Studies comparing adult living donor liver transplantation to deceased donor liver transplantation have focused on post-transplant survival. Our aim was to focus on the impact of living donor liver transplant on waiting time mortality and overall mortality. We analyzed the affect of living donor liver transplantation on waiting time mortality and overall mortality (from listing until last follow up) in a cohort of 116 transplant candidates. Fifty-eight candidates who had individuals present as potential living donors (volunteer group) were matched by MELD score to 58 liver transplant candidates who did not have individuals present as a potential living donor (no volunteer group). Twenty-seven percent of candidates in the no volunteer group and 62% of candidates in the volunteer group underwent liver transplantation, p = 0.0003. One-year waiting list mortality for the volunteer group and no volunteer group was 10% and 20%, respectively, p = 0.03. Patient survival from the time of listing to last follow up was similar between the two groups. In our study group, living donor liver transplantation is associated with a higher rate of liver transplantation and lower waiting time mortality. In the era of living donor liver transplantation, estimates of patient survival should incorporate waiting time mortality.
Increasing waiting time for liver transplantation owing to the shortage of organs has led to the development of alternative strategies to treat patients with end stage liver disease. More than 16 000 patients are listed for liver transplantation and only approximately 30% of these individuals will be transplanted (1). Thus, there is an immediate need to find alternative treatment strategies for the majority of individuals awaiting transplantation.
Adult living donor liver transplantation (LDLT) evolved, in part, as a response to the organ shortage and mortality associated with waiting for liver transplantation. Individuals with cirrhosis and ascites have up to 50% 1-year mortality and those with hepatocellular carcinoma have a mortality of up to 20% (2). Living donor liver transplantation offers patients the opportunity to undergo transplantation before developing complications associated with cirrhosis and portal hypertension.
The introduction of LDLT in pediatrics over a decade ago was accompanied by controversy, which has been rekindled by the highly publicized death of an adult-to-adult living donor (3). Proper comparison of LDLT to deceased donor transplantation must include the risk of waiting list mortality, which is a complication inherent in deceased donor transplantation, but is unlikely in living donation. On the other hand, it is likely that a whole liver is superior to part of a liver in most cases, and therefore justification of the extra difficulty of LDLT must quantify the advantage gained in both timing of and access to transplantation using this modality.
The past 5 years have represented the earliest phase of adult-to-adult LDLT, and the primary focus has been to determine the clinical efficacy and safety of the procedure (4). While numerous reports have catalogued post-transplant outcomes and surgical complications for recipients of LDLT (5–11), the impact of LDLT on mortality while patients are listed (waiting time mortality) is not known. Studies that analyzed the fate of patients on the waiting list failed to correlate waiting time with mortality because patients listed preemptively for transplant with early liver disease do not die while on the list (12). Nonetheless, for any given patient with decompensated cirrhosis, increased waiting time for liver transplantation is associated with increased mortality (13). A better measure of the overall impact of LDLT on survival may be measured from the time of listing instead of from the time of transplant.
A measure of efficacy of LDLT on waiting time mortality would be to follow patients with similar severity of illness from the time of listing. An intention-to-treat estimate could be determined by categorizing patients into those with living donor volunteers, but who may not necessarily donate, and patients without living donor volunteers. A study from Hong Kong of 152 liver transplant candidates reported lower waiting time mortality and a greater transplantation rate in their group of patients with potential living donor volunteers compared with patients without donor volunteers (14). However, because significant differences in organ allocation and patient population, such as etiology of liver disease and prevalence of hepatocellular carcinoma, exist between the United States and Hong Kong these results may not be generalizeable to a U.S. cohort.
Based on our hypothesis that patients with LDLT experience a decreased risk of dying before transplant, we studied a cohort of adult patients at our center that were listed for liver transplantation. We compared waiting time survival between liver transplant candidates with and without living donor volunteers. We suggest that to capture the full benefits of LDLT survival estimates should start from the time of listing to last follow up. This approach would capture any benefits on waiting time mortality. The current standard is to report post-transplant survival, which would not capture benefits on waiting time survival.
Materials and Methods
The cohort of adult liver transplant candidates in our study were listed for liver transplantation at our center between March 1, 1999 and March 30, 2002. All patients met minimal listing criteria and were listed on the United Network for Organ Sharing (UNOS) waiting list with regular follow up at the transplant center during the pretransplant period. Transplant candidates who were UNOS status 1 or 2A and in the intensive care unit at the time of initial presentation were not presented with the option of adult living donor liver transplantation because of poor outcomes associated with critically ill liver patients after LDLT. As a result, all subjects in this study were either UNOS status 2B or 3 at the time of listing. All patients in this study were presented with the option of LDLT at the time of listing for liver transplantation by the hepatologist who was caring for the patient. To be considered for LDLT potential donors had to be at least 21 years old. Potential living donors were not solicited and had to initiate contact by calling our center requesting to be seen for LDLT evaluation as a donor for a specific candidate. All LDLTs were right lobe donations. Follow up was complete through December 2002.
Liver transplant candidates who had to individually call our center to be evaluated for living donor liver transplantation formed the volunteer group. Liver transplant candidates who did not have to individually call our center for living donor liver transplantation formed the no volunteer group. We matched patients in the volunteer group and no volunteer group by MELD score at the time of listing, as MELD score correlates with mortality in patients with end stage liver disease.
Definitions:‘Waiting time mortality’ was defined for these analyses as mortality that occurred while on the liver transplant list. Patients were censored at the time of transplant or last follow up.
‘Post-transplant mortality’ was defined as mortality that occurred from the time of transplantation to last follow up. Patients who were alive were censored at last follow up.
‘Overall mortality’ was defined as mortality that occurred from the time of listing to last follow up. Patients who were alive were censored at last follow up.
Proportions were compared with Fisher's exact test and means were compared with an unpaired Student's t-test. Kaplan-Meier curves were constructed and compared with the log-rank test. Multivariate analysis was performed using logistic regression to test the association between mortality on the waiting list and having a living donor volunteer in the presence of other variables. Data were analyzed in STATA 6.0 (STATA Corporation, College Station, TX).
Table 1 shows the clinical features of liver transplant candidates with and without living donor volunteers. The mean MELD score in both groups was 11. The two groups were similar in age, etiology of liver disease, and time on the waiting list. Living donor liver transplantation recipients were more likely to be Hispanic compared with individuals without potential living donors (p = 0.03). None of the individuals with alcohol-related liver disease had a living donor volunteer. There were more deaths while on the waiting list in the group without volunteers, p = 0.04 (Table 1).
Table 1. Characteristics of liver transplant candidates with and without living donor volunteers
No volunteers (n = 58)
Volunteers (n = 58)
1p = 0.03 for the difference between living donor recipients and deceased donor recipients.
2p = 0.04 for the difference between the groups with and without donor volunteers.
Etiology of liver disease
Hepatitis C (%)
Hepatitis B (%)
UNOS status at listing
MELD score at listing
Hepatocellular carcinoma (%)
Time on waiting list (days)
290 + 241
301 + 247
Deaths on waiting list2
Outcomes of candidates in the volunteer group and no volunteer group are shown in Figure 1. Forty-five percent of candidates who had potential living donor volunteers have had living donor liver transplantation and 17% have undergone deceased donor liver transplantation after a mean of 479 days on the waiting list. Twenty-seven percent of candidates in the no volunteer group have undergone deceased donor liver transplantation after a mean of 388 days on the waiting list. Twenty-seven percent of patients in the no volunteer group and 62% of patients in the volunteer group underwent liver transplantation, p = 0.0003. Mean MELD score at the time of transplant in living donor recipients and deceased donor recipients was 12 and 20, respectively, p = 0.0006.
Eleven patients in the no volunteer group and three patients in the volunteer group died on the waiting list. One-year waiting time mortality for the volunteer group and no volunteer group was 10% and 20%, respectively, p = 0.03, by the log rank test (Figure 2). The causes of death are listed in Table 2. The most common causes of death were liver failure and variceal bleeding.
Table 2. Causes of death in transplant candidates with and without living donor volunteers
No volunteer group (n = 58)
Volunteer group (n = 58)
We analyzed survival from the time of listing to last follow up to determine if the survival benefit in the volunteer group persisted through liver transplantation. One-year survival in the volunteer group and no volunteer group was 90% and 83%, respectively, p = 0.58. One year survival in recipients who underwent living donor liver transplantation and deceased donor liver transplantation was 85% and 91%, p = 0.58. The mean MELD score at the time of transplant in the group that underwent LDLT (n = 26) and deceased donor liver transplant (n = 26) was 12 and 20, respectively, p = 0.001.
Results of the multivariate analysis are shown in Table 3. Patients with hepatocellular carcinoma were excluded from the analysis because there were no deaths in patients with a living donor volunteer who had hepatocellular carcinoma. The risk of death on the waiting list was increased in the group without living donor volunteers, OR = 6.3 [1.3, 31.0], p = 0.025. Age and MELD score at the time of listing were also associated with waiting time mortality.
Table 3. Multivariate analysis of factors associated with mortality on the waiting list
Odds ratio [95% CI]
No living donor volunteer
6.3 [1.3, 31.0]
1.1 [1.0, 1.2]
MELD at listing
1.2 [1.0, 1.4]
1.8 [0.5, 7.3]
0.6 [0.1, 3.9]
Diagnosis: hepatitis C
2.3 [0.5, 10.5]
Blood type O
0.8 [0.2, 2.9]
1.8 [0.3, 9.5]
Our study demonstrates that after controlling for MELD score at the time of listing, candidates with potential living donor volunteers have a lower waiting time mortality compared with candidates without potential living donor volunteers. A major benefit of LDLT compared with deceased donor transplantation is the ability to control the time of transplantation. Thus, LDLT may be performed before patients become too ill for transplant, develop advanced hepatocellular carcinoma or die. For these reasons we analyzed survival from the time of listing, in addition to that from the time of transplant. Studies have analyzed post-transplant survival, comparing outcomes after living donor and deceased donor liver transplantation (5–11), but benefits derived while awaiting liver transplantation may not be captured if waiting time mortality is not analyzed. Although waiting time mortality was lower in the volunteer group, we did not detect a significant difference in mortality from time of listing to last follow up between the volunteer and no volunteer groups. There are several reasons why we may have found a higher, but not significant, survival advantage in the volunteer group. Our sample size may have been too small or follow up too short to have adequate power to detect a significant difference. Additionally, our cohort includes patients who underwent living donation at the inception of the program. Our analysis incorporated the early experience with LDLT, ‘learning curve’, and including these early cases would bias our results against LDLT. We had too few patients to conduct a meaningful subgroup analyses, excluding patients from the early part of our program. As technical advances and patient selection improve with LDLT we may observe better survival from the time of listing through follow up between the candidates with and without potential living donors. As more experience is gained with LDLT future studies should analyze changes in complication and survival rates in LDLT recipients.
Deceased donor recipients were sicker at the time of transplant, as measured by MELD score, compared with living donor recipients. However, severity of illness did not seem to confer a post-transplant survival disadvantage when compared with LDLT. Recipients of living donation receive partial grafts, and thus, although they may be less ill at the time of transplant, they receive less hepatic mass compared with deceased donor recipients. Furthermore, the increased complexity and surgical morbidity of LDLT has been well described in most series. These increased surgical risks inevitably offset some of the benefits LDLT confers on waiting time mortality. The ultimate fulfilment of the potential benefit of LDLT will depend upon improvements in surgical and medical outcomes of LDLT. In addition, other factors may come into play after liver transplantation that offset any pretransplant survival advantage, such as reduced graft size or recurrent hepatitis C (15). Furthermore, our results provide an intention-to-treat estimate. That is, the donor volunteer group included patients who had an individual present for living donor evaluation, but may not have followed through with donation for medical or psychosocial reasons. If donor selection improves and the proportion of donor volunteers who undergo right donor hepatectomy increases, there should be an even greater survival advantage in the donor volunteer group.
As waiting times increase for liver transplantation so does waiting time mortality, although others have not found an association between waiting time and mortality (12,16). A study of data from the waiting list of three liver transplantation programs demonstrated a twofold increase in waiting time mortality, as waiting time increased from 44 days to 108 days (13). In the U.S. the average waiting time is 468 days with waiting time mortality as high as 20% (2). Hispanic-Americans wait longer than Caucasian patients (16). We have previously shown that Hispanic patients are more likely to have a potential living donor and to undergo LDLT (17). Thus, LDLT may offer more pronounced survival benefits to subgroups of patients who have long waiting times on the transplant list. Finally, our study does not address quality of life advantages that recipients in the LDLT group may have, as the duration of time spent with end stage liver disease is shortened compared with those awaiting deceased donor transplantation.
A study of 152 liver transplant candidates reported that survival rates were higher in the group of patients with donor volunteers (14). The waiting time mortality in the no volunteer group and volunteer group was 30% and 15%, respectively; slightly higher than what we reported. Three-year survival was 21% higher in the group with volunteers, which may partly be owing to the high rate of living donor transplantation in the group with volunteers (68% of volunteers underwent LDLT). The high volunteer transplantation rate seen in their study was higher than the rate we found in our group. Other important differences between our study (and other U.S. transplant populations) and theirs is that most of their patients were transplanted for end stage liver disease from hepatitis B and a greater proportion of their patients had hepatocellular carcinoma at the time of listing. Therefore, results from their study may not be generalizeable to U.S. populations owing to differences at the time of listing, etiology of liver disease, and high rates of donors who proceed with LDLT. However, despite differences in study populations and organ allocation between our study and the above study, the findings from both studies support survival benefits in the group with living donor volunteers.
Most of the patients in our study were transplanted before the MELD system was instituted. However, since MELD has been implemented patients at our center are sicker at the time of transplant, supporting that our results would be applicable under the MELD allocation system. The average MELD score at transplant for our recipients of deceased donor organs is 27 since MELD has been implemented and the average MELD score at transplant in our study cohort was 20 in deceased donor recipients. Thus, since MELD has been implemented our patients have had more advanced liver disease with higher MELD scores, supporting that the benefits of having a living donor volunteer on waiting time mortality should be greater since MELD implementation. In the era of MELD, the group of patients who may benefit from LDLT is those with malnutrition and refractory ascites with preserved synthetic function whose MELD score may not adequately reflect their severity of illness. Studies comparing waiting time mortality from before and after MELD implementation should provide an estimate of the impact of changes in organ allocation on survival rates in patients awaiting liver transplantation.
In summary, individuals listed for liver transplantation who have individual volunteers as potential donors for LDLT have lower waiting time mortality than individuals listed for liver transplantation without living donor volunteers. This survival benefit did not persist through transplantation, which may be owing to slightly lower post-transplant survival in living donor recipients, small sample with insufficient power, or insufficient length of follow up in our cohort. In the era of LDLT, we propose that survival analyses of liver transplantation incorporate analyses starting from the time of listing to capture any potential survival advantage during the waiting period. As technical advances improve post-transplant outcomes after LDLT, additional survival and quality of life benefits may be seen, especially in groups of patients with long waiting times.
LDLT, adult living donor liver transplantation; MELD, model for end stage liver disease.