- Email: [email protected]
Endovascular treatment of hepatic artery stenosis after liver transplantation
From the Society for Clinical Vascular Surgery
Endovascular treatment of hepatic artery stenosis after liver transplantation Blake A. Hamby, MD,a Daniel E. Ramirez, MD,a George E. Loss, MD, PhD,b Hernan A. Bazan, MD,a Taylor A. Smith, MD,a Edward Bluth, MD,c and W. Charles Sternbergh III, MD,a New Orleans, La Background: Hepatic artery stenosis (HAS) after orthotopic liver transplantation is a significant risk factor for subsequent hepatic artery thrombosis (HAT). HAT is associated with a 30%-50% risk of liver failure culminating in retransplantation or death. Traditional treatment of hepatic artery complications has been surgical, with hepatic artery revision or retransplantation. Endovascular therapy of HAS, described primarily in the interventional radiology literature, may provide a less-invasive treatment option. Methods: This was a retrospective review of all endovascular interventions performed for HAS after orthotopic liver transplantation over a 31-month period (August 2009 to January 2012). Patients with duplex ultrasound imaging evidence of severe main HAS (peak systolic velocity of >400 cm/s, resistive index of <.5) underwent endovascular treatment with either primary stent placement or percutaneous transluminal angioplasty (PTA) alone. Patients were followed with serial ultrasound imaging to assess for treatment success and late restenosis. Reintervention was performed if significant restenosis occurred. Results: Thirty-five hepatic artery interventions were performed in 23 patients. Over the 31-month study period, 318 orthotopic liver transplantations were performed, yielding a 7.4% (23/318) rate of hepatic artery intervention. Primary technical success was achieved in 97% (34/35) of cases. Initial treatment was with PTA alone (n [ 10) or primary stent placement (n [ 13). The initial postintervention ultrasound images revealed improvements in hepatic artery peak systolic velocity (267 ± 118 [posttreatment] vs 489.9 ± 155 cm/s [pretreatment]; P < .0001) and main hepatic artery resistive index (0.61 ± 0.08 [posttreatment] vs 0.41 ± 0.07 [pretreatment]; P < .0001). At a mean follow-up of 8.2 ± 1.8 months (range, 0-29), there were 12 reinterventions in 10 patients for recurrent HAS. Thirty-one percent (n [ 4/13) of patients undergoing initial stent placement required reintervention (at 236 ± 124 days of follow-up) compared with 60% (n [ 6/10) of patients undergoing initial PTA (at 62.5 ± 44 days of follow-up). Primary patency rates (Kaplan-Meier) after primary stent placement were 92%, 85%, and 69% at 1, 3, and 6 months, respectively, compared with 70%, 60%, and 50% after PTA (P [ .17). Primary-assisted patency for the entire cohort was 97% at 6 and 12 months. Major complications were one arterial rupture managed endovascularly and one artery dissection that precipitated HAT and required retransplantation. The overall rate of HAT in the entire cohort was 4.3% (1/23). Conclusions: Endovascular treatment of HAS can be performed with high technical success, excellent primary-assisted patency, and acceptable morbidity. Initial use of a stent may improve primary patency when compared with PTA. The need for reintervention is common, placing particular importance on aggressive surveillance. Longer follow-up and a larger cohort are needed to confirm these encouraging early results. (J Vasc Surg 2013;57:1067-72.)
Hepatic artery thrombosis (HAT) is the most common vascular complication after orthotopic liver transplantation, with an incidence estimated to be between 4% and 11% in adult transplants and as high as 11%-26% in the pediatric population.1-3 Hepatic artery stenosis (HAS) is seen in 5%-11% of transplants and, if left untreated, has up to a 65% chance of developing into HAT within 6 months.4 From the Sections of Vascular and Endovascular Surgery,a Transplant Surgery,b and Diagnostic Radiology,c Ochsner Clinic Foundation. Author conflict of interest: none. Presented at the Fortieth Annual Symposium of the Society of Clinical Vascular Surgery, Las Vegas, Nev, March 14-17, 2012. Reprint requests: W. Charles Sternbergh III, MD, Sections of Vascular and Endovascular Surgery, Ochsner Clinic Foundation, 1514 Jefferson Highway, New Orleans, LA 70121 (e-mail: [email protected]). The editors and reviewers of this article have no relevant financial relationships to disclose per the JVS policy that requires reviewers to decline review of any manuscript for which they may have a conflict of interest. 0741-5214/$36.00 Copyright Ó 2013 by the Society for Vascular Surgery. http://dx.doi.org/10.1016/j.jvs.2012.10.086
Patients with HAT have a 30%-50% incidence of eventual liver failure, leading to retransplantation or death. Hence, maintaining hepatic artery patency is likely an important variable in minimizing graft loss and mortality. Treatment of both HAS and HAT has traditionally involved either open surgical revision of the hepatic artery or liver retransplantation. There are several reports of alternative approaches to this dilemma, including vein graft interposition, endovascular thrombolysis, and mechanical thrombectomy.5-16 Endovascular techniques have also been used as an alternative to open surgical repair with less morbidity and mortality and with similar rates of overall patient survival at 5 years (78% vs 80%, respectively).4,9-11,14,16 Additionally, percutaneous transluminal angioplasty (PTA) with or without stenting has been found to prolong graft survival and delay the need for open surgical revascularization or retransplantation.16 Recently, enhanced collaboration between transplant and vascular surgeons in our institution has resulted in a more aggressive policy of endovascular treatment of 1067
1068 Hamby et al
JOURNAL OF VASCULAR SURGERY April 2013
Fig 1. Hepatic angiogram with an allograft coming from the main abdominal aorta before (left) and after (right) angioplasty.
HAS after orthotopic liver transplantation. This report describes the initial outcomes of this experience in treating HAS, which began in mid-2009. METHODS This is a retrospective review of all endovascular interventions performed for HAS in a single tertiary care institution between August 1, 2009 and January 31, 2012. A single vascular surgeon (W.C.S.) performed all interventions. Institutional review board approval was obtained for this study. Definitions. Significant HAS considered for intervention was defined by a main hepatic artery resistive index (RI) <.5 and a peak systolic velocity (PSV) >400 cm/s as detected by duplex ultrasound imaging. Tardus parvus waveforms were usually observed in these highly stenotic hepatic arteries. In instances where the PSV and/or RI were on the borderline of meeting these criteria, the presence of tardus parvus waveforms generally led to intervention. Significant tortuosity or kinking of the hepatic artery was defined as a >90 acute bend in the vessel, as measured on computed tomography angiogram (CTA) or digital subtraction angiogram. Initial technical success was defined as <30% residual stenosis of the treated hepatic artery by visual estimation of the final arteriogram. Neither intravascular ultrasound imaging nor pressure gradients were used in this series. Diagnostic evaluation. Most patients with significant HAS identified by ultrasound imaging subsequently underwent a CTA to provide additional anatomic information. This study was found to be valuable in case planning. In 22% of our cases, the hepatic artery had an aberrant recipient takeoff, either as a replaced right hepatic arising from the superior mesenteric artery (n ¼ 2/23) or as an accessory left hepatic artery with a direct takeoff from the aorta (n ¼ 1/23) or from an infrarenal allograft (2/23; Fig 1). A steep caudal orientation of the celiac axis was occasionally seen, prompting a change from femoral to brachial access. Finally, the CTA provided excellent information regarding the relative tortuosity of the hepatic artery.
The decision to intervene was based solely on the imaging studies. Liver transaminase levels were frequently normal in patients with significant HAS and were therefore not helpful in patient selection or surveillance. Technique. Femoral access was favored unless there was significant caudal orientation of the celiac artery on the CTA; in these cases, left brachial access was used. Most femoral approach cases were performed using a 6F renal double-curve (RDC) guiding catheter (Boston Scientific, Natick, Mass) to gain access to the hepatic artery. A 4F, 5F, or 6F multipurpose sheath (Cook, Bloomingtom, Ill) was generally used with brachial access. A 0.014-inch wire was used to cross the lesion in most cases. Depending on the degree of tortuosity and need for trackability, a variety of wires were used, from fairly flexible wires (Balanced Medium Weight; Abbott, Abbott Park, Ill), to increasingly stiff hydrophilic wires (Choice PT, Platinum Plus; Boston Scientific), to extrastiff wires (Balanced Heavy Weight; Abbott). The stiffer wires sometimes required placement through a quick-cross catheter (Spectranetics, Colorado Springs, Colo), placed previously over a softer wire. Occasionally, a second “buddy wire” was required to further straighten a tortuous vessel to allow delivery of a stent. Low-profile coronary angioplasty balloons were preferentially used, as their trackability was much better in these tortuous vessels. Angioplasty balloons (2.5-5.0 mm in diameter) were sized to the nominal vessel reference size and inflated for a minimum of 60 seconds. Low-profile self-expanding stents (X-pert; Abbott, 4-6 mm in diameter) or coronary balloon-expandable stents (2.5-5.0 mm in diameter) were used. Drug-eluting coronary stents were selectively placed for recurrent stenosis in two patients. Self-expanding stents were preferentially chosen when there was a major size mismatch of the vessel proximal and distal to the target legion. Early in our experience, we also attempted to use self-expanding stents in lesions located in highly angulated/tortuous hepatic vessels. It was thought that these would conform better and be less apt to create a new “kink” distal to the stent, as a more rigid balloon-
JOURNAL OF VASCULAR SURGERY Volume 57, Number 4
expandable stent might. However, the trackability of selfexpanding stents in highly tortuous vessels was found to be poor, even with stiff wires and the use of a “buddy” wire. The coronary profile balloon-expandable stents were found to be much more trackable in these anatomic situations; their utilization increased as we gained more experience. Lesions were routinely predilated with PTA prior to placement of a stent. Early in our experience, PTA alone was used if a <30% residual stenosis of the hepatic artery was achieved. Provisional stenting was performed if there was a greater than 30% residual narrowing. Despite confirmation of significant improvement in HAS by ultrasound imaging after initial PTA, we frequently observed rapid restenosis. On the basis of these early results, we shifted to primary stenting when technically possible. Postintervention protocol. Patients underwent ultrasound imaging at 1-7 days postintervention, then typically at 6 weeks, 3, 6, and 12 months and yearly thereafter. Patients with evidence of moderate recurrent stenosis were followed more frequently. Treated patients typically received dual-antiplatelet therapy (acetylsalicylic acid [ASA] 81 mg and clopidogrel [Plavix] 75 mg) for 3-6 months postintervention. Clopidogrel was usually stopped after 6 months if there was no evidence of recurrent stenosis. Statistics. Pre- and postintervention mean hepatic artery PSV and RI were compared using a two-tailed ttest. The Fischer exact test was used to compare reintervention rates. Primary patency with angioplasty alone compared with angioplasty and stent, freedom from reintervention, and assisted primary patency were analyzed using Kaplan-Meier curve log analysis. Data are presented as means 6 standard errors. RESULTS Thirty-five endovascular interventions for HAS were performed in 23 patients during the 31-month study period. Of note, 13% (3/23) of treated patients underwent a second orthotopic liver transplantation; all had lost their first liver transplant secondary to HAT. From August 2009 through January 2012, 318 liver transplants were performed, with 7.2% (23/318) of patients undergoing endovascular treatment for significant HAS. No open repairs of HAS were performed during the study period. The mean time to initial intervention after transplantation was 92.5 6 62 days (range, 12-236). Patients had a mean age of 49.4 6 9 years (range, 15-63), and 65% (15/23) were male. Ultrasound findings. Preintervention ultrasound imaging revealed a mean main hepatic artery PSV of 490 6 34.9 cm/s and an RI of 0.41 6 0.07. Initial postintervention PSV significantly decreased to 267 6 49 cm/s, and the RI increased to 0.61 6 0.08 (P < .0001). These salutary changes were sustained at 1, 3, and 6 months (Table I). Technical success. Primary technical success was achieved in 97% (34/35) of cases. Initial treatment was PTA (n ¼ 10) or primary stent placement (n ¼ 13). There
Hamby et al 1069
Table I. Ultrasound findings
Preintervention Postintervention 1-month follow-up 3-month follow-up 6-month follow-up
n
Peak systolic velocity(cm/s)
Resistive index
Tardus parvus
35 34 29 27 17
490 257 234 230 243
0.41 0.61 0.6 0.62 0.68
83% 13% 16% 25% 42%
Fig 2. Kaplan-Meier curve log analysis comparing primary patency in hepatic artery angioplasty (n ¼ 10) and stenting (n ¼ 13) (P ¼ .14). PTA, Percutaneous transluminal angioplasty.
were 12 reinterventions in 10 patients for recurrent HAS. Sixty percent (n ¼ 6/10) of patients undergoing initial PTA required reintervention compared with 31% (n ¼ 4/13) of patients undergoing initial stent placement (P ¼ .17). Reintervention. Time to initial reintervention was 62.5 6 44 days (range, 7-182) in those first treated with PTA compared with 236 6 124 days (range, 80-608) in those initially treated with a stent (P ¼ .016). In the initial PTA group, reinterventions were stent placements in all patients (n ¼ 6). In the initial stent group, reinterventions were performed with PTA alone (n ¼ 2) or with a second stent (n ¼ 3). Although restenosis after stent placement was most commonly observed within the stent, the culprit lesion was occasionally adjacent to the stent. Patency. Primary patency rates at 1, 3, and 6 months by Kaplen-Meier life-table analysis were 92%, 85%, and 69%, respectively in the patients initially treated with a stent, compared with 70%, 60%, and 50% in those treated with PTA (P ¼ .17) (Fig 2; Table II). Primary patency rates for
JOURNAL OF VASCULAR SURGERY April 2013
1070 Hamby et al
Table II. Primary patency rates of angioplasty vs angioplasty þ stent and overall patency rates at 1-, 3-, and 6-month follow-ups Primary patency ratea Follow-up
PTA (n ¼ 10)
PTA þ stent (n ¼ 13)
Overall primary patency rate (n ¼ 23)
1 month 3 months 6 months
70% (n ¼ 7) 60% (n ¼ 6) 50% (n ¼ 5)
92% (n ¼ 12) 85% (n ¼ 11) 69% (n ¼ 9)
83% (n ¼ 19) 74% (n ¼ 17) 61% (n ¼ 14)
PTA, Percutaneous transluminal angioplasty. a P ¼ .14.
Fig 4. Kaplan-Meier curve log analysis comparing primary patency of hepatic arteries with kinks (n ¼ 14) with that of hepatic arteries with no kinks (n ¼ 9) (P ¼ .5). HAS, Hepatic artery stenosis.
Fig 3. Kaplan-Meier curve log analysis of assisted primary patency (n ¼ 12) in patients with reinterventions; two patients had three reinterventions.
the entire cohort were 83%, 74%, and 61% at 1, 3, and 6 months, respectively. Overall primary-assisted patency was 97% at 6 and 12 months (Fig 3). Mean follow-up of the entire patient cohort was 8.2 6 1.8 months (range, 0-29). Sixty-one percent (14/23) of patients had significant tortuosity or kink in the treated hepatic artery. Fig 4 illustrates the primary patency of patients with (n ¼ 14; Fig 5) and without (n ¼ 9) tortuosity or kinks. There was not a significance difference between the groups at 1 and 3 months, but a trend toward inferior patency rates at 6 months in the group with significant tortuosity or kinking was noted. Complications. There were two major complications in this treatment cohort (5.7%, 2/35) and no periprocedural mortality. One hepatic artery rupture occurred in a patient being retreated for HAS after initial hepatic PTA 6 months earlier. This contained rupture was treated with placement of a bare-metal self-expanding stent and prolonged balloon tamponade. The patient required a blood transfusion but did not need open surgical repair. A follow-up CTA 4 days after the rupture revealed no hepatic artery pseudoaneurysm, and serial ultrasound imaging showed resolution of her high-grade HAS. Another patient
had a dissection of the hepatic artery that precipitated HAT. This patient went on to progressive graft failure and underwent a second successful liver transplant 1 month later. This was the sole patient in the series who went on to HAT, giving a rate of 4.3% (1/23) of treated patients. Mortality. Thirty-day mortality was zero. At a mean follow-up of 8.2 months, there was one death, yielding a mortality rate of 4.3% (1/23). This patient had an anoxic brain injury caused by a respiratory event 38 days after hepatic artery intervention. In this case, the hepatic artery was patent without significant stenosis at the time of death. DISCUSSION HAS, with an incidence ranging from 4.8% to 13%, has been recognized as a risk factor for subsequent HAT. HAT can double the rate of biliary complications and potentially lead to graft dysfunction.4-6,17 Appearance of a stenosis at or near the anastomosis usually occurs secondary to operative technique or vascular clamp injury. Other causes include allograft rejection and microvascular injury associated with cold preservation, which are more commonly found in intraparenchymal vessels.18-20 HAS and HAT are diagnosed both clinically and with Doppler ultrasound imaging. Patients presenting with elevated liver transaminases, biliary leak, cholangitis, or signs of rejection are often referred for evaluation by Doppler ultrasound imaging. Routine post-transplant Doppler evaluations may also identify asymptomatic HAS. Vit et al21 found the presence of tardus parvus waveform to be the most accurate individual indicator of HAS or HAT, with a sensitivity of 91% and specificity of 99.1%. Our series achieved an overall initial technical success rate of 97% (34/35), which compared favorably with
JOURNAL OF VASCULAR SURGERY Volume 57, Number 4
Hamby et al 1071
Fig 5. Hepatic angiograms of a tortuous (kink) hepatic artery before (left) and after (right) angioplasty.
reported technical success rates in other series ranging from 81% to 100%.5,7,9,10,14,16 The major complication rate of 5.7% (2/35 interventions) in this series was comparable to those in prior reports. Sabri et al1 reported an 8% (2/ 25) need for surgical intervention after treatment for HAS, whereas Saad et al8 reported a 4.8% (2/42) major complication rate in patients treated with PTA. We showed a strong trend toward improved primary patency with initial stent placement when compared with PTA. Because of the relatively small number of patients, these differences did not reach statistical significance and likely represent a type II statistical error. Early in our experience, we frequently observed rapid restenosis of the hepatic artery after treatment with PTA alone. These cases typically had no residual stenosis at the completion of the intervention, and the initial ultrasound images confirmed major improvements in lowered PSV and improved RIs. More patients undergoing PTA alone required retreatment compared with those receiving a primary stent (60% vs 31%), and the time to reintervention was much shorter in PTA patients (62.5 days vs 236 days). On the basis of these initial results, we have evolved to preferential use of primary stenting in most instances. Our findings are concordant with some prior reports, suggesting a benefit to stenting over angioplasty alone for HAS.4,5,10 However, the literature is not uniform in this area. In a series of 26 patients undergoing stent placement for HAS, Ueno et al reported that 45% of patients required reintervention for restenosis.6 A recent study by Maruzzelli et al comparing initial PTA with stent placement found no significant differences in recurrent stenosis or thrombosis and mortality.11 Saad et al opined that PTA provided better patency rates than stent placement in properly selected patients, although
they did not use any stents in their series of 42 treated patients.8 The overall primary patency rates in this series were 83%, 74%, and 61% at 1, 3, and 6 months, respectively. Prior studies have demonstrated variable primary patency rates. In a series of 20 patients, Chen et al reported higher primary patency rates at 3, 6, and 12 months of 94%, 87%, and 79%, respectively.14 However, there was a 30% need for repeat transplantation at a median follow-up of 14.4 months. Sabri et al reported a primary patency of 52% at 4 months in 25 patients treated for HAS.1 Saad et al reported a 1-year primary patency rate after angioplasty alone (n ¼ 42) of 44%, which improved to 65% for lesions not associated with hepatic arterial kinks.8 Significant tortuosity and/or kinks in the hepatic artery were seen in the majority of our patient cohort (Fig 5). These cases were considerably more technically challenging as detailed in under Methods. In the series from Saad et al, PTA resulted in a 14% technical success rate in patients with a kink, compared with 94% in those without a kink. In his series, 5 of 57 patients were treated with primary open surgical revision without attempt at endovascular treatment.8 Our results in patients with significant tortuosity/kinks were not significantly worse at 1 and 3 months, although there was a trend toward inferior primary patency at 6 months (Fig 6). Although speculative, the better performance in the present series may be because of stent utilization; Saad et al8 treated exclusively with PTA. The 4.3% (1/23) risk of HAT in our patient cohort compared favorably with the literature. Saad et al8 reported a 19% rate of HAT at 6 months after PTA in 42 patients. Aggressive postintervention duplex ultrasound surveillance and timely reintervention may have contributed to our low
1072 Hamby et al
JOURNAL OF VASCULAR SURGERY April 2013
Fig 6. Hepatic artery angiogram before (left) and after (right) angioplasty and stent placement.
rate. Both these series suggest that endovascular treatment of significant HAS can improve the natural history. Up to 65% of patients with untreated significant HAS will go on to HAT within 6 months.8 CONCLUSIONS Endovascular treatment of HAS can be performed with high technical success and low morbidity. Initial use of a stent may improve primary patency when compared with PTA. Even with stent placement, the need for reintervention is common, placing particular importance on aggressive surveillance. Timely reintervention can provide excellent primary-assisted patency. This aggressive policy of endovascular treatment of HAS appears to improve on the natural history of HAS, when compared with historical controls. Longer follow-up and a larger cohort are needed to confirm these encouraging initial findings. AUTHOR CONTRIBUTIONS Conception and design: WS, BH, DR Analysis and interpretation: WS, BH, DR, EB Data collection: BH, DR Writing the article: WS, DR, BH Critical revision of the article: WS, TS, HB, GL, EB Final approval of the article: WS Statistical analysis: BH, DR Obtained funding: WS Overall responsibility: WS BH and DR contributed equally to this work. REFERENCES 1. Sabri SS, Saad WEA, Schmitt TM, Turba UC, Kumer SC, Park AW, et al. Endovascular therapy for hepatic artery stenosis following liver transplantation. Vasc Endovasc Surg 2011;45:447-52. 2. Stagne BJ, Glanemann M, Nuessler NC, Settmacher U, Steinmuller T, Neuhaus P. Hepatic artery thrombosis after adult liver transplantation. Liver Transpl 2003;9:612-20. 3. Busuttil R, Colonna JO, Hiatt JR, Brems JJ, el Khoury G, Goldstein LI, et al. The first 100 liver transplants at UCLA. Ann Surg 1987;206:387-99. 4. Orons PD, Sheng R, Zajko AB. Hepatic artery stenosis in liver transplant recipients: prevalence and cholangiographic appearance of associated biliary complications. AJR Am J Roentgenol 1995;165:1145-9. 5. Cotroneo AR, Di Stasi C, Cina A, De Gaetano AM, Evangelisti R, Paloni F, et al. Stent placement in four patients with hepatic artery stenosis or thrombosis after liver transplantation. J Vasc Interv Radiol 2002;13:619-23.
6. Ueno T, Jones G, Martin A, Ikegami T, Sanchez EQ, Chinnakotla S, et al. Clinical outcomes from hepatic artery stenting in liver transplantation. Liver Transpl 2006;12:422-7. 7. Huang M, Shan H, Jiang Z, Li Z, Zhu K, Guan S, et al. The use of coronary stent in hepatic artery stenosis after orthotopic liver transplantation. Eur J Radiol 2006;10:1016. 8. Saad WEA, Davies MG, Sahler L, Lee DE, Patel NC, Kitanosono T, et al. Hepatic artery stenosis in liver transplant recipients: primary treatment with percutaneous transluminal angioplasty. J Vasc Interv Radiol 2005;16:795-805. 9. Boyvat F, Aytekin C, Karakayali H, Haberal M. Interventional radiology in liver transplant. Exp Clin Transplant 2008;6:105-12. 10. Denys AL, Qanadli SD, Durand F, Vilgrain V, Farges O, Belghiti J, et al. Feasibilty and effectiveness of using coronary stents in the treatment of hepatic artery stenoses after orthotopic liver transplant: preliminary report. AJR Am J Roentgenol 2002;178:1175-9. 11. Maruzzelli L, Miraglia R, Caruso S, Milazzo M, Mamone G, Gruttadauria S, et al. Percutaneous endovascular treatment of hepatic artery stenosis in adult and pediatric patients after liver transplantation. Cardiovasc Interv Radiol 2010;33:1111-9. 12. Zhu KD, Meng XC, Huang MS, Qian JS, Guan SH, Jiang ZB, et al. The role of early hepatic artery ischemia on biliary complications after liver transplantation and hepatic arterial interventional therapy. Zhonghua Yi Xue Za Zhi 2009;89:2195-8. 13. Singhal A, Stokes K, Sebastian A, Wright HI, Kohli V. Endovascular treatment of hepatic artery thrombosis following liver transplantation. Transpl Int 2010;23:245-56. 14. Chen GH, Wang GY, Yang Y, Li H, Lu MQ, Cai CJ, et al. Single-center experience of the therapuetic management of hepatic artery stenosis after orthotopic liver transplantation. Eur Surg Res 2009;42:21-7. 15. Li ZW, Wang MQ, Zhou NX, Liu Z, Huang ZQ. Interventional treatment of acute hepatic artery occlusion after liver transplantation. Hepatobiliary Pancreat Dis Int 2007;6:474-8. 16. Jeon GS, Won JH, Wang HJ, Kim BW, Lee BM. Endovascular treatment of acute arterial complications after living donor liver transplantation. Clin Radiol 2008;63:1099-105. 17. Abbasoglu O, Levy MF, Brkic BB, Testa G, Jeyarajah DR, Goldstein RM, et al. Ten years of liver transplantation: an evolving understanding of late graft loss. Transplantation 1997;64:1801-7. 18. Samuel D, Gillet D, Castaing D, Reynes M, Bismuth H. Portal and arterial thrombosis in liver transplantation: a frequent event in severe rejection. Transplant Proc 1989;21(1 pt 2):2225-7. 19. Payen DM, Fratacci MD, Dupuy P, Gatecel C, Vigouroux C, Ozier Y, et al. Portal and hepatic arterial blood flow measurements of human transplanted liver by implanted Doppler probes: interest for early complications and nutrition. Surgery 1990;107:417-27. 20. Mor E, Scwartz ME, Sheiner PA, Menesses P, Hytiroglou P, Emre S, et al. Prolonged preservation in University of Wisconsin solution associated with hepatic artery thrombosis after orthotopic liver transplantation. Transplantation 1993;56:1399-402. 21. Vit A, De Candia A, Como G, Del Frate C, Marzio A, Bazzocchi M. Doppler evaluation of arterial complications of adult orthotopic liver transplantation. J Clin Ultrasound 2003;31:339-45. Submitted Jun 10, 2012; accepted Oct 12, 2012.
Endovascular treatment of hepatic artery stenosis after liver transplantation Blake A. Hamby, MD,a Daniel E. Ramirez, MD,a George E. Loss, MD, PhD,b Hernan A. Bazan, MD,a Taylor A. Smith, MD,a Edward Bluth, MD,c and W. Charles Sternbergh III, MD,a New Orleans, La Background: Hepatic artery stenosis (HAS) after orthotopic liver transplantation is a significant risk factor for subsequent hepatic artery thrombosis (HAT). HAT is associated with a 30%-50% risk of liver failure culminating in retransplantation or death. Traditional treatment of hepatic artery complications has been surgical, with hepatic artery revision or retransplantation. Endovascular therapy of HAS, described primarily in the interventional radiology literature, may provide a less-invasive treatment option. Methods: This was a retrospective review of all endovascular interventions performed for HAS after orthotopic liver transplantation over a 31-month period (August 2009 to January 2012). Patients with duplex ultrasound imaging evidence of severe main HAS (peak systolic velocity of >400 cm/s, resistive index of <.5) underwent endovascular treatment with either primary stent placement or percutaneous transluminal angioplasty (PTA) alone. Patients were followed with serial ultrasound imaging to assess for treatment success and late restenosis. Reintervention was performed if significant restenosis occurred. Results: Thirty-five hepatic artery interventions were performed in 23 patients. Over the 31-month study period, 318 orthotopic liver transplantations were performed, yielding a 7.4% (23/318) rate of hepatic artery intervention. Primary technical success was achieved in 97% (34/35) of cases. Initial treatment was with PTA alone (n [ 10) or primary stent placement (n [ 13). The initial postintervention ultrasound images revealed improvements in hepatic artery peak systolic velocity (267 ± 118 [posttreatment] vs 489.9 ± 155 cm/s [pretreatment]; P < .0001) and main hepatic artery resistive index (0.61 ± 0.08 [posttreatment] vs 0.41 ± 0.07 [pretreatment]; P < .0001). At a mean follow-up of 8.2 ± 1.8 months (range, 0-29), there were 12 reinterventions in 10 patients for recurrent HAS. Thirty-one percent (n [ 4/13) of patients undergoing initial stent placement required reintervention (at 236 ± 124 days of follow-up) compared with 60% (n [ 6/10) of patients undergoing initial PTA (at 62.5 ± 44 days of follow-up). Primary patency rates (Kaplan-Meier) after primary stent placement were 92%, 85%, and 69% at 1, 3, and 6 months, respectively, compared with 70%, 60%, and 50% after PTA (P [ .17). Primary-assisted patency for the entire cohort was 97% at 6 and 12 months. Major complications were one arterial rupture managed endovascularly and one artery dissection that precipitated HAT and required retransplantation. The overall rate of HAT in the entire cohort was 4.3% (1/23). Conclusions: Endovascular treatment of HAS can be performed with high technical success, excellent primary-assisted patency, and acceptable morbidity. Initial use of a stent may improve primary patency when compared with PTA. The need for reintervention is common, placing particular importance on aggressive surveillance. Longer follow-up and a larger cohort are needed to confirm these encouraging early results. (J Vasc Surg 2013;57:1067-72.)
Hepatic artery thrombosis (HAT) is the most common vascular complication after orthotopic liver transplantation, with an incidence estimated to be between 4% and 11% in adult transplants and as high as 11%-26% in the pediatric population.1-3 Hepatic artery stenosis (HAS) is seen in 5%-11% of transplants and, if left untreated, has up to a 65% chance of developing into HAT within 6 months.4 From the Sections of Vascular and Endovascular Surgery,a Transplant Surgery,b and Diagnostic Radiology,c Ochsner Clinic Foundation. Author conflict of interest: none. Presented at the Fortieth Annual Symposium of the Society of Clinical Vascular Surgery, Las Vegas, Nev, March 14-17, 2012. Reprint requests: W. Charles Sternbergh III, MD, Sections of Vascular and Endovascular Surgery, Ochsner Clinic Foundation, 1514 Jefferson Highway, New Orleans, LA 70121 (e-mail: [email protected]). The editors and reviewers of this article have no relevant financial relationships to disclose per the JVS policy that requires reviewers to decline review of any manuscript for which they may have a conflict of interest. 0741-5214/$36.00 Copyright Ó 2013 by the Society for Vascular Surgery. http://dx.doi.org/10.1016/j.jvs.2012.10.086
Patients with HAT have a 30%-50% incidence of eventual liver failure, leading to retransplantation or death. Hence, maintaining hepatic artery patency is likely an important variable in minimizing graft loss and mortality. Treatment of both HAS and HAT has traditionally involved either open surgical revision of the hepatic artery or liver retransplantation. There are several reports of alternative approaches to this dilemma, including vein graft interposition, endovascular thrombolysis, and mechanical thrombectomy.5-16 Endovascular techniques have also been used as an alternative to open surgical repair with less morbidity and mortality and with similar rates of overall patient survival at 5 years (78% vs 80%, respectively).4,9-11,14,16 Additionally, percutaneous transluminal angioplasty (PTA) with or without stenting has been found to prolong graft survival and delay the need for open surgical revascularization or retransplantation.16 Recently, enhanced collaboration between transplant and vascular surgeons in our institution has resulted in a more aggressive policy of endovascular treatment of 1067
1068 Hamby et al
JOURNAL OF VASCULAR SURGERY April 2013
Fig 1. Hepatic angiogram with an allograft coming from the main abdominal aorta before (left) and after (right) angioplasty.
HAS after orthotopic liver transplantation. This report describes the initial outcomes of this experience in treating HAS, which began in mid-2009. METHODS This is a retrospective review of all endovascular interventions performed for HAS in a single tertiary care institution between August 1, 2009 and January 31, 2012. A single vascular surgeon (W.C.S.) performed all interventions. Institutional review board approval was obtained for this study. Definitions. Significant HAS considered for intervention was defined by a main hepatic artery resistive index (RI) <.5 and a peak systolic velocity (PSV) >400 cm/s as detected by duplex ultrasound imaging. Tardus parvus waveforms were usually observed in these highly stenotic hepatic arteries. In instances where the PSV and/or RI were on the borderline of meeting these criteria, the presence of tardus parvus waveforms generally led to intervention. Significant tortuosity or kinking of the hepatic artery was defined as a >90 acute bend in the vessel, as measured on computed tomography angiogram (CTA) or digital subtraction angiogram. Initial technical success was defined as <30% residual stenosis of the treated hepatic artery by visual estimation of the final arteriogram. Neither intravascular ultrasound imaging nor pressure gradients were used in this series. Diagnostic evaluation. Most patients with significant HAS identified by ultrasound imaging subsequently underwent a CTA to provide additional anatomic information. This study was found to be valuable in case planning. In 22% of our cases, the hepatic artery had an aberrant recipient takeoff, either as a replaced right hepatic arising from the superior mesenteric artery (n ¼ 2/23) or as an accessory left hepatic artery with a direct takeoff from the aorta (n ¼ 1/23) or from an infrarenal allograft (2/23; Fig 1). A steep caudal orientation of the celiac axis was occasionally seen, prompting a change from femoral to brachial access. Finally, the CTA provided excellent information regarding the relative tortuosity of the hepatic artery.
The decision to intervene was based solely on the imaging studies. Liver transaminase levels were frequently normal in patients with significant HAS and were therefore not helpful in patient selection or surveillance. Technique. Femoral access was favored unless there was significant caudal orientation of the celiac artery on the CTA; in these cases, left brachial access was used. Most femoral approach cases were performed using a 6F renal double-curve (RDC) guiding catheter (Boston Scientific, Natick, Mass) to gain access to the hepatic artery. A 4F, 5F, or 6F multipurpose sheath (Cook, Bloomingtom, Ill) was generally used with brachial access. A 0.014-inch wire was used to cross the lesion in most cases. Depending on the degree of tortuosity and need for trackability, a variety of wires were used, from fairly flexible wires (Balanced Medium Weight; Abbott, Abbott Park, Ill), to increasingly stiff hydrophilic wires (Choice PT, Platinum Plus; Boston Scientific), to extrastiff wires (Balanced Heavy Weight; Abbott). The stiffer wires sometimes required placement through a quick-cross catheter (Spectranetics, Colorado Springs, Colo), placed previously over a softer wire. Occasionally, a second “buddy wire” was required to further straighten a tortuous vessel to allow delivery of a stent. Low-profile coronary angioplasty balloons were preferentially used, as their trackability was much better in these tortuous vessels. Angioplasty balloons (2.5-5.0 mm in diameter) were sized to the nominal vessel reference size and inflated for a minimum of 60 seconds. Low-profile self-expanding stents (X-pert; Abbott, 4-6 mm in diameter) or coronary balloon-expandable stents (2.5-5.0 mm in diameter) were used. Drug-eluting coronary stents were selectively placed for recurrent stenosis in two patients. Self-expanding stents were preferentially chosen when there was a major size mismatch of the vessel proximal and distal to the target legion. Early in our experience, we also attempted to use self-expanding stents in lesions located in highly angulated/tortuous hepatic vessels. It was thought that these would conform better and be less apt to create a new “kink” distal to the stent, as a more rigid balloon-
JOURNAL OF VASCULAR SURGERY Volume 57, Number 4
expandable stent might. However, the trackability of selfexpanding stents in highly tortuous vessels was found to be poor, even with stiff wires and the use of a “buddy” wire. The coronary profile balloon-expandable stents were found to be much more trackable in these anatomic situations; their utilization increased as we gained more experience. Lesions were routinely predilated with PTA prior to placement of a stent. Early in our experience, PTA alone was used if a <30% residual stenosis of the hepatic artery was achieved. Provisional stenting was performed if there was a greater than 30% residual narrowing. Despite confirmation of significant improvement in HAS by ultrasound imaging after initial PTA, we frequently observed rapid restenosis. On the basis of these early results, we shifted to primary stenting when technically possible. Postintervention protocol. Patients underwent ultrasound imaging at 1-7 days postintervention, then typically at 6 weeks, 3, 6, and 12 months and yearly thereafter. Patients with evidence of moderate recurrent stenosis were followed more frequently. Treated patients typically received dual-antiplatelet therapy (acetylsalicylic acid [ASA] 81 mg and clopidogrel [Plavix] 75 mg) for 3-6 months postintervention. Clopidogrel was usually stopped after 6 months if there was no evidence of recurrent stenosis. Statistics. Pre- and postintervention mean hepatic artery PSV and RI were compared using a two-tailed ttest. The Fischer exact test was used to compare reintervention rates. Primary patency with angioplasty alone compared with angioplasty and stent, freedom from reintervention, and assisted primary patency were analyzed using Kaplan-Meier curve log analysis. Data are presented as means 6 standard errors. RESULTS Thirty-five endovascular interventions for HAS were performed in 23 patients during the 31-month study period. Of note, 13% (3/23) of treated patients underwent a second orthotopic liver transplantation; all had lost their first liver transplant secondary to HAT. From August 2009 through January 2012, 318 liver transplants were performed, with 7.2% (23/318) of patients undergoing endovascular treatment for significant HAS. No open repairs of HAS were performed during the study period. The mean time to initial intervention after transplantation was 92.5 6 62 days (range, 12-236). Patients had a mean age of 49.4 6 9 years (range, 15-63), and 65% (15/23) were male. Ultrasound findings. Preintervention ultrasound imaging revealed a mean main hepatic artery PSV of 490 6 34.9 cm/s and an RI of 0.41 6 0.07. Initial postintervention PSV significantly decreased to 267 6 49 cm/s, and the RI increased to 0.61 6 0.08 (P < .0001). These salutary changes were sustained at 1, 3, and 6 months (Table I). Technical success. Primary technical success was achieved in 97% (34/35) of cases. Initial treatment was PTA (n ¼ 10) or primary stent placement (n ¼ 13). There
Hamby et al 1069
Table I. Ultrasound findings
Preintervention Postintervention 1-month follow-up 3-month follow-up 6-month follow-up
n
Peak systolic velocity(cm/s)
Resistive index
Tardus parvus
35 34 29 27 17
490 257 234 230 243
0.41 0.61 0.6 0.62 0.68
83% 13% 16% 25% 42%
Fig 2. Kaplan-Meier curve log analysis comparing primary patency in hepatic artery angioplasty (n ¼ 10) and stenting (n ¼ 13) (P ¼ .14). PTA, Percutaneous transluminal angioplasty.
were 12 reinterventions in 10 patients for recurrent HAS. Sixty percent (n ¼ 6/10) of patients undergoing initial PTA required reintervention compared with 31% (n ¼ 4/13) of patients undergoing initial stent placement (P ¼ .17). Reintervention. Time to initial reintervention was 62.5 6 44 days (range, 7-182) in those first treated with PTA compared with 236 6 124 days (range, 80-608) in those initially treated with a stent (P ¼ .016). In the initial PTA group, reinterventions were stent placements in all patients (n ¼ 6). In the initial stent group, reinterventions were performed with PTA alone (n ¼ 2) or with a second stent (n ¼ 3). Although restenosis after stent placement was most commonly observed within the stent, the culprit lesion was occasionally adjacent to the stent. Patency. Primary patency rates at 1, 3, and 6 months by Kaplen-Meier life-table analysis were 92%, 85%, and 69%, respectively in the patients initially treated with a stent, compared with 70%, 60%, and 50% in those treated with PTA (P ¼ .17) (Fig 2; Table II). Primary patency rates for
JOURNAL OF VASCULAR SURGERY April 2013
1070 Hamby et al
Table II. Primary patency rates of angioplasty vs angioplasty þ stent and overall patency rates at 1-, 3-, and 6-month follow-ups Primary patency ratea Follow-up
PTA (n ¼ 10)
PTA þ stent (n ¼ 13)
Overall primary patency rate (n ¼ 23)
1 month 3 months 6 months
70% (n ¼ 7) 60% (n ¼ 6) 50% (n ¼ 5)
92% (n ¼ 12) 85% (n ¼ 11) 69% (n ¼ 9)
83% (n ¼ 19) 74% (n ¼ 17) 61% (n ¼ 14)
PTA, Percutaneous transluminal angioplasty. a P ¼ .14.
Fig 4. Kaplan-Meier curve log analysis comparing primary patency of hepatic arteries with kinks (n ¼ 14) with that of hepatic arteries with no kinks (n ¼ 9) (P ¼ .5). HAS, Hepatic artery stenosis.
Fig 3. Kaplan-Meier curve log analysis of assisted primary patency (n ¼ 12) in patients with reinterventions; two patients had three reinterventions.
the entire cohort were 83%, 74%, and 61% at 1, 3, and 6 months, respectively. Overall primary-assisted patency was 97% at 6 and 12 months (Fig 3). Mean follow-up of the entire patient cohort was 8.2 6 1.8 months (range, 0-29). Sixty-one percent (14/23) of patients had significant tortuosity or kink in the treated hepatic artery. Fig 4 illustrates the primary patency of patients with (n ¼ 14; Fig 5) and without (n ¼ 9) tortuosity or kinks. There was not a significance difference between the groups at 1 and 3 months, but a trend toward inferior patency rates at 6 months in the group with significant tortuosity or kinking was noted. Complications. There were two major complications in this treatment cohort (5.7%, 2/35) and no periprocedural mortality. One hepatic artery rupture occurred in a patient being retreated for HAS after initial hepatic PTA 6 months earlier. This contained rupture was treated with placement of a bare-metal self-expanding stent and prolonged balloon tamponade. The patient required a blood transfusion but did not need open surgical repair. A follow-up CTA 4 days after the rupture revealed no hepatic artery pseudoaneurysm, and serial ultrasound imaging showed resolution of her high-grade HAS. Another patient
had a dissection of the hepatic artery that precipitated HAT. This patient went on to progressive graft failure and underwent a second successful liver transplant 1 month later. This was the sole patient in the series who went on to HAT, giving a rate of 4.3% (1/23) of treated patients. Mortality. Thirty-day mortality was zero. At a mean follow-up of 8.2 months, there was one death, yielding a mortality rate of 4.3% (1/23). This patient had an anoxic brain injury caused by a respiratory event 38 days after hepatic artery intervention. In this case, the hepatic artery was patent without significant stenosis at the time of death. DISCUSSION HAS, with an incidence ranging from 4.8% to 13%, has been recognized as a risk factor for subsequent HAT. HAT can double the rate of biliary complications and potentially lead to graft dysfunction.4-6,17 Appearance of a stenosis at or near the anastomosis usually occurs secondary to operative technique or vascular clamp injury. Other causes include allograft rejection and microvascular injury associated with cold preservation, which are more commonly found in intraparenchymal vessels.18-20 HAS and HAT are diagnosed both clinically and with Doppler ultrasound imaging. Patients presenting with elevated liver transaminases, biliary leak, cholangitis, or signs of rejection are often referred for evaluation by Doppler ultrasound imaging. Routine post-transplant Doppler evaluations may also identify asymptomatic HAS. Vit et al21 found the presence of tardus parvus waveform to be the most accurate individual indicator of HAS or HAT, with a sensitivity of 91% and specificity of 99.1%. Our series achieved an overall initial technical success rate of 97% (34/35), which compared favorably with
JOURNAL OF VASCULAR SURGERY Volume 57, Number 4
Hamby et al 1071
Fig 5. Hepatic angiograms of a tortuous (kink) hepatic artery before (left) and after (right) angioplasty.
reported technical success rates in other series ranging from 81% to 100%.5,7,9,10,14,16 The major complication rate of 5.7% (2/35 interventions) in this series was comparable to those in prior reports. Sabri et al1 reported an 8% (2/ 25) need for surgical intervention after treatment for HAS, whereas Saad et al8 reported a 4.8% (2/42) major complication rate in patients treated with PTA. We showed a strong trend toward improved primary patency with initial stent placement when compared with PTA. Because of the relatively small number of patients, these differences did not reach statistical significance and likely represent a type II statistical error. Early in our experience, we frequently observed rapid restenosis of the hepatic artery after treatment with PTA alone. These cases typically had no residual stenosis at the completion of the intervention, and the initial ultrasound images confirmed major improvements in lowered PSV and improved RIs. More patients undergoing PTA alone required retreatment compared with those receiving a primary stent (60% vs 31%), and the time to reintervention was much shorter in PTA patients (62.5 days vs 236 days). On the basis of these initial results, we have evolved to preferential use of primary stenting in most instances. Our findings are concordant with some prior reports, suggesting a benefit to stenting over angioplasty alone for HAS.4,5,10 However, the literature is not uniform in this area. In a series of 26 patients undergoing stent placement for HAS, Ueno et al reported that 45% of patients required reintervention for restenosis.6 A recent study by Maruzzelli et al comparing initial PTA with stent placement found no significant differences in recurrent stenosis or thrombosis and mortality.11 Saad et al opined that PTA provided better patency rates than stent placement in properly selected patients, although
they did not use any stents in their series of 42 treated patients.8 The overall primary patency rates in this series were 83%, 74%, and 61% at 1, 3, and 6 months, respectively. Prior studies have demonstrated variable primary patency rates. In a series of 20 patients, Chen et al reported higher primary patency rates at 3, 6, and 12 months of 94%, 87%, and 79%, respectively.14 However, there was a 30% need for repeat transplantation at a median follow-up of 14.4 months. Sabri et al reported a primary patency of 52% at 4 months in 25 patients treated for HAS.1 Saad et al reported a 1-year primary patency rate after angioplasty alone (n ¼ 42) of 44%, which improved to 65% for lesions not associated with hepatic arterial kinks.8 Significant tortuosity and/or kinks in the hepatic artery were seen in the majority of our patient cohort (Fig 5). These cases were considerably more technically challenging as detailed in under Methods. In the series from Saad et al, PTA resulted in a 14% technical success rate in patients with a kink, compared with 94% in those without a kink. In his series, 5 of 57 patients were treated with primary open surgical revision without attempt at endovascular treatment.8 Our results in patients with significant tortuosity/kinks were not significantly worse at 1 and 3 months, although there was a trend toward inferior primary patency at 6 months (Fig 6). Although speculative, the better performance in the present series may be because of stent utilization; Saad et al8 treated exclusively with PTA. The 4.3% (1/23) risk of HAT in our patient cohort compared favorably with the literature. Saad et al8 reported a 19% rate of HAT at 6 months after PTA in 42 patients. Aggressive postintervention duplex ultrasound surveillance and timely reintervention may have contributed to our low
1072 Hamby et al
JOURNAL OF VASCULAR SURGERY April 2013
Fig 6. Hepatic artery angiogram before (left) and after (right) angioplasty and stent placement.
rate. Both these series suggest that endovascular treatment of significant HAS can improve the natural history. Up to 65% of patients with untreated significant HAS will go on to HAT within 6 months.8 CONCLUSIONS Endovascular treatment of HAS can be performed with high technical success and low morbidity. Initial use of a stent may improve primary patency when compared with PTA. Even with stent placement, the need for reintervention is common, placing particular importance on aggressive surveillance. Timely reintervention can provide excellent primary-assisted patency. This aggressive policy of endovascular treatment of HAS appears to improve on the natural history of HAS, when compared with historical controls. Longer follow-up and a larger cohort are needed to confirm these encouraging initial findings. AUTHOR CONTRIBUTIONS Conception and design: WS, BH, DR Analysis and interpretation: WS, BH, DR, EB Data collection: BH, DR Writing the article: WS, DR, BH Critical revision of the article: WS, TS, HB, GL, EB Final approval of the article: WS Statistical analysis: BH, DR Obtained funding: WS Overall responsibility: WS BH and DR contributed equally to this work. REFERENCES 1. Sabri SS, Saad WEA, Schmitt TM, Turba UC, Kumer SC, Park AW, et al. Endovascular therapy for hepatic artery stenosis following liver transplantation. Vasc Endovasc Surg 2011;45:447-52. 2. Stagne BJ, Glanemann M, Nuessler NC, Settmacher U, Steinmuller T, Neuhaus P. Hepatic artery thrombosis after adult liver transplantation. Liver Transpl 2003;9:612-20. 3. Busuttil R, Colonna JO, Hiatt JR, Brems JJ, el Khoury G, Goldstein LI, et al. The first 100 liver transplants at UCLA. Ann Surg 1987;206:387-99. 4. Orons PD, Sheng R, Zajko AB. Hepatic artery stenosis in liver transplant recipients: prevalence and cholangiographic appearance of associated biliary complications. AJR Am J Roentgenol 1995;165:1145-9. 5. Cotroneo AR, Di Stasi C, Cina A, De Gaetano AM, Evangelisti R, Paloni F, et al. Stent placement in four patients with hepatic artery stenosis or thrombosis after liver transplantation. J Vasc Interv Radiol 2002;13:619-23.
6. Ueno T, Jones G, Martin A, Ikegami T, Sanchez EQ, Chinnakotla S, et al. Clinical outcomes from hepatic artery stenting in liver transplantation. Liver Transpl 2006;12:422-7. 7. Huang M, Shan H, Jiang Z, Li Z, Zhu K, Guan S, et al. The use of coronary stent in hepatic artery stenosis after orthotopic liver transplantation. Eur J Radiol 2006;10:1016. 8. Saad WEA, Davies MG, Sahler L, Lee DE, Patel NC, Kitanosono T, et al. Hepatic artery stenosis in liver transplant recipients: primary treatment with percutaneous transluminal angioplasty. J Vasc Interv Radiol 2005;16:795-805. 9. Boyvat F, Aytekin C, Karakayali H, Haberal M. Interventional radiology in liver transplant. Exp Clin Transplant 2008;6:105-12. 10. Denys AL, Qanadli SD, Durand F, Vilgrain V, Farges O, Belghiti J, et al. Feasibilty and effectiveness of using coronary stents in the treatment of hepatic artery stenoses after orthotopic liver transplant: preliminary report. AJR Am J Roentgenol 2002;178:1175-9. 11. Maruzzelli L, Miraglia R, Caruso S, Milazzo M, Mamone G, Gruttadauria S, et al. Percutaneous endovascular treatment of hepatic artery stenosis in adult and pediatric patients after liver transplantation. Cardiovasc Interv Radiol 2010;33:1111-9. 12. Zhu KD, Meng XC, Huang MS, Qian JS, Guan SH, Jiang ZB, et al. The role of early hepatic artery ischemia on biliary complications after liver transplantation and hepatic arterial interventional therapy. Zhonghua Yi Xue Za Zhi 2009;89:2195-8. 13. Singhal A, Stokes K, Sebastian A, Wright HI, Kohli V. Endovascular treatment of hepatic artery thrombosis following liver transplantation. Transpl Int 2010;23:245-56. 14. Chen GH, Wang GY, Yang Y, Li H, Lu MQ, Cai CJ, et al. Single-center experience of the therapuetic management of hepatic artery stenosis after orthotopic liver transplantation. Eur Surg Res 2009;42:21-7. 15. Li ZW, Wang MQ, Zhou NX, Liu Z, Huang ZQ. Interventional treatment of acute hepatic artery occlusion after liver transplantation. Hepatobiliary Pancreat Dis Int 2007;6:474-8. 16. Jeon GS, Won JH, Wang HJ, Kim BW, Lee BM. Endovascular treatment of acute arterial complications after living donor liver transplantation. Clin Radiol 2008;63:1099-105. 17. Abbasoglu O, Levy MF, Brkic BB, Testa G, Jeyarajah DR, Goldstein RM, et al. Ten years of liver transplantation: an evolving understanding of late graft loss. Transplantation 1997;64:1801-7. 18. Samuel D, Gillet D, Castaing D, Reynes M, Bismuth H. Portal and arterial thrombosis in liver transplantation: a frequent event in severe rejection. Transplant Proc 1989;21(1 pt 2):2225-7. 19. Payen DM, Fratacci MD, Dupuy P, Gatecel C, Vigouroux C, Ozier Y, et al. Portal and hepatic arterial blood flow measurements of human transplanted liver by implanted Doppler probes: interest for early complications and nutrition. Surgery 1990;107:417-27. 20. Mor E, Scwartz ME, Sheiner PA, Menesses P, Hytiroglou P, Emre S, et al. Prolonged preservation in University of Wisconsin solution associated with hepatic artery thrombosis after orthotopic liver transplantation. Transplantation 1993;56:1399-402. 21. Vit A, De Candia A, Como G, Del Frate C, Marzio A, Bazzocchi M. Doppler evaluation of arterial complications of adult orthotopic liver transplantation. J Clin Ultrasound 2003;31:339-45. Submitted Jun 10, 2012; accepted Oct 12, 2012.