Rupture of a Stenotic Hepatic Artery After Liver Transplantation: Endovascular Salvage Using a Covered Stent

Transplant Hepatic Artery Stenosis: Endovascular Treatment and Complications

Hepatic artery stenosis (HAS) is an infrequent complication of liver transplant; if left untreated, it can lead to hepatic artery thrombosis with high risk of biliary necrosis and graft loss. HAS is diagnosed with screening Doppler ultrasound, together with computed tomography angiography and magnetic resonance angiography. Endovascular treatment with angioplasty ± stent placement is safe and effective with infrequent major complications; however, when complications occur, they can devastate long-term graft survival. Herein, we present two cases of HAS treated with balloon angioplasty with resultant major complications requiring operative intervention.

Complications after endovascular treatment of hepatic artery stenosis after liver transplantation

BACKGROUND

Hepatic artery stenosis (HAS) after liver transplantation can progress to hepatic artery thrombosis (HAT) and a subsequent 30% to 50% risk of graft loss. Although endovascular treatment of severe HAS after liver transplantation has emerged as the dominant method of treatment, the potential risks of these interventions are poorly described.

METHODS

A retrospective review of all endovascular interventions for HAS after liver transplantation between August 2009 and March 2016 was performed at a single institution, which has the largest volume of liver transplants in the United States. Severe HAS was identified by routine surveillance duplex ultrasound imaging (peak systolic velocity >400 cm/s, resistive index <0.5, and presence of tardus parvus waveforms).

RESULTS

In 1129 liver transplant recipients during the study period, 106 angiograms were performed in 79 patients (6.9%) for severe de novo or recurrent HAS. Interventions were performed in 99 of 106 cases (93.4%) with percutaneous transluminal angioplasty alone (34 of 99) or with stent placement (65 of 99). Immediate technical success was 91%. Major complications occurred in eight of 106 cases (7.5%), consisting of target vessel dissection (5 of 8) and rupture (3 of 8). Successful endovascular treatment was possible in six of the eight patients (75%). Ruptures were treated with the use of a covered coronary balloon-expandable stent graft or balloon tamponade. Dissections were treated with placement of bare-metal or drug-eluting stents. No open surgical intervention was required to manage any of these complications. With a median of follow-up of 22 months, four of eight patients (50%) with a major complication progressed to HAT compared with one of 71 patients (1.4%) undergoing a hepatic intervention without a major complication (P < .001). One patient required retransplantation. Severe vessel tortuosity was present in 75% (6 of 8) of interventions with a major complication compared with 34.6% (34 of 98) in those without (P = .05). In the complication cohort, 37.5% (3 of 8) of the patients had received a second liver transplant before intervention compared with 12.6% (9 of 71) of the patients in the noncomplication cohort (P = .097).

CONCLUSIONS

Although endovascular treatment of HAS is safe and effective in most patients, target vessel injury is possible. Severe tortuosity of the hepatic artery and prior retransplantation were associated with a twofold to threefold increased risk of a major complication. Acute vessel injury can be managed successfully using endovascular techniques, but these patients have a significant risk of subsequent HAT and need close surveillance.

The use of Dacron graft for cavo-cavostomy in orthotopic liver transplantation

The end-to-end "interposition" technique and end-to-side "piggyback" technique are standard approaches to in situ anastomosis during orthotopic liver transplantation. We demonstrate that anastomosis of liver allograft to a Dacron vena cava graft can be a feasible solution if traditional anastomoses cannot be used. A 55-year-old man with end-stage liver failure from alcoholic cirrhosis underwent orthotopic liver transplantation; however, an intraoperative complication during recipient hepatectomy rendered the native vena cava unsalvageable. In addition, the donor vena cava was too short to bridge the caval defect for interposition. We therefore used Dacron for an in situ graft to span the gap, with subsequent anastomosis of the allograft to the prosthetic graft in piggyback fashion. The patient did well postoperatively; his only major complication was late anastomotic stenosis, which was treated successfully with balloon dilatation. Unfortunately the patient became recidivous and expired ten months posttransplant, despite indications of satisfactory allograft function.

Development of in vivo tissue-engineered autologous tissue-covered stents (biocovered stents)

Biocovered stents, which are stents covered with autologous membranous tissues, were developed by applying a novel concept based on in vivo tissue engineering. Balloon-expandable stents crimped on silicone rods as a mold (diameter: 2 mm) were embedded into dorsal subcutaneous pouches in rabbits. After 1 month, the struts of the stents were fully encapsulated with membranous connective tissues formed around the silicone rods. Upon removing the silicone rod, stents covered with tubular connective tissues, in which the struts were completely impregnated, were obtained as biocovered stents. These tissues were composed mainly of collagen and fibroblasts and had a thickness of less than approximately 200 microm with an excellent high burst strength of approximately 1000 mmHg. The luminal surface of the tissues was extremely flat and smooth. The stents could be mounted on balloon catheters with a hand crimping tool and could be expanded by inflation with little damage to the tissues. It is anticipated that these novel stents may greatly enhance early normal vascular reconstruction with high reliability, thereby reducing the rate of in-stent restenosis.

Development of self-expandable covered stents

We newly developed self-expandable covered stents by combining two of our original technologies. Of these, the first is the dip-coating covering method that was developed previously for balloon-expandable stents; the other is the newly developed self-expandable Nitinol stents, namely, Sendai stents. The three types of covered stents with the expansion diameter of 4.5, 5.0, or 6.0 mm thus obtained had a laser-processed microporous elastomeric cover film (pore diameter: 100 microm, interpore distance: 250 microm). Although the film was extremely thin (approximately 15 microm), the film could be expanded without causing any damage, the strut was completely embedded within the film, and the luminal surface of the film was smooth and flat. Mechanical properties such as ideal flexibility to follow the shapes of arteries were almost retained even after covering. As appropriate drugs, the blood-contacting inner and tissue-contacting outer surfaces of the film were differentially coated with argatroban for antithrombogenicity or FK506 for anti-inflammation, respectively. The preliminary in vivo study indicated that the covered stents mounted in the delivery catheter were navigated and placed to appropriate position in the arteries, and permissible neointimal thickening after 1-month implantation was observed similarly in noncovered stents.