Transjugular intrahepatic portosystemic shunts in liver transplant recipients


  • Alexander R. Bonnel,

    1. Department of Medicine, Baylor College of Medicine, Houston, TX
    Search for more papers by this author
  • Chalermrat Bunchorntavakul,

    1. Division of Gastroenterology and Hepatology, Department of Medicine, Rajavithi Hospital, Bangkok, Thailand
    Search for more papers by this author
  • K. Rajender Reddy

    Corresponding author
    1. Division of Gastroenterology/Hepatology, Department of Medicine, University of Pennsylvania, Philadelphia, PA
    • Address reprint requests to K. Rajender Reddy, M.D., Division of Gastroenterology/Hepatology, Department of Medicine, University of Pennsylvania, 3400 Spruce Street, 2 Dulles, Philadelphia, PA 19104. Telephone: 215-662-4311; Fax: 215-349-5915; E-mail:

    Search for more papers by this author

  • There was no grant support for this project, and the authors have no conflicts of interest to declare.


The insertion of a transjugular intrahepatic portosystemic shunt (TIPS) is a minimally invasive procedure used to relieve the signs and symptoms of portal hypertension in patients with liver disease. The most common indications for placement are refractory ascites and variceal hemorrhage. In properly selected candidates, TIPS placement can serve as a bridge to liver transplantation. Expertise in TIPS placement after transplantation has significantly increased, which has allowed the procedure to become a viable option for retransplant candidates suffering the consequences of recurrent portal hypertension due to portal vein thrombosis, recurrent liver disease, or hepatic venous outflow obstruction (HVOO). However, TIPSs in liver transplant recipients are associated with a lower clinical response rate and a higher rate of complications in comparison with patients with native liver disease, and they are, therefore, generally reserved for patients with a Model for End-Stage Liver Disease (MELD) score ≤ 15 and ≤ 12 in patients with HCV. The role of TIPS placement in nonliver transplant recipients has been well studied in large trials, and it translates well into clinical applicability to candidates for orthotopic liver transplantation (OLT). However, the experience with OLT recipients is heterogeneous and restricted to small series. Thus, we focus here on reviewing the current literature and discussing the proper use of TIPSs in liver transplant recipients. Liver Transpl 20:130-139, 2014. © 2013 AASLD.


cystic fibrosis




end-stage liver disease


hepatitis B virus


hepatocellular carcinoma


hepatitis C virus


hepatic encephalopathy


hepatic venous outflow obstruction


Model for End-Stage Liver Disease


nonalcoholic steatohepatitis


orthotopic liver transplantation


primary biliary cirrhosis


primary sclerosing cholangitis


portosystemic gradient


portal vein thrombosis


refractory ascites


spontaneous bacterial peritonitis


sinusoidal obstruction syndrome


transjugular intrahepatic portosystemic shunt


variceal hemorrhage


veno-occlusive disease

The insertion of a transjugular intrahepatic portosystemic shunt (TIPS) is used in the management of portal hypertension because this nonsurgical procedure significantly reduces pressure in the portal vein by creating a channel between the portal vein and a hepatic vein branch. Decreasing the portal pressure reduces the risk of variceal bleeding and is an acceptable treatment modality for refractory ascites (RA) as well. Additional less frequent indications include portal hypertension events related to portal vein thrombosis (PVT) and hepatic venous outflow obstruction (HVOO). TIPS complications that occur to variable degrees include inadvertent puncture of the liver capsule, shunt dysfunction, liver failure, and hepatic encephalopathy (HE). Survival after TIPS placement is known to be related to the severity of liver disease as indicated by the Model for End-Stage Liver Disease (MELD) score at the time of shunt placement, and it is, therefore, used with caution in patients with high MELD scores.[1]

TIPSs are created in candidates for orthotopic liver transplantation (OLT) to manage complications of portal hypertension and to preserve their candidacy for transplantation. Special considerations for TIPSs in OLT candidates are taken into account in order to minimize technical challenges during OLT. The vast majority of studies of TIPS placement in transplant candidates have found the procedure to have no impact on intraoperative complications, transplant outcomes, or overall patient survival[2-5]; recent but limited data suggest that it may in fact promote early graft function by increasing portal flow to the graft, possibly through the effect of reducing collateral blood flow.[6]

After OLT, a TIPS can be placed to treat portal hypertension–related complications of recurrent end-stage liver disease (ESLD) and venous obstructions. TIPS placement in OLT recipients was once thought to be more technically challenging because of their altered anatomy; however, recent experience suggests otherwise.[7, 8] Because a TIPS in the post-OLT state is often required for patients who have significant liver disease and portal hypertension, the clinical response and survival rates after TIPS placement have been noted to be lower than those for pretransplant patients.[9, 10] Furthermore, immunosuppression increases the risk of infections, renal failure, and HE after TIPS placement in OLT recipients.[10, 11]


After liver transplantation, portal hypertension and related complications may evolve because of recurrent disease, allograft failure/rejection, and vascular or biliary complications.[12] Indications for a TIPS in the transplant recipient mostly match those for a TIPS in pretransplant patients (ie, variceal bleeding and RA[13-16]; Table 1). However, the hepatic venous pressure gradient is known to be lower after liver transplantation, so variceal bleeding may not be as common as it is in the pretransplant setting.[7] The retrospective nature, small sample sizes, lengths of follow-up, and patient selection of available studies are challenges to determining the success of TIPSs in these patients. Several reports suggest that TIPS placement is an effective treatment for addressing early complications of liver transplantation such as PVT, HVOO, and delayed graft function due to ongoing portal hypertension (Fig. 1).[6, 7, 11, 16-19, 23]

Figure 1.

Summary of the considerations for and outcomes of TIPS placement in the early and late posttransplant periods.

Table 1. Summary of Studies of TIPS Placement in OLT Recipients
StudyPatients (n): Indications (n)Etiology of Recurrent Portal Hypertension (n)CP Class (n) or MELD ScoreClinical ResponseComplications/Survival
  1. a

    Two pediatric patients [RA (1) and VH (1)] were included.

  2. b

    For 3 of the 4 adults (75%), RA was resolved.

Lerut et al.[14] (1999)8: RA (5), VH (1), hydrothorax + RA (1), and redo biliary surgery (1)Recurrent HCV (5), recurrent HBV (1), lymphoproliferative disease (1), and VOD (1)CP class B (5) and CP class C (2)8/8 (100%) responded to TIPS.6/8 (75%) had new or worse HE. The 1-year survival rate was ∼50%. 3 died from liver failure.
Amesur et al. (1999)12: RA (6) and VH (6)Recurrent HCV (10), recurrent PBC (1), and unknown (1)CP class A (1), CP class B (3), and CP class C (2)4/6 (67%) with RA recovered. For 4/6 (66%), VH was controlled.2/6 (33%) RA patients died of liver failure within 1 week.
Abouljoud et al. (1999)8: VH + RA (2) and RA (6)Not reportedFor 7/8 (88%), RA and portal hypertension were resolved.For 1/8 (12.5%), TIPS was occluded within 3 months. 3/8 (37.5%) died.
Stewart et al.[15] (2004)7: RA (7)HCV (7)Not reported4/7 (57%) died. TIPS placement was associated with a high risk of death (relative risk = 2.86). There was no difference in the MELD score between patients with a TIPS who died and patients with a TIPS who survived.
Patel et al.[7] (2005)4: RA (4)HCV (2), HVOO (1), and rejection (1)CP class B (4)Not reportedNot reported 1/4 (25%) died.
Vasta et al.[16] (2005)5: RA (5)Delayed graft function (5)5/5 (100%) responded. 1/5 (20%) had complete resolution.0 had HE. 0 had stenosis.
Van Ha et al.[17] (2006)6: RA (5) and VH (1)aChronic rejection (1), biliary cirrhosis (1), acute rejection (1), HVOO (1), and chronic hepatitis of unknown cause (2)RA was resolved for 4/5 (80%).bFor VH, 1/1 (100%) had a response.2/6 (33%) died.
Senzolo et al.[18] (2006)2: SOS (2)SOS (2)Portal hypertension was resolved for 2/2 (100%). 1/2 (50%) had a histological resolution of SOS.None
Kim et al.[11] (2008)11: VH (4) and RA (7)Recurrent HCV (7), HVOO (2), PVT (2), chronic rejection (1), recurrent PSC (1), and unknown (1)MELD score = 16.4 ± 6.24/7 (57%) with RA responded (nonresponders had chronic renal insufficiency). For VH, 2/4 (50%) had a response.4/11 (36%) developed infections. 9/11 (81%) had new or worse HE. 4/8 (50%) had acute renal failure. N=14 of the study and not the N=11 relevant to our review.
Choi et al.[12] (2009)18: VH (2) and RA (16)Recurrent HCV (11), de novo HCV (1), HBV (1), small-for-size syndrome (1), recurrent sarcoidosis (1), and unknown (3)MELD score = 168/16 (50%) with RA had symptomatic relief. For 2/2 (100%), VH was controlled.3/18 (17%) had to undergo a second attempt. 9/12 (75%) died after 3 months.
Finkenstedt et al.[13] (2009)10: RA (7), VH (1), and hydrothorax (2)Ductopenic rejection (5), recurrent HCV (4), and PVT (1)MELD score = 205/7 (71%) with RA responded: 3/7 (43%) had a complete remission. 1/1 (100%) with VH had a response. 0/2 (0%) with hydrothorax had a response.7/10 (70%) had HE. 1 had TIPS dysfunction. 9/10 (90%) died. 5/9 (56%) died of sepsis. The median survival time was 3.3 months.
Saad et al.[9] (2010)19: RA (19)Recurrent HCV or HBV (15)MELD score = 16 ± 4.83/19 (16%) responded to TIPS.5/19 (26%) had new or worsening HE. The 30- and 90-day mortality rates were 16% and 21%. The 1-, 3-, and 6-month graft survival rates were 79%, 58%, and 47%.
Feyssa et al.[10] (2011)26: RA (26)MELD score = 1515/26 (58%) were considered clinically successful.10/26 (38%) had HE. 6/26 (23%) had acute renal failure. 5/26 (19%) needed TIPS revisions. The 1-year survival rate was 13/26 (50%).
King et al.[19] (2011)22: RA (14) and VH (8)Vascular abnormalities (7), HVOO (4), recurrent HCV (3), recurrent PBC (2), recurrent PSC (2), chronic rejection (1), VOD (1), unknown (1), and cholangiopathy (1)MELD score = 13.4 ± 5.115/22 (68%) had a reduction in PSG to <12 mm Hg. 11/14 (79%) with RA had a clinical response. For VH, the rate of prevention was 6/8 (75%). There was a trend toward increased rates of shunt insufficiency in OLT recipients.4/15 (27%) deaths were due to infections (sepsis and SBP). A pre-TIPS MELD score > 15 was significantly associated with decreased survival.
Campos-Varela et al.[20] (2012)1: SOS (1)HBV/alcohol (1)RA was resolved completely, and the patient remained asymptomatic at 20 months.None
Ghinolfi et al.[21] (2012)19: RA (11), hydrothorax (6), and RA + hydrothorax (2)HCV (19)MELD score = 12.4 ± 3.2For 7/19 (37%), RA was resolved. For 4/6 (67%), hydrothorax was resolved.1/19 (5%) died because of sepsis after 35 days. 6/19 (32%) had HE within 1 month. Survival for patients with a pre-TIPS MELD score ≤ 12 did not differ from survival for a control group.
El Atrache et al.[22] (2012)15: RA (12), hydrothorax (2), and VH (1)HCV (9), HCC (3), alcohol (2), and PBC (1)MELD score = 15.8 ± 5.87/12 (58%) had a complete resolution of RA.There was 1/15 (7%) post-TIPS graft failure. 2 /15 (13%) had HE. 3 /15 (20%) had a shunt occlusion requiring revision. 1 /15 (7%) died from sepsis, 1 died from lung cancer, and 1 /15 (7%) died from congestive heart failure. Earlier allograft dysfunction was more likely with partial or no resolution of RA, but there was no significant difference in survival between complete and partial responders to TIPS.
Saad et al.[8] (2013)38: RA (34) and VH (4)HCV (21), alcohol (6), HCV/alcohol (1), HBV (3), cryptogenic (3), NASH (1), CF (1), PSC (1), and sarcoidosis (1)MELD score = 17 ± 5Whole graft OLT recipients did not pose additional technical difficulties. For 11/34 (32%), RA was resolved.7/38 (18%) had HE. The mortality rate at 30 days was 7/38 (18%). The mortality rate at 90 days was 10/38 (26%).

Special Technical Considerations

When it was first introduced for OLT recipients, TIPS placement was thought to be more challenging because of the altered anatomy of the hepatic vessels after OLT. Unlike the standard OLT approach of end-to-end anastomoses at the suprahepatic and infrahepatic vena cava, the piggyback and cavocaval techniques are not uncommon and require a different approach during TIPS placement.[8, 13] A Cochrane Database review found the piggyback technique, which involves the preservation of the recipient's retrohepatic vena cava, to be a safe and effective technique for transplantation.[24] In this case, the left internal jugular vein, rather than the right, is used for TIPS placement.[7] After cavocaval OLT with a side-to-side anastomosis of the donor's retrohepatic vena cava, a gun-sight approach is used to create a direct cava-to-portal vein TIPS.[7, 25] Past studies reported multiple attempts at creating a TIPS in these patients because of technical inability or failed cannulation of the portal or hepatic veins.[11, 12, 19] However, the most recent data from interventional radiologists indicate that transplanted livers do not pose any additional technical difficulty in comparison with native livers.[8] Although these data remain limited to whole graft recipients, they suggest that technical expertise has significantly improved for TIPS placement in OLT recipients. This supports previous data showing that the reported technical complication rate for this procedure has dropped below the accepted 5% threshold, which is a quality measure determined by the Society of Interventional Radiology.[17, 26] Successful TIPS placement in increasingly challenging cases, such as patients with small-for-size syndrome after living donor transplantation, has also been described and indicates continued progress in the field of TIPS creation in transplant recipients.[12, 27]

Complications Following Transjugular Intrahepatic Portosystemic Shunt Placement

A lower rate of shunt stenosis has been reported after posttransplant TIPS placement versus pretransplant TIPS placement. Approximately 10% to 20% of the stents in OLT recipients require revisions because of the development of strictures or occlusions,[9, 10, 12-14, 22, 28] whereas up to 70% do in nontransplant patients.[1] However, the largest case-control study thus far reported no such complications in its posttransplant group.[19] It must be noted that most of the data come from the era of uncovered-stent use and may not be applicable to the current days of polytetrafluoroethylene-covered stents. Immunosuppressive therapies may play a role in reducing rates of shunt stenosis. Experimental studies suggest that tacrolimus and mammalian target of rapamycin inhibitors (everolimus and sirolimus) have inhibitory effects on both smooth muscle cell proliferation and endothelial progenitor cells.[29-31] Furthermore, this is supported by the experience with drug-eluting stents in vascular interventions, for which immunosuppressive therapies have been known to significantly reduce stenosis.[32, 33] Although there are reasons to believe that there are factors in liver transplant recipients that are favorable for shunt patency, it needs to be recognized that sirolimus use has been associated with hepatic artery thrombosis, which potentially presents a risk for precipitating liver failure after TIPS placement.[34] Thus, research is needed on shunt patency, specifically after TIPS in the posttransplant setting, and particularly on the mechanistic role of various immunosuppressants.

Transplant recipients face a high risk of infection, renal failure, and neurological complications after TIPS placement. Sepsis is the most common complication and cause of death and occurs in approximately 20% to 50% of reported cases.[10, 11, 13, 19] The risk of infection in OLT recipients is attributable to their immunosuppressive therapies. However, recent data did not show the rate of infectious complications among OLT recipients to be significantly higher than the rate among pretransplant patients,[19] in contrast to the findings of a previous case-control study.[11] Patients with ESLD are often in a state of cirrhosis-associated immune dysfunction.[35] Therefore, the conflicting results of recent studies may be explained by differences in the extent of cirrhosis-associated immune dysfunction within each group of TIPS patients. Before periprocedural prophylaxis is routinely recommended, as some authors have suggested,[11, 13] a further evaluation of the risk of infection after TIPS placement among OLT recipients is needed.

Acute renal failure directly following TIPS placement has been encountered in approximately 20% of OLT recipients.[10, 11] Renal dysfunction in OLT recipients is attributable to a number of factors such as long-term calcineurin inhibitor use, and pretransplant renal dysfunction is due to 1 or more factors, including hepatorenal syndrome, diabetes, and hypertension.[36] As a result, radiographic contrast during TIPS placement may precipitate the development of renal failure after the procedure.

Calcineurin inhibitors are known to cause neurological complications such as encephalopathy, seizures, psychological disorders, stroke, tremor, and posterior reversible encephalopathy syndrome.[37] In OLT recipients, new or worsening HE after TIPS has been documented in 25% to 80% of cases.[9-11, 13, 14, 19] Previous findings suggest that OLT recipients may be prone to encephalopathy because of the added neurotoxicity of immunosuppressive drugs.[11, 14] However, recent data did not show a higher rate of HE after TIPS placement in OLT recipients versus pre-OLT TIPS patients.[19] Therefore, the association between immunosuppression and the extent of HE in OLT recipients after TIPS placement remains controversial.

Clinical Response to a Transjugular Intrahepatic Portosystemic Shunt

The therapeutic response to TIPSs in OLT recipients is highly variable and is influenced by various factors, such as pretransplant conditions, posttransplant complications, and the severity of the recurrent liver disease.35-37 Despite some conflicting data, the overall clinical response rate to TIPSs in liver transplant recipients appears to be lower than that in pretransplant patients.[9, 15, 17, 21, 38] This difference appears to be more significant for the control of ascites versus the prevention of portal hypertension–related bleeding. Larger studies of liver transplant recipients have reported RA remission rates in the range of 15% to 79%,[8-10, 19, 22] whereas a success rate > 90% has been reported for pretransplant patients.[39] Although data on the prevention of portal hypertension–related bleeding in liver transplant recipients are very limited (<30 patients), current rates of prevention among these patients (50%-75%)[19] are similar to those among pretransplant patients.[39]

The poor control of RA with TIPSs in transplant recipients suggests an entirely different pathophysiology behind RA in a subset of patients after transplantation. The rate of RA after transplantation is in the range of 5% to 10%, and it has been attributed to causes such as RA before transplantation, cold ischemia, hepatitis C virus (HCV) recurrence, posttransplant renal dysfunction, outflow obstructions secondary to the surgical technique, bacterial peritonitis, and graft rejection.[15, 38, 40] The cold ischemia time reflects the degree of ischemia/reperfusion injury, and cold ischemia likely damages the sinusoidal endothelium and increases portal pressures.[41]

Chronic kidney disease, a known predictor of a poor response to TIPS placement, is more common in OLT recipients and is primarily determined by pretransplant renal dysfunction.[42] Interestingly, the duration of the creatinine elevation before transplantation—not the presence of hepatorenal syndrome—is an independent predictor of renal dysfunction after transplantation.[43] In the posttransplant setting, chronic kidney disease has been attributed to calcineurin inhibitor toxicity; however, the degree of the contribution may be overestimated in light of the multiple comorbidities in OLT recipients.[44] Nevertheless, the cumulative incidence of chronic renal failure among transplant patients increases from 8% at 1 year to approximately 14% at 3 years and 18% at 5 years after OLT.[45] It is, therefore, likely that multiple comorbidities in OLT recipients (particularly renal dysfunction) largely explain their poor response to TIPSs.

In a retrospective study of OLT recipients,[38] the etiology of RA could not be identified in close to one-third of the cases. The etiology was more often unknown in the early posttransplant period (<3 months) and for patients who did not receive a venovenous bypass. Ascites more often resolved in those with early recurrence and may have been a reflection of the continuum of the process of portal hypertension and unaccounted perioperative factors.[15] RA that does not respond to interventions such as TIPS placement has a dismal outcome, and it has been the focus of much investigation, particularly for patients with HCV recurrence.[15, 46] Often, the more aggressive disease seen in OLT recipients with recurrent HCV disease leads to bridging fibrosis and, subsequently, portal hypertension. However, there is evidence showing that this subset of patients is particularly susceptible to developing portal hypertension unrelated to architectural distortion, ie, a form of noncirrhotic portal hypertension.[46] Nearly two-thirds of transplant recipients with HCV recurrence had high hepatic venous pressure gradients in the absence of significant fibrosis, and they tended to present later (≥3 months) in the posttransplant period. As a result, a hepatic venous pressure gradient > 6 mm Hg has proven to be a more accurate prognostic feature than the fibrosis stage in determining the likelihood of clinical decompensation secondary to HCV recurrence.[47] Thus, unique pathophysiological mechanisms for portal hypertension in patients with HCV recurrence may be a potential reason for the development of RA in this population even in the absence of overt fibrosis and cirrhosis.

Survival and Patient Selection

Overall, when matched by liver disease and age, OLT recipients were found to have a markedly lower rate of short-term survival after TIPS in comparison with pretransplant patients.[11] The 1-year survival rate after TIPS placement in OLT recipients was approximately 14%, whereas it was 58% in pretransplant patients, and the rate was particularly low for nonresponders.[11, 12, 48] Recent data did not show a significant difference in cumulative mortality but did show a trend toward lower survival in a group of OLT recipients.[19] Complications of ESLD account for the majority of deaths after TIPS placement in both pretransplant and posttransplant patients.[10, 19, 28] Recent data suggest that a TIPS may accelerate decompensation in OLT recipients, although this claim needs further validation. Patients with RA after transplantation have been noted to have a mortality rate 8.6 times higher than the rate for patients without RA, and this may partly be a reflection of the overall status of the recipients and not due to RA alone.[38] There are promising data showing that OLT recipients who have ascites that is later resolved face no greater risk than those who never develop ascites.[38] This observation cannot be translated to patients with RA because such a clinical course reflects a greater degree and longer duration of portal hypertension in comparison with those who may have ascites for a shorter term that is then resolved.

The major challenge at this point is identifying which OLT recipients are most likely to respond to and benefit from TIPSs. There are significant risks associated with TIPSs in OLT recipients with a MELD score > 15 in comparison with pretransplant patients with such a MELD score.[17, 21] In OLT recipients with RA, a 6-month transplant-free survival rate of only 12.5% has been reported.[9, 10] A significant reason for such poor outcomes is likely related to HCV recurrence. It has recently been suggested that TIPS placement in HCV patients with RA should be reserved for those with a MELD score ≤ 12.21 This cutoff likely reflects the extent of HCV disease because the most common cause of death in the study group was progression of HCV. A previous study found TIPS placement in HCV patients to be associated with a high risk of death but did not note a difference in the MELD scores of those who died and those who survived with a TIPS in place.[15] Understanding the various mechanisms behind portal hypertension in the setting of recurrent HCV remains a barrier to more refined TIPS candidate selection and warrants caution in these patients. In all OLT recipients with a MELD score > 15 and specifically in OLT recipients with HCV and a MELD score > 12, TIPS placement has been associated with greater failure rates and is, therefore, unlikely to benefit the patient.


Portal Vein Thrombosis

PVT is a rare but severe complication that typically occurs early after OLT and is often related to abnormal venous reconstruction during surgery in patients with preexisting PVT.[23] Interventions for PVT after OLT include surgical thrombectomy, percutaneous thrombolysis, angioplasty, and stent placement, and they are associated with good long-term outcomes.[49-52] Several case reports have shown a combination of a TIPS and local thrombolysis with or without thrombectomy to be an effective method of transhepatic recanalization of the portal vein.[14, 53, 54] The use of a TIPS in PVT is associated with a reduced risk of bleeding and technical complications in comparison with transhepatic puncture for portal vein interventional procedures.[53]

Hepatic Venous Outflow Obstruction

HVOO after OLT is uncommon and occurs in less than 2% of cases, but it is life-threatening if it is left untreated.[55, 56] The most common causes of hepatic vein thrombosis are anastomotic complications following transplantation (30%) and disease recurrence (25%).[55] TIPS is an effective treatment for hepatic vein thrombosis refractory to medical therapy in the pretransplant setting; however, data on its use in transplant recipients remain limited.[7, 11] HVOO has been noted to be an independent predictor of mortality after TIPS placement in transplant recipients, and it has been suggested that the pathophysiology of portal hypertension in patients with HVOO is different from the pathophysiology of ESLD-related portal hypertension.[19] Yet, a TIPS appears to reduce the complications of portal hypertension and may serve as a bridge to retransplantation in OLT recipients with HVOO.[57] Because of the small sample size behind the current data on TIPSs in patients with HVOO, concrete recommendations cannot be made.

Sinusoidal obstruction syndrome (SOS), formerly known as veno-occlusive disease, includes both central vein and sinusoidal occlusion and has been attributed to acute allograft dysfunction.[58, 59] Current guidelines do not recommend TIPS placement for SOS1; however, some have suggested that TIPS placement be tried in carefully selected patients with SOS after OLT, particularly if retransplantation is not an option.[57, 60] A recent case report[20] presents TIPS placement as a potentially useful treatment for severe SOS refractory to medical treatment.


TIPS placement is used in the management of portal hypertension–related complications and decreases the risks associated with variceal bleeding, RA, PVT, and HVOO. However, this procedure is not without complications such as transcapsular puncture, shunt thrombosis or stenosis, liver failure, and HE. After OLT, recurrent liver disease and conditions such as venous obstruction may lead to portal hypertension–related complications. Because of the post-OLT altered anatomy, TIPS placement may be challenging and can result in inadvertent puncture of the liver capsule and bleeding complications. However, advances in interventional radiology expertise have made the procedure safe and effective. Immunosuppression may increase the risk of infection, renal failure, and HE after the procedure, although further data in this area are needed. TIPS placement in OLT recipients, in contrast to pretransplant patients, has a lower rate of clinical response, especially in those who undergo the procedure for RA. This finding may be attributable to the increased prevalence of comorbidities among OLT recipients but is likely related to the differences in the pathophysiology underlying RA in this patient population. Recent case-control data suggest that a TIPS may accelerate the progression of advanced liver disease in OLT recipients versus pretransplant patients. Therefore, TIPS placement in general is not recommended for patients with a MELD score > 15 and specifically for HCV patients with a MELD score > 12. In patients with PVT after transplantation, TIPS placement is an important nonsurgical intervention that is associated with good outcomes. TIPS placement in OLT recipients with HVOO should be reserved as a bridge to retransplantation.