importance of portal inflow to the liver graft, the major objective in the management of PVT in candidates on the LT wait list is to achieve at least partial recanalization of the portal vein to facilitate physiological anastomosis during LT, which has been associated with improved outcomes. If appropriate portal inflow cannot be achieved, portal hypoperfusion may occur because of either insufficient thrombectomy and/or a steal syndrome from spontaneous portosystemic shunts (SPSs). The ability to achieve adequate portal inflow in patients with extensive thrombosis relies heavily on surgical planning and the technical approach. Figure 1 illustrates surgical options for portal inflow at LT depending on the extent of thrombus (Yerdel grade). Experience from pediatric LT recipients with atretic PV in the setting of biliary atresia and living donor LT has aided in the development of portal vein flow measurement and systematic portosystemic shunt identification and ligation with portal modulation at the time of LT. This portal vein flow evaluation can help determine the need for systematic portosystemic shunt ligation to prevent steal syndrome in the setting of PVT. Individual shunt ligation is optimal, but it can be tedious and difficult. Ligation of the left renal vein (LRV) will reverse the steal syndrome from splenorenal shunts, but it can result in inferior renal function post‐LT.

Fig. 1.

Surgical options for portal inflow at LT depending on extent of thrombus. Yerdel grade 1/2 represents partial or complete main portal thrombosis. This can usually be removed using an eversion thrombectomy technique with restoration of conventional portal flow, using a standard end‐to‐end technique. Grade 3 represents thrombosis into the distal portion of the SMV. Portal flow is restored to the liver using a VJG to the proximal SMV, which bypasses the thrombosis. Grade 4, complete portomesenteric occlusive thrombosis, poses a more difficult problem. Obtaining adequate portal inflow and sufficient drainage of portal varices requires different techniques in different settings. A large left GV can be used as portal inflow if present and accessible. If large SRSs or other abdominal portosystemic shunts are present, then the LRV or the IVC itself can be used as portal inflow.

In the management of PVT prior to LT, if recanalization cannot be achieved, then the objective shifts to preventing further extension of the PVT during wait time, particularly to the superior mesenteric vein (SMV). Screening for PVT in wait‐listed LT candidates is critical to facilitate early detection and initiate management strategies. Doppler ultrasound (US) is the first‐line technique for PVT detection, with sensitivity of 90% for complete PVT and 50% for partial PVT. ² Doppler US can accurately detect thrombosis in the portal vein trunk and intrahepatic branches, and provide information regarding portal venous flow and direction. ¹ Multiphase computed tomography (CT) allows for improved definition of PVT extension and assessment of the SMV, SPSs, renal veins, and inferior vena cava (IVC). Magnetic resonance imaging (MRI) is an alternative diagnostic modality that can be used in candidates with impaired renal function, but PVT characterization is less optimal than that of CT. ¹ In candidates without PVT during initial evaluation, Doppler US should be repeated every 3 to 6 months for continued screening. In LT candidates found to have PVT, the available management strategies include anticoagulation and transjugular intrahepatic portosystemic shunt (TIPS). Anticoagulation results in portal vein recanalization rates ranging between 42% and 100% and a low rate of thrombus progression of 0% to 15%. ⁸ , ⁹ It is important to note that most patients included in studies of anticoagulation had partial PVT, with only a small proportion of complete PVT. TIPS can achieve two endpoints: (1) to recanalize the thrombosed portal vein using endovascular strategies, including balloon angioplasty, stent placement, thrombectomy, and thrombolysis; and (2) to resolve portal hypertension and prevent thrombus recurrence or extension via portosystemic shunting. Portal vein recanalization rates after successful TIPS may reach up to 80%. ⁸

PVT is a common complication in patients with cirrhosis who are being evaluated for LT. In transplant centers with experience performing LT in candidates with PVT, the use of a multidisciplinary approach can lead to excellent short‐ and long‐term outcomes, even with high‐grade PVT. Figure 2 provides an algorithm for management of LT candidates with PVT. Given the ability of most patients with PVT to achieve post‐LT outcomes comparable with those without PVT, we should not withhold lifesaving LT from patients with PVT.

Fig. 2.

Algorithm for management of PVT in LT candidates.