Interventional treatment for portal venous occlusion after liver transplantation: long-term follow-up results.

Portal vein (PV) occlusion after liver transplant is an uncommon clinical situation, and percutaneous interventional treatment for this condition has not been widely described. The aim of this study was to evaluate the long-term treatment effect of interventional treatment for PV occlusion after liver transplantation (LT). Follow-up data of 13 patients who received interventional treatment for PV occlusion after LT between July 2007 and April 2013 were analyzed. Of these, 10 patients had portal hypertension-related signs and symptoms. Percutaneous balloon angioplasty and stent placement were performed, with percutaneous thrombolysis treatment as appropriate. Embolization therapy was required for significant collateral circulation. Technical and clinical success, complications, and patency of PV were analyzed. Both technical and clinical success was achieved in 11 of the 13 patients (84.6%). Direct portogram showed limited PV occlusion in 7 patients and extensive PV occlusion in 4 patients. The former underwent balloon angioplasty followed by stent placement, while the latter underwent balloon angioplasty followed by stent placement and additional percutaneous thrombolysis treatment. Embolization therapy for collateral circulation was performed in all 4 patients with extensive PV occlusion and 1 patient with limited PV occlusion. All stents remained patency during the follow-up (28.5 ± 6.8 months). No portal hypertension-related symptoms reoccurred during follow-up. In conclusion, interventional treatment for PV occlusion after LT showed a high success rate and good long-term results. Comprehensive interventional treatment should be used for extensive PV occlusion.

INTRODUCTION

Liver transplantation (LT) is an effective treatment for end-stage liver disease.[1] Portal vein (PV) complications after LT that include PV stenosis and PV thrombosis are rare,[2-4] but can be devastating and lead to graft failure. If not promptly treated, PV stenosis and PV thrombosis may lead to PV occlusion. PV occlusion is mainly manifested with portal hypertension-related symptoms. In less serious cases, patients may not present any clinical symptoms, whereas in more severe cases, patients may present life-threatening symptoms such as upper gastrointestinal hemorrhage. Early diagnosis and treatment of PV occlusion may ensure optimal graft function and good recipient survival. In the past, surgical treatments such as thrombectomy, surgical revision, and retransplantation were the main approaches to manage the PV complications after LT.[5,6] However, surgical treatments were limited because of technical difficulties and invasiveness. Because of minimally invasive and more effective, currently, studies have focused mainly on percutaneous interventional treatment for PV complications after LT.[7,8] However, to our knowledge, percutaneous interventional treatment for PV occlusion after LT, as well as long-term follow-up, has not been widely described. Our purpose was to review the long-term results of comprehensive percutaneous interventional treatment in 13 patients with PV occlusion after LT.

MATERIALS AND METHODS

This study was approved by the Institutional Review Board, Chaoyang Hospital, Beijing, China.

General Information

Between July 2007 and April 2013, 13 patients (9 male and 4 female patients) underwent percutaneous interventional treatment for PV occlusion. The patients ranged in age from 25 to 65 years (median age 46.8 years). Twelve patients had received orthotopic LT and 1 patient had received a living donor LT. The average interval between LT and interventional procedures was 127 days ± 120 (range, 14,901 days). The interval between diagnosis of PV occlusion and interventional procedure was within 1 month in 3 patients, within 3 months in 6 patients, and >3 months in 4 patients. The underlying diseases included posthepatitic cirrhosis (n = 6), congenital postpolycystic liver cirrhosis (n = 1), and primary hepatic carcinoma (n = 6).

Ten patients mainly presented with portal hypertension-related symptoms including abdominal distension and ascites in 6 patients, esophageal variceal bleeding in 2 patients, and splenomegaly in 2 patients. Three patients were asymptomatic but exhibited elevated liver function test results. The demographic information is shown in the Table 1.

TABLE 1

Demographic Characteristics of Patients

Diagnosis of PV occlusion was based on Doppler ultrasonography (US), magnetic resonance angiography (MRA), or computed tomography angiography (CTA). Occlusion was diagnosed if Doppler US showed no blood flow in the PV or MRA and CTA demonstrated discontinuity of the PV.

Percutaneous Intravascular Interventional Treatment for PV Occlusion

Informed consent was obtained from each patient prior to all procedures. All procedures were performed under local anesthesia. The liver was punctured with a 22-gange Chiba needle (MReye, Cook, Bllmington, United States) under fluoroscopic or ultrasound guidance. When a PV branch was entered, the needle was exchanged for a 4-F coaxial dilator and a 7-F sheath (Pinnacle, Terumo, Tokyo, Japan) over a 0.018-inch guide wire (Cook) or a 0.035-inch angled hydrophilic guide wire (Radiofocus, Terumo, Tokyo, Japan). Then a 5-F KMP angiographic catheter (VanSchie Beacon, Cook, Bllmington, United States) was introduced into the intrahepatic PV along the guide wire. A direct portogram was obtained, and then the 0.035-inch guide wire and 5-F catheter were used to traverse occlusion of the PV. After entering the superior mesenteric vein or the splenic vein, venograms were obtained to determine the length of the occlusion segment and the surrounding collateral circulation.

The occlusion segment was dilated with a balloon (Advance, Cook, Bjaeverskov, Denmark) followed by stent (Smart Control, Cordis, Miami Lakes, United States) placement. Transcatheter embolization with coils (Tornado, Cook Medical, Bjaeverskov, Denmark) was required for serious collateral circulation or low hepatopetal blood flow. The percutaneous thrombolysis treatment was determined based on the venous blood flow and thrombosis situation after stent placement (Figures 1 and 2A–D). The catheter with multiple side holes was retained in PV stent with a loading dosage of urokinase (3000 U/kg) followed by a continuous maintenance dosage (600 U/kg/h). Portograms were reviewed every 24 hours, and thrombolytic therapy was sustained until no obvious filling defects were observed in PV (Figure 2E–G). The transhepatic tracts were embolized with gelfoam routinely.

FIGURE 1

CTA examination showed occluded main PV (arrowhead) in a patient after LT. CTA = computed tomography angiography, PV = portal vein, LT = liver transplantation.

FIGURE 2

(A) Direct portogram showed normal branches of PV (arrowhead); the main PV was not displayed. (B) A catheter was introduced to traverse the occlusion segment, and entered the superior mesenteric vein (arrowhead); venogram showed rich collateral circulations (arrow), but the main PV was not displayed. (C) A balloon catheter (arrowhead) was used to dilate occluded main PV. (D) Portogram following balloon angioplasty showed multiple filling defect (arrowhead) in the main and right branch of PV, as well as rich collateral circulations. (E) Portogram following stent placement and transcatheter embolization showed multiple filling defect in the stent and little hepatopetal blood flow (arrowhead). (F) Portogram following 24 hours of percutaneous thrombolysis showed normal hepatopetal blood flow (arrowhead) in stent without filling defect; the proximal stent located in the right branch of PV (arrow); the right branch was clearly displayed, while the left branch was not displayed. (G) Intrahepatic portal venogram showed patency of the left branches (arrowhead) and right branches (arrow). PV = portal vein.

On the first day after procedure, patients underwent systemic anticoagulation with low-molecular-weight heparin calcium (100 U/kg) twice daily for 3 days, and oral administration of warfarin for at least 6 months, to maintain the international normalized ratio (INR) of 2 to 2.5.

Analysis and Follow-Up

The technical and clinical success rate, incidence of complications, clinical improvement, and stent patency were analyzed. Technical success was defined as successful stent placement and restored hepatopetal blood flow. Clinical success was defined as subsequent normalization of liver function and disappearance of portal hypertension-related clinical signs and symptoms. In successful procedures, routine clinical evaluation and Doppler US surveillance was performed on postprocedural day 1, and 1, 3, and 6 months, and every 6 months thereafter. A CTA was required in case of abnormal US results and a direct portogram was carried out if necessary.

RESULTS

The outcomes of percutaneous interventional treatment are shown in the Table 2. Initial technical success was achieved in 11 of 13 patients (84.6%). For the 11 patients in whom initial technical success was achieved, the mean length of follow-up was 28.5 ± 6.8 months, with a maximum length of follow-up of 38 months. Clinical success was achieved in all 11 patients who underwent successful procedures (100%). Portal hypertension-related symptoms, such as abdominal distension, ascites, and esophageal variceal bleeding, all disappeared without recurrence during the follow-up. Suspected PV restenosis was observed by US in 1 patient at 23 months postprocedure; however, the following direct portogram revealed patency of the PV.

TABLE 2

Outcomes of Percutaneous Interventional Treatment in 13 Patients

Balloon angioplasty followed by stent placement was performed in all 11 patients. Portograms showed extensive PV occlusion (>50% of the whole length of the main PV) in 4 patients and limited occlusion (<50% of the whole length of the main PV) in 7 patients. In 4 patients with extensive PV occlusion, balloon angioplasty followed by stent placement was not able to restore the hepatopetal blood flow, and additional indwelling catheter local thrombolysis and collateral circulation embolization were performed. Transcatheter thrombolysis sustained for 24 hours in 2 patients, 48 hours in 1 patient, and 72 hours in 1 patient. In 7 patients with limited PV occlusion, balloon angioplasty followed by stent placement was able to restore hepatopetal blood flow in 6 patients, and additional collateral circulation embolization was performed in 1 patient. A total of 16 stents were deployed. For the 4 patients with extensive PV occlusion, 2 stents per patient were placed in 3 patients, and 3 stents were placed in 1 patient. For the 7 patients with limited PV occlusion, 1 stent per patient was placed. The interventional treatment failed in 2 patients because of being unable to traverse the occluded PV. Intrahepatic portal venogram showed occlusion of the left and right PV branches in the 2 patients. The guide wire could not pass the occlusion and entered the main PV. The 2 patients were given anticoagulant therapy. Both of the patients were lost to follow-up after being discharged.

Procedure-related complications occurred in 2 patients. One patient complained of respiratory chest pain after procedure. Routine blood test indicated no significant decrease in hemoglobin. Chest x-ray showed a small amount of right pleural effusion. Diagnostic thoracocentesis showed uncoagulated blood, which was gradually absorbed after symptomatic treatment. The other patient, a young female, exhibited progressively increased postprocedure heart rate. Urgent blood tests showed significant decrease in hemoglobin (from 135 to 102 g/L). Computed tomography revealed a large amount of right pleural effusion that was considered to be due to active chest hemorrhage. Emergency surgery was performed, and numerous blood vessels on the surface of the diaphragm and pulsating bleeding from a small artery were observed; the bleeding was stopped by local cauterization.

DISCUSSION

Since PV angioplasty and stent placement was first reported by Olcott et al[9] in 1990, it has subsequently been established as a widely accepted, safe, and effective procedure for treating posttransplantation PV stenosis.[7,10-12] Funaki et al[7] reported that PV angioplasty for the treatment of PV stenosis showed a 50% recurrence rate at an average time of 6.3 months, while stent deployment showed a good patency of 100% over a follow-up of 47 months. To our knowledge, few study focused on the percutaneous interventional treatment for the PV occlusion after LT. Successful use of percutaneous thrombolysis and stent placement for treatment of posttransplantation PV occlusion with long-term follow-up is rare.

Cherukuri et al[13] reported 2 patients with posttransplantation PV occlusion due to thrombosis. The patients were treated successfully with percutaneous thrombolysis followed by stent placement, and the PV patency was maintained for 2.5 and 4 years without secondary PV thrombosis. In our study, multiple interventional techniques, including balloon angioplasty, stent placement, percutaneous thrombolysis, and transcatheter embolization, were used to treat extensive PV occlusions in order to restore hepatopetal blood flow in 4 patients. The percutaneous thrombolysis was followed by stent placement in our study, which was different from the study by Cherukuri et al. This was because of the occlusion time of PV that was much longer in the 4 patients in our study, which might be unresponsive to initial treatment with thrombolysis. In our experience, in-stent thrombosis would easily develop after stent placement in patients with extensive PV occlusion, which would necessitate percutaneous thrombolysis treatment. Besides, significant collateral venous was more common in the patients with extensive PV occlusion than patients with limited PV occlusion, which would reduce the hepatopetal blood flow after placement of the stents. In that case, transcatheter embolization could further increase the hepatopetal blood flow. Our results suggest that comprehensive interventional treatment is required for patients with extensive PV occlusion. On the contrary, balloon angioplasty followed by stent placement can restore the hepatopetal blood flow in most patients with limited PV occlusion. However, transcatheter embolization should be performed for patients with rich collateral circulation.

The initial technique success of our study was 84.6%, which was much higher than 50% reported by Cheng et al.[14] A total of 10 patients with PV occlusion posttransplantation were included in the study by Chen et al. The technical success was achieved in 5 of 10 patients (50%). However, a success rate of 83.3% (5/6) was achieved in patients who were diagnosed with PV occlusion within 1 year. They failed to traverse the occluded PV in 4 patients who were diagnosed with PV occlusion >1 year. In our study, technical success was achieved in all patients with limited PV occlusion, and in 4 of 6 patients with extensive PV occlusion (66.7%). PV occlusion time >1 year occurred in just 1 patient; however, technical success was achieved in this patient. In our experience, extensive PV occlusion seemed to be a factor causing technical failure.

Percutaneous transhepatic approach is the most commonly used route for PV stent placement. However, the transsplenic approach is less injurious to the transplanted liver graft and is an alternative method for patients who failed to enter PV by transhepatic approach; it is especially recommended in pediatric patients because of the upper location of the graft in the left subphrenic region.[14-17] Transjugular intrahepatic portosystemic shunt (TIPS) approach also has been reported for the treatment of PV occlusion, and is especially recommended in patients with significant coagulation disorders and ascites.[18,19] In our experience, PV can be entered by transhepatic approach in most cases. Familiarity with the technique of vascular puncture under ultrasound guidance can be a good help for the procedure. Gentle and delicate manipulation of the puncture needle and guide wire is important to minimize injury to the transplanted liver graft.

The primary patency of the stents in this series was 100% (11 of 11) at the mean 28.5-month follow-up examination. The patency rates were extremely high in our study; however, this degree of patency with PV stents is not unique. According to most reports and our previous study,[4,7,11] long-term patency of stent was maintained in patients with PV stenosis after stent placement and appropriate anticoagulation therapy. The findings of this study indicate the similarity between PV occlusion and PV stenosis in which anticoagulation can maintain long-term patency after stent placement. However, there is no unified standard for maintaining anticoagulation states after PV stent placement at present.[11,14,20,21] In our experience, systemic anticoagulation with low-molecular-weight heparin calcium for 3 days and oral administration of warfarin for at least 6 months can keep long-term patency of the stent. However, the dosage should be adjusted promptly to maintain the INR of 2 to 2.5.

The most common postprocedure complications of percutaneous transhepatic interventional treatment included bleeding and biliary injury. Postprocedure pleural cavity bleeding occurred in 2 patients in our study, possibly because of vessels injury during the puncture. Rich local collateral circulation may be easily induced because of the adhesion of diaphragm and liver membrane after LT. Appropriate puncture site or using TIPS approach may reduce the risk of bleeding.

Our study has some limitations. First, the pressure gradients were not obtained during procedures. Second, this was a retrospective design. Third, the number of patients was not enough.

In conclusion, percutaneous interventional procedure is a safe and effective treatment with long-term patency for PV occlusion after LT. Comprehensive interventional treatment, including balloon angioplasty, stent placement, percutaneous thrombolysis, and transcatheter embolization, should be used for extensive PV occlusion.