EXPERIMENTAL AND CLINICAL TRANSPLANTATION

Objectives: Hepatic artery thrombosis remains a major complication after orthoptic liver transplant. Treatment of hepatic artery thrombosis is complex and requires a multidisciplinary approach. Retransplant is the procedure of choice. In nonsurgical candidates, endovascular options are evolving.

Materials and Methods: Based on our experience at a busy transplant center, we discuss 4 representative cases to explain the potential role of endovascular treatment beyond just attempts at recanalization. From our experience, as well as a review of the literature, we propose a clinical practice algorithm for optimal treatment of hepatic artery thrombosis after orthoptic liver transplant.

Results: The primary traditional endovascular interventional options remain thrombectomy, balloon angioplasty, and use of stents with the aim of revascularization. However, these methods have not proven to be effective. Ultrasonography-assisted thrombolysis, which has thus far been relatively less described in the hepatic vasculature, has the potential of producing the same angiographic results but at lower doses of the thrombolytic agent, thus decreasing the potential for hemorrhagic complications. The adjunctive use of splenic artery embolization and prompt treatment of biliary complications are in our opinion useful in “buying time” to allow adequate development of collateral “neovascularization of the liver,” thus preventing further ischemia.

Conclusions: Although surgical retransplant still remains the standard treatment for hepatic artery thrombosis, organ shortages and high mortality still exist. Endovascular techniques are rapidly evolving, but these techniques are dependent on expertise available and, even in the best hands, have not proven to be effective at reversing hepatic artery thrombosis. The use of a multimodality endovascular approach could salvage the liver allografts, thereby preventing retransplant or facilitating transplant at a more elective setting.

Key words : Posttransplant hepatic artery thrombosis, Endovascular options, Neovascularization of the liver, Ultrasonography-assisted thrombolysis

Hepatic artery thrombosis (HAT) remains a major complication after orthoptic liver transplant. The treatment of HAT is complex and requires a multidisciplinary approach by both interventional radiology and transplant services. Although surgical transplant still remains the standard treatment for HAT, organ shortage and high mortality related to retransplant still exist. The role of endovascular techniques is rapidly evolving. However, these techniques require expertise and, even in the best hands, have not proven to be effective in reversing thrombosis.^1-3

Based on our experience at a transplant center, we describe 4 representative patient cases to explain a potential role of endovascular treatment.

Patient 1
Patient 1 (Figure 1) was a 6-year-old boy at day 2 after orthoptic liver transplant who presented with HAT on routine posttransplant ultrasonography. Angiography showed diffuse irregular narrowing of the hepatic artery, confirming HAT. Intra-arterial thrombolysis was performed using a Micromewi Multi-Sidehole Infusion Catheter (ev3, Inc., Plymouth, MN, USA), and tissue plasminogen activator (tPA) was infused at 0.5 mg/h for 16 hours. Follow-up ultrasonography after tPA infusion demonstrated a recanalized hepatic artery. The infusion catheter was then removed, and ultrasonography was repeated, revealing a significant change in the color flow in the hepatic artery with loss of flow within a few minutes. Angiography revealed diffuse spasm of the vessel. After reinsertion of the infusion catheter, 200 μg of nitroglycerin was administered. Papaverine (vasodilator) was infused at 6.25 mg/h overnight. The patient was also placed on intravenous heparin for anticoagulation. After the patient received a loading dose of 40 U/kg, the dose was titrated at 2 U/kg/h to maintain partial thromboplastin time of between 60 and 80 minutes. These attempts to recanalize the hepatic artery showed no significant improvement. The study was terminated, and retransplant was performed

Patient 2
Patient 2 (Figure 2) was a 57-year-old Hispanic woman who presented with pain in the hypochondrium 12 days after orthoptic liver transplant. Baseline serum liver enzymes were elevated (aspartate aminotransferase 128 U/L and alanine aminotransferase 67 U/L). Ultrasonography and angiography confirmed HAT. Intra-arterial thrombolysis was performed using a Cragg-McNamara catheter (Micro Therapeutics Inc., Irvine, CA, USA) loaded with Katzen infusion wire (Boston Scientific, Marlborough, MA, USA), with tPA infused at a rate of 0.5 mg/h for 20 hours. There was minimal angiographic improvement. Mechanical thrombectomy (AngioJet, MEDRAD Inc, Warrendale, PA, USA) using pulse-spray thrombolysis was attempted but was unsuccessful. Ultrasonography-assisted thrombolysis was performed using the EkoSonic Endovascular System (EKOS Corporation, Bothell, WA, USA) infusion catheter, with tPA infused at rate of 0.5 mg/h for 15 hours. After overnight infusion, the patient’s serum hepatic enzymes decreased (aspartate aminotransferase 80 U/L, alanine aminotransferase 56 U/L). Follow-up angiography demonstrated complete resolution of clot. However, there was sluggish flow in the recanalized hepatic artery. Embolization of the splenic artery was performed with an 8 mm × 7 mm Amplatzer vascular plug (St. Jude Medical, St. Paul, MN, USA) to avoid visceral steal phenomenon and direct flow toward the liver. A postembolization ultrasonography examination demonstrated a doubling of arterial blood flow to the allograft versus that shown before embolization. A biloma developed 3 weeks later, which was not seen on prior imaging. This was presumed to be secondary to HAT rather than a posttransplant complication. The biloma was successfully treated by percutaneous biliary drainage with an 8F drainage catheter. A routine ultrasonographic examination 57 days after thrombolysis detected recurrent HAT. However, there appeared to be development of collateral vessels, preserving the liver allograft. Retransplant was avoided, and the patient remained asymptomatic even at 57 weeks

Patient 3
Patient 3 (Figure 3) was a 56-year-old man who presented with lower limb edema 9 years after orthoptic liver transplant. Hepatic enzymes were elevated (aspartate aminotransferase 271 U/L, alanine aminotransferase 369 U/L). Ultrasonography and later computer tomography angiogram both showed HAT. The hepatic artery was patent on a computed tomography angiogram 4 weeks earlier. On retrospective examination, there appeared to be a small area of narrowing that had not been noticed. Intra-arterial thrombolysis was initiated with Cragg-McNamara catheter loaded with a Katzen infusion wire, with tPA infused at a rate of 1.0 mg/h for 12 hours. A postinfusion angiography showed significant residual clot burden, which led to the patient having mechanical thrombectomy, resulting in marginal improvement. Ultrasonography-assisted thrombolysis was then pursued, with tPA infused at a rate of 1.0 mg/h for 21 hours resulting in significant improvement in clot burden. However, this treatment uncovered an underlying critical stenosis at the surgical anastomosis. The stenosis was treated with balloon angioplasty, during which there was recurrence of HAT. Ultrasonography-assisted thrombolysis was again performed, with tPA infused at 1.5 mg/h for > 23 hours. Hepatic enzymes significantly decreased (aspartate aminotransferase 112 U/L, alanine aminotransferase 89 U/L), and angiography demonstrated complete clot resolution. A moderate hepatic artery stenosis persisted. This was addressed by placement of a 6-mm iCast covered stent (Atrium Medical Corp., Hudson, NH, USA). The patient showed good resolution of the lower limb edema. Routine follow-up ultrasonography at 52 weeks demonstrated a patent hepatic artery with brisk, arterial flow. The patient continued to do well with no complications

Patient 4
Patient 4 (Figure 4) was a 63-year-old woman. She had received 2 liver transplants, with the first graft failing due to HAT. After retransplant, her early postoperative period was complicated by peritonitis and a colonic perforation, resulting in re-exploration. On postoperative day 18, a routine screening ultrasonography demonstrated HAT. Hepatic enzymes were normal. Ultrasonography-assisted thrombolysis was pursued, with tPA infused at 0.5 mg/h for 13 hours. Follow-up angiography demonstrated complete resolution of clot. Ultrasonography and computer tomography angiogram at 12 weeks showed patent vessels. A biloma that developed 16 weeks later was successfully treated with percutaneous biliary drainage. The patient’s course was complicated by 2 intrahepatic mycotic pseudoaneurysms, which were thought to develop due to the prior infected graft and underlying sepsis. These were successfully treated with a combination of intravenous antibiotics and endovascular techniques. Axium detachable coils (Ev3/Covidien, Plymouth, MN, USA) and liquid embolic Onyx34 (Ev3/Covidien) were used to embolize the pseudoaneurysms in staged procedures. The patient remained asymptomatic. Computer tomography angiogram at 52 weeks demonstrated patent hepatic arterial vasculature and a surviving liver allograft Patient 1 to 4 timelines have been illustrated in Figure 5.

Treatment of hepatic artery thrombosis
Hepatic artery complications have been reported in 4% to 25% of patients4 after liver transplant, with life-threatening HAT shown in 3% to 9%.^2,3,5 Hepatic artery complications can be divided into 2 categories depending on the time of onset: early and late. Early HAT often presents asymptomatically within the first 4 weeks after liver transplant. If it is untreated, this complication carries a significant morbidity and mortality with 75% rate of allograft failure.1-3 In the early period after orthoptic liver transplant, the collateral supply to the liver is poor and hepatic parenchyma and biliary tree rely mainly on the blood vessels derived from hepatic arteries. Therefore, early HAT causes bile duct necrosis followed by uncontrollable sepsis in the already immunocompromised patient.⁶

The most effective treatment approach to HAT is controversial and includes retransplant, surgical revascularization, and endovascular revascularization. Although retransplant has been the first choice of therapy, it is associated with higher morbidity than primary transplant.⁷ Surgical options for acute HAT have traditionally included surgical revascularization and open thrombectomy.⁷ With major advancements in technology, endovascular management has emerged as a less invasive alternative treatment option. Hepatic artery thrombosis has been reported to be successfully treated with multiple endovascular techniques, including transcatheter intra-arterial thrombolysis, percutaneous transluminal angioplasty, stenting, or a combination of these.^3,8 Although presently there are no published randomized clinical trials on the use of thrombolytic therapy,⁹ selective thrombolysis via the hepatic artery has several advantages such as small thrombolytic dose, high localized concentration, and little influence on systemic coagulation. It is thought to be safe and effective if the infusion catheter is placed inside the thrombus.^10-12 Despite its local effect, hemorrhage is the most common complication of intra-arterial thrombolysis; hence, it is necessary to monitor closely by evaluating parameters such as fibrinogen, prothrombin time, or activated partial thromboplastin time.8 Percutaneous mechanical thrombectomy, which has been successfully used in other vascular beds,¹³ also can be used as an alternative treatment for HAT when intra-arterial thrombolysis is contraindicated.¹⁴ Restoration of flow using thrombolysis can uncover underlying anatomic defects, including kinking, anastomotic stenosis, or stricture, which if left untreated can lead to rethrombosis.^12,15-17Hence, the combined use of thrombolysis with percutaneous transluminal angioplasty and/or stenting has been shown to have better patency and survival rates than thrombolysis alone.^18,19 Angioplasty is useful in treatment of first-time stenosis, with stenting reserved for resistant stenosis.²⁰

In 2009, Singhal and associates8 reviewed 69 published cases from 16 reports in the literature on endovascular therapy for HAT. In their review, they estimated the success rate of thrombolysis to be 68% (47 of 69 cases), of which 29 patients (62%) underwent further intervention including percutaneous transluminal angioplasty (n = 4, with 2 remaining patent at 1 and 8 months), stent placement (n = 20, with 5 remaining patent at 2 years), or both (n = 5, with 1 remaining patent). Minor bleeding was observed in 15 patients (22%), and fatal intra-abdominal hemorrhage occurred in 3 patients. Among patients who had successful endovascular treatment (thrombolysis with or without percutaneous transluminal angioplasty plus stenting), either the need of retransplant or mortality was reported in only 2 patients. Among patients in the failed group, subsequent treatment was contemplated in the form of retransplant and surgical revascularization. None of the reports discussed the long-term results of endovascular therapy. This and other reports have concluded that, although there are encouraging results of endovascular interventions, the efficacy and risk of complications of these procedures are controversial, with need for resorting to surgical intervention in many cases.8,21 In addition to hemorrhage, procedure-related complications include rethrombosis, vascular dissection, or even rupture.²²

Ultrasound-accelerated thrombolysis is emerging as a promising tool in many vascular beds, including in peripheral arterial,²³ venous,²⁴ portal,²⁵ pulmonary,^25,26 and even in intracerebral vasculature.²⁷ High-frequency ultrasound waves synergistically enhance the fibrinolytic effects of the thrombolytic due to direct-mechanical acoustic streaming, improving clot permeability via creation of micropores and thus potentially reducing tPA doses and infusion rates. In addition, the thrombolytic agent reversibly alters fibrin structure, increasing the number of accessible binding sites.^28-30

In recent years, the role of splenic artery embolization in hypersplenism associated with portal hypertension has increased. Besides this primary role, splenic artery embolization has shown added advantages in effectively increasing hematologic indexes and improving responses to liver function tests. Specifically, in the transplant setting, splenic artery embolization has been effective in the management of splenic artery steal syndrome, a condition in which increased splenic blood flow due to low splenic arterial resistance, secondary to splenomegaly, siphons blood from the already decreased blood flow due to increased hepatic arterial resistance in the transplant liver, thereby decreasing liver blood flow.^31-34 Complications like bilomas are seen after HAT in orthoptic liver transplant. These can be successfully managed using percutaneous drainage techniques.^35,36

Interruption of the hepatic arterial blood flow to the new allograft is not necessarily associated with bile duct necrosis. This may be explained by the presence of collaterals (mainly derived from the phrenic arteries), which have been shown angiographically as early as 2 weeks after liver transplant.¹⁹ Dydynski and associates described “collateral transformation of the hepatic artery” as the development of multiple collaterals in the perihilar region.³⁷ Vaidya and associates38 reported collaterals in 19.4% of patients undergoing angiography performed for a suspected hepatic artery abnormality. Panaro and associates³⁹ also reported collaterals in 36.7% of long-term survivors after HAT, terming this a “neovascularized liver.” Fouzas and colleagues described their experience with 3 patients in which development of collateral arterial flow was beneficial and saved patients from transplant.⁴⁰ The phenomenon of a “neovascularized liver” is poorly understood. It has been attributed to angiogenesis, which may involve the omentum and is believed to display angiogenic potential to promote neovascularization in chronically ischemic organs.⁴¹ Collaterals may prevent a disastrous outcome in patients with late HAT.^42-44

Our experience in treating hepatic artery thrombosis
In this study, we selected 4 patients who presented with HAT after orthoptic liver transplant. In patients 1, 2, and 3, endovascular options were sought due to lack of availability of a donor for retransplant. Patient 4 had a previous transplant; hence, a third transplant would have increased morbidity.

In patient 1, intra-arterial thrombolysis was successful in restoring the patency of the hepatic artery. However, this was short lived, and the vessel went into spasm, unrelieved even with overnight transcatheter infusion of a vasodilator and systemic anticoagulation. Retransplant was performed. In patient 2, intra-arterial thrombolysis and mechanical thrombectomy failed to recanalize the hepatic artery. Left with limited options, our group decided to do perform an ultrasound-accelerated thrombolysis. Ultrasound-accelerated thrombolysis has so far not been frequently described in the hepatic vasculature; however, for our patient, the procedure resulted in resolution of a clot. The hepatic artery was patent even at 57-week follow-up. Patient 3 presented in a similar way. Both intra-arterial thrombolysis and mechanical thrombectomy failed to recanalize the hepatic artery. Because of our successful experience with ultrasound-accelerated thrombolysis in patient 2, we decided to explore this treatment option. Ultrasound-accelerated thrombolysis was performed by injecting tPA at 1 mg/h for 21 hours, which resulted in clot resolution. However, this uncovered a stenosis, with angioplasty resulting in recurrent HAT. Ultrasound-accelerated thrombolysis was again attempted, with success. This time, we decided to stent the hepatic artery, resulting in a patent hepatic artery at 52-week follow-up. Patient 4 presented a few months after our experience with our 2 cases involving ultrasound-accelerated thrombolysis. We decided to perform ultrasound-accelerated thrombolysis as a first-line approach. Injection of tPA at 0.5 mg/h for 13 hours resulted in complete resolution of clot with no rethrombosis at 52-week follow-up.

As in patients 1, 2, and 3, our other experiences with intra-arterial thrombolysis have not been very encouraging. Ultrasound-accelerated thrombolysis was successful in eliminating thrombus in patients 2, 3, and 4. Adjunctive procedures like splenic artery embolization were performed to maintain patency and possibly avoid retransplant. In patient 2, even an adjunctive procedure such as splenic artery embolization was not useful as the hepatic artery reclotted 57 weeks later. However, by this time, collaterals had already developed, which probably saved the liver from further ischemia. As shown in Figure 2e and 2f, the right inferior phrenic artery appeared relatively hypertrophied and appeared to collateralize with branches from the hepatic artery. In patient 3, HAT developed 9 years after transplant and collateral formation was already demonstrated in the transplanted liver. These collaterals were distinctly seen and documented during the preintervention ultrasonography and angiogram (Figure 3a, 3b, and 3c). In patient 4, although ultrasound-accelerated thrombolysis resulted in complete recanalization of the hepatic artery, she developed mycotic pseudoaneurysms. These were successfully treated using endovascular coil and liquid embolization, thus preventing a third retransplant. Two of the patients developed biliary complications, which were treated successfully with percutaneous intervention.In our patients, ultrasound-accelerated thrombolysis was performed using a 5.2F EkoSonic Endovascular System catheter (EKOS Corporation, Bothell, WA, USA). Placement of a relatively large diameter (5.2F) EKOS catheter in a small caliber hepatic artery can potentially increase risk of iatrogenic injury, such as vessel spasm, as well as more serious injuries, such as dissection and thrombosis. In our limited experience, however, we successfully placed the catheter in the thrombosed allograft vessels without complications. Tissue plasminogen activator infusion rates as low as 0.5 to 1.5 mg/h resulted in satisfactory angiographic results, with a lower total dose to the patient and prevention of potential hemorrhagic complications.

Hepatic artery thrombosis remains a major complication after liver transplant. Management of HAT is complex and requires a multidisciplinary approach.^1-3 If retransplant is not possible, an endovascular approach is now evolving as the next best step. After thrombolysis, the use of traditional endovascular techniques like balloon angioplasty and stent to treat the underlying cause is crucial. However, in line with a literature review,8 our experience at a large transplant center with endovascular HAT recanalization after liver transplant has not been encouraging.

In the 4 representative cases at a busy transplant center, we describe the use of a multimodality endovascular approach with the aim to preserve the liver allograft rather than aggressive attempts at recanalization. Although our experience is limited, the use of ultrasound-accelerated thrombolysis, hitherto relatively less well known in the hepatic vasculature, may have an important role in the delivery of high local concentration of thrombolytic, thereby avoiding major hemorrhagic complications. We suggest a clinical practice algorithm in treatment of HAT after orthoptic liver transplant based on our experience (Figure 6).

An important phenomenon of “neovascularization of the liver”39 due to development of collateral blood supply beginning as early as 12 weeks after transplant has been described in a few reports. The use of adjunctive measures like splenic artery embolization and prompt treatment of biliary and other complications may be useful in buying time to allow development of collaterals, thus resulting in liver salvage. Furthermore, evaluation for collaterals, during initial angiographic examinations, may help plan the type of intervention and aggressiveness of therapy. More studies are essential to evaluate the role of these measures in avoiding retransplant or facilitating the transplant to a more elective setting.

Volume: 14 Issue: 5 October 2016

FULL TEXT