Objectives: Liver transplant for patients with hepatocellular carcinoma involves 3 main types of donor allografts: donation after brain death, donation after cardiac death, and donation after brain and cardiac death. Data on this topic are limited, and controversies exist regarding liver transplant outcomes in hepatocellular carcinoma patients who have received these allografts. Materials and Methods: Data from 490 hepatocellular carcinoma patients who received liver transplant from 2015 to 2021 at the Shulan (Hangzhou) Hospital were retrospectively analyzed. Participants were divided into 3 cohorts according to allograft type: donation after brain death, donation after cardiac death, and donation after brain and cardiac death. Kaplan-Meier and Cox regression methods were used to evaluate patient survival, graft survival, and recurrence-free survival rates after liver transplant. Results: Kaplan-Meier analysis revealed that 3-year patient survival rates were 69.2% for donations after brain death, 69.2% for donations after cardiac death, and 46.6% for donations after brain and cardiac death (P = .42); the 3-year graft survival rates were 53.3% for donations after brain death, 56.4% for donations after cardiac death, and 46.6% for donations after brain and cardiac death (P = .44); and 3-year recurrence-free survival rates were 55% for donations after brain death, 56.6% for donations after cardiac death, and 39.5% for donations after brain and cardiac death (P = .46). Complications were also similar across the 3 cohorts (P = .36). Multivariable analysis showed that intraoperative red blood cell transfusion (hazard ratio: 1.820; P = .042) and early allograft dysfunction (hazard ratio: 3.240; P = .041) were independent risk factors for graft survival. Conclusions: Similar outcomes can be achieved for hepatocellular carcinoma patients who undergo liver transplant with donations after brain death, donations after cardiac death, or donations after brain and cardiac death allografts, especially when strict donor selection criteria are applied.
Key words : Donation after brain death, Donation after brain and cardiac death, Donation after circulatory death, Graft survival, Patient survival
Liver transplantation (LT) is a curative therapy for hepatocellular carcinoma (HCC), which is a main etiology for cancer-related mortality internationally.1,2 Liver transplant currently involves 3 types of donor allografts: donation after brain death (DBD), donation after cardiac death (DCD), and donation after brain and cardiac death (DBCD).3,4 Donation after brain death is the strategy of choice for LT. However, because of organ shortages, new strategies on the use of DCD and DBCD allografts have been implemented. Donations after brain and cardiac death are similar to the Maastricht criteria class type IV.3 One of the reasons for using DBCD donors in China is because of cultural beliefs, as Chinese people do not commonly accept brain death as a valid criterion for death, but instead only accept donor death when the heart has irreversibly stopped beating.5
When DCD LT was first implemented, it was associated with inferior outcomes compared with DBD LT, mainly due to ischemic cholangiopathy, early graft dysfunction, and higher graft failure and mortality rates.6-8 Recently, outcomes of DCD LT have been comparable to those of DBD LT as a result of robust donor and recipient selection and surgical technique optimization, especially during procure-ment, as well as insights into the effects of warm ischemia time (WIT) and cold ischemia time (CIT) and the use of machine perfusion.9-13
Patients with HCC are considered ideal candi-dates for DCD allografts because they tend to have preserved liver function and to be in better physical condition when listed for LT, making them able to tolerate potential complications associated with the use of marginal livers.14-16 Moreover, compared with DBD LT , DCD LT procedures have been associated with a lower dropout rate and shorter time on the waitlist. Nevertheless, in the few studies on this topic,17-23 controversy exists regarding the outcomes of HCC patients receiving DBD LT and DCD LT. Some studies have shown that DCD LT were associated with lower patient survival and graft survival rates and higher HCC recurrence after LT,17-19 whereas other studies found similar patient survival, graft survival, and recurrence-free survival rates between the 2 groups.20-22 In yet another study, higher recurrence rate and similar patient survival and graft survival rates were observed among DCD LT compared with DBD LT.23
Although DBCD LT is commonly performed in many Chinese transplant centers, no study has yet, to our knowledge, evaluated outcomes of DBCD in HCC patients after LT. Hence, we aimed to evaluate and compare the outcomes of DBD, DCD, and DBCD LTs in HCC patients at a high-volume center performing LT.
Materials and Methods
Before start of this retrospective study, the study protocol was approved by the Institutional Review Board (IRB) of Shulan (Hangzhou) hospital (IRB No. K202203021) in accordance with Regulations on Human Organ Transplantation, national legal requirements, and the 2000 Declaration of Helsinki. Patient consent was obtained, and all donors were civilians in China, with no organs coming from executed prisoners.
The study population consisted of 490 adult patients with HCC who underwent primary LT at our center between January 1, 2015, and December 31, 2021. The primary outcomes of interest were patient survival, graft survival, and recurrence-free survival after LT. These outcomes were assessed at 1 and 3 years after surgery for the entire cohort, as well as for subgroups based on donor type (DBD, DCD, and DBCD).
Patient survival was calculated as the time interval in months from the time of surgery to the time of death or last known follow-up, and graft survival was calculated as the time interval in months from the time of surgery to the time of death or retransplant. Recurrence-free survival was calcula-ted as the time interval in months from the time of surgery to the time of HCC recurrence or last known follow-up.
We also recorded WIT, CIT, and postoperative complications. During follow-up after surgery, imaging (mainly computed tomography scan and/or magnetic resonance imaging) and serum tumor biomarkers were obtained from patients. Data on donor, recipient, and operative factors were collected from our center’s database, including donor age, sex, height, weight, body mass index (BMI), blood group type, blood group match, WIT, CIT, relevant serum biochemistries, and medical history. We also obtained operative factors, which included anhepatic time, intraoperative red blood cell (RBC) transfusion, and blood loss. Obtained recipient factors included age, male sex, BMI, blood cohort type, tumor burden, past medical history, pretransplant ?-fetoprotein (AFP), relevant serum biochemistries, tumor characteristics, Model for End-Stage Liver Disease (MELD) score, Child-Pugh score, Milan criteria, retransplant status, and early graft dysfunction.
We used Stata version 16 MP for statistical analysis. Continuous variables are shown as median and interquartile ranges, with nonparametric Kruskal-Wallis test applied for comparisons. Categorical variables are shown as numbers and percentages, with and the chi-square test or the Fisher exact test used for comparison. We calculated the rates for patient survival, graft survival, and recurrence-free survival at 1 and 3 years using Kaplan-Meier analysis, with comparisons using the log-rank test. To determine the risk factors for graft survival after LT, we used a Cox regression model, with results expressed as hazard ratio (HR) with 95% confidence intervals. The statistical significance threshold for this study was P < .05.
Of 490 participants in this retrospective study, 204 (41.6%) received DBD LT, 228 (46.5%) received DCD LT, and 58 (11.8%) received DBCD LT. Most patients were male (91.7%), and the median age was 53.2 years (range, 54.8-59.4 y). The main etiology of liver disease was the hepatitis B virus (80.8%). Most patients had solitary nodule cirrhosis (88.6%) with a median tumor burden of 5 cm. Among 490 patients, 344 (70.2%) met the Milan criteria. The median MELD and Child-Pugh scores were 35.8 (range, 28-40) and 11 (range, 10-11), respectively.
Regarding recipient factors, the DBCD cohort had a higher rate of diabetes (P = .052) and shorter stature (P = .091) than the other cohorts. Regarding donor factors, the median WIT was longest in the DCD cohort (P < .001). We observed that aspartate amino-transferase and alanine aminotransferase levels were lowest in the DBCD cohort (P = .026 and P = .07, respectively). Intraoperative RBC transfusion was much lower in the DCD cohort than in the other cohorts (P = .036). The other donor and recipient factors were equivalent for the 3 cohorts (Table 1).
The median follow-up time was 18.6 months (range, 8-35 months) after LT. The overall 1- and 3-year patient survival rates after LT were 77.4% and 61.6%, respectively. When analyzed by donor allograft type, the 1- and 3-year patient survival rates were 83.3% and 69.2%, respectively, for the DBD cohort, 84.6% and 69%, respectively, for the DCD cohort, and 64.4% and 46.6%, respectively, for the DBCD cohort (P = .42) (Figure 1A).
The overall 1- and 3-year graft survival rates were 71.7% and 52%, respectively. When analyzed by donor allograft type, the 1- and 3-year graft survival rates were 75.1% and 53.3%, respectively, for the DBD cohort, 75.5% and 56.4%, respectively, for the DCD cohort, and 64.4% and 46.6%, respectively, for the DBCD cohort (P = .44) (Figure 1B).
The overall 1- and 3-year recurrence-free survival rates were 68.4% and 50.4%, respectively. When analyzed by donor allograft type, 1- and 3-year recurrence-free survival rates were 76.6% and 55%, respectively, for the DBD cohort, 76.4% and 56.6%, respectively, for the DCD cohort, and 52.1% and 39.5%, respectively, for the DBCD cohort (P = .46) (Figure 1C).
The overall complication rates in the DBD, DCD, and DBCD cohorts were similar (13.2%, 15.4%, and 8.6%, respectively; P = .36) (Table 2). The DCD cohort had a trend toward highest incidence of abdominal bleeding (P = .081), whereas the DBCD cohort had a trend toward highest peritoneal effusion and or abscess rate (P = .06) (Table 2).
Univariable analysis of donor factors showed that weight, BMI, total bilirubin, and blood group match were risk factors for graft survival (Table 3). Univariable analysis of the operative factors revealed that anhepatic time, blood transfusion, and blood loss were risk factors for graft survival. Univariable analysis of the recipient factors showed that blood group type, pretransplant AFP level, MELD score, Milan criteria, tumor number, tumor burden, and early graft dysfunction were risk factors for graft survival (Table 3). Using multivariable analysis, we observed that intraoperative blood transfusion (HR: 1.820; P = .042) and early graft dysfunction (HR: 3.240; P = .041) were the only factors associated with poor graft survival.
In light of cultural barriers to acceptance of DBD donation, the Chinese government and the Chinese Society of Organ Transplantation developed guide-lines for “Donation after Citizen’s Death” program, which defined 3 types of donations, namely, DBD, DCD, and DBCD, to be medically, legally, and ethically accepted.3,4
Compared with DBD LT, DCD LT procedures have been linked to higher incidences of biliary complications and graft dysfunction, leading to inferior outcomes.7,11,24-26 However, amid growing knowledge, it has been recognized that DCD LT can lead to long-term patient survival and graft survival rates that are comparable to rates with DBD LT.27 These findings can be attributed to robust donor and recipient selection and an understanding of relevant risk factors, which could improve the outcomes along the entire course. For example, graft survival rates were similar for DBD and DCD LTs from donors younger than 40 years old with CIT of less than 8 hours and WIT of less than 30 minutes that were transplanted into recipients who were admitted from their home, were younger than 60 years old, and who had serum creatinine levels of <2 mg/dL.11,26 In our center, we adhere strictly to the aforementioned donor and recipient factors, which allow for a better outcome.
In addition, machine perfusion has shown the potential to improve DCD function and reduce the risk of biliary complications. Two main types of machine perfusion are currently used in clinical settings: hypothermic machine perfusion and normot-hermic machine perfusion, which has a special form called normothermic regional perfusion.28 Many animal studies of both approaches have revealed their advantages over traditional cold storage in terms of reduced hepatocellular injury, superior graft function, lower sinusoidal endothelial dysfunction, better bile production, and superior survival after LT. In the past few years, more clinical studies on machine perfusion in humans have been published, and the strongest clinical evidence has been based on cohort studies.28 Although few clinical trials have been completed, more trials are on their way to being completed. The first clinical trial on normothermic machine perfusion showed that it led to a reduction in graft injury with less early allograft dysfunction and much greater organ utilization compared with traditional cold storage.29 Likewise, the first clinical trial on hypothermic machine perfusion showed that it led to a reduction in biliary complications using DCD allografts.30
Patients with HCC are more likely to be ideal candidates for DCD LT.14-16,31 However, controversy remains in the literature regarding the posttransplant outcomes of HCC patients who receive DCD LT. The results of our study showed that the 3-year patient survival rates for the DCD and DBD cohorts were similar. Likewise, the 3-year graft survival and recurrence-free survival rates for these cohorts were not significantly different. These results are in accor-dance with earlier studies that found no differences in patient survival,20-22,31-34 graft survival,22,31 or recurrence-free survival20-23,31,32 rates for DCD and DBD LTs. Postoperative complications were also comparable in both cohorts. Only 3 studies skewed away from these findings and found that DCD LT compared with DBD LT had inferior patient survival and graft survival rates and a higher rate of postoperative complications, such as ischemic cho-langiopathy, for which retransplant is required.17,23,25
Croome and colleagues17 showed that severe ischemia-reperfusion injury, which occurs more quickly in patients who receive DCD allografts than those who receive DBD allografts, could promote HCC growth and increase HCC recurrence rates. Silverstein and colleagues23 found that DCD LT compared with DBD LT was linked with a lower patient survival rate in those with higher risk of recurrence and a lower graft survival rate; however, HCC recurrence rates were similar for both DCD LT and DBD LT. The inferior patient survival and graft survival rates in these 2 studies could be related to poor graft quality and unreported complications in the first 6 months after LT.
Some authors, however, have attributed dif-ferences in DCD and DBD LTs outcomes to the oncological features of HCC rather than donor allograft factors.31 For example, Wallace and colleagues35 found that survival rates in HCC and non-HCC patients did not differ in the first months after surgery. After that, the survival rate of HCC recipients declined, and their graft dysfunction and mortality rates were higher than those of non-HCC recipients. They concluded that HCC-related death was the most important explanation and that HCC recurrence was the most important factor.
Interestingly, the present study’s multivariable analysis results showed that the only independent risk factors for graft survival were intraoperative RBC transfusion (HR = 1.820; P = .042) and early graft dysfunction (HR = 3.240; P = .041). Similarly, Jiménez-Romero and colleagues36 found that intraoperative RBC transfusion was an independent risk factor for patient survival and graft survival.
The concept of DBCD was developed to increase the donor pool, and it has been widely used in many Chinese transplant centers.5 In a DBCD porcine model, pathological characteristics, including well-organized hepatocyte cords, mild hepatocyte edema and vacuolization, and less parenchymal necrosis, were found to be comparable to those of DBD allografts.37 To the best of our knowledge, the present study is the first to compare DBCD, DBD, and DCD allograft outcomes in HCC patients who underwent LT. The results showed that survival and complication rates of DBCD LT and DBD LT are comparable. Similar findings have been reported in previous studies.38,39
The present study had some limitations. First, the retrospective nature of the data makes this study prone to the typical bias associated with this type of design. In addition, because the data were collected from a single center and involved shorter follow-up than follow-up in other studies, prospective studies from multiple centers are highly recommended to confirm our findings.
Our present study showed that similar outcomes can be achieved for DBD, DCD, and DBCD LTs in HCC patients who underwent LT, especially when strict donor selection is applied.
Volume : 21
Issue : 8
Pages : 664 - 670
DOI : 10.6002/ect.2023.0119
From the 1Department of Hepatobiliary and Pancreatic Surgery, The First Affiliated Hospital, Zhejiang University School of Medicine, Hangzhou, China; the 2Department of Hepatobiliary and Pancreatic Surgery, Shulan (Hangzhou) Hospital, Hangzhou, China; and the 3NHC Key Laboratory of Combined Multi-organ Transplantation, the 4Key Laboratory of the Diagnosis and Treatment of Organ Transplantation, Research Unit of Collaborative Diagnosis and Treatment For Hepatobiliary and Pancreatic Cancer, Chinese Academy of Medical Sciences, and the 5Key Laboratory of Organ Transplantation, Research Center for Diagnosis and Treatment of Hepatobiliary Diseases, Zhejiang Province, Hangzhou China
Acknowledgements: The authors have not received any funding or grants in support of the presented research or for the preparation of this work and have no declarations of potential conflicts of interest.
Corresponding author: Shusen Zheng, Department of Hepatobiliary and Pancreatic Surgery, The First Affiliated Hospital, Zhejiang University School of Medicine, Hangzhou 310003, China
Phone: +86 571 87236570
Table 1. Donor and Recipient Characteristics According to Liver Transplant Donor Allograft Type
Figure 1. Patient, Graft, and Recurrence-Free Survival After Liver Transplant According to Allograft Type
Table 2.Complications after Liver Transplant According to Donor Allograft Type
Table 3. Univariable Cox Regression for Risk Factors of Graft Survival After Liver Transplant