The incidence of hepatic artery thrombosis (HAT) varies from 1% to 26%1-4 and contributes to increased morbidity and mortality after liver transplantation (LT). In particular, HAT remains one of the main causes of graft loss after LT.5
Many factors increase the incidence of HAT, such as a reduced size of the arterial anastomosis,3 technical problems,6,7 and hepatic malignancies.8 Urgent revascularization may prevent the need for retransplantation,9 but early diagnosis is an essential prerequisite for the success of this strategy.6
Centers that perform pediatric living donor LT (LDLT) must aim at reducing the occurrence of HAT to avoid the severe complications of this condition. Improvements in medical care and refinements in surgical techniques can accomplish these goals, together with an improved understanding of the risk factors involved and the current surgical practices adopted in each institution. Our group previously demonstrated that the occurrence of HAT and patient body weight (BW) under 10 kg were independently associated with worse patient and graft survivals in a cohort of 413 primary pediatric LDLT patients.10 The present report is a reflection of 20 years of practice in pediatric LDLT and describes the risk factors for HAT and outcomes after LT. Some simple technical refinements that were introduced over the years are also described.
A total of 656 primary LDLTs were performed in patients under 18 years of age at Hospital Sírio-Libanês and Hospital A. C. Camargo Cancer Center, in São Paulo, Brazil from October 1995 to June 2015. The same medical group in both hospitals managed the patients. Data on transplant recipients were collected through a retrospective examination of medical records and a prospectively collected database. The hospitals' ethics committees approved this study.
Recipient and donor selection was based on ABO blood group compatibility. Pediatric patients with end-stage liver disease, who were accepted by the transplant team for LT, were initially placed on the waiting list for deceased donor LT. Assessments of the LT recipients included the investigation of liver disease etiology and severity, imaging studies when applicable, and dental, psychological, and nutritional evaluations.
The donors' preoperative evaluations and surgical techniques were reported in previous publications11 and followed the principles described by Yamaoka et al.12 The age limit and highest body mass index accepted for donation was 50 years and 28 kg/m2, respectively. Doppler ultrasound was performed to evaluate vascular anatomy, liver echogenicity, and liver volumetry for left lateral segment (LLS) donation. Magnetic resonance imaging and cholangio-magnetic resonance imaging were used to assess the anatomy for left lobe (LL) and right lobe (RL) donation.
The liver segments used were the LLS, LL, LL with segment I, RL, and monosegments. The grafts were implanted using a “piggyback technique.” The graft's hepatic vein was anastomosed to the recipient's vena cava/hepatic vein as previously described.13 The graft's portal vein (PV) was anastomosed in an end-to-end fashion, either to the recipient’s PV trunk or using an interposition vascular graft.14 The hepatic artery (HA) was always reconstructed using microvascular techniques with 9-0 or 10-0 nylon sutures (Ethicon, Edinburgh, UK) since the beginning of this series. The OPMI PENTERO 900 and OPMI Vario S88 from ZEISS (Jena, Germany) were the surgical microscopes used in the majority of the cases in this series. Two orthopedic hand surgeons and 1 plastic surgeon were responsible for the HA reconstructions during the study period. The microscope setup took 10 minutes to 15 minutes, and it was done during the hepatic veins and PV reconstruction. Twenty to 30 minutes were required to perform an arterial anastomosis, which was done under 10 times magnification. A variable number of arteries, from 1 to 3 arterial stumps, supplied the left liver grafts, which constituted the majority of the transplants in this series. The arteries supplying the left liver grafts were: left HA, segment IV artery or middle HA, and left HA arising from the left gastric artery. One or 2 microsurgical anastomosis were performed during the surgery. The only mandatory circumstances where all efforts were performed to proceed with a double arterial anastomosis occurred when a back flow was not obtained after the completion of the first anastomosis in the presence of 2 arteries to the graft. All patients who presented intraoperative HAT underwent immediate thrombectomy with the use of a Fogarty catheter when necessary, heparinization and reanastomosis. The diagnosis of intraoperative HAT was always confirmed with an intraoperative Doppler ultrasound. Intraoperative portal pressure (PP) measurements were undertaken via catheterization of the inferior mesenteric vein when indicated. Biliary anastomosis was performed either as a duct-to-duct or Roux-in-Y bilioenteric reconstruction.
Postoperative Image Studies, Anticoagulation and Management
Doppler US scans were routinely performed on postoperative day (POD) 1 to evaluate vascular patency. After the first POD, there were no protocols for Doppler ultrasound during the follow-up period, and further examinations were performed when clinically indicated (eg, elevation of liver enzymes, coagulopathy, and fever). Patients who presented altered HA signal findings on Doppler ultrasound underwent upper-abdominal angiocomputed tomography for further investigation.
Postoperative anticoagulation throughout the study period was restricted to patients who presented intraoperative vascular thrombosis or Budd-Chiari syndrome. Anticoagulation was based on an intravenous infusion of heparin at a starting dose of 10 units/kg per hour. The heparin infusion was tapered for a partial thromboplastin time between 2 and 3 times the normal range. Patients with Budd-Chiari syndrome were maintained on oral anticoagulation with a target international normalized ratio between 2 and 3. All patients with platelet counts greater than 50 000/mm3 were maintained on dipyridamole (1 mg/kg per day) for 3 months after the transplant.
Tacrolimus (FK 506, Prograf) and steroids were used for immunosuppression. Details on postoperative clinical management were previously described.10,13 Portal vein thrombosis (PVT) was considered early (e-PVT) when it occurred within 30 days of the transplant and late (l-PVT) when it was established after this period. The same time spans were considered for early and late HAT (e-HAT <30 days, l-HAT ≥30 days).
The following variables indicative of the pretransplant clinical status and those related to technical aspects during the surgery were investigated: diagnosis (biliary atresia [BA] versus other diseases), recipients' BW, age, sex, pediatric end-stage liver disease (PELD) score, Z score weight/age, Z score height/age, ascites, previous surgery, graft-to-recipient weight ratio (GRWR), type of living donor liver graft (eg, LLS, and LL), volume of packed red blood cell transfusion (mL/kg), cold ischemia time and warm ischemia time, total ischemia time, the occurrence of intraoperative HAT, and the number of HA anastomosis (1 or 2).
The impact of HAT on patient and graft survival and the occurrence of associated complications after the transplant were investigated. A historical analysis of the data was also performed. The incidence of HAT and the number of arterial anastomosis were analyzed in 7 different periods since the beginning of the series: 6 periods with 100 patients each (P1 to P6) and the last period with 56 patients (P7). The management strategies adopted in cases of HAT were shown.
Means and medians were calculated to summarize continuous effects, and the results were compared using t tests or appropriate nonparametric tests when distributional assumptions were in doubt. Categorical variables are expressed as numbers and percentages. Differences between groups were assessed using χ2 or Fisher exact tests, when needed.
Binary logistic regression analysis was performed to identify the variables that were independently associated with HAT, and the variables that were significant at P less than 0.10 were selected for the multivariate analysis.
Survival analysis was conducted according to the Kaplan-Meier product-limit estimates, and patient subgroups were compared using a 2-sided log-rank test.
All analyses were performed using the SPSS 21.0 statistical package (IBM, Inc., Chicago, IL).
Demographics and Variables Associated With HAT
There were 656 primary pediatric LDLTs performed in the study period. The following indications were used for liver replacement: 406 (61.8%) BA, 31 (4.7%) choledocal, 29 (4.4%) cryptogenic cirrhosis, 19 (2.8%) α1-antitrypsin deficiency, 17 (2.3%) type-1 tyrosinemia, 26 (3.9%) primary liver tumors, 11 (1.6%) fulminant liver failure, 21 (3.2%) Alagille syndrome, 18 (2.7%) progressive intrahepatic familial cholestasis, 20 (3.0%) other metabolic diseases, 29 (4.4%) chronic cholestasis, 13 (2%) Budd-Chiari syndrome, 8 (1.2%) autoimmune hepatitis, and 8 (1.2%) other causes. There were 359 female recipients, and the median age, BW, and PELD score at the time of transplant were 13 months, 8.4 kg, and 15, respectively. Approximately half of the patients had ascites, and 36.4% underwent a previous surgery before the transplant. Twenty-one (3.2%) patients presented HAT. Table 1 shows the demographics of the entire cohort and the differences between patients who presented with or without HAT. The initial analyses revealed no differences in the type of diagnosis (BA versus others), BW, age, nutritional status, PELD score, presence of ascites or previous surgery in relation to pretransplant variables. Only the occurrence of intraoperative HAT and the use of grafts with a GRWR less than 1.1% were different between the groups. Binary logistic regression analysis was performed to determine the variables that were independently associated with HAT (Table 2). The multivariate model demonstrated that intraoperative HAT (odds ratio, 62.63; 95% confidence interval, 12.64-310.19; P < 0.001) and the use of grafts with a GRWR less than 1.1% (odds ratio, 24.46; 95% confidence interval, 4.55-131.56, P < 0.001) retained statistical significance.
Associated Complications and Patient and Graft Survival
Table 3 shows the association of HAT and other technical complications that occurred after LT. Nearly 30% of the patients with e-PVT (6/22) also presented HAT (P < 0.001). Five patients with combined HAT and PVT died, and 1 patient received a new liver graft. The occurrence of bile leak was 14.2% (3/21) in patients with HAT versus 7.1% (45/635) in patients without HAT (P = 0.21).
Figures 1 and 2 show the 24-month Kaplan-Meier patient and graft survival curves for patients with and without HAT. The 1-, 3-, and 24-month patient survival rates in the group without HAT were 97.5%, 95.4%, and 89.7%, respectively, versus 66.7%, 57.1%, and 51.9%, respectively, for patients with HAT (P < 0.001). The 1-, 3-, and 24-month graft survival rates for patients without HAT were 97.2%, 95.1%, and 89.6%, respectively, versus 45%, 20%, and 15%, respectively, for patients with HAT (P < 0.001).
The cases were divided into 7 periods (P1 to P7), and each period included 100 patients, except for P7, which included 56 patients (Table 4). For example, it took almost 10 years (1995 to 2004) to complete 100 LDLTs in pediatric recipients. Table 4 also shows the incidence of HAT and the use of double anastomosis in each period. During P1, the incidence of HAT was 2%, and the rate of 2-arterial anastomosis was 15%. The use of 2-arterial anastomosis dropped to an average rate of 6% for the next 300 patients, and the average incidence of HAT was also 6% (P2, P3, and P4). There were 2 cases of HAT (0.7%) in the last 3 periods of the study (256 patients), and the average use of double anastomosis was 19.4%. The overtime trend analysis was statistically significant and exhibited a decrease in the incidence of HAT and an increased use of 2-arterial anastomosis.
The available data (583 cases) demonstrated that there was only 1 artery supplying the graft in 378 (64.8%) cases, 2 arteries in 191 (32.7%), and 3 arteries in 14 (2.4%) live donor grafts. Among the 21 cases that developed HAT, 17 presented only 1 artery to the graft, and 4 patients presented 2 arteries. Double anastomosis was only performed in 1 instance. Only 2 recipients required the use of interposition vascular grafts for HA reconstruction, and the recipient's inferior mesenteric vein was used in both cases.
Eleven transplants before 2010 were performed with a GRWR of 1.1% or less, and 3 of these patients developed HAT. Only 1 patient underwent a portocaval shunt and splenic artery ligation (SAL) to modulate the portal inflow after the reperfusion. The initial PP after reperfusion was 25 mm Hg in this patient, and it dropped to 19 mm Hg after the 2 procedures. There were records of PP measurements in 2 other patients. One patient with a PP of 18 mm Hg did not receive inflow modulation, and the PP in the other patient was 11 mm Hg. There were 10 transplants with a GRWR of 1.1% or less after 2010, and 5 (50%) of these patients underwent inflow modulation via SAL (5 cases). There were records of PP measurements in 2 patients. An initial recorded PP of 19 mm Hg dropped to 15 mm Hg after SAL in the first patient. The PP was 9 mm Hg in the other patient, and no procedures were performed. None of the patients transplanted after 2010 developed HAT.
Therapeutic Strategies and Outcomes in Patients with HAT
Nineteen patients presented e-HAT, and 2 patients exhibited l-HAT (Figure 3). Eleven patients with e-HAT were retransplanted, and 5 patients underwent a prior surgery to recanalize the HA anastomosis. Only 2 patients (2/8) with e-HAT survived without retransplantation. The first patient underwent another surgery on the first POD, and arterial patency was achieved. The other patient exhibited a late spontaneous HA recanalization. The remaining 6 patients died, and surgical reanastomosis was attempted in 2 patients. Two patients with l-HAT were retransplanted, but they did not survive. The causes of death included liver failure (5 patients), sepsis (2 patients), and intracranial bleeding (1 patient).
Fifty years have passed since the first pediatric LT, and the success of these transplants still largely depends on our ability to maintain artery patency.15 A high percentage of arteries continue to become occluded despite medical and technical improvements, which compromises patient outcomes.16 Reoperation and retransplantation are generally performed after e-HAT, and the identification of risk factors associated with HAT can help to achieve better results. Physicians who are directly involved with patient care know that long cold ischemia time,17 severe hypotension,9 high posttransplant hematocrit,18 cytomegalovirus infection,19,20 and acute cellular rejection21 are associated with HAT, and these conditions must be properly managed. Technical issues and the small size of arterial anastomosis3,6,7 are also implicated in the occurrence of HAT. Most transplant centers introduce technical improvements with time and experience to overcome these problems.
Uchida et al22 published the largest report on pediatric LDLT with a focus on HAT. Female sex and the use of larger grafts (higher GRWR) were associated with an increased rate of HAT (HAT = 6.7%, 382 patients, 402 liver transplants, patients younger than 12 years), which is in contrast to the results obtained in our series. Frequent Doppler ultrasound assessments (twice a day for 2 weeks, and daily thereafter) allowed for the salvage of most grafts. There was not a significant difference in patient and graft survival between patients with or without HAT during follow-up. However, patients with HAT who underwent reoperation presented an increased rate of biliary complications. There was an attempt to reestablish the arterial flow in only 8 (44.4%) cases of e-HAT in our series because the surveillance was not as frequent as in the Kyoto series. Consequently, there was a negative impact on patient and graft survivals. Differences in policies may partially explain these differences between studies, because LDLT recipients in Asian countries do not have the option for retransplantation with deceased donor grafts. Therefore, countermeasures against HAT are necessary to prevent graft loss. Other reports focused on graft salvage have shown variable results. Warnaar et al23 only studied pediatric data, and 32 of 232 (13.7%) recipients developed HAT. Urgent recanalization was accomplished in 37.5% (6/16) of the patients. A total of 24 patients received early retransplants. Duffy et al24 observed 5% occurrence of HAT (203/4234 transplants) in adults and children, and the success rate of surgical revision was 10.5% (9/86).
Kiuchi et al25 defined small grafts as grafts with a GRWR less than 0.8%. The utilization of these grafts determined the occurrence of small-for-size syndrome, which is characterized by deficient bile production, cholestasis, coagulopathy, ascites, synthetic liver dysfunction and death, generally secondary to infectious complications. Another aspect encountered in small-for-size syndrome is portal hyperflow to the liver graft with an automatic decrease in arterial inflow (ie, “arterial buffer response”26), which increases the risk of HAT. The surgical strategies that are used to modulate portal inflow to the graft include proximal SAL,27 portosystemic shunts and splenectomy, or a combination of these procedures. A GRWR of 1.1% or less was independently associated with HAT in the present series. We have performed SAL more recently (P5 to P7) to modulate the portal inflow (5/10). Hepatic artery thrombosis was not observed in these patients, although records of PP changes before and after SAL were not available/recorded, except in 1 patient (19 mm Hg to 15 mm Hg). As a result, it was not possible to draw conclusions from the presented data because the PP values were missing. The observed improvements in results (eg, decreased HAT incidence with a GRWR ≤1.1% after 2010) followed the trends of the cohort and may only reflect that better surgical techniques were achieved with time. Nevertheless, these inflow modulation strategies must be considered when transplanting teenagers with small liver grafts.
Microvascular anastomosis with surgical microscopes was used throughout our experience. Some groups have not encountered differences in results between microvascular techniques and other reconstruction strategies.28 Heffron et al29 obtained low HAT rates (2.7%) without the use of microsurgery. However, Shackleton et al30 reported that not using microvascular techniques was associated with an increased incidence of HAT. In particular, the increased risk of HAT in cases that presented intraoperative HAT (persistent HAT) indicates that technical problems were not properly addressed in the operating room. Several subjective strategies, which were difficult to quantify, were introduced over time to avoid these situations, such as careful arterial perfusion on the back table to prevent intimal dissection. Furthermore, avoiding redundancy, anastomotic kinking, and anastomosis under tension are warranted for the recipient surgery. Special attention must also be taken during the biliary reconstruction because the maneuvers that are used to expose the bile duct for the anastomosis may compromise the patency of the arterial anastomosis (intraoperative kink or stretch).
Most centers performing LDLT follow the recommendation from Ikegami et al.31 They suggested “in LDLT, it is not always necessary to anastomose all the hepatic arteries supplying the graft if it is possible to confirm pulsatile arterial outflow from the nonanastomosed cut stumps of the arteries.” Indeed, we followed this recommendation until we analyzed our own data on 413 primary pediatric LDLT in 2012.10 We demonstrated that HAT and BW under 10 kg were markers of worse patient and graft survivals. Thus, 2-arterial anastomosis was set as the standard whenever possible (2 or 3 arteries to the graft) to reduce the incidence of HAT. Two other reports have analyzed the use of 2-arterial anastomosis and HAT. Julka et al32 compared the occurrence of HAT and other complications in a cohort of 87 pediatric recipients. The incidence of HAT was 6.9%, and these authors did not detect differences between patients who received 1- or 2-arterial anastomosis. Uchida et al22 did not demonstrate any protective effect of double anastomosis in a multivariate analysis in a Japanese cohort, which is similar to the present results. However, our analysis revealed a statistically significant trend for a decreased incidence of HAT and an increased use of 2-arterial anastomosis over time.
Rigorous statistical analysis suggested that the use of grafts with a GRWR of 1.1% or less and the occurrence of intraoperative HAT are important factors related to HAT, although it is difficult to establish the precise dimension of this relationship because the confidence intervals were very wide. However, it is possible to apply the double anastomosis policy to both groups of patients (GRWR ≤1.1% and intraoperative HAT), and this policy may provide added protection for these at-risk patients. In parallel to this policy, inflow modulation strategies should be applied in cases of GRWR of 1.1% or less if the measured PP is high (>15 mm Hg).
The association of e-PVT and e-HAT was present in 6 cases. Five of these patients died, and 1 was retransplanted. It is difficult to define a cause-and-effect relationship in these instances. Furthermore, these patients were very sick and presented infectious complications, sepsis, and/or septic shock, which are unfortunate events with high mortality rates.
In conclusion, this 20-year experience with pediatric LDLT demonstrated that a GRWR of 1.1% or less and the occurrence of intraoperative HAT were independently associated with HAT. Patients who developed HAT exhibited worse patient and graft survival compared to those without HAT. Furthermore, trend analysis revealed a significant reduction in the incidence of HAT over time, which was observed in parallel to the increased use of 2 hepatic arteries for anastomosis during graft implantation. The double artery anastomosis may represent an extra protection to pediatric recipients undergoing LDLT.
1. Tiao GM, Alonso MH, Ryckman FC. Pediatric liver transplantation. Semin Pediatr Surg
. 2006; 15: 218–227.
2. Broering DC, Kim JS, Mueller T, et al. One hundred thirty-two consecutive pediatric liver transplants without hospital mortality: lessons learned and outlook for the future. Ann Surg
. 2004; 240: 1002–1012.
3. Mazzaferro V, Esquivel CO, Makowka L, et al. Hepatic artery thrombosis after pediatric liver transplantation—a medical or surgical event? Transplantation
. 1989; 47: 971–977.
4. Lallier M, St-Vil D, Dubois J, et al. Vascular complications after pediatric liver transplantation. J Pediatr Surg
. 1995; 30: 1122–1126.
5. Ikegami T, Hashikura Y, Nakazawa Y, et al. Risk factors contributing to hepatic artery thrombosis following living-donor liver transplantation. J Hepatobiliary Pancreat Surg
. 2006; 13: 105–109.
6. Langnas AN, Marujo W, Stratta RJ, et al. Hepatic allograft rescue following arterial thrombosis. Role of urgent revascularization. Transplantation
. 1991; 51: 86–90.
7. Yanaga K, Makowka L, Starzl TE. Is hepatic artery thrombosis after liver transplantation really a surgical complication? Transplant Proc
. 1989; 21: 3511–3513.
8. Sanchez SE, Javid PJ, Lao OB, et al. Hepatic artery thrombosis and liver malignancy in pediatric liver transplantation. J Pediatr Surg
. 2012; 47: 1255–1260.
9. Markmann JF, Markowitz JS, Yersiz H, et al. Long-term survival after retransplantation of the liver. Ann Surg
. 1997; 226: 408–418.
10. Neto JS, Pugliese R, Fonseca EA, et al. Four hundred thirty consecutive pediatric living donor liver transplants: variables associated with post-transplant patient and graft survival. Liver Transpl
. 2012; 18: 577–584.
11. Seda-Neto J, Godoy AL, Carone E, et al. Left lateral segmentectomy for pediatric live-donor liver transplantation: special attention to segment IV complications. Transplantation
. 2008; 86: 697–701.
12. Yamaoka Y, Ozawa K, Tanaka A, et al. New devices for harvesting a hepatic graft from a living donor. Transplantation
. 1991; 52: 157–160.
13. Neto JS, Carone E, Pugliese V, et al. Living donor liver transplantation for children in Brazil weighing less than 10 kilograms. Liver Transpl
. 2007; 13: 1153–1158.
14. Neto JS, Fonseca EA, Feier FH, et al. Analysis of factors associated with portal vein thrombosis in pediatric living donor liver transplant recipients. Liver Transpl
. 2014; 20: 1157–1167.
15. Gallo A, Esquivel CO. Late hepatic artery thrombosis in pediatric liver transplantation: an incomplete story. Liver Transpl
. 2014; 20: 512–513.
16. Ackermann O, Branchereau S, Franchi-Abella S, et al. The long-term outcome of hepatic artery thrombosis after liver transplantation in children: role of urgent revascularization. Am J Transplant
. 2012; 12: 1496–1503.
17. Buckels JA, Tisone G, Gunson BK, et al. Low haematocrit reduces hepatic artery thrombosis after liver transplantation. Transplant Proc
. 1989; 21(1 Pt 2): 2460–2461.
18. Mor E, Schwartz ME, Sheiner PA, et al. Prolonged preservation in University of Wisconsin solution associated with hepatic artery thrombosis after orthotopic liver transplantation. Transplantation
. 1993; 56: 1399–1402.
19. Gunsar F, Rolando N, Pastacaldi S, et al. Late hepatic artery thrombosis after orthotopic liver transplantation. Liver Transpl
. 2003; 9: 605–611.
20. Madalosso C, de Souza NF Jr, Ilstrup DM, et al. Cytomegalovirus and its association with hepatic artery thrombosis after liver transplantation. Transplantation
. 1998; 66: 294–297.
21. Samuel D, Gillet D, Castaing D, et al. Portal and arterial thrombosis in liver transplantation: a frequent event in severe rejection. Transplant Proc
. 1989; 21(1 Pt 2): 2225–2227.
22. Uchida Y, Sakamoto S, Egawa H, et al. The impact of meticulous management for hepatic artery thrombosis on long-term outcome after pediatric living donor liver transplantation. Clin Transplant
. 2009; 23: 392–399.
23. Warnaar N, Polak WG, de Jong KP, et al. Long-term results of urgent revascularization for hepatic artery thrombosis after pediatric liver transplantation. Liver Transpl
. 2010; 16: 847–855.
24. Duffy JP, Hong JC, Farmer DG, et al. Vascular complications of orthotopic liver transplantation: experience in more than 4,200 patients. J Am Coll Surg
. 2009; 208: 896–903.
25. Kiuchi T, Kasahara M, Uryuhara K, et al. Impact of graft size mismatching on graft prognosis in liver transplantation from living donors. Transplantation
. 1999; 67: 321–327.
26. Smyrniotis V, Kostopanagiotou G, Kondi A, et al. Hemodynamic interaction between portal vein and hepatic artery flow in small-for-size split liver transplantation. Transpl Int
. 2002; 15: 355–360.
27. Ito T, Kiuchi T, Yamamoto H, et al. Changes in portal venous pressure in the early phase after living donor liver transplantation: pathogenesis and clinical implications. Transplantation
. 2003; 75: 1313–1317.
28. Guarrera JV, Sinha P, Lobritto SJ, et al. Microvascular hepatic artery anastomosis in pediatric segmental liver transplantation: microscope vs loupe. Transpl Int
. 2004; 17: 585–588.
29. Heffron TG, Welch D, Pillen T, et al. Low incidence of hepatic artery thrombosis after pediatric liver transplantation without the use of intraoperative microscope or parenteral anticoagulation. Pediatr Transplant
. 2005; 9: 486–490.
30. Shackleton CR, Goss JA, Swenson K, et al. The impact of microsurgical hepatic arterial reconstruction on the outcome of liver transplantation for congenital biliary atresia. Am J Surg
. 1997; 173: 431–435.
31. Ikegami T, Kawasaki S, Matsunami H, et al. Should all hepatic arterial branches be reconstructed in living-related liver transplantation? Surgery
. 1996; 119: 431–436.
32. Julka KD, Lin TS, Chen CL, et al. Reconstructing single hepatic artery with two arterial stumps: biliary complications in pediatric living donor liver transplantation. Pediatr Surg Int
. 2014; 30: 39–46.