Effect of Elderly Donors on Long-Term Outcomes of Liver Transplantation: A Retrospective Cohort Study
Objectives: Transplant centers are increasingly using livers from donors aged >60 years, but long-term outcomes remain unclear.
Materials and Methods: We collected data from 957 adult deceased donor liver transplant recipients (conducted at our center from January 2015 through October 2023). The cohort was divided into early, middle, and late transplant era groups, each covering a 3-year interval based on transplant date. We compared 183 donors aged ≥60 years (older group [average age 64.9 ± 3.4 y]) versus 774 donors aged 18 to 59 years (young group), which we also compared after propensity score matching. We compared outcomes of older organs transplanted into older recipients versus older organs transplanted into young recipients. We conducted multivariate logistic regression analysis to identify potential factors influencing liver transplant outcomes.
Results: Over the time period, average donor age increased significantly, cold ischemia time and warm ischemia time decreased, postoperative complications decreased among older donor liver transplant recipients, and cumulative recipient and graft survival rates improved. The older donor group had fewer male donors, more donors with diabetes, and lower average donor liver weight versus the young donor group. Recipients in the older donor group had significantly lower body mass index and higher prevalence of alcoholic cirrhosis versus recipients in the young donor group. Three-month postoperative mortality rate was significantly higher in the older donor group. Subgroup analysis showed significant differences in outcomes between older donors assigned to older recipients and older donors assigned to young recipients, including higher incidence of major postoperative complications and lower 3-month mortality rate. Recipient-donor age differences, early transplant era, and postoperative peak bilirubin level may affect differences.
Conclusions: Livers from older donors can yield positive outcomes for young and old recipients. Use of organs from older donors is a viable approach to mitigate organ shortages, including for younger recipients.
Key words : Older donors, Organ, Transplant
Introduction
Liver transplantation is the gold standard treatment for patients with end-stage liver disease, with a 5-year survival rate of 73.6% and more than half of recipients living for >20 years.1,2 Severe organ shortages have led to increasing reliance on marginal donors. In liver transplant, there is no standard definition for marginal donors, but previous studies have identified donors over 60 years old as high-risk because of their poorer prognoses, classifying them as marginal donors.3 Although livers from older donors are more susceptible to damage during transplant, the ongoing organ shortage has driven the increased use of these donors, particularly in Europe.4 With growing experience in transplant with older donors, liver transplant centers worldwide have reported positive outcomes.5-10 China, a country with a high incidence of liver disease, still lacks sufficient research data on the use and effects of livers from older donor on transplant outcomes. We aimed to summarize recent use of older donors at our center, analyze the characteristics and outcome of older donor liver transplants, explore factors affecting poor outcome, and discuss strategies for optimizing allocation of organs from older donors.
Materials and Methods
Study objectives: We collected data on 1019 adult patients who underwent deceased donor liver transplant at Beijing You’an Hospital affiliated with Capital Medical University from January 1, 2015, to October 31, 2023. We excluded 62 patients because they were aged ≤18 years, had split liver transplants, retransplants, or multiple organ transplants. Ultimately, we had 957 patients in the initial cohort, which we categorized into different groups based on study objectives.
Research design: To investigate the effects of older donors on liver transplant outcomes across different eras, we established cohorts based on 3-year intervals: 2015-2017 (early transplant era), 2018-2020 (middle transplant era), and 2021-2023 (late transplant era). To examine the effects of donor age on transplant outcomes, recipients were divided into 2 groups: the older donor group (donor age ≥60 y; n = 183) and the young donor group (donor age 18-59 y; n = 774). We compared clinical characteristics and outcomes between groups. To minimize potential confounding factors in prognostic analysis, propensity score matching (PSM) was performed, resulting in 106 patients in both the older and young donor groups. To study the effects of older donors, recipients were divided into 2 subgroups: elderly (older) donor to young recipient (EY group) and elderly (older) donor to elderly (older) recipient (EE group), based on whether the recipient was over 60 years old. The study aimed to guide the optimal allocation of organs from older donors based on our findings. Finally, we used multivariate logistic regression analysis to identify potential factors influencing the outcomes of liver transplants from older donors.
Data collection: For recipients, we collected the following preoperative variables: demographics, clinical history, laboratory tests, Model for End-Stage Liver Disease (MELD) score, and Child-Pugh score. For surgical variables, we collected cold ischemia time (CIT), warm ischemia time (WIT), liver transplant operation time, duration of the anhepatic phase, intraoperative blood loss, and blood transfusion volume. For postoperative data, we collected length of hospital stay, intensive care unit stay, ventilator support, renal replacement therapy, major complications, and in-hospital mortality. For donors, we collected demographic characteristics, body mass index (BMI), past medical history, and laboratory test results. We also collected liver graft weight and calculated graft-to-recipient weight ratio. All grafts had been biopsied to assess for steatosis. Liver transplant recipients at our center had been routinely followed up after discharge, including having assessments of general conditions, relevant laboratory tests, and postoperative complications. We collected information on patient and graft survival, with cause and time of death noted for deceased patients. Overall survival was defined as the time from liver transplant to the patient’s death or end of study.
Statistical analyses: We used SPSS version 24.0 (IBM SPSS Inc) statistical software for statistical analysis. We expressed quantitative data as mean ± SD. We used independent sample t test and one-way analysis of variance to compare quantitative data. We presented qualitative data as frequency and rate and used chi-square tests and Fisher exact tests to compare qualitative data. We used the Kaplan-Meier method followed by log-rank test to calculate survival rate and draw survival curves. We used multivariate logistic regressive analysis to determine risk factors associated with survival in older donor liver transplant. Propensity score matching was performed using a nearest-neighbor matching algorithm. Variables used to calculate the propensity score included recipient age, recipient sex, transplant era, MELD score, Child-Pugh score, liver graft weight, donor sex, moderately to severely steatotic liver grafts, CIT, and WIT; unmatched patients were excluded from the analysis. P < .05 was considered significant.
Results
Comparison of use and outcomes of older donor liver transplant in different transplant eras: This study included 957 liver transplant cases, with 183 cases (19.1%) involving donors aged ≥60 years old (older donors). Over the transplant eras, the mean age of donors significantly increased over time (46.2 ± 12.9 vs 49.2 ± 11.1 vs 50.3 ± 11.7 y; P = .002). The utilization rate of older donors aged ≥60 years also increased (14.9% vs 17.5% vs 21.4%; P = .167), although this increase was not significant. However, the proportion of liver transplants using donors aged 70 and older increased significantly (0.0% vs 0.3% vs 3.5%; P = .001). As the transplant eras progressed, we noted significant reductions in CIT, WIT, anhepatic phase, liver transplant operation time, and intraoperative blood loss for older donor liver transplants (Table 1). The incidence of postoperative complications also significantly decreased in these cases (Table 1).
During long-term follow-up of 183 older donor liver transplant recipients, 2 cases were lost, with a median follow-up time of 22.3 months. During this period, 49 recipients died. The 1-year recipient and graft survival rates for all older donor liver transplant recipients were 79.7% and 81.4%, respectively, and the 3-year recipient and graft survival rates were 70.0% and 72.0%, respectively. Kaplan-Meier survival analysis showed significant improvements in cumulative recipient and graft survival rates over different transplant eras (Figure 1). In each era, the survival rates for liver transplant recipients and grafts in the older donor group were comparable to those in the young donor group (Figure 2, Figure 3, Figure 4).
Comparison of characteristics and outcomes between older donor and young donor groups: Characteristics of donors and recipients in the 2 groups are presented in Table 2. The proportion of male donors was significantly lower in the older donor group compared with the young donor group. Mean recipient-donor age were significantly different between the groups. Donors in the older group had a significantly higher prevalence of diabetes mellitus than those in the young donor group. Mean weight of liver grafts was significantly lower in the older donor group, although the graft-to-recipient weight ratio did not significantly differ between groups.
For recipients, the BMI was significantly lower in the older donor group compared with the young donor group. Prevalence of alcoholic cirrhosis was significantly higher in the older donor group. Baseline hemoglobin levels were slightly lower in the older donor group than in the young donor group.
Surgical parameters did not significantly differ between the 2 groups. Peak values of major postoperative laboratory tests, incidence of major postoperative complications, and in-hospital mortality were not significantly different between the groups. However, the 3-month recipient mortality rate was significantly higher in the older donor group compared with the young donor group (Table 3).
During follow-up of 957 liver transplant recipients after discharge, 10 cases were lost (2 in the older donor group and 8 in the young donor group). Median follow-up time was 25.9 months. There were 49 deaths (27.1%) in the older donor group and 180 deaths (23.5%) in the young donor group, with no significant difference in total mortality (P = .313) or in distribution of causes of death between the 2 groups (P = .196). The most common causes of death in the older donor group were multiple organ failure (59.2%) and sepsis (16.3%), with multiple organ failure (46.1%) and recurrence of the primary disease (24.4%) being the most common in the young donor group. Kaplan-Meier survival analysis showed no significant differences in cumulative recipient survival and graft survival between the 2 groups (Figure 5).
Comparison of characteristics and outcomes between the older donor and the young donor groups after propensity score matching: To minimize potential confounding factors in the prognostic analysis, we used PSM to match the older donor group with the young donor group. After PSM, we had 106 cases in each group. Comparisons of baseline characteristics between the 2 groups after PSM showed that no variables significantly different before matching remained significantly different after matching, except for the prevalence of hepatitis C virus, which was lower in the older donor group compared with the younger donor group (Table 4). After PSM, no significant differences were shown between older and young donor groups in peak values of postoperative laboratory tests, duration of postoperative ventilator support, length of intensive care unit stay, total length of hospital stay, incidence of major complications, in-hospital mortality, and 3-month mortality (Table 5). Kaplan-Meier survival analysis indicated no significant difference in recipient and graft survival rates between the older donor group and the young donor group after PSM (Figure 6).
Results of subgroups matched by recipient age: In subgroup analysis, we examined the distinct prognoses when older donors were allocated to either older recipients (EE group) or young recipients (EY group). The findings revealed a notably higher incidence of postoperative major complications and 3-month mortality after surgery in the EE group compared with the EY group (Table 6). Kaplan-Meier survival analysis indicated a significantly lower cumulative recipient survival rate in the EE group compared with the EY group, although no significant difference in cumulative graft survival rate was shown between the EY and EE group (Figure 7).
Prognostic factors for older donor liver transplant: Multivariate logistic regression analysis was conducted, incorporating 16 factors with P < .05 from the univariate analysis. Analysis confirmed that early transplant era, larger recipient-donor age difference value (recipient-donor age difference value = recipient age minus donor age), and postoperative peak bilirubin level were independent predictors of death in older donor liver transplant recipients (Table 7). Receiver operating characteristic curve analysis (Figure 8 and Figure 9) indicated that recipient-donor age difference value had predictive value for death of older donor liver transplant recipients (P = .015). The predictive cut-off value was -8.50 years, with a predictive specificity of 68.9% and sensitivity of 55.1%. Similarly, postoperative peak bilirubin level had predictive value for death of older donor liver transplant recipients (P < .001). The predictive cut-off value for postoperative peak bilirubin was 89.15 μmol/L, with predictive specificity of 68.9% and sensitivity of 77.6%.
Discussion
As the shortage of donor livers persists, the once-considered high-risk practice of liver transplant from older donors is increasingly prevalent in transplant centers worldwide. Data from a study in the United States indicated a steady rise in the utilization rate of donors over aged >60 years since 1990, stabilizing over the past decade.11 In Europe, where experience with older donor liver transplant is more extensive, the proportion of donors aged >60 years has surged from 1% in 1989 to 29% in 2009.4,12 However, limited research is available on use and outcomes of older donor liver transplant among Asian populations.
A study in China revealed a yearly increase in the utilization rate of donors aged >60 years, rising from 9.3% in 2015 to 14.5% in 2018.13 Concurrent with these findings, our study demonstrated a similar upward trend in the utilization rate of donors aged >60 years over the past decade. The proportion of older donors at our center increased from 14.9% in the early era to 21.4% in the later era, although these numbers are still lower compared with data from Europe. In addition, the average age of donors significantly rose with transplant era progression, and there was a notable increase in the proportion of liver transplants performed by donors aged >70 years. This trend suggested a growing inclination toward expanding the age limit for donor selection, mirroring reports from foreign transplant centers. Indeed, in recent years, numerous transplant centers have reported positive outcomes with donors over >70 years or even 80 years.14-17
Compared with young donors, the selection, maintenance, and operation of older donors have always been approached with greater caution. We have observed that, as the transplant era progressed, CIT and operation-related parameters were significantly reduced in liver transplants that had older donors, along with a notable decrease in the incidence of major postoperative complications. These findings suggest substantial advancements in transplant surgery techniques occurred, not only within our center but also across China in recent years. In our study, cumulative survival in both recipients and grafts showed significant improvements over time. Similar trends were observed at Duke Medical Center, where significant enhancements in graft and patient 5-year survival were noted over the past 15 years and during which time the gap in recipient survival between older and young grafts was considerably narrowed.11 Although our study did not observe a reduction in the survival gap between older and young donor liver transplant recipients, we found that recipient and graft survival in older donor liver transplants were comparable to those of young donor liver transplant across all transplant eras.
Effects of donor age on liver transplant outcomes have been evaluated for decades, with ongoing debate regarding the safety and feasibility of use of older donors. Some studies have indicated that increasing donor age poses a risk factor and can lead to decreased recipient survival rates, whereas other studies have reported that liver transplant using older donors can yield survival benefits similar to those from young donors. In recent years, there have been successful instances of use of donors >70 years old or even >80 years old for liver transplants.5-10,14-19 From a pathophysiological perspective, the primary changes in the liver from an aged donor involve a reduction in tissue mass and blood flow.20,21 However, fundamental liver function typically remains intact in older livers, with the effects of aging on liver function not as pronounced as those shown in kidney or heart.21
In our study, we observed that, although the average weight of donor livers in the older donor group was significantly lower than that in the young donor group, the proportion of severely steatotic liver grafts was similar between the 2 groups. This finding may be linked to the increased prevalence of dyslipidemia and nonalcoholic fatty liver disease among the young population in recent years. Moreover, the average age of older donors in our study was only 64.9 years old, with a relatively small number of donors >70 years old; thus characteristics related to liver aging were not prominently observed. In addition, both our study and previous research have identified common characteristics among older donors that differ from those of young donors, such as a relatively lower proportion of males, a higher prevalence of diabetes, and cerebrovascular accidents as the primary cause of death.15,22 Prior studies have suggested that older donors often exhibit a higher prevalence of diabetes and obesity, which could potentially have adverse effects on transplant outcomes using their livers.22
We observed that CIT and surgery-related parameters in the older donor group were comparable to those in the young donor group, and the incidence of major complications and length of intensive care unit stay in the older donor group did not significantly differ from those in the young donor group. This conclusion held even after PSM. Despite previous studies that indicated that organs from older donors were associated with a higher incidence of ischemic biliary tract complications and hepatic artery thrombosis, numerous studies have supported the notion that use of older donors may not increase the risk of postoperative complications.23,24 For example, a study by Caso-Maestro and colleagues demonstrated that donors over aged >70 years did not elevate the risk of postoperative complications in recipients, particularly primary nonfunctional complications, vascular complications, and biliary complications.15 Similarly, Hu and colleagues found that the incidence of primary nonfunction, delayed graft function, biliary tract and vascular complications, rejection, and postoperative infection after liver transplant in the older donor group (>60 years old) was comparable to incidence among young donors.13 Researchers have suggested that the overall graft function of older donors remained well preserved in the early stages after liver transplant,22,25 which may partially explain why the effect of donor age on major complications after liver transplant is not significant.25
Previous studies have indicated higher failure and death rates among older donor transplant recipients.26,27 However, our study revealed that, under conditions of similar disease severity between the older donor group and the young donor group, the long-term outcomes of recipients in the older donor group were comparable to outcomes of the young donor group, despite a higher 3-month death rate in the older donor group. The graft survival rate and recipient survival rates did not significantly differ from those of the young donor group. This conclusion remained consistent even after PSM, which reduced interference from confounding factors in both groups. In fact, this finding aligns with an increasing number of recent studies on liver transplant that examined donors aged >60 or even >70 years, indicating that older donor liver transplants can achieve favorable long-term outcomes similar to those of young donor liver transplant. Rigorous recipient selection criteria have been crucial in achieving these outcomes, and even donors aged >80 years have demonstrated results comparable to those obtained from young donors in liver transplant.5-10,14-19,28,29
Research from the Chinese organ donation system has also shown that overall postoperative survival and disease-free survival rates of patients with hepatocellular carcinoma receiving older donor liver transplant were not significantly different from those of young donors.13 Several potential factors may have contributed to these outcomes, including stringent donor selection, reduced cold ischemia time, optimized recipient selection, improvements in surgical techniques, and advancements in perioperative care. Some studies have suggested that the use of marginal donors such as older donors in liver transplant recipients with high MELD scores could negatively effect recipient outcomes, particularly when MELD scores exceed 20.30 However, our study found that the average MELD score of recipients in both the older donor group and the young donor group exceeded 20 points, indicating that the favorable outcomes of the older donor group were not attributed to the recipient MELD score and the high MELD score did not significantly affect the outcome of liver transplant in the older donor group. In addition, research has indicated that high MELD score of recipients does not adversely affect the outcome of liver transplants using older donors.31 Another study that examined the effect of donor age on graft survival rates of recipients stratified by MELD score found no significant differences in graft survival rates among recipients with different MELD scores categorized by donor age.18
Although some studies have indicated that donor age independently contributes to poor outcomes in liver transplant, there is considerable debate regarding whether donor age serves as a predictive factor for outcomes in liver transplant. In our study, we did not find donor age to be a determinant of outcome in older donor liver transplant.12,22,30,32 However, we did observe that the recipient-donor age difference value may influence the outcome of older donor liver transplant. Specifically, a positive association was noted between larger recipient-donor age difference value (recipient-donor age difference value = recipient age minus donor age) and the risk of death in recipients of older donor liver transplant. We identified a predictive cut-off value for recipient-donor age difference value at -8.50 years, indicating an increased risk of death when the recipient-donor age difference value exceeded this threshold. Nonetheless, it is worth noting that the specificity and sensitivity of this index were relatively low. Models have been developed by prior research for stratifying transplant risks based on donor-related factors, such as the donor risk index, the donor and recipient age model (which predicts transplant risk when the sum of donor and recipient age is ≥120 years), and the DR-MELD model (which predicts transplant risk through the multiplication of donor and recipient age and MELD score).21,33,34 Going forward, it is imperative to develop multifactorial models that can predict transplant outcomes by incorporating donor criteria, recipient characteristics, and novel functional biomarkers, thereby facilitating the optimization of recipient selection.
In addition to age, multivariate logistic regression analysis revealed that early transplant era and postoperative peak bilirubin levels were factors influencing the outcome of liver transplant in older donors. The predictive cut-off value for postoperative peak bilirubin level was determined to be 89.15 μmol/L, albeit with relatively low sensitivity and specificity for this predictive index. Several studies have highlighted CIT as a predictive factor for the outcome of older donor liver transplant, recommending CIT be kept within 8 hours.35,36 However, studies have had conflicting findings, suggesting that CIT, along with donor body mass index and graft steatosis, does not predict graft survival in older donor liver transplant.15 Our study indicated that CIT is not an independent risk factor for the outcome of liver transplant using older donors. This finding may be attributed to the relatively stringent control of CIT in our transplant center. Specifically, the CIT in the older donor group was relatively short, with a mean of 4.6 hours, and even slightly lower than that in the young donor group. Although prolonged CIT should not serve as a contraindication for the use of older grafts, given the heightened susceptibility of older liver grafts to damage from cold storage, we advocate for minimizing CIT when older grafts are used.
With regard to allocation of organs from older donors, some researchers have argued that recipient factors might exert a more significant negative influence on postoperative survival than donor factors. Therefore, marginal donors, such as older donors, have been suggested to be restricted to young recipients without additional relevant risk factors.35,37 Does donor age genuinely affect outcomes of liver transplant in recipient populations of varying ages? Our study indicated that older donors may exert a greater effect on the long-term outcome of older recipients than young recipients. The incidence of postoperative major complications and 3-month mortality in group EE was notably higher than in group EY, and the cumulative recipient survival rate was significantly lower than in group EY. Previous studies have suggested that the overall postoperative complication rate of older donor liver transplant is similar between older and young recipients, yet older recipients are more susceptible to sepsis than their younger counterparts.21,38
No standardized criterion is yet available for allocation of organs from older donors in liver transplant. Following the principle of donor allocation based on overall survival benefit, the effect of older donors on the overall life expectancy of young recipients is often considered of greater importance than the life expectancy of older recipients. Hence, in clinical practice, older donors are typically preferentially allocated to older recipients.11 Alternatively, a “disease-first” donor allocation policy, centered on the MELD score, prioritizes candidates who are of worst health regardless of donor age. Although some researchers have proposed use of organs from older donors solely for “low-risk” recipients with mild disease and no risk factors,30,39 our study, along with others, did not find a negative association between MELD score and the survival of older donor liver transplant recipients.14,21,40 Our study showed that older donors did not adversely affect long-term outcome compared with young donors. Considering the high mortality rate of patients on wait lists without prompt transplant, the use of older donors for liver transplant is deemed reasonable and can aid in mitigating the organ shortage to some extent. Our research also indicated that older donors may derive greater benefit from young recipients than older recipients. Therefore, we advocate that allocation of organs from older donors should not be confined to older recipients alone. Rather, the donor-recipient allocation policy in liver transplant should be optimized, considering various factors such as the interaction between donors and recipients and the risk of prolonged wait time for transplant.21
This study had several limitations. First, it was a single-center retrospective study. The extensive time span covered in the study introduced potential variations in samples, surgical techniques, and follow-up, leading to concerns regarding sample homogeneity. In addition, there were discrepancies in follow-up duration between the first and last cases. Second, the sample size of older donor liver transplant was relatively small, potentially resulting in data bias. To address these limitations, further data collection and larger prospective randomized controlled studies are warranted in the future to enhance the utilization of marginal organs.
References:
Volume : 23
Issue : 2
Pages : 120 - 132
DOI : 10.6002/ect.2024.0241
From the Department of general surgery, Capital Medical University Xuanwu Hospital, Beijing, 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: Dong-Dong Lin, Department of general surgery, Capital Medical University Xuanwu Hospital, Beijing, China
E-mail: ldd1231@ccmu.edu.cn
Table 1. Comparison of Surgical Variables and Short-Term Outcomes of Older Donor Liver Transplants in Different Transplant Eras
Figure 1. Overall Recipient and Graft Survival Rates of Older Donor Liver Transplant in Different Transplant Eras
Figure 2. Early Transplant Era: Comparison of Overall Recipient and Graft Survival Rates According to Donor Age
Figure 3. Middle Transplant Era: Comparison of Overall Recipient and Graft Survival Rates According to Donor Age
Table 2. Comparison of Baseline Characteristics Between Older and Young Donor Groups
Figure 4. Late Transplant Era: Comparison of Overall Recipient and Graft Survival Rates According to Donor Age
Table 3. Comparison of Surgical Variables and Short-Term Outcomes Between Older and Young Donor Groups
Figure 5. Comparison of Overall Recipient and Graft Survival Rates in Young and Older Donor Groups
Figure 6. Comparison of Overall Recipient and Graft Survival Rates in Young and Older Donor Groups After Propensity Score Matching
Table 4. Comparison of Characteristics Between Older and Young Donor Groups After Propensity Score Matching
Table 5. Comparison of Short-Term Outcome Between Older and Young Donor Groups After Propensity Score Matching
Table 6. Short-Term Prognoses of Subgroups Matched by Recipient Age
Figure 7. Comparison of Overall Recipient and Graft Survival Rates in EY and EE Groups
Table 7. Results of Univariate and Multivariate Analysis of Predictors for Recipient Survival in Older Donor Liver Transplant
Figure 8. Receiver Operating Characteristic Curve Analysis of Recipient-Donor Age Difference Value for Prediction of Death in Older Donor Liver Transplant Recipients
Figure 9. Receiver Operating Characteristic Curve Analysis of Postoperative Peak Bilirubin Level for Prediction of Death in Older Donor Liver Transplant Recipients