Journal Logo

Original Clinical Science—Liver

Assessing the Impact of Suboptimal Donor Characteristics on Mortality After Liver Transplantation: A Time-dependent Analysis Comparing HCC With Non-HCC Patients

Wallace, David MSc1,2; Walker, Kate PhD1; Charman, Susan MSc1; Suddle, Abid MD2; Gimson, Alex MD3; Rowe, Ian PhD4,5; Callaghan, Chris PhD6; Cowling, Tom PhD1; Heaton, Nigel FRCS2; van der Meulen, Jan PhD1

Author Information
doi: 10.1097/TP.0000000000002559

Abstract

INTRODUCTION

The rising incidence of hepatocellular carcinoma (HCC) has led to a marked increase in the number of patients with HCC receiving a liver transplant.1 This has put pressure on transplantation services in many countries as they struggle to cope with transplanting patients with HCC in an acceptable oncological time frame given the limited availability of donor organs.1 In response, livers with suboptimal donor characteristics are increasingly being used.2

It is unknown to what extent the increase in the number of liver transplants for HCC and the related increased use of marginal livers have affected posttransplantation outcomes. International consensus recommendations only indicate that posttransplant outcomes of patients transplanted for HCC should be comparable to those transplanted for non-HCC indications.3

A study including patients transplanted between 1988 and 2003 in a number of European countries suggested that posttransplant survival immediately after transplantation is often better in patients transplanted for HCC compared with those who had liver transplant for other reasons.4,5 However, survival in HCC patients can deteriorate later during follow-up, most likely as a result of tumor recurrence. It has been argued that the introduction of the Milan criteria—a set of tumor characteristics introduced in the late 1990s to identify HCC patients in whom liver transplantation may provide curative treatment (1 lesion with a diameter ≤5 cm, or alternatively 3 lesions each with a diameter ≤3 cm)—will have reduced tumor recurrence and in that way will have canceled the reversal of HCC’s impact on posttransplant outcomes.5-7 There has been no recent large-scale study that has empirically tested this assertion.

In the United Kingdom, the Milan criteria for listing patients with HCC for liver transplantation were expanded in response to studies that suggested that less restrictive criteria would not negatively affect cancer recurrence rates and posttransplant survival.8,9 As a result, a set of expanded criteria were formally accepted in the United Kingdom in 2008 (1 lesion with a diameter ≤5 cm, or up to 5 tumors each with diameter ≤3 cm, or 1 lesion with a diameter >5 and ≤7 cm with no evidence of tumor progression, extrahepatic spread, or new nodule formation over a 6-month period).10,11

Our aim was to examine the prognostic impact of HCC over different time periods (epochs) after liver transplantation using recent data from the Standard National Liver Transplant Registry. To correlate with the introduction of expanded selection criteria, our analysis focused on a cohort of patients who received a liver transplant between 2008 and 2016. We investigated whether the impact of HCC varied over 3 epochs of follow-up: patient survival up to 90 days was chosen to reflect the occurrence of surgical complications, primary nonfunction and acute rejection,12 survival between 90 days and 2 years and between 2 and 5 years to reflect tumor recurrence and chronic rejection.3,7,12,13 These results were first adjusted for recipient characteristics and in a second step also for donor characteristics to investigate the impact that the use of livers with suboptimal donor characteristics has on differences in posttransplant survival between HCC and non-HCC recipients. In a series of sensitivity analyses, we also tested whether the effect of HCC on mortality differed according to a previous diagnosis of hepatitis C (HCV) and, more specifically, whether mortality from tumor recurrence differed according to the use of donation after circulatory death (DCD) donors.

MATERIALS AND METHODS

Standard National Liver Transplant Registry

Since 1968, the Standard National Liver Transplant Registry contains information about all liver transplants done in the 6 liver transplant centers in England and 1 center in Scotland. The data set is managed by National Health Service (NHS) Blood and Transplant,14 and regular checks indicate that the data are consistently >93% complete and accurate and results from several studies confirm the validity of the data set.14-17

Study Population

The study population included all recipients aged 16 years or older who received the first elective orthotopic liver transplant in the United Kingdom between January 1, 2008, and December 31, 2016. The diagnostic category of each patient was identified from the 3 diagnostic fields available in the Standard National Liver Transplant Registry and patients were categorized into 2 groups, patients transplanted with HCC and patients transplanted with other liver disease diagnoses according to their primary liver diagnosis at the time of transplantation (non-HCC patients). In the event of multiple diagnoses, patients were considered to have HCC if HCC was mentioned in any of 3 diagnosis fields. There was no information in the UK transplant registry on explant pathology.

To limit heterogeneity of the study cohort, patients who underwent transplantation for types of liver cancer other than HCC and those who underwent multivisceral, superurgent, domino, or living-related liver transplantations were excluded as well as those who received a liver transplant for acute liver failure (including auxiliary transplantation). We also excluded patients whose survival data were missing.

Donor and recipient characteristics and primary cause of death were compared between HCC and non-HCC recipients. Recipient’s lifestyle activity was assessed using a 5-point scale ranging from able to carry out normal activity without restriction to completely reliant on nursing/medical care,17 and United Kingdom Model for End-Stage Liver Disease (UKELD) was used to score the severity of the liver disease.18 Cold ischemic time was defined as the duration between the start of cold perfusion in the donor to the start of blood flow through the organ in the recipient.19 Values for ethnicity were grouped into white and non-white groups.

Statistical Analysis

To describe the prognostic impact of HCC, we included patients who received a liver transplant over a 9-year period between January 1, 2008, and December 31, 2016. Categorical variables were presented as proportions and compared using chi-squared tests and continuous variables presented as means with SDs and compared using t tests. Patients transplanted for non-HCC indications who were subsequently found to have an HCC on explant pathology were analyzed on an intention-to-treat basis and remained in the non-HCC cohort.

The Kaplan-Meier method was used to compare posttransplant patient and graft survival in HCC and non-HCC recipients and to compare posttransplant patient and graft survival in patients with HCC who were transplanted within the Milan criteria and those transplanted within the expanded criteria. Follow-up data were available until December 31, 2016. Patients with a functioning graft or alive at their last follow-up visit were considered to be censored observations. Graft loss was defined as either retransplantation or patient death. Differences in survival were assessed with the log-rank test.

We used multifactorial Cox regression to build 3 separate models. All models were designed to examine the prognostic impact of HCC status on patient survival in 3 separate epochs of follow-up time: up to 90 days after transplantation, between 90 days and 2 years, and between 2 and 5 years. In the first model, hazard ratios (HRs) comparing posttransplant survival in liver transplant recipients with and without HCC were estimated without adjustment for the donor and recipient characteristics. In the second model, HRs were estimated with adjustment for recipient factors only, and in the final model, HRs were estimated after adjustment for both donor and recipient factors. We performed a series of sensitivity analysis that first explored the effect of partitioning the epochs into posttransplantation time periods that included 90 days to 1 year and 1 to 2 years and second determined whether the effect of HCC on mortality differed according to HCV status by testing the interaction between HCC and HCV.

In all Cox models, adjustment for specific tumor characteristics was not included as comparisons of posttransplantation survival in HCC patients were made with a cohort of non-HCC patients. All donor and recipient factors were selected on the basis of their clinical plausibility of being a risk factor for posttransplant survival.16 The time dependency of HCC as a risk factor for posttransplant survival and the interaction effect between HCC and HCV were tested with Wald tests.

In the regression models in which we adjusted for donor and recipient characteristics, we also explored possible nonlinear relationships between the recipient and donor characteristics measured as continuous variables and posttransplant survival, by including these as both linear and quadratic terms in the model. Missing patient and donor characteristics were imputed using chained equations creating 10 complete data sets.20 The Cox regression results for each of these data sets were pooled using Rubin rules.20 No patient or donor characteristic had >15% of missing values.

Stata V15 (StataCorp, College Station, TX) was used for all statistical analyses. A P <0.05 was considered significant for each statistical analysis.

RESULTS

A total of 4927 first adult elective liver transplants were performed between 2008 and 2016, of which 1270 liver transplants were for HCC recipients and 3657 for non-HCC recipients (Figure 1). Compared with non-HCC recipients, those who received a liver transplantation for HCC between 2008 and 2016 were more likely to be male, from non-white ethnic backgrounds, and positive for HCV infection (Table 1). Despite being significantly older at the time of transplantation, HCC patients were physically more active (according to their recorded lifestyle activity), had better liver function (exhibited by lower UKELD scores), and were less likely to show signs of end-stage liver disease (varices, encephalopathy, and ascites). They were also less likely required ventilation or hospital admission immediately before transplantation and less likely to have undergone previous abdominal surgery. Patients with HCC received more grafts from organs DCD or grafts in which the appearance had been documented as abnormal or steatotic. Cold ischemic time was marginally lower in HCC recipients, and there were only small differences between the cohorts in the frequency of capsular damage in the donor organ. Of the 1270 HCC recipients who were included in our study, only 81 (6.4%) had tumor characteristics that were beyond the Milan but within the expanded criteria at the time of registration on the transplant waiting list.

TABLE 1.
TABLE 1.:
Donor and recipient patient characteristics (N = 4927)
FIGURE 1.
FIGURE 1.:
Flow chart detailing selection of study population (2008-2016). NHSBT, National Health Service Blood and Transplant.

Kaplan-Meier survival curves comparing outcomes in HCC and non-HCC patients showed that patient and graft survival in the first months following liver transplantation is very similar (Figure 2). After about 3 to 4 months, HCC patients seem to have progressively worse patient survival, resulting in a 5-year patient survival of 74.5% (95% CI, 71.2%–77·5%) for HCC patients and 84.6% (95% CI, 83.0%–86.1%, P < 0.001) for non-HCC patients. A similar time pattern was observed for prognostic impact of HCC on graft survival with corresponding 5-year estimates of 70.2% (95% CI, 66.8%–73.3%) for HCC patients and 79.1% (95% CI, 77.4%–80.7%, P < 0.001) for non-HCC patients.

FIGURE 2.
FIGURE 2.:
Five-year patient and graft survival stratified by hepatocellular carcinoma (HCC) status 2008-2016 (N = 4927).

We did not find a difference in the 5-year patient survival between the 1189 HCC patients who met the Milan criteria (74.6%; 95% CI, 71.1%–77.7%) and the 81 who did meet the expanded criteria (74.5%; 95% CI, 58.6%–85.0%; P = 0.76 (Figure 3). Neither did we find differences in graft survival between these patient groups (70.4%; 95% CI, 67.0%–73.6% and 67.8%; 95% CI, 55.2%–79.3%, respectively; P = 0.81).

FIGURE 3.
FIGURE 3.:
Five-year patient and graft survival for hepatocellular carcinoma (HCC) patients stratified by type of selection criteria 2006-2016 (N = 1270).

The first Cox regression model, comparing HCC and non-HCC patients without adjustment for donor or recipient characteristics, did not find a statistically significant difference in survival in the first 90 days after transplantation (HR, 0.88; CI, 0.63–1.23; Table 2). In the subsequent 2 epochs of follow-up time, patients with HCC had a significantly poorer survival (HR, 2.27 between 90 days and 2 years and HR, 2.00 between 2 and 5 years). In the second Cox regression model, only adjusting for recipient characteristics did not dramatically change the impact of HCC on survival in either the first 90 days after transplantation (adjusted HR, 0.76; CI, 0.53–1.09) or in the 2 later epochs of follow-up time (adjusted HR, 1.99 between 90 days and 2 years and adjusted HR, 1.77 between 2 and 5 years). In the third Cox model, additional adjustment for donor characteristics also had little effect on the impact of HCC in each of the epochs of follow-up time (adjusted HR, 0.74 between 0 and 90 days; adjusted HR, 1.96 between 90 days and 2 years; and adjusted HR, 1.74 between 2 and 5 years). The results of the Cox regression analysis of graft survival (Table 3) closely mirrored the results found for patient survival (Table 2).

TABLE 2.
TABLE 2.:
Impact of HCC on posttransplant patient survival in 3 separate epochs of follow-up time (N = 4927)
TABLE 3.
TABLE 3.:
Impact of HCC on post-transplant graft survival in 3 separate epochs of follow-up time (N = 4927)

In the sensitivity analysis that explored the impact of HCC in 4 separate epochs, we found that it was highest between 90 days and 1 year after transplantation (adjusted HR, 2.10; 95% CI, 1.47–3.00) and that it remained at a very similar level thereafter (Table S5, SDC, http://links.lww.com/TP/B665). The sensitivity analysis testing the interaction between HCC and HCV status did not show that the effect of HCC on mortality differed significantly according to HCV status (HCV+ve HR, 1.16; 0.87–1.56, and HCV−ve HR, 0.86; 0.64–1.15; P for interaction = 0.10).

In the first 90 days after transplantation, there were no statistically significant differences in the distribution of cause of death between HCC and non-HCC recipients and no patient died from tumor recurrence (recurrence of malignant primary disease; Table 4). In the subsequent posttransplant epochs, tumor recurrence in HCC recipients became a more frequent cause of death accounting for 23 of the 101 deaths (22.7%) between 90 days and 2 years and 12 of the 77 deaths (15.6%) between 2 and 5 years. When splitting cause of death into 4 epochs, we found that from 90 days onward the number of patients dying from tumor recurrence remained more or less constant (Table S6, SDC, http://links.lww.com/TP/B665).

TABLE 4.
TABLE 4.:
Primary cause of death following liver transplantation for HCC and non-HCC patients in 3 separate epochs of follow-up time (n = 620)

Of the 35 HCC recipients who died of tumor recurrence, 9 (25.7%) had received a DCD liver compared with 396 of the 1235 other HCC recipients (32.1%; P = 0.43) which demonstrates that there is no evidence that the use of DCD livers is linked to an HCC recurrence risk. The proportion of patients who died from malignancies other than tumor recurrence was higher in the HCC recipients (2.9% or 37/1270) than in non-HCC recipients (1.1% or 41/3657), and this difference was most prominent in deaths from nonlymphoid malignancies (Table 4). Overall, recurrence of benign primary disease, which includes HCV, was infrequently reported as a cause of death (Table 4 and see Table S6, SDC, http://links.lww.com/TP/B665), irrespective of HCC status at the time of transplant or epoch of follow-up (Table 4 and see Table S6, SDC, http://links.lww.com/TP/B665).

DISCUSSION

Summary of Results

At the time of transplantation, HCC patients were on average in a better physical condition and had less signs of end-stage liver disease than non-HCC patients, but they received more often suboptimal grafts. We found that the survival of HCC and non-HCC recipients was similar in the first months after transplantation. Then the survival of HCC recipients deteriorated with the rate of mortality and graft failure being at least 50% higher than in non-HCC recipients, with tumor recurrence as the most important explanation. The difference in survival could not be explained by HCC recipients receiving a higher proportion of livers from DCD donors or from donors with other suboptimal characteristics.

Methodological Limitations

The key limitation of our analysis is that we used predefined posttransplant epochs (up to 90 days, between 90 days and 2 years, and between 2 and 5 years) to investigate the time dependency of the impact of HCC on patient and graft survival. This approach assumes that the prognostic impact of HCC on survival is constant within each of these epochs.21 The advantage of this approach is that the HRs can be estimated using standard Cox regression methods and, more importantly, that the results are relatively easy to interpret. Its disadvantage is that the partitioning of the survival time in distinct epochs needs to be chosen in advance and that the number of separate epochs as well as their duration is arbitrary.

In our analysis, we compared HCC patients with a heterogeneous cohort of non-HCC patients. This approach may have masked specific survival patterns of individual primary liver diseases. However, the dichotomy in HCC and non-HCC patients reflects the difference in how HCC and non-HCC patients were selected for transplantation in the United Kingdom. While for most non-HCC patients, the urgency of transplantation was taken from their liver function according to the UKELD score, the urgency for HCC patients came from the need to avoid cancer progression before transplantation.

Comparison With Other Studies

We studied the prognostic impact of HCC on posttransplant survival in 3 distinct epochs, aiming to capture on the one hand that HCC patients are in a better physical condition at the time of transplantation—which may give them better surgical outcomes—but on the other that tumor recurrence may deteriorate survival in the later stages. Already 30 years ago, the importance of analyzing liver transplant outcomes in epochs of follow-up time was recognized, but this statistical approach is very rarely practised.4,22 Our study is an example of how important it is to analyze posttransplant outcomes in distinct epochs of follow-up time, guided by the understanding of the relevant underlying clinical mechanisms. For example, risk factors for immediate surgical outcomes are predominantly linked to the recipients’ physical condition and risk factors for longer-term outcomes to recurrence of the original disease that was the reason for transplantation.

It was expected that the introduction of the Milan criteria would lead to a decrease in recurrence rates in patients transplanted for HCC.4 However, our study, which reflects the outcomes of modern liver transplantation practice, including a national population-based cohort of patients transplanted between 2008 and 2016, indicates that tumor recurrence remains an important risk factor for survival in the later stages after liver transplantation, which corresponds with earlier reports of posttransplant survival.1-3

Despite a formal adoption of expanded HCC selection criteria in 2008, we found that only 6.4% of HCC recipients were selected for transplantation within these expanded criteria and we could not demonstrate differences in posttransplant outcomes compared with those who were selected according to the Milan criteria. Reasons for why only a very small minority of HCC recipients were transplanted beyond the Milan criteria are difficult to explain and we must acknowledge that this analysis does not specifically address this question. However, a tendency for radiological assessment to understage some HCC patients before transplantation may have prohibited the aggressive use of the extended selection criteria, especially when other studies have indicated a linear relationship between tumor burden and posttransplantation survival.23,24

Explanation of Results

Our study found that HCC patients were more likely to receive suboptimal donor organs with characteristics previously proven to have poorer posttransplant outcomes.2 This included donated livers that were either steatotic, abnormal in appearance, or that were from DCD donors. However, our analysis was specifically designed to test the impact of donor characteristics on posttransplant survival, and we observed that additional adjustment for donor characteristics had little effect on the differences in survival between HCC and non-HCC recipients in any of the epochs after transplantation.

The incidence of HCV recurrence after transplantation is also an unlikely explanation for the observed differences in survival. Previous studies have reported that, irrespective of HCC status at the time of transplantation, survival between those with and without HCV is similar up to 5 years after transplantation and worse thereafter.25,26 In our own analysis, we did not find the effect of HCC on mortality to differ significantly according to whether the patient had a previous diagnosis of HCV nor did we find HCV recurrence to be frequently reported as a cause of death in the first 5 years after transplantation.

Similarly, differences in the incidence of acute rejection do not explain the differences in the survival patterns of HCC and non-HCC recipients. In efforts to reduce the risk of tumor recurrence, HCC recipients can be subjected to more conservative immunosuppression protocols,27 and therefore, they may be at an increased risk of acute rejection. However, we have found that 1-year readmissions for acute rejection in patients transplanted in the United Kingdom between 2008 and 2016 occurred less frequently in HCC recipients (2.8% or 35/1270) than in non-HCC recipients (3.1% or 112/3657; P = 0.57), while acute rejection recorded as a cause of death was not identified at all within the study cohort (London School of Hygiene and Tropical Medicine, unpublished data, 2018).

We identified some differences in the proportion of HCC and non-HCC recipients who died of malignancies other than tumor recurrence. This cause of death, particularly nonlymphoid-related malignancies, were more frequent in HCC recipients and consistent with the existing literature there was a high incidence between 3 months and 2 years after transplantation.28 However, the differences in the overall number of HCC and non-HCC recipients who died from malignancies other than tumor recurrence were too small to fully explain the differences in survival between the 2 cohorts.

Beyond 90 days, differences in survival are best explained by differences in deaths due to tumor recurrence and this remained so even when we further partitioned the follow-up period to include survival from 90 days to 1 year and from 1 to 2 years. Of the HCC patients who were recorded to have died of tumor recurrence within 1 year, only one was preoperatively staged according to the extended criteria with other early deaths potentially explained by aggressive tumor biology or radiological understaging of the HCC before transplantation.13 In further analysis, we did not find the use of DCD livers to be associated with an increased risk of death from tumor recurrence.

In the past, HCC patients were found to have 90-day outcomes that were statistically significantly better than non-HCC patients.5 Our results suggested that 90-day outcomes of HCC patients were better, but the difference with non-HCC patients was not statistically significant. One important explanation for not finding a significant difference is the substantial improvement in posttransplant outcomes in the last 30 years which considerably reduces the statistical power to detect differences.29 Another explanation is that the impact of recipients’ frailty at the time of transplantation has decreased given the improvements in perioperative care and the high dependency care immediately after transplantation.30

Implications of Findings

Our results demonstrate that outcomes in patients transplanted for HCC are worse than in those transplanted for non-HCC indications. This is not explained by the fact that we are using more DCD donors in HCC patients or that we are transplanting a significant proportion of patients who, at the time of transplantation, are beyond the Milan criteria. Instead, we must acknowledge that even with the stringent adoption of the Milan criteria in the United Kingdom, we are still selecting for transplantation a significant proportion of patients with HCC who are at risk of tumor recurrence. Therefore, until we can add to our selection criteria new parameters that better predict tumor recurrence, the poorer survival of HCC patients after liver transplantation will remain.

Until recently, many guidelines stipulated that patients with HCC should only receive a liver transplant if their predicted outcomes are comparable to non-HCC patients.3 However, in the last decade, this has never been the case. This has been recognized by the service providers, and donor liver allocation schemes are now moving toward using criteria based on transplant benefit—in which they aim to maximize the net life years gained from the point of registration on the waiting rather than providing the greatest chance of surviving after transplantation.31,32 However, the decision to offer HCC patients a liver transplant is further complicated as other treatments, including resection and ablation, have to be considered which is all the more important considering the impact that an increased use of liver transplantation in HCC patients will have on outcomes for non-HCC patients on the waiting list for transplantation given the ongoing donor organ shortage.1,3

CONCLUSIONS

Between 2008 and 2016, almost all HCC patients who received a liver transplant in the United Kingdom met the Milan criteria. Nevertheless, 1 in 4 HCC recipients died within 5 years compared with only 1 in 6 non-HCC patients, with tumor recurrence being the most likely explanation for this difference. These differences could not be explained by the increased use of poorer-quality donor organs in HCC patients. Donor allocation schemes based on transplant benefit schemes are likely to accommodate the poor posttransplant survival of HCC patients given their greater net gain in posttransplanted expected life years.

ACKNOWLEDGMENTS

The authors would like to thank all liver transplant centers for providing data to the Standard National Liver Transplant Registry. The authors would also like to thank all those involved in collecting and handling liver transplant data at National Health Service Blood and Transplant. The UK Liver Transplant Audit is supported by the NHS National Specialised Commissioning Group and NHS England.

REFERENCES

1. Bruix J, Gores GJ, Mazzaferro V. Hepatocellular carcinoma: clinical frontiers and perspectives. Gut. 2014;63:844–855.
2. Callaghan CJ, Charman SC, Muiesan P, et al.; UK Liver Transplant Audit. Outcomes of transplantation of livers from donation after circulatory death donors in the UK: a cohort study. BMJ Open. 2013;3:e003287.
3. Clavien PA, Lesurtel M, Bossuyt PM, et al.; OLT for HCC Consensus Group. Recommendations for liver transplantation for hepatocellular carcinoma: an international consensus conference report. Lancet Oncol. 2012;13:e11–22.
4. Bird SM, Calne RY, Sharples LD. Analyse transplant outcomes in distinct epochs of follow-up. Lancet. 2006;367:1816.
5. Burroughs AK, Sabin CA, Rolles K, et al. 3-month and 12-month mortality after first liver transplant in adults in Europe: predictive models for outcome. Lancet. 2006;367:225–232.
6. Mazzaferro V, Regalia E, Doci R, et al. Liver transplantation for the treatment of small hepatocellular carcinomas in patients with cirrhosis. New Engl J Med. 1996;334:693–699.
7. Davis E, Wiesner R, Valdecasas J, et al. Treatment of recurrent hepatocellular carcinoma after liver transplantation. Liver Transpl. 2011;17:162–166
8. Yao FY, Ferrell L, Bass NM, et al. Liver transplantation for hepatocellular carcinoma: expansion of the tumor size limits does not adversely impact survival. Hepatology. 2001;33:1394–1403.
9. Mazzaferro V, Llovet JM, Miceli R, et al. Predicting survival after liver transplantation in patients with hepatocellular carcinoma beyond the Milan criteria: a retrospective, exploratory analysis. Lancet Oncol. 2009;10:35–43.
10. Menon KV, Hakeem AR, Heaton ND. Review article: liver transplantation for hepatocellular carcinoma: a critical appraisal of the current worldwide listing criteria. Aliment Pharmacol Ther. 2014;40:893–902.
11. Available at http://odt.nhs.uk/pdf/advisory_group_papers/LAG/HCC_recommendations_IR_TS_b_NAS_Work_in_Progress.pdf. Accessed January 12, 2018.
12. Tovikkai C, Charman SC, Praseedom RK, et al. Time-varying impact of comorbidities on mortality after liver transplantation: a national cohort study using linked clinical and administrative data. BMJ Open. 2015;5:e006971.
13. Zimmerman MA, Ghobrial RM, Tong MJ, et al. Recurrence of hepatocellular carcinoma following liver transplantation: a review of preoperative and postoperative prognostic indicators. Arch Surg. 2008;143:182–188.
14. NHS Blood and Transplant. Organ Donation and Transplantation Directorate. Liver Advisory Group. Provision of Standard Data Sets for Liver Transplant. Available at http://odt.nhs.uk/pdf/advisory_group_papers/LAG/Provision_of_Standard_Data_Set_for_Liver_Transplant_v4.pdf.
15. Dawwas MF, Gimson AE, Lewsey JD, et al.; on behalf of the UK and Ireland Liver Transplant Audit. Survival after liver transplantation in the United Kingdom and Ireland compared with the United States. Gut. 2007;56:1606–1613.
16. Charman SC, Copley L, Tovikkai C, et al. UK Liver Transplant Audit in patients who received a liver transplant between 1st March 1994 and 31st March 2012. 2012. The Royal College of Surgeons of England, Available at: http://www.rcseng.ac.uk/surgeons/research/surgical-research/docs/liver-transplant-audit-report-2012. Accessed June 10, 2018.
17. Jacob M, Copley LP, Lewsey JD, et al. Functional status of patients before liver transplantation as a predictor of posttransplant mortality. Transplantation. 2005;80:52–57.
18. Barber K, Madden S, Allen J, et al. Elective liver transplant list mortality: development of a United Kingdom end-stage liver disease score. Transplantation. 2011;92:469–476.
19. Pan ET, Yoeli D, Galvan NTN, et al. Cold ischemia time is an important risk factor for post-liver transplant prolonged length of stay. Liver Transpl. 2018;24:762–768.
20. White I, Royston P, Wood A. Multiple imputation by chained equations: issues and guidance for practice. Stat Med. 2011;30:377–399.
21. Lehr S, Schemper M. Parsimonious analysis of time-dependent effect in the Cox model. Stat Med. 2007;26:2686–2698.
22. Gore SM, Barroso E, White DJG. Calne R. Risk factors in orthotopic first liver transplantation. In: Liver Transplantation19872nd edLondon: Grune and Stratton; 513–530.
23. Germani G, Gurusamy K, Garcovich M, et al. Which matters most: number of tumours, size of the largest tumour, or total tumour volume. Liver Transpl. 2011;17:S58–S66.
24. Marelli L, Grasso A, Pleguezuelo M, et al. Tumour size and differentiation in predicting recurrence of hepatocellular carcinoma after liver transplantation: external validation of a new prognostic score. Ann Surg Oncol. 2008;15:3503–3511.
25. Gane EJ, Portmann BC, Naoumov NV. Long-term outcome of hepatitis C infection after liver transplantation. N Engl J Med. 1996;334:815–821.
26. Vinaixa C, Rubin A, Aguilera V, et al. Recurrence of hepatitis C after liver transplantation. Ann Gastroenterol. 2013;26:304–313.
27. Khorshandi SE, Heaton N. Optimization of immunosuppressive medication upon liver transplantation against HCC recurrence. Transl Gastroenterol Hepatol. 2016;125.
28. Collet D, Mumford L, Banner NR, et al. Comparison of the incidence of malignancy in recipients of different types of organ: A UK Registry audit. Am J Transplant. 2010;10:1889–1896.
29. Adam R, Karam V, Delvart V, et al. Evolution of indications and results of liver transplantation in Europe. A report from the European Liver Transplant Registry (ELTR). J Hepatol. 2012;57:675–688.
30. Singal AK, Guturu P, Hmoud B, et al. Evolving frequency and outcomes of liver transplantation based on etiology of liver disease. Transplantation. 2013;95:755–760.
31. Schaubel DE, Guidinger MK, Biggins SW, et al. Survival benefit-based deceased-donor liver allocation. Am J Transplant. 2009;9:970–81.
32. NHS Blood and Transplant. Organ Donation and Transplantation Directorate. Liver Advisory Group. Fixed Term Working Unit - Organ Allocation October 2014. Available at http://odt.nhs.uk/pdf/advisory_group_papers/LAG/Allocation_System.pdf. Accessed June 1, 2018.

Supplemental Digital Content

© 2019 The Author(s). Transplantation Direct. Published by Wolters Kluwer Health, Inc.