This study aimed to evaluate the effectiveness of different treatments for hepatic artery thrombosis (HAT) and hepatic artery stenosis (HAS) after pediatric liver transplantation. We systematically reviewed studies published since 2000 that investigated the management of HAT and/or HAS after pediatric liver transplantation. Studies with a minimum of 5 patients in one of the treatment methods were included. The primary outcomes were technical success rate and graft and patient survival. The secondary outcomes were hepatic artery patency, complications, and incidence of HAT and HAS. Of 3570 studies, we included 19 studies with 328 patients. The incidence was 6.2% for HAT and 4.1% for HAS. Patients with an early HAT treated with surgical revascularization had a median graft survival of 45.7% (interquartile range, 30.7%-60%) and a patient survival of 61.3% (interquartile range, 58.7%-66.9%) compared with the other treatments (conservative, endovascular revascularization, or retransplantation). As for HAS, endovascular and surgical revascularization groups had a patient survival of 85.7% and 100% (interquartile range, 85%-100%), respectively. Despite various treatment methods, HAT after pediatric liver transplantation remains a significant issue that has profound effects on the patient and graft survival. Current evidence is insufficient to determine the most effective treatment for preventing graft failure.

To the editor, Hepatic venous outflow obstruction (HVOO) caused by various reasons is one of the more difficult complications after liver transplantation. Despite the introduction of various innovative techniques in different liver transplantation procedures, the incidence of HVOO has not been significantly reduced. How to deal with HVOO is still a topic that needs clinical attention. Balloon angioplasty and/or stent placement have become the primary treatment for HVOO after liver transplantation. The recent study by Sambommatsu et al,1 published in Liver Transplantation, analyzing the clinical data from 16/324 (4.9%) patients with HVOO, underscored the morphology of HVOO and corresponding treatment plan. They recommended repeat balloon angioplasty be considered for simple anastomotic stenosis, but stent placement was recommended for kinks or intrahepatic stenosis. In addition, even if treatment was successful, patients with late-onset HVOO should be closely followed. We firmly appreciate the contributions of the authors on the subject and would like to discuss some important aspects of this study. First, although the authors described that vascular access was obtained by means of the right femoral vein or the right internal jugular vein, they did not provide specific details regarding the selection of access for interventional radiology. The choice of access appears to significantly influence the success rate.2 Both the right femoral vein and the right internal jugular vein access possess their own inherent limitations. For instance, when performing transjugular vein puncture, potential complications may arise, including incorrect puncture of the carotid artery, local hematoma, hemothorax, pneumothorax, and so forth. Besides, the comfort level of patients and doctors is poor, patients need to lie on their side head for a long time, and doctors have little operating space. The femoral vein access can avoid the above shortcomings, but it requires skilled interventional operation techniques to overcome the acute angle between the hepatic vein and the IVC. Consequently, it is highly expected that the authors could give more information about the selection of interventional radiology access to guide clinical practice. Further, we concur with the author’s perspective that the utilization of balloon angioplasty should be prioritized in cases of simple anastomotic stenosis. The stent is easy to fall off and has the risk of thrombosis, and retransplantation will make the operation more difficult. Additionally, there is currently a lack of multicenter clinical data on long-term patency and stent-related complications in HVOO patients after liver transplantation, so stent placement should be carefully considered. However, for HVOO patients who do not respond to repeated balloon angioplasty, prompt stent placement is warranted. Nonetheless, the optimal timing of stent placement is unclear. Further study of indications for early stent placement would help to avoid graft failure as well as retransplantation.

To the editor, We thank Dr. Guo and colleagues for their interest in our manuscript published in Liver Transplantation regarding hepatic venous outflow obstruction after adult living donor liver transplantation.1 They discussed 2 different choices of venous access (femoral and internal jugular) and highlighted the differences in procedure-related risks and successful rates of these 2 accesses. They suggested that we should provide more information about the selection of interventional radiology access to guide clinical practice. We agree with the opinion that femoral vein access is theoretically safer than internal jugular vein access. However, catheter canulation to the hepatic vein is sometimes difficult in femoral vein access due to the acute angle between the hepatic vein and the IVC (especially for right liver grafts), as Dr. Guo mentioned. This also stands true for stent insertion, and there is a risk of unstable stent positioning or stent migration. Currently, we have no uniform criteria for the choice of venous access, and it is decided on a case-by-case basis. We preferred the use of femoral access during the early era (2000–2010), but we have experienced several cases in which stent insertion was difficult. Recently, internal jugular vein access is more often selected because it is technically easier. In fact, most (4/5) of the stent placement in our cohort was performed by internal jugular access. In addition, when repeated interventional radiological procedures are needed, it is important to select suitable venous access based on the angiographic findings of the initial session. In fact, in many patients in our study, the first angiogram and balloon angioplasty were performed with femoral access, but the second procedure (either repeat balloon angioplasty or stent placement) was performed by internal jugular access. There is no need to adhere to one approach. Regarding success rates, we believe there is no difference between the two approaches if an experienced interventional radiologist selects the appropriate venous access and carefully carries out the procedure. Regarding the risks associated with internal jugular vein access, we believe it is safe if the venous puncture is performed under both ultrasound and fluoroscopic guidance. Among the 15 patients in our study, we did not experience any serious complications related to venous puncture. As mentioned by Dr. Guo, the working space tends to be smaller in jugular vein access; however, there is still enough space for 2 interventional radiologists to safely and efficiently perform the procedures. In conclusion, based on our experience, both accesses are feasible and have similar success rates. It is paramount to select the access that is favorable from the venous anatomy standpoint and the type of procedure according to each individual patient.

Chairman, Max center for liver and biliary sciences, Max Superspeciality hospital, Saket, Delhi, India 110017 Director, Max center for liver and biliary sciences, Max Superspeciality hospital, Saket, Delhi, India 110017 Abbreviations: DDLT, deceased donor liver transplant; EAD, early allograft dysfunction; LDLT, living donor liver transplant; SFSS, small for size syndrome Correspondence ANG Centre for Liver and Biliary Sciences, Department of Surgery. Email: [email protected]

Transarterial chemoembolization (TACE) and transarterial radioembolization (TARE) are the 2 most used modalities for patients with HCC while awaiting liver transplant. The purpose of this study is to perform a cost-effectiveness analysis comparing TACE and TARE for downstaging (DS) patients with HCC. A cost-effectiveness analysis was performed comparing TACE and TARE in DS HCC over a 5-year time horizon from a payer's perspective. The clinical course, including those who achieved successful DS leading to liver transplant and those who failed DS with possible disease progression, was obtained from the United Network for Organ Sharing. Costs and effectiveness were measured in US dollars and quality-adjusted life years (QALYs). Probabilistic and deterministic sensitivity analyses were performed. TARE achieved a higher effectiveness of 2.51 QALY (TACE: 2.29 QALY) at a higher cost of $172,162 (TACE: $159,706), with the incremental cost-effectiveness ratio of $55,964/QALY, making TARE the more cost-effective strategy. The difference in outcome was equivalent to 104 days (nearly 3.5 months) in compensated cirrhosis state. Probabilistic sensitivity analyses showed that TARE was more cost-effective in 91.69% of 10,000 Monte Carlo simulations. TARE was more effective if greater than 48.2% of patients who received TACE or TARE were successfully downstaged (base case: 74.6% from the pooled analysis of multiple published cohorts). TARE became more cost-effective when the cost of TACE exceeded $4,831 (base case: $12,722) or when the cost of TARE was lower than $43,542 (base case: $30,609). Subgroup analyses identified TARE to be the more cost-effective strategy if the TARE cohort required 1 fewer locoregional therapy than the TACE cohort. TARE is the more cost-effective DS strategy for patients with HCC exceeding Milan criteria compared to TACE.

1Department of Surgery, Division of Abdominal Transplant, Stanford University Medical Center, Stanford, California, USA 2Department of Transplant Surgery, Mita Hospital, International University of Health and Welfare, Tokyo, Japan Abbreviations: AI, artificial intelligence; HBOT, hyperbaric oxygen therapy; LT, liver transplantation. The paper is not based on previous communication with a society or meeting. Correspondence Kazunari Sasaki, Department of General Surgery, Division of Abdominal Transplant, Stanford University School of Medicine, Stanford, CA 94305, USA. Email: [email protected]

The long-term (>5y) outcomes following liver transplantation (LT) have not been extensively reported. The aim was to evaluate outcomes of LT recipients who have survived the first 5 years. A multicenter retrospective analysis of prospectively collected data from 3 high volume LT centers (Dallas-USA, Birmingham-UK, and Barcelona-Spain) was undertaken. All adult patients, who underwent LT since the inception of the program to December 31, 2010, and survived at least 5 years since their LT were included. Patient survival was the primary outcome. A total of 3682 patients who survived at least 5 years following LT (long-term survivors) were included. Overall, median age at LT was 52 years (IQR 44-58); 53.1% were males; and 84.6% were Caucasians. A total of 49.4% (n=1820) died during a follow-up period of 36,828 person-years (mean follow-up 10y). A total of 80.2% (n=1460) of all deaths were premature deaths. Age-standardized all-cause mortality as compared to general population was 3 times higher for males and 5 times higher for females. On adjusted analysis, besides older recipients and older donors, predictors of long-term mortality were malignancy, cardiovascular disease, and dialysis. Implementation of strategies such as noninvasive cancer screening, minimizing immunosuppression, and intensive primary/secondary cardiovascular prevention could further improve survival.

Correspondence Christopher J. Sonnenday, Section of Transplantation, Department of Surgery, University of Michigan, F6686 UH South, 1500 E Medical Center Dr Ann Arbor, MI 48109. Email: [email protected]

1Division of GI/Hepatology, University of California, San Francisco, San Francisco, California, USA 2Department of Obstetrics and Gynecology, University of California, San Francisco, California, USA Abbreviations: AASLD, American Association for the Study of Liver Diseases; LT, liver transplant; MELD, Model for End-Stage Liver Disease. Correspondence Monika Sarkar, Division of Gastroenterology and Hepatology, University of California, San Francisco, 513 Parnassus Avenue, Room S-357, San Francisco, CA 94143-0358, USA. Email: [email protected]

Verma, Sapana; Sakamoto, Seisuke; Abdelwahed, Mohamed Sami; Shimizu, Seiichi; Uchida, Hajime; Okada, Noriki; Nakao, Toshimasa; Kodama, Tasuku; Komine, Ryuji; Fukuda, Akinari; Rela, Mohamed; Kasahara, Mureo Author Information