SEARCH

SEARCH BY CITATION

Keywords:

  • Extended criteria donors;
  • graft acceptance;
  • liver transplantation;
  • transplant centers

Abstract

  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and Methods
  5. Results
  6. Discussion
  7. Acknowledgments
  8. Disclosure
  9. References

Organ shortage has led to increased utilization of higher risk liver allografts. In kidneys, aggressive center-level use of one type of higher risk graft clustered with aggressive use of other types. In this study, we explored center-level behavior in liver utilization. We aggregated national liver transplant recipient data between 2005 and 2009 to the center-level, assigning each center an aggressiveness score based on relative utilization of higher risk livers. Aggressive centers had significantly more patients reaching high MELDs (RR 2.19, 2.33 and 2.28 for number of patients reaching MELD > 20, MELD > 25 and MELD > 30, p < 0.001), a higher organ shortage ratio (RR 1.51, 1.60 and 1.51 for number of patients reaching MELD > 20, MELD > 25 and MELD > 30 divided by number of organs recovered at the OPO, p < 0.04), and were clustered within various geographic regions, particularly regions 2, 3 and 9. Median MELD at transplant was similar between aggressive and nonaggressive centers, but average annual transplant volume was significantly higher at aggressive centers (RR 2.27, 95% CI 1.47–3.51, p < 0.001). In cluster analysis, there were no obvious phenotypic patterns among centers with intermediate levels of aggressiveness. In conclusion, highwaitlist disease severity, geographic differences in organ availability, and transplant volume are the main factors associated with the aggressive utilization of higher risk livers.


Abbreviations
AUC

area under the curve

BMI

body mass index

CF

clustering factor

CIT

cold ischemia time

CDC

centers for disease control

DCD

donation after cardiac death

GLM

generalized linear models

HBV

hepatitis B virus

HCV

hepatitis C virus

LFT

liver function tests

OPO

Organ Procurement Organization

PSR

program specific reports

SSDMF

Social Security Death Master File

SRTR

Scientific Registry of Transplant Recipients

UNOS

United Network for Organ Sharing

Introduction

  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and Methods
  5. Results
  6. Discussion
  7. Acknowledgments
  8. Disclosure
  9. References

The liver transplantation waiting list has grown continuously over the past decade in the context of a profound organ shortage [1, 2]. Liver transplantation provides the best survival and quality of life for those with end-stage liver disease and is the only definitive therapy [3]. In efforts to increase transplantation rates and decrease waitlist mortality, providers have expanded their criteria for donor selection [4, 5]. Some centers have reported significant increases in their transplant volumes with comparable outcomes by increasing the utilization of marginal or higher risk grafts [6-9].

Although there is no universal definition of a higher risk liver donor, various donor characteristics have been considered risky. These factors include, but are not limited to, advanced donor age, donors classified by the CDC as high infectious risk (CDC), evidence of hepatitis B (HBV) or hepatitis C (HCV) infection, elevated liver functions tests (LFTs), donation after cardiac death (DCD), imported grafts, or donors with elevated body mass indexes (BMI) [6, 9-20]. However, it is unclear if utilization of these higher risk livers varies by transplant center. Although there are reports outlining the increased utilization of higher risk livers and their effect on graft and patient survival, little is known about what center-level factors account for the variation seen in the use of higher risk livers [6, 7, 9, 18, 21, 22].

We recently explored center-level patterns in the utilization of higher risk kidney allografts using novel methods for characterizing these patterns and found that many center-level factors, including waitlist size, organ shortage and waiting times, were associated with the aggressive utilization of higher risk kidneys. Furthermore, we also found phenotypic patterns in aggressive behavior, suggesting that aggressive use of one type of higher risk kidney was associated with the aggressive use of other types of higher risk kidneys [23]. This motivated an extension of these analyses to liver transplant centers to determine if a liver aggressive phenotype exists and if center-level factors are associated with that phenotype.

Materials and Methods

  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and Methods
  5. Results
  6. Discussion
  7. Acknowledgments
  8. Disclosure
  9. References

Study population

We analyzed center-level UNOS data for 25 812 adult deceased donor liver transplants between January 1, 2005 and December 31, 2009. Initial exploration of center-level clustering involved 123 centers. However, for the sake of stability of subsequent regression modeling and phenotyping, we excluded centers if they had more than 2 years with no deceased donor transplants or fewer than 10 transplants per year for all 5 years; in these analyses, we ultimately analyzed data from 101 US transplant centers.

Defining aggressiveness by organ type

Based on the available literature and clinical judgment, eight categories of higher risk livers were selected for exploration (Table 1) [6, 9-14, 17-20]We used similar methodology to our previously published research in higher risk kidneys [23]. In brief, for each category, we quantified the degree of center-level clustering by the area under the cumulative distribution curve (AUC; Figure 1). We referred to the AUC as the clustering factor (CF) and selected CF's to allow the necessary amount of discrimination to identify phenotypic patterns.

Table 1. Clustering factors (CF) for various cut-points of categories potentially considered aggressive in deceased donor liver transplantation
FactorLess evidence of clusteringEvidence of clusteringToo much clustering
  1. Cumulative distribution functions were plotted for each cut-point (Figure 1), and area under these curves is reported as the clustering factor. CF < 0.7 were considered not clustered enough for defining a center's aggressiveness, and CF ≥ 0.85 were considered too clustered. The cut-points shown in the middle column were used to define aggressiveness for each metric, which were then decile-scored, averaged and assigned on a center level.

Age≥55≥65≥75
 0.650.730.86
CDC CDC infectious risk 
  0.74 
HBV HBV Core-onlyActive HBV
  0.780.93
HCV Positive 
  0.81 
Donor LFTs≥100>500 
 0.630.80 
DCD DCD 
  0.81 
Share type Regionally importedNationally imported
  0.740.91
Donor BMI≥30>40>50
 0.620.750.90
image

Figure 1. Cumulative distribution of deceased donor liver transplants performed by age group. Centers are sorted by decreasing proportion of transplants with each of the specified characteristics (i.e. age exceeding a given cut-point). Each line represents the cumulative distribution of transplants performed at a given center that are from donors of that age group. As the age cut-point increases, a smaller number of centers contribute to the cumulative volume of transplants for that age group. Therefore, as the age cut-point increases, the curve gets shifted to the upper left hand corner of the graph.

Download figure to PowerPoint

Defining the aggressiveness phenotype

Category-specific aggressiveness scores were then determined for each center (the degree to which the center was aggressive in using higher risk organs of a given category) according to the relative proportion of center volume comprised by grafts of that category. For each of the eight categories, the centers were ordered by the proportion of their deceased donor transplants composed of livers in that category and were then assigned a decile score (1 = least aggressive, 10 = most aggressive). A final aggressiveness score was calculated for each center by averaging the individual decile scores for each category at that center; this score was treated as a continuous score in most analyses. However, for the purposes of a few dichotomous analyses, a score of 6.5 or greater was empirically selected as “aggressive” so that approximately 30% of centers would be selected (Figure 2); as a sensitivity analysis, a more stringent score of 7.6 was tested, so that approximately 15% of centers would be selected.

image

Figure 2. Distribution of aggressiveness score across all centers. Each center was given decile-scores based on its relative use of organs of various categories. These decile scores were averaged, and a center was classified as “aggressive” if its average score was 6.5 or higher (so that as close to 30% of centers as possible were marked aggressive). For sensitivity analysis, a more stringent cut-point (7.6) was chosen so that as close to 15% of centers as possible were marked aggressive.

Download figure to PowerPoint

Center-level characteristics

Waitlist size was defined as the number of active registrants on the waitlist on the last day of the study period. To account for differences in listing practice by disease severity, we calculated the number of candidates added to the waiting list with a model for end-stage liver disease (MELD) score >15 during the study period, as well as the total number of patients at each center whose MELD scores reached or exceeded 20, 25 and 30 during the study period. Organ shortage ratio was defined as the number of patients who reached or exceeded the specified MELD cutpoint within an OPO divided by the number of livers recovered by that OPO (annually). Median calculated MELD score at transplant and average annual transplant volume were determined for each center across the study period. The distributions of all continuous center characteristics except organ shortage were skewed so medians were used to compare characteristics at aggressive versus nonaggressive centers. The distribution of organ shortage ratio across transplant centers was pseudonormally distributed so mean organ shortage ratio was compared at aggressive versus nonaggressive centers. All continuous center characteristics were categorized into tertiles for the purposes of regression analysis.

Statistical analysis

Univariate comparisons of center characteristics for aggressive and nonaggressive centers were performed using t-tests for pseudonormally distributed continuous variables, rank-sum tests for non-normally distributed continuous variables, and chi-squared tests or Fisher's Exact tests for categorical variables. Generalized linear models (GLM) were used to model the association between aggressiveness and center-level characteristics. To obtain estimates of relative risk, a GLM was fit using a Poisson model with robust variance estimation as previously described [24]. Potential phenotypes were determined using hierarchical agglomerative cluster analysis and illustrated using a linked heat map and dendrogram, as previously described in our kidney aggressive phenotype study, using the hclust() function in the stats package and graphics from the qplots package for R version 2.15.1 (Vienna, Austria)[23]. Other than cluster analysis, all other analyses were performed using STATA 12.1/MP for Linux (College Station, TX, USA).

Results

  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and Methods
  5. Results
  6. Discussion
  7. Acknowledgments
  8. Disclosure
  9. References

Center demographics

There were 25 812 adult deceased donor liver transplants during the study period, with a mean annual volume per center of 51.4 (median 40.2, interquartile range [IQR] 25.0–73.6). Among all centers, median utilization of each category of higher risk liver ranged from 1.5–18.5% (Table 2). Among aggressive centers, median utilization of higher risk liver allografts ranged from 3.4–37.3%.

Table 2. Percent of transplant volume comprised by various categories of higher-risk livers, stratified by center aggressiveness
 Median percentage (Range, IQR)a
Donor characteristicAggressive centersNonaggressive centersAll centers
  1. a

    Column 3 reports the median percentage, range and interquartile range (IQR) of each category of higher-risk organ transplanted from among all deceased donor transplants at a given center. For example, the median percentage of grafts from donors over the age of 65 transplanted at any given center was 7.0% (range 0–35%, IQR 7.0).

Age > 6511.4% (3.6–34.9%, 5.9)5.3% (0–27.3%, 4.4)7.0% (0–34.9%, 6.8)
CDC9.3% (4.5–25%, 5)6.0% (0.8–32.8%, 5.1)7.7% (0.8–32.8%, 5.9)
HBcAb7.4% (2.1–18.8%, 4.9)2.5% (0–13.2%, 3.9)4.2% (0–18.8%, 5.3)
HCV+4.1% (0.7–20.8%, 3.7)1.1% (0–13%, 2)1. 7% (0–20.8%, 2.9)
LFTs > 5003.4% (0.6–12.5%, 2.4)1.1% (0–4.8%, 1.6)1.5% (0–12.5%, 2.2)
DCD6.1% (0.9–14.9%, 6.4)1.7% (0–39.2%, 4.7)3.4% (0–39.2%, 5.9)
Regionally imported37.3% (9.1–91.3%, 30.2)14.6% (0–78.8%, 13.8)18.5% (0–91.3%, 20.4)
BMI > 404.4% (0.7–7.6%, 2.1)2.0% (0–6%, 2.1)2.6% (0–7.6%, 2.7)

Aggressive center characteristics

Center volume had a positive linear association with aggressiveness (Figure 3). Aggressive centers had larger waitlist size (153 vs. 92, RR 1.49, p = 0.041) and significantly more candidates listed with MELD >15 (320 vs. 179, RR 2.27, p < 0.001; Table 3). Aggressive centers also had significantly more candidates reach a MELD of 20 or higher (RR 2.19, p < 0.001), a MELD of 25 or higher (RR 2.33, p < 0.001), or a MELD of 30 or higher (RR 2.28, p < 0.001). Aggressive centers appeared to be clustered within various geographic regions, particularly in UNOS regions 2, 3 and 9, with region 6 having no aggressive centers at all. Only 12.9% of aggressive centers were in a single-center OPO (defined as having one center perform >80% of an OPO's deceased-donor volume), compared with 30% of nonaggressive centers. There were on average 2.03 candidates who reached a MELD of 20 or higher waiting for each recovered liver within an OPO at aggressive centers versus 1.34 at nonaggressive centers (RR 1.51, p = 0.035); similar trends were seen with candidates who reached a MELD of 25 or higher (RR 1.60, p = 0.016), and a MELD of 30 or higher (RR 1.51, p = 0.022). Median MELD at transplant was similar between aggressive and nonaggressive centers (24 vs. 23, p = 0.6). However, the average annual transplant volume (RR 2.27, p < 0.001) was significantly higher at aggressive centers.

Table 3. Center-level characteristics of aggressive versus nonaggressive centers
CharacteristicAggressive centers (n = 31)Nonaggressive centers (n = 70)Relative riska (95% CI)p-Value
  1. a

    Univariate; relative risks were not calculated for region because there were an insufficient number of centers in some regions (making the model unstable). Continuous metrics were categorized into tertiles for regression analysis.

  2. b

    Dominated (>80% of OPO transplant volume) by one center.

  3. c

    Number of candidates added to the waitlist with a MELD >15 during the study period.

  4. d

    Defined as the ratio of patients who reached the specified MELD cutpoint within an OPO to livers recovered annually by that OPO.

  5. e

    Defined as the number of patients who reached the specified MELD cutpoint.

UNOS Region  p = 0.023 
16.5%7.1%  
219.4%10.0%  
316.1%11.4%  
43.2%14.3%  
512.9%15.7%  
60%5.7%  
76.5%11.4%  
89.7%4.3%  
916.1%0%  
106.5%7.1%  
113.2%12.9%  
Single-center OPOb12.9%30.0%0.45 (0.17–1.17)0.1
Median waitlist size153921.49 (1.02–2.18)0.041
Median # listed with MELD>15c3201792.27 (1.56–3.31)<0.001
Organ Shortage Ratio (MELD>20)d2.031.341.51 (1.03–2.22)0.035
Organ shortage ratio (MELD>25)d1.320.821.60 (1.09–2.33)0.016
Organ shortage ratio (MELD>30)d0.820.541.51 (1.06–2.16)0.022
Median MELD at transplant24231.02 (0.96–1.09)0.6
Avg. annual transplant volume73.236.12.27 (1.47–3.51)<0.001
Median # reaching MELD>20e3101822.19 (1.48–3.24)<0.001
Median # reaching MELD>25e1941142.33 (1.55–3.52)<0.001
Median # reaching MELD>30e12576.52.28 (1.52–3.40)<0.001
image

Figure 3. Aggressiveness score and its correlation with deceased-donor volume. The horizontal (red) line identifies the cut-point (a score of 6.5) for classifying a center as “aggressive” or “nonaggressive” based on the average aggressiveness score for all categories. The diagonal (black) line represents the locally weighted scatterplot smoother (lowess), showing a fairly linear association between transplant volume and aggressiveness.

Download figure to PowerPoint

Aggressive center characteristics: sensitivity analysis

Using a more restrictive definition of “aggressive” that selected only half of the centers as the above more permissive definition (approximately 15% instead of 30%), differences between aggressive and nonaggressive centers were somewhat more pronounced. For example, there was a more pronounced difference in ratio of candidates listed with MELD >15 (RR 2.86 with restrictive definition vs. RR 2.27 with permissive definition), candidates reaching higher MELDs (RR 2.57, 2.95 and 2.86 vs. RR 2.19, 2.33 and 2.28 for patients reaching MELD of 20 or higher, MELD of 25 or higher, and MELD of 30 or higher), and annual transplant volume (RR 2.86 vs. RR 1.51).

The aggressive phenotype

The most aggressive centers (shown under the red colorbar on the far right of Figure 4) were, in general, aggressive in all categories. Similarly, the least aggressive centers (shown under the blue colorbar in Figure 4) were, in general, nonaggressive in all categories. There were no obvious phenotypic patterns among centers with intermediate aggressiveness (shown under the green, yellow and orange colorbars in Figure 4). Annual transplant volume was highest for the most aggressive cluster. At the most aggressive centers, the average annual transplant volume was approximately 77 deceased donor liver transplants per year, compared with 37 at the least aggressive centers.

image

Figure 4. Heat map and cluster analysis of aggressive metrics across centersa. aEach column represents an individual center, and each row represents a category of organs. Each “sliver” (in varying darkness) represents the aggressiveness score for a given center within a given category; higher aggressiveness scores are darker. The top/bottom colorbar represents empirically determined phenotypic patterns of aggressiveness, based on cluster analysis. Column 1(Red) = Aggressive in all measures; Column 5(Blue) = Nonaggressive in all measures; Columns 2 through 4(Green, Yellow and Orange) = Centers with intermediate aggressiveness. bThis row represents the median number of candidates added to the waiting list with a MELD >15 over the course of the entire study period at each cluster of centers.

Download figure to PowerPoint

Discussion

  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and Methods
  5. Results
  6. Discussion
  7. Acknowledgments
  8. Disclosure
  9. References

Based on methods that we previously developed to study the aggressive kidney phenotype, we now present utilization patterns of higher risk liver allografts at transplant centers in the United States. We found wide variations in utilization of these allografts; for example, regionally imported allografts comprised between 0% and 91.3% of a given center's transplant volume. Aggressive centers had statistically significantly larger waitlist size, more candidates listed with MELD >15, more candidates reaching higher MELDs, more candidates waiting per organ recovered, and higher annual transplant volume, but similar MELD scores at transplantation. Finally, cluster analysis demonstrated that while the most aggressive centers were aggressive in all metrics, and vice versa, there were no obvious phenotypic patterns among centers of intermediate aggressiveness.

Geographic clustering of aggressive liver transplant centers is consistent with our findings in kidney transplantation [23]. As in kidney transplantation, where aggressiveness was associated with waitlist size (RR 1.69, p = 0.002), overall waitlist size at liver transplant centers was also associated with aggressive practices (RR 1.49, p = 0.041). Furthermore, aggressiveness was associated with waitlist disease burden, specifically the number of candidates waiting with a MELD >15, the number of candidates reaching higher MELDs, and the number of high-MELD candidates versus the number or local organs available for transplantation. These measures are likely a more appropriate assessment of center demand than waitlist size alone given the rise in waitlist mortality associated with increasing MELD, the increase in survival benefit at this MELD cutpoint, and center-level variations in listing practices [25-30].

The association between annual transplant volume and aggressiveness has many possible explanations. First, surgeon comfort and experience or institutional experience and protocols for certain higher risk organs may lead to increased utilization. Additionally, given the current regulatory climate, centers with lower volume, whose outcomes reports might be more significantly affected by the lower graft and patient survival associated with higher risk grafts, may be less inclined to transplant these organs [31]. The increased volume at aggressive centers could therefore potentially buffer the differences in patient and graft survival associated with higher risk allografts. Alternatively, disparities in organ allocation and sharing could preferentially shunt higher risk organs to higher volume centers [28, 29]. Furthermore, it is possible that there is a desire to increase volume, potentially driven by institutional, departmental, or personal incentives to increase revenue, and that this desire contributes to the aggressive phenotype and the relationship between “aggressive” and volume. Finally, we recognize that there are other possible factors associated with aggressiveness that we were unable to measure, specifically those that are not amenable to registry analysis.

There are interesting policy implications for the novel method we have created and the findings of our study. From a societal perspective, aggressive use of higher risk organs is consistent with the Organ Donation Breakthrough Collaborative goals of making use of every possible organ. After all, it is likely that almost every available liver can offer some survival benefit to some patient on the waiting list, so the challenge is to (1) find the right patient who can benefit and (2) develop protocols and systems to make these higher risk transplants as successful as possible. We have purposefully considered here only aggressiveness in terms of donor characteristics, leaving it up to the center to find the right recipient for the organ. Effect modification analyses can help guide these decisions; for example, we showed that recipients of a certain phenotype do equally well with livers from donors over 70 as they do with livers from much younger donors, whereas recipients of other phenotypes do much worse with older versus younger livers (in other words, there are some recipient characteristics that attenuate the otherwise adverse consequences of older livers; Refs. [32, 33]). From a patient perspective, the advantage of a center's aggressive organ use is lower waiting time (i.e. lower risk of dying on the waiting list), whereas of course the risk is a potentially worse outcome with a higher risk organ; the latter depends on the center's ability to meet the above challenges, of finding the right patient and developing protocols to make higher risk transplants as successful as possible. It is possible that combined reporting of all sides of the equation, namely the organ shortage in the geographic area, the aggressiveness of the centers in that area, and the outcomes of these centers with higher risk organs, will provide patients the information they need to choose a center that is right for them.

There are several limitations to our methodology. First, our methodology is subject to all of the limitations previously reported in our analysis of aggressive kidney transplantation behaviors, including misclassification bias and unmeasured confounding [23]. Again, sensitivity analyses were performed to verify the robustness of the empiric cutpoints utilized in our aggressive center definition. Additionally, aggressiveness was again defined by metrics that describe center utilization patterns but do not account for recovery practices within a given OPO. Finally, listing and organ recovery practices for patients with end-stage liver disease are quite varied at centers across the United States and difficult to identify in registry analysis, which could therefore lead to residual confounding.

This study provides a framework to understand and compare center practices regarding utilization of higher risk liver allografts in the United States. These results could inform policy and individual transplant center practice. These results could also help inform decision-making regarding organ allocation. Finally, transplant candidates could potentially utilize this information, along with other center characteristics, to inform their decision where to list for a transplant.

Acknowledgments

  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and Methods
  5. Results
  6. Discussion
  7. Acknowledgments
  8. Disclosure
  9. References

As a study of the United Network for Organ Sharing database, this work was supported in part by Health Resources and Services Administration contract 234-2005-370011C. The content is the responsibility of the authors alone and does not necessarily reflect the views or policies of the Department of Health and Human Services.

Disclosure

  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and Methods
  5. Results
  6. Discussion
  7. Acknowledgments
  8. Disclosure
  9. References

The authors of this manuscript have no conflicts of interest to disclose as described by the American Journal of Transplantation.

References

  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and Methods
  5. Results
  6. Discussion
  7. Acknowledgments
  8. Disclosure
  9. References
  • 1
    Brown RS, Rush SH, Rosen HR, et al. Liver and intestine transplantation. Am J Transplant 2004; 4 (Suppl 9): 8192.
  • 2
    Organ Procurement and Transplantation Network: Liver Transplantation Waiting LIst. January 1, 2000-December 31, 2010. Available at: http://optn.transplant.hrsa.gov/latestData/rptData.asp. Accessed November 11, 2011.
  • 3
    Hoofnagle JH, Carithers RL, Jr., Shapiro C, Ascher N. Fulminant hepatic failure: Summary of a workshop. Hepatology 1995; 21: 240252.
  • 4
    Markmann JF, Markmann JW, Markmann DA, et al. Preoperative factors associated with outcome and their impact on resource use in 1148 consecutive primary liver transplants. Transplantation 2001; 72: 11131122.
  • 5
    Alexander JW, Vaughn WK. The use of “marginal” donors for organ transplantation. The influence of donor age on outcome. Transplantation 1991; 51: 135141.
  • 6
    Tisone G, Manzia TM, Zazza S, et al. Marginal donors in liver transplantation. Transplant Proc 2004; 36: 525526.
  • 7
    Tector AJ, Mangus RS, Chestovich P, et al. Use of extended criteria livers decreases wait time for liver transplantation without adversely impacting posttransplant survival. Ann Surg 2006; 244: 439450.
  • 8
    Gruttadauria S, Cintorino D, Mandala L, et al. Acceptance of marginal liver donors increases the volume of liver transplant: Early results of a single-center experience. Transplant Proc 2005; 37: 25672568.
  • 9
    Renz JF, Kin C, Kinkhabwala M, et al. Utilization of extended donor criteria liver allografts maximizes donor use and patient access to liver transplantation. Ann Surg 2005; 242: 556563; discussion 563–555.
  • 10
    Rinella ME, Alonso E, Rao S, et al. Body mass index as a predictor of hepatic steatosis in living liver donors. Liver Transpl 2001; 7: 409414.
  • 11
    Hoofnagle JH, Lombardero M, Zetterman RK, et al. Donor age and outcome of liver transplantation. Hepatology 1996; 24: 8996.
  • 12
    Alkofer B, Samstein B, Guarrera JV, et al. Extended-donor criteria liver allografts. Semin Liver Dis 2006; 26: 221233.
  • 13
    Feng S, Goodrich NP, Bragg-Gresham JL, et al. Characteristics associated with liver graft failure: The concept of a donor risk index. Am J Transplant 2006; 6: 783790.
  • 14
    Busuttil RW, Tanaka K. The utility of marginal donors in liver transplantation. Liver Transpl 2003; 9: 651663.
  • 15
    Gastaca M, Valdivieso A, Pijoan J, et al. Donors older than 70 years in liver transplantation. Transplant Proc 2005; 37: 38513854.
  • 16
    Scotte M, Dousset B, Calmus Y, Conti F, Houssin D, Chapuis Y. The influence of cold ischemia time on biliary complications following liver transplantation. J Hepatol 1994; 21: 340346.
  • 17
    Cassuto JR, Patel SA, Tsoulfas G, Orloff MS, Abt PL. The cumulative effects of cold ischemic time and older donor age on liver graft survival. J Surg Res 2008; 148: 3844.
  • 18
    Yu L, Koepsell T, Manhart L, Ioannou G. Survival after orthotopic liver transplantation: The impact of antibody against hepatitis B core antigen in the donor. Liver Transpl 2009; 15: 13431350.
  • 19
    Reich DJ, Hong JC. Current status of donation after cardiac death liver transplantation. Curr Opin Organ Transplant 2010; 15: 316321.
  • 20
    Mangus RS, Fridell JA, Vianna RM, et al. No difference in clinical transplant outcomes for local and imported liver allografts. Liver Transpl 2009; 15: 640647.
  • 21
    Schemmer P, Nickkholgh A, Hinz U, et al. Extended donor criteria have no negative impact on early outcome after liver transplantation: A single-center multivariate analysis. Transplant Proc 2007; 39: 529534.
  • 22
    Nickkholgh A, Weitz J, Encke J, et al. Utilization of extended donor criteria in liver transplantation: A comprehensive review of the literature. Nephrol Dial Transplant 2007; 22 (Suppl 8): viii29viii36.
  • 23
    Garonzik-Wang JM, James NT, Weatherspoon KC, et al. The aggressive phenotype: Center-level patterns in the utilization of suboptimal kidneys. Am J Transplant 2012; 12: 400408.
  • 24
    Grams ME, Womer KL, Ugarte RM, Desai NM, Montgomery RA, Segev DL. Listing for expanded criteria donor kidneys in older adults and those with predicted benefit. Am J Transplant 2010; 10: 802809.
  • 25
    Bambha K, Kim WR, Kremers WK, et al. Predicting survival among patients listed for liver transplantation: An assessment of serial MELD measurements. Am J Transplant 2004; 4: 17981804.
  • 26
    Kremers WK, van IM, Kim WR, et al. MELD score as a predictor of pretransplant and posttransplant survival in OPTN/UNOS status 1 patients. Hepatology 2004; 39: 764769.
  • 27
    Narayanan Menon KV, Nyberg SL, Harmsen WS, et al. MELD and other factors associated with survival after liver transplantation. Am J Transplant 2004; 4: 819825.
  • 28
    Schaffer RL, 3rd, Kulkarni S, Harper A, Millis JM, Cronin DC, 2nd. The sickest first? Disparities with model for end-stage liver disease-based organ allocation: One region's experience. Liver Transpl 2003; 9: 12111215.
  • 29
    Trotter JF, Osgood MJ. MELD scores of liver transplant recipients according to size of waiting list: Impact of organ allocation and patient outcomes. JAMA 2004; 291: 18711874.
  • 30
    Wiesner R, Edwards E, Freeman R, et al. Model for end-stage liver disease (MELD) and allocation of donor livers. Gastroenterology 2003; 124: 9196.
  • 31
    Edwards EB, Roberts JP, McBride MA, Schulak JA, Hunsicker LG. The effect of the volume of procedures at transplantation centers on mortality after liver transplantation. N Engl J Med 1999; 341: 20492053.
  • 32
    Segev DL, Kucirka LM, Nguyen GC, et al. Effect modification in liver allografts with prolonged cold ischemic time. Am J Transplant 2008; 8: 658666.
  • 33
    Segev DL, Maley WR, Simpkins CE, et al. Minimizing risk associated with elderly liver donors by matching to preferred recipients. Hepatology 2007; 46: 19071918.