Hepatic artery stenosis is one of the most common vascular complications after liver transplantation. Most commonly the stenosis occurs in the donor arteries. Hepatic artery stenosis may cause graft ischemia, with deterioration of liver function and formation of biliary strictures. Surgical reconstruction has traditionally been the first choice for treatment, but improving interventional radiologic technique makes it possible to repair the stenosis without surgery. However, reports of stenting of hepatic artery stenosis are few and do not evaluate the clinical outcome. The aim of this paper is to assess current outcome and complications of hepatic artery stenting.
Hepatic artery stenosis after liver transplantation may affect liver function and result in hepatic artery thrombosis. Surgical reconstruction has been the first choice for treatment. Interventional radiologic technique can be used, but there is no report on long-term outcome. The aim of this paper is to assess current outcome and complications of hepatic artery stenting. Twenty-six adult patients were stented for hepatic artery stenosis between 1998 and 2003. Nine patients had previous surgical reconstruction for hepatic artery stenosis. Seventeen patients suffered newly developed hepatic artery stenosis. Three patients were retransplanted. After stenting, the patients were followed by Doppler ultrasound at day 1, 1 month, and 6 months. Angiography was scheduled in 6 months. Four patients died within 2 months. The other 22 patients were followed for mean 31 ± 14 months (8-71 months). One of 22 patients died from renal failure 2 years later. Twelve patients' hepatic arteries looked normal after stenting. Restenosis was seen in 8 patients (36%). Other complications were artery thrombosis (n = 1) and long segment stricture (n = 1). In 2 patients (25%) restenosis resulted in thrombosis. Six of the 8 patients who developed recurrent stenosis were successfully treated interventionally: restent (n = 5) and balloon dilation (n = 3). However, 3 patients (38%) restenosed. Kaplan-Meier complication-free survival was 54% at 1 year after stenting. In conclusion, hepatic artery stenting is a viable treatment for hepatic artery stenosis with reasonable results. Stenting is useful as adjuvant treatment after surgical revision. Liver Transpl 12:422–427, 2006. © 2006 AASLD.
PATIENTS AND METHODS
Sixty-five patients were diagnosed with hepatic artery stenosis that needed treatment after liver transplantation between 1998 and 2003 at our institution. Twenty-six patients underwent hepatic artery stenting for hepatic artery stenosis. All patients were successfully stented. Fifteen patients were male and 11 were female. The mean age was 49 years (range, 21-69).
Twenty-three patients underwent primary liver transplantation and another three underwent retransplantation. Patient demographic details are shown in Table 1.
|Patient||Age||Sex||Diagnosis||Retransplant||Prestenting Arterial Procedure|
|1||49||M||Hepatitis C||PSC||HA reconstruction|
|2||57||M||Hepatic artery thrombosis||α antitrypsin deficiency||None|
|4||47||F||Hepatitis C||HA reconstruction with saphenous graft patch|
|11||21||F||Alagille syndrome||HA reconstruction with aortic conduit|
|13||47||M||Hepatitis C||HA reconstruction with saphenous graft interposition|
|14||49||F||Hepatitis C||HA reconstruction with aortic conduit|
|15||47||F||Polycystic liver disease||HA reconstruction with saphenous graft patch|
|16||48||M||Hepatitis C/Laennec's||HA reconstruction with aortic conduit|
|17||62||M||Laennec's||HA reconstruction with aortic conduit|
|18||48||M||Hepatitis C/Laennec's||HA reconstruction with aortic conduit|
|26||37||F||R lobe infarction||CBD stricture||None|
Nine patients had a history of surgical hepatic artery reconstruction for hepatic artery stenosis. The mean duration between the arterial revision surgery and subsequent stenting was 73 days (range, 13-221 days). Seventeen patients presented newly developed hepatic artery stenosis after liver transplantation.
For this study all angiograms were reviewed by one of the authors (G.J.). All other patient data were retrieved from our prospectively maintained research database and analyzed through chart review.
All patients had orthotopic liver transplantation between 1998 and 2003 in our institution. Donor hepatectomy and recipient operation were performed using standard techniques. Preservation of all donor livers was accomplished using the University of Wisconsin solution. The arterial anastomosis was done with running 6-0 or 7-0 polypropylene sutures. The level of the arterial anastomosis depended on the size and quality of the recipient hepatic artery. The anastomoses were in the donor: aortic patch (n = 1), celiac artery (n = 16), common hepatic artery (n = 8), and superior mesenteric artery (n = 1); recipient site: proper hepatic artery (n = 17), common hepatic artery (n = 5), and infra renal aorta using a donor iliac arterial conduit (n = 4). Ten patients (38%) had a history of previous hepatic artery stenosis after transplantation. Nine patients had undergone hepatic artery reconstruction surgically using a donor iliac artery conduit to the aorta (n = 4), Gore-Tex graft as an aortic conduit (n = 1), saphenous vein patch (n = 2), saphenous vein interposition (n = 1), or primary reanastomosis (n = 1).
After liver transplantation, liver enzymes, including aminotransferase, were obtained as a part of routine lab work. All patients underwent Doppler ultrasonography (DUS) day 1, 1 year, 2 years, 5 years, and every 5 years thereafter. Additional DUS is scheduled when a patient showed liver enzymes elevated from baseline, which suggests arterial insufficiency. In case of DUS abnormalities of hepatic artery flow such as high velocity flow or low resistance, hepatic artery angiography was performed to verify the lesion and to establish a treatment plan.
Immunosuppressive medications were given according to our standard protocol. The immunosuppression regimen was recorded at the time of stenting. Tacrolimus (Prograf) or cyclosporine (Neoral) based regimens were used with mycophenolate mofetil (CellCept) or sirolimus (Rapamune) as a third agent. Prednisolone was given in all cases on a tapering protocol. Mycophenolate mofetil was given 1,000 mg twice a day. Cyclosporine level was kept at 200-300 ng/mL for the first 3 months then maintained at 100-200 ng/mL. Tacrolimus level was initially kept at 10-15 ng/mL for the first 2 weeks then maintained at 5-10 ng/mL.
Hepatic Artery Stenting
Hepatic artery stenting was performed by the Department of Interventional and Vascular Radiology at Baylor University Medical Center in Dallas, TX. Techniques of hepatic artery stenting were as follows. Single wall puncture technique was used to gain access to the common femoral artery. A guide wire was advanced and a 5 French Cobra catheter (Cook, Inc., Bloomington, IN) was used for selective hepatic artery catheterization via sheath. Selective hepatic arteriogram was then performed. The catheter was exchanged for a 5 French sheath, which was placed in the proximal hepatic artery. The 5 French sheath was used for all cases. The sheath was placed in proximal conduit selectively for those patients who had a conduit. A transcend wire was used to select a hepatic artery branch distal to the hepatic artery stenosis. A coronary stent was then positioned and deployed. S70 coronary stents (Medtronic AVE, Santa Rosa, CA) were used for hepatic artery stenosis. Size was selected from 3.5 mm to 4.5 mm depending on peristenosis luminal diameter. No stent was oversized. A transcend wire was used for all cases.
Patients received 5,000 units of heparin during the procedure. Following the procedure, anticoagulants (Heparin or Lovenox) and/or antiplatelet (acetylsalicylic acid or clopidogrel bisulfate (Plavix, Bristol-Myers Squibb, New York, NY) were given1 for at least 3 months.
Follow-up after stenting
Serum liver enzymes were obtained routinely. Patients were followed by DUS on day 1, at 1 month, 6 months, and 12 months after stenting, and a routine hepatic angiography was obtained 6 months after stenting. Resistive indices of the hepatic artery were used to screen for restenosis at DUS.2 If the resistive index was between 0.3 and 0.5 or there was a trend of decreasing resistive index, restenosis was suspected. When a patient showed liver enzyme elevation, DUS was also performed. All suspicious DUS exams prompted hepatic angiography for final diagnosis.
Outcome of hepatic artery stenting was evaluated for patient and graft survival, length of hospital stay for the procedure, complications of the procedure, and further hepatic artery complications.
Complications that derived from the vascular procedure itself as well as general complications were evaluated. General complications were recorded before and after stenting.
Statistics were based on t test. P < 0.05 was considered significant. This study was reviewed and approved by the Institutional Review Board of Human Protection at Baylor University Medical Center and Baylor Healthcare System in Dallas, TX.
Twenty-two out of 26 patients survived more than 2 months after the hepatic artery stenting procedure. Four patients died within 2 months from liver failure (n = 1), multiorgan failure (n = 1), sepsis (n = 1), and hepatic artery pseudo aneurysm (n = 1). Twenty-two patients were followed for mean 31 ± 14 months (8-71 months). One of 22 patients died from renal failure 2 years after hepatic artery stenting. Kaplan-Meier patient survival was 85% at 1 and 2 years, and 78% at 5 years (Fig. 1).
The site of hepatic artery stenosis was assessed by angiogram. The type of anastomosis included: aortic patch of donor to common hepatic artery of recipient (n = 1), celiac artery of donor to aortic conduit of recipient (n = 3), celiac artery of donor to proper hepatic artery of recipient (n = 13), common hepatic artery of donor to common hepatic artery of recipient (n = 4), common hepatic artery of donor to proper hepatic artery of recipient (n = 3), common hepatic artery of donor to aortic conduit of recipient (n = 1), and superior mesenteric artery of donor to proper hepatic artery of recipient (n = 1).
Initial Finding of Stenosis
The presence of an arterial stenosis was first suspected because of elevation of liver enzymes (n = 14) at routine follow-up angiogram after hepatic artery revision (n = 5), finding of a DUS performed for general purpose (n = 3), and DUS done at yearly routine follow-up (n = 4).
Aminotransferases were recorded at the time of diagnosis. Mean aspartate aminotransferase was 150 U/L (range, 21-1,109) and mean alanine aminotransferase was 191 U/L (range, 11-918). Seven of the 26 patients had aminotransferases within normal limits.
Procedure and Complications
Stenting was performed at a mean interval of 203 days (range, 34-784 days) after liver transplantation and a mean of 161 days (9-784 days) in the cases that previously underwent hepatic artery reconstruction. An angiogram of a hepatic artery stenosis is shown in Figure 2. The length of hospital stay for the hepatic artery stenting procedure was 2.7 ± 1.3 days (n = 20). Six patients underwent hepatic artery stenting while hospitalized for other reasons. Immediate complications were hematoma at the arterial puncture site (n = 2), puncture site bleeding (n = 2), hepatic artery aneurysm (n = 1), and arterio-venous fistula at the insertion site (n = 1). The complications were treated successfully except for the patient with hepatic artery aneurysm who died 44 days after procedure. This patient's hepatic artery was ligated.
Twenty-two patients entered follow-up exceeding two months. Twelve of them did not have arterial complications after stenting. However, the other 10 patients (45%) developed recurrent hepatic arterial complications. The observed complications were restenosis (n = 8), hepatic artery thrombosis (n = 1), and long segment stricture (n = 1). The time between the stenting and the diagnosis of a new arterial complication was 136 ±104 days (16-314 days). The Kaplan-Meier hepatic artery complication-free graft survival among 22 patients is shown in Figure 3. One- and 2-year arterial complication-free graft survival was 54%, and 5-year complication-free graft survival rate was 47%.
Eight patients who developed recurrent hepatic arterial stenosis were successfully treated interventionally with restent (n = 5) and balloon dilation (n = 3). However, 2 out of the 8 patients with recurrent stenosis (25%) experienced subsequent hepatic artery thrombosis. Three patients (38%) developed another stenosis; 2 of them were treated with balloon dilatation, and the third required surgical repair. An angiogram for interventional repair is shown in Figure 4.
All patients received intravenous bolus of heparin 5,000 U at the beginning of the procedure. Poststenting coagulation prophylaxis was given with a combination of aspirin and clopidogrel bisulfate (Plavix) (n = 12), aspirin (n = 9), clopidogrel bisulfate (Plavix) alone (n = 4), and none (n = 1).
Patients (n = 10) who developed arterial complication received a combination of aspirin and clopidogrel bisulfate (Plavix) (n = 8) and aspirin (n = 2). Only 4 patients who did not develop arterial complication (n = 12) received a combination of aspirin and clopidogrel bisulfate (Plavix).
Two patients required hepatic artery reconstruction surgery. The hepatic artery was successfully repaired in one patient, but the left hepatic artery occluded. The other patient was operated when he developed repeat stenosis after an attempt at balloon dilatation of a restenosis failed. In the latter patient, the stenotic segment was resected and reconstruction was performed using a saphenous vein interposition graft. The metallic stent was found deep in the thickened endothelial tissue. Pathological examination showed intimal proliferation without any signs of acute cellular rejection.
Nonvascular complications observed before stenting were bile leak (n = 4), bile duct stricture (n = 2), splenic rupture (n = 1), duodenal ulcer (n = 1), and gastrointestinal bleeding (n = 1). Nonvascular complications observed after stenting were bile duct stricture (n = 7), chronic rejection (n = 2), stomach perforation (n = 1), ileus (n = 1), pancytopenia and diffuse intravascular coagulation (n = 1), and biloma (n = 1). One patient with bile duct stricture before stenting developed multiple strictures poststenting. Overall, bile duct complications were observed in 12 of the 22 patients (54%).
At the time of stenting, 21 patients were on tacrolimus and five patients were on cyclosporine-based immunosuppression. Twenty patients were on steroids. In 12 patients, mycophenolate mofetil (n = 11) or sirolimus (n = 1) was added as a third agent prior to stenting.
After stenting, steroid was discontinued in six patients. Additionally, sirolimus was started in five patients to attempt prevention of restenosis after stent insertion.
Hepatic artery stenosis is an uncommon but serious vascular complication after liver transplantation. Its incidence is estimated to be between 1.6 and 8% in adults.3–6 The progression of hepatic artery stenosis may result in hepatic artery thrombosis, which is associated with high morbidity, graft loss and mortality.
Hepatic artery stenosis at or near the site of the hepatic artery anastomosis may be due to operative technique or vascular clamp injury and predisposes to subsequent hepatic artery thrombosis. Other etiologies may include allograft rejection7, 8 or microvascular injury associated with cold preservation injury.9 The clinical presentation is usually graft dysfunction or biliary tract complication related to the decreased hepatic blood flow.10
Interventional vascular procedures are used increasingly as a therapeutic alternative for the treatment of hepatic artery stenosis. Fibrinolysis for early hepatic artery thrombosis was reported.11–13 Several series of balloon dilatation with fibrinolysis have been reported for hepatic artery stenosis. However, the success rate of the balloon dilatation is questionable. Fibrinolysis and percutaneous transluminal angioplasty have a high early success rate in recanalizing the hepatic artery with relatively few complications compared to surgery. However, when performed alone, they do not ensure adequate mid- to long-term patency, which is needed to ensure a good long-term outcome.14–17
Abbasoglu et al. reported that normal liver function was obtained in 67% of patients after hepatic artery revision by either surgical revision or endovascular intervention.15
The superiority of vascular stenting over balloon angioplasty alone in patients with coronary artery stenosis after cardiac transplantations has been reported.18 This technology can be applied to hepatic artery stenosis. The progressive improvements of materials and stent design have produced stents with higher flexibility. Stenting together with improved percutaneous interventional techniques may improve the outcome of arterial complications after liver transplantation. However, only a few cases have been reported so far.19–24
In our series, the overall patient survival rate was 78%, similar to the 80% survival rate after surgical hepatic artery revision after liver transplantation reported in the literature,15 To define the adequate indication, a prospective study should be required for all arterial stenosis cases. Based on our experience, we can only conclude that the survival rate after arterial stenting is a satisfactory alternative to surgical reconstruction. The invasive radiology approach can be considered for short stenosis, for intrahepatic stenosis not accessible by surgery, and for patients with hostile abdomen following multiple surgeries. In this study, we found that hepatic artery stenting was useful to correct recurrent stenosis after surgical hepatic artery revision. Our routine after all surgical repair is to obtain a hepatic angiogram 3 to 5 days after the procedure. If residual stenosis is found, surgical revision or stenting is performed.
In most patients the initial suggestion of hepatic artery stenosis was liver enzyme elevation. However, 27% of the patients did not show liver enzyme elevation. DUS is an important screening tool for hepatic artery stenosis but it is not diagnostic.25 An angiogram must be performed for accurate diagnosis and to plan the treatment of a hepatic artery stenosis. Stenting requires minimal lengths of hospital stay and carries a relatively low risk of complications. We experienced one serious complication of hepatic artery aneurysm just after stenting. However, that case had undergone surgical arterial reconstruction for a stenosis only 13 days prior to the stenting. We now require at least a three-week interval between a surgical reconstruction and subsequent stenting. All patients are treated with acetylsalicylic acid or clopidogrel bisulfate (Plavix) following the stenting procedure to prevent thrombosis. Nevertheless, restenosis develops in some patients. Combinations of acetylsalicylic acid and clopidogrel bisulfate (Plavix) treatment were seen more often in the patients who had subsequent arterial complication than in uncomplicated patients. However, this treatment combination was initiated for particularly difficult cases. Pathology of resected stenotic arterial segments demonstrates growing endothelial tissue. Whether sirolimus-coated stents or oral sirolimus can prevent the endothelial growth is not clear at this point. However, such coated stents were not available at the time of this study at our institution.
Restenosis after hepatic artery revision was reported in up to 26% of patients.15 In our experience, hepatic artery stenting has excellent short- and long-term effectiveness. There are fewer complications compared with angioplasty alone, and the patients require shorter admissions. Percutaneous stenting can also be used as an adjunct to surgical repair for hepatic artery stenosis. Two of our patients developed hepatic artery thrombosis. One patient developed hepatic artery thrombosis 18 days after restenting of a stenosed hepatic artery conduit. The other patient thrombosed 63 days after first stenting. No predictive factors for thrombosis could be found in these two patients. However, restenosis is a common outcome poststenting. More than 25% of the patients developed restenosis. Most stenoses were found at the stent site itself. Restenosis was treated with restenting or angioplasty with short-term success. Such attempts had a high incidence of hepatic artery thrombosis (25% in our series). Therefore, we advocate surgical repair for arterial stenosis recurring after stenting. Biliary complications such as intrahepatic strictures, bile duct necrosis, and intrahepatic bile leaks commonly occur in patients with hepatic artery thrombosis.10 Patients with severe arterial stenosis have a tendency to develop biliary complications such as strictures or biloma even after successful stenting.
In conclusion, percutaneous stenting is an effective treatment of hepatic artery stenosis after liver transplantation. Stenting is well tolerated. It requires short admissions and has excellent short-term success rates. Hepatic artery stenting is useful not only for primary stenosis but also as an adjunct treatment after surgical hepatic artery revision.