Journal Logo

Original Clinical Science—Liver

Predicting Liver Allograft Discard

The Discard Risk Index

Rana, Abbas MD1; Sigireddi, Rohini R. BA1; Halazun, Karim J. MD2; Kothare, Aishwarya1; Wu, Meng-Fen MS3; Liu, Hao PhD3; Kueht, Michael L. MD1; Vierling, John M. MD1; Sussman, Norman L. MD1; Mindikoglu, Ayse L. MD1; Miloh, Tamir MD4,5; Galvan, N. Thao N. MD1; Cotton, Ronald T. MD1; O’Mahony, Christine A. MD1; Goss, John A. MD1

Author Information
doi: 10.1097/TP.0000000000002151
  • Free

The Donor Risk Index (DRI) was published in 2006 and helped refine the concepts of the quality in liver allografts.1 The DRI used donor factors and cold ischemia time to predict graft failure. Over the years, the DRI has demonstrated poor predictive ability for allograft failure and accordingly has had limited clinical relevance.2-5 A major contributing factor to the acceptance of marginal donor livers is the subjective assessment of the donor allograft by the surgeon. This assessment, combined with liver biopsy results, often results in acceptance of an allograft despite adverse donor demographic and laboratory test results.5,6 This weight of the surgical assessment results in reduced predictive values for donor demographics and laboratory values.

Instead of using graft failure as the primary endpoint as was done with the DRI, it was our aim to look at allograft discard as the endpoint of failure. A resulting Discard Risk Index (DSRI) that accurately predicts liver donor discard at the time of DonorNet assessment would have several direct clinical uses. First, the index would stratify allografts based on the outcome of interest in allocation, risk of allograft discard. Second, it would enable a policy of preferential allocation of marginal donors to aggressive centers. We would speculate that aggressive centers would be more likely to use marginal donors and thereby increase the net donor yield.7-9

We hypothesized that information available on DonorNet at the time of an organ offer can be used to accurately predict the risk that the donor allograft will be discarded. Thus, the aim of this study was to use the identified donor risk factors for discard to construct an accurate index or direct clinical application using DonorNet information.


Study Population

We retrospectively analyzed the United Network for Organ Sharing (UNOS) database to identify all deceased donors where at least 1 organ was procured for the purpose of transplantation between January 1, 2002, and September 30, 2016. Our analysis used data from the deceased donor registry collected by the Organ Procurement and Transplantation Network (OPTN). We included all deceased donors older than 10 years to focus on those deceased donors potentially capable of providing liver allografts to adults. Because the focus of this analysis is on the deceased donor's clinical characteristics and laboratory values and their potential to predict discard rates, we excluded eligible donors where consent was not requested or obtained (n = 1329). A total of 109 540 deceased donors were included in this analysis. The liver allografts in 84 548 (77.2%) deceased donors were transplanted into recipients. The liver allografts in 24 992 (22.8%) deceased donors were discarded (Table 1).

Demographic and clinical characteristics

Statistical Analysis

Data were analyzed using a standard statistical software package, Stata 12.1 (Stata Corp, College Station, TX). Continuous variables were reported as a mean ± standard deviation and compared using the Student t test. Nonparametric variables were compared using the Mann-Whitney U test. Contingency table analysis was used to compare categorical variables. Results were considered significant at a P value less than 0.05. All reported P values were two-sided. The primary outcome measure was liver allograft discard. In this analysis, the two potential outcomes for deceased donors were that the liver allograft was either transplanted or discarded. Deceased donor clinical characteristics and laboratory values were the independent variables, whereas liver allograft discard was the dependent variable in logistic regression analysis. Risk factors that were significant in univariate analysis (P < 0.05) were included in the multivariate logistic regression analysis. We confirmed the stability of the model by performing backward elimination and stepwise elimination using elimination based on a P value greater than 0.05. The entire cohort was randomized into two distinct cohorts. The training set (n = 72 297) included 66% of the cohort and was used to derive a predictive index for discard. The validation set (n = 37 243) included 34% of the cohort and was used to internally validate the index.

Risk Factors

Deceased donor risk factors for liver allograft discard are listed in Table 2 and only include factors available in the DonorNet presentation of the initial organ offer. Laboratory values are terminal before recovery. Hepatitis B core antibody (ab) and hepatitis C ab had a high-entry completion percentage, these covariates were used instead of nucleic acid testing, which had a low-entry completion percentage. Continuous variables were categorized using clinically relevant groupings. Donation after circulatory death was included to create an index that can be universally applied to deceased donors. Height was reported in centimeters. Cold time was reported in hours.

Discard risk factors considered in univariate analysis using the training set

Liver Biopsies

We did not incorporate liver biopsies into the DSRI because they are not consistently available at the DonorNet presentation of the initial organ offer. Furthermore, the data collected on liver biopsies in the OPTN database is not adequate for this analysis. The timing of the liver biopsies (intraoperatively vs preoperatively) is not recorded, nor is the degree of fibrosis or necrosis. Only macrosteatosis and microsteatosis are recorded in 78% of the biopsies (30 160/38 339).

Because this is likely to be a contentious issue, we incorporated biopsy into a sensitivity analysis.

Missing Variables

Missing variables were imputed using predictive mean matching imputation method for incomplete predictors in the OPTN database (Table 2).

Development of DSRI

Logistic regression analysis determined the predictors of liver allograft discard using the training set. Deceased donor variables analyzed using univariate analysis are listed in Table 2. Variables identified as statistically significant in univariate analysis were subjected to multivariate analysis. An index was created as a continuous risk score for donor discard using all available deceased donor clinical characteristics and laboratory values. In our DSRI, similar to the Kidney Donor Risk Index, each deceased donor start with a DSRI of 1.00, and significant risk factors are subsequently factored in.10

Using the validation set, we created 10 risk groups using increments of 10 percentiles. Because donation after brain death (DBD) and donation after circulatory death (DCD) deceased donors had distinctive discard rates and DSRI profiles, we created separate risk groups for DBD and DCD deceased donors. Model discrimination was assessed using the area under the receiver operating characteristic (ROC) curve.

Deceased donors included in this study were donors where at least 1 organ was procured for the purpose of transplantation. We speculate that elevated creatinine in marginal donors would rule out kidney allograft procurement and focus practicality of procurement on the liver allograft. For this reason, donor creatinine was not included in the DSRI.

Allograft Failure

We applied the DSRI and DRI to predict liver allograft failure in the validation set. The following equation was used for the DRI: DRI= exp[(0.154 if 40 ≤ age <50) + (0.274 if 50 ≤ age <60) + (0.424 if 60 ≤ age <70) + (0.501 if 70 ≤ age) + (0.079 if cause of death [COD] = anoxia) + (0.145 if COD = cerebrovascular accident [CVA]) + (0.184 if COD = other) + (0.176 if race = African American) + (0.126 if race = other) + (0.411 if DCD) + (0.422 if partial/split) + (0.066 ((170 − height)/10)) + (0.105 if regional share) + (0.244 if national share) + (0.010 × cold time)]. The indices were applied in 84 548 patients who underwent liver transplantation from January 1, 2002, and September 30, 2016. We also performed a retrospective analysis of UNOS deidentified patient-level data. Donor and recipient characteristics were reported at the time of transplant. Follow-up information was collected at 6 months and then yearly after transplantation. All recipients were followed up from the time of transplantation to last known follow-up (mean follow up time, 4.4 ± 3.9 years). The primary outcome was liver allograft failure (n = 26 784, 31.3%). Time to graft failure was assessed as the date of liver transplantation to the date of allograft failure. Kaplan-Meier analysis was used for time-to-event analysis. Model discrimination was assessed using the area under the ROC curve.

Geographic Variation of Allograft Discard

We analyzed liver allograft discard by region and deidentified donor-specific area. The procuring center that declined the allograft was not available in the database; therefore, we could not provide center-specific discard rates.

Defining Aggressive Centers

We used the DSRI to define marginal allografts. We arbitrarily defined a DSRI 80 to 100 or the top 20% of allografts most likely for discard as marginal. We then quantified the degree of center aggressiveness by analyzing the percentage of marginal liver transplanted to the total number of livers transplanted.


Study Population

Among the study population of 109 540 deceased donors, 84 548 liver allografts (77.2%) were transplanted, whereas 24 992 were discarded (22.8%). Demographic and clinical characteristics are summarized in Table 1. The rate of liver allograft discard has been relatively constant over the multiple years (Figure 1). As expected, discarded liver allografts predominantly came from donors who were older, heavier, and had greater laboratory derangements (Table 1).

Brain dead donor liver allograft discard by year. Abscissa—percent discarded. Ordinate—year.

Data Entry Rate

Data entry completion for variables is listed in Table 2. Most variables were well populated. Missing variables were imputed using the predictive mean matching imputation method.

Univariate and Multivariate Analyses

Table 2 lists all the risk factors that were analyzed. Risk factors that were significant in univariate logistic regression analysis were subjected to multivariate logistic regression analysis. The risk factors that were significant in multivariate analysis are presented in Table 3. The most significant risk factors for liver allograft discard in multivariate analysis were total bilirubin level greater than 10 mg/dL (odds ratio [OR], 25.23; confidence interval [CI], 17.32-36.77), donation after circulatory death (OR, 14.13; CI, 13.30-15.01), total bilirubin of 5 to 10 mg/dL (OR, 7.57; CI, 6.32-9.05), aspartate transaminase (AST) of 500 IU/L or greater (OR, 4.88; CI, 4.20-5.66), total bilirubin levels of 4 to 5 mg/dL (OR, 4.14; CI, 3.40-5.04), and hepatitis C ab-positive (OR, 3.43; CI, 3.16-3.74).

Discard risk factors included in multivariate analysis using the training set

Relative Weight of Risk Factors

We performed relative weighting among risk factors by assigning a risk point for every 10% increase in the OR (Table 3). These were reported to give the clinician a sense of the relative weight of each risk factor but were not used in the DSRI.

Risk Index

Table 4 presents the risk equation used to calculate the DSRI. The DSRI calculates the risk of liver allograft discard compared to the reference deceased donor with the following characteristics: male; donor age, 31 to 45 years; alanine transaminase (ALT), less than 50; AST, less than 50 IU/mL; total bilirubin, less than 1 mg/dL; body mass index (BMI), 20 to 30 kg/m2; serum Na, 135 to 155 mmol/L; COD not CVA; not African-American; not CDC high risk; not DCD; anti-HBc ab-negative; anti-Hepatitis C (HCV) ab-negative; no history of diabetes or hypertension.


Using the DSRI, we created risk groups using 10 percentile increments (Table 4). Separate risk groups were created for DBD and DCD donors. In DBD donors, the discard percentage increases from 3% to 50% (P < 0.001) (Figure 2). Liver biopsies were performed in 8% of DSRI 10 allografts compared with 57% of DSRI 90 allografts (P < 0.001). Only 2% of DSRI 10 allografts were shared nationally compared with 14% of DSRI 90 allografts (P < 0.001). In DCD donors, the discard percentage increases from 43% to 92% (P < 0.001). Only 5% of DSRI 10 allografts were shared nationally compared with 14% of DSRI 90 allografts (P < 0.001).

Discard rates and the DSRI. Abscissa—percent discarded. Ordinate—DSRI by 10 percentile increments. Each group is significant (P < 0.05) compared with DSRI 10.

Using the 10th percentile group as the reference, the DBD DSRI 20th percentile group had an OR of 1.32 (1.13-1.54) for discard, DSRI 30 was 2.31 (2.02-2.67), DSRI 40 was 3.30 (2.89-3.77), DSRI 50 was 4.63 (4.06-5.27), DSRI 60 was 5.95 (5.24-6.76), DSRI 70 was 8.01 (7.06-9.08), DSRI 80 was 10.17 (8.98-11.51), DSRI 90 was 14.31 (12.67-16.19), and DSRI greater than 90 was 32.34 (28.63-36.53).

Using the 10th percentile group as the reference, the DCD DSRI 20th percentile group had an OR of 1.33 (1.13-1.57) for discard, DSRI 30 was 1.55 (1.31-1.83), DSRI 40 was 2.35 (1.98-2.77), DSRI 50 was 2.91 (2.46-3.44), DSRI 60 was 3.72 (3.11-4.45), DSRI 70 was 4.43 (3.71-5.30), DSRI 80 was 5.88 (4.88-7.07), DSRI 90 was 6.67 (5.51-8.05), and DSRI greater than 90 was 16.15 (12.72-20.50).

The area under the ROC curve (C-statistic) for liver allograft discard using the DSRI is 0.804 (0.800-0.808) for the training set and 0.802 (0.797-0.808) for the validation set (Figures 3 and 4). When bilirubin alone was used as a continuous variable it predicted discard with a c-statistic of 0.548 (0.544-0.552).

Histogram of DSRI and brain dead donors and donation after circulatory death donors. Abscissa—percentage of deceased donors. Ordinate—DSRI.
The area under the ROC curve when DSRI is used to predict liver allograft discard.

Liver Biopsies

Out of the 24 992 liver allografts that were discarded, 9923 were biopsied (39.7%). Only 8% of low-risk allograft (DSRI 10) were biopsied compared to 60% of high risk allografts (DSRI, 90) (P < 0.001) (Table 4). Among the highest risk group (DSRI > 90), only 40% of liver allografts were discarded when biopsies were performed compared to a 60% discard rate when biopsies were not performed (P < 0.001).

Liver biopsies were not included in the index because they are not readily available at the time of DonorNet offer presentation. Furthermore, the data collected on liver biopsies in the OPTN database are not adequate for this analysis. The timing of the liver biopsies (intraoperatively vs preoperatively) is not recorded, nor is the degree of fibrosis or necrosis.

Because this is likely to be a contentious issue, we forced biopsy into a sensitivity analysis. We included obtaining the biopsy as a covariate in addition to the degree of macro and microsteatosis. There was no statistical improvement in the accuracy of the model which incorporated biopsies. The C-statistic in the validation set was 0.781 (0.774-0.789).

Donor Risk Index

Using the Donor Risk Index to predict liver allograft discard, the index achieves a C-statistic of 0.690 (0.686-0.694). Cold time, regional and national sharing are significant factors in the DRI, but are unknown for discarded livers and could not be used. In the original publication of the DRI, a DRI greater than 2.0 or the worst 6th percentile had a discard percentage of 12.5%.1

Liver Allograft Failure

Even though the DSRI was designed to predict liver allograft discard, it also correlated with liver allograft failure after transplantation (Table 4). Despite this correlation, the suboptimal C-statistic of 0.542 (0.538-0.546) precludes clinical use. On the other hand, the DRI was designed to predict allograft failure. However, is also not an inaccurate predictor with a C-statistic of 0.569 (0.565-0.573).

Reason for Discard

A reason for discard was coded for only 36.7% of the discarded liver allografts. Despite the limitations created by the poor entry completion, we did identify interesting patterns. For example, biopsy findings were the reason for discard in 21% of DSRI 10 discards compared to 57% of DSRI greater than 90 discards (P < .001). Vascular damage, organ damage, and anatomical abnormalities were the reason for discard in 18.2% of DSRI 10 discards compared with 9.2% of DSRI > 90 discards (P = 0.007). “Other” was listed as the discard reason for 31.8% of DSRI 10 discards compared with 11.9% of DSRI greater than 90 discards (P < 0.001).

Geographic Variation of Allograft Discard

Allograft discard varied from 16% in region 3 to 29.8% in region 2 (P < 0.001). The remaining UNOS regions had the following discard percentage: region 1, 27.9%; region 4, 21.1%; region 5, 24.6%; region 6, 29.3%; region 7, 22.3%; region 8, 21.8%; region 9, 21.3%; region 10, 24.6%; and region 11, 20.1%. The allograft discard by each individual organ procurement organization (OPO) varied from 11% to 39% (P < 0.001) (Figure 5A).

A, Allograft discard by OPO. Abscissa—percent allograft discard. Ordinate—deidentified OPO number. B, Defining aggressive centers using the DSRI. Abscissa—percent marginal liver used marginal livers transplanted/total livers transplanted. Marginal livers were defined as DSRI 80 percentile or above. Ordinate—deidentified transplant center.

Defining Aggressive Centers

Our definition of marginal donors was a DSRI 80 to 100 percentile or 20% of allografts most likely to be discarded. Fifteen (11%) of 135 centers used more than 25% marginal donors. Nineteen (14%) centers used no marginal donors. Forty-seven (35%) centers used under 10% marginal donors. Figure 5B demonstrates the wide spectrum of center aggressiveness.


In this study, we conducted an extensive multivariate analysis of predictors of liver allograft discard using information available from DonorNet at the initial organ offer. The most significant risk factors for discard were as follows: total bilirubin level, greater than 10 mg/dL (OR, 23.39; CI, 17.39-31.46); donation after circulatory death (OR, 13.39; CI, 12.80-14.01); total bilirubin, 5 to 10 mg/dL (OR, 8.21; CI, 7.09-9.49); AST, of 500 IU/L or greater (OR, 4.52; CI, 4.01-5.09); total bilirubin levels, 4 to 5 mg/dL (OR, 4.32; CI, 3.68-5.06); and hepatitis C ab-positive (OR, 3.35; CI, 3.13-3.59).

Using all of the significant risk factors, we constructed a weighted index, the DSRI, to predict liver allograft discard from only information available at the initial organ offer. The DSRI had a very good ability to predict liver allograft discard with a c-statistic of 0.8. However, the highest-risk group (90th percentile), as defined by the DSRI, still only had a discard percentage for DBD donors of 50%. When a biopsy was performed in this high-risk group (90th percentile), the allograft was used 60% of the time compared with 40% when a biopsy was not used. Unfortunately, there is likely bias in the use of liver biopsies and significant limitations in the quality of the data. Given the influential role biopsies play in clinical practice, we would suggest that the OPTN increase their efforts to collect complete and quality data on biopsies. At the very least, we capture the reason of biopsy and degree of necrosis and fibrosis. If validated with better quality data, this analysis may support a new policy proposal that every liver allograft be biopsied before deciding to discard. It may even be prudent to require a bedside biopsy on all high discard risk deceased donors before allocation and procurement; however, there may be logistical and financial issues that may prevent execution of these policies.

Our extensive multivariate analysis identified 15 significant risk factors for liver allograft discard. Donor age, donor race, donor height, cause of donor death, and donation after circulatory death are factors that the DSRI shares with DRI. CDC high risk, sex, hepatitis B core ab, hepatitis C ab, history of diabetes, history of hypertension, AST, ALT, total bilirubin, serum sodium, and BMI are unique to the DSRI. It is clear why a majority of the covariates are risk factors for liver allograft discard. Laboratory derangements may demonstrate underlying liver dysfunction and ischemic injury to the liver. A history of diabetes and elevated BMI correlated with liver steatosis, an established risk factor for live allograft dysfunction.11-15 A history of hypertension might correlate with the degree of atherosclerosis in the hepatic artery. Finally, although some center may decline CDC high-risk allografts, there are many centers who widely accept CDC high-risk allografts.16 This factor may simply be a marker for younger allografts.

To the best of our knowledge, our study is the first direct analysis of risk factors for liver allograft discard designed to create a clinically useful DSRI to predict the probability of discard based on information available at the time of a donor offer. In liver transplantation, the original publication of the DRI demonstrated that discard rates correlated with increasing DRI.1 There has been some exploration of this idea in other solid organ transplants, with the importance of trending allograft discard rates.2,17 A publication in the early period of kidney transplantation identified risk factors predisposing kidney allografts to discard.18 More recently, Tanriover et al19 correlated Kidney Donor Profile Index (KDPI) with kidney allograft discards. Maglione et al20 explored the influence of demographic factors, surgical assessment, and cold ischemia time on pancreas allograft discards. In a testament to the importance of allograft discard, the National Kidney Foundation has convened a Task Force of experts and key stakeholders for a consensus conference to identify practical solutions to reduce discard in kidney transplantation.

The DRI does not accurately predict liver allograft failure after transplantation. In this analysis, we demonstrate a C statistic of 0.5. The DRI may have limited accuracy because the donor surgeon's subjective assessment in conjunction with a liver biopsy may be the dominant predictor of failure. It may also be that graft failure is too rare. The DSRI, similarly, also does not have an ability to predict graft failure after transplantation and should not be used for this purpose or for the purpose of evaluating allografts for transplantation. The real contribution of the DSRI is an objective measure to grade allografts with the outcome that matters, discard. We further contend that the DSRI, using 15 weighted factors to predict donor discard with a c statistic of 0.80, will likely outperform the current clinical judgement. If we take isolated factors, such as bilirubin, as a continuous variable to predict allograft discard, it has a c statistic of 0.55. Donation after circulatory death predicted discard with a c statistic of 0.63. HCV positivity predicted discard with a c statistic of 0.52.

There is significant room to improve our discard rates because our analysis demonstrates significant geographic variation in discard. We will pursue this idea further in a future analysis by determining whether the discard rate is lower when an aggressive program is the procuring team. There is, however, indirect evidence that bypassing nonaggressive centers (expedited allocation) may increase the donor yield. Lai et al21 analyzed the patterns of nationally placed liver and found that a high proportion of acceptances was on the first national offer. Hayashi et al22 identified region 9 as being the most aggressive UNOS region. They estimated that approximately 300 discarded livers may have been used if region 9 criteria were broadly used.

Many have pointed out that the current state of expedited allocation of marginal livers lacks transparency and uniformity. There is significant regional variation, as 1 analysis demonstrates that region 1 used 33% of all expedited allocated liver allografts.23 In another analysis, 73% of expedited livers were transplanted in 1 of 6 aggressive centers.21 There is also great variation in the use of expedited allocation by donor-specific areas, suggesting significant discretion and a lack of uniformity.22 The DSRI can address these issues by providing objective means of identifying marginal allografts (ie, DSRI < 10th percentile) and aggressive centers (>10% usage of marginal allografts in the past). We envision a transparent policy for expedited allocation where any marginal allograft (DSRI < 10th percentile) after local refusal would be allocated to aggressive centers (>10% usage of marginal allografts as defined by DSRI) by distance. In this scheme, the objective grading of both allografts and centers using the DSRI is based on the outcome that matters, allograft discard.


The OPTN deceased donor database only includes candidates who were taken to the OR with the intent to procure organs for transplantation. Eligible donors who were outright refused for all solid organ transplants were not included in this analysis. A more complete analysis would include all eligible donors but these data are not available in the deceased donor OPTN database. Another significant limitation is that this analysis does not account for center behavior or recipient factors. If a risk adverse center procures the allograft and refuses, we can assume that it is less likely that allograft will be reallocated compared with the scenario where an aggressive center procures. Despite these limitations, the DSRI did achieve a c-statistic of 0.80 when applied to over 100 000 deceased donors.

Since the passage of the National Transplantation Act of 1984, data entry has been mandatory for all U.S. transplant centers. Nevertheless, all patient registries often suffer from variability in data entry. The findings from this study were based on large cohorts of patients and are unlikely to be significantly impacted by small amounts of missing data. We attempted to account for missing data with multiple imputation analyses.


The DSRI uses 15 significant risk factors, all available at initial DonorNet assessment of a liver allograft, to accurately predict risk of liver allograft discard with a C-statistic of 0.80 in over 100 000 deceased donors. The use of the DSRI can help predict liver allograft discard. The DSRI can be used by to preferentially allocate marginal allografts to aggressive centers to maximize the donor yield and expedite allocation.


1. 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;6:783–790.
2. Renz JF. A critical analysis of liver allograft utilization from the US deceased donor pool. Liver Transpl. 2010;16:543–547.
3. Mangus RS, Fridell JA, Vianna RM, et al. No difference in clinical transplant outcomes for local and imported liver allografts. Liver Transpl. 2009;15:640–647.
4. Akkina SK, Asrani SK, Peng Y, et al. Development of organ-specific donor risk indices. Liver Transpl. 2012;18:395–404.
5. Yersiz H, Lee C, Kaldas FM, et al. Assessment of hepatic steatosis by transplant surgeon and expert pathologist: a prospective, double-blind evaluation of 201 donor livers. Liver Transpl. 2013;19:437–449.
6. McCormack L, Dutkowski P, El-Badry AM, et al. Liver transplantation using fatty livers: always feasible? J Hepatol. 2011;54:1055–1062.
7. Ozhathil DK, Li YF, Smith JK, et al. Impact of center volume on outcomes of increased-risk liver transplants. Liver Transpl. 2011;17:1191–1199.
8. 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;242:556–563. discussion 563–5.
9. Ghinolfi D, Marti J, De Simone P, et al. Use of octogenarian donors for liver transplantation: a survival analysis. Am J Transplant. 2014;14:2062–2071.
10. Rao PS, Schaubel DE, Guidinger MK, et al. A comprehensive risk quantification score for deceased donor kidneys: the Kidney Donor Risk Index. Transplantation. 2009;88:231–236.
11. Browning JD, Szczepaniak LS, Dobbins R, et al. Prevalence of hepatic steatosis in an urban population in the United States: impact of ethnicity. Hepatology. 2004;40:1387–1395.
12. Marchesini G, Moscatiello S, Di Domizio S, et al. Obesity-associated liver disease. J Clin Endocrinol Metab. 2008;93(11 Suppl 1):S74–S80.
13. Marchesini G, Bugianesi E, Forlani G, et al. Nonalcoholic fatty liver, steatohepatitis, and the metabolic syndrome. Hepatology. 2003;37:917–923.
14. Uemura T, Randall HB, Sanchez EQ, et al. Liver retransplantation for primary nonfunction: analysis of a 20-year single-center experience. Liver Transpl. 2007;13:227–233.
15. Brokelman W, Stel AL, Ploeg RJ. Risk factors for primary dysfunction after liver transplantation in the University of Wisconsin solution era. Transplant Proc. 1999;31:2087–2090.
16. Kucirka LM, Namuyinga R, Hanrahan C, et al. Formal policies and special informed consent are associated with higher provider utilization of CDC high-risk donor organs. Am J Transplant. 2009;9:629–635.
17. Ojo AO, et al. Organ donation and utilization in the USA. Am J Transplant. 2004;4(Suppl 9):27–37.
18. Lucas BA, Vaughn WK, Spees EK, et al. Identification of donor factors predisposing to high discard rates of cadaver kidneys and increased graft loss within one year posttransplantation—SEOPF 1977-1982. South-Eastern Organ Procurement Foundation. Transplantation. 1987;43:253–258.
19. Tanriover B, Mohan S, Cohen DJ, et al. Kidneys at higher risk of discard: expanding the role of dual kidney transplantation. Am J Transplant. 2014;14:404–415.
20. Maglione M, Ploeg RJ, Friend PJ. Donor risk factors, retrieval technique, preservation and ischemia/reperfusion injury in pancreas transplantation. Curr Opin Organ Transplant. 2013;18:83–88.
21. Lai JC, Feng S, Vittinghoff E, et al. Offer patterns of nationally placed livers by donation service area. Liver Transpl. 2013;19:404–410.
22. Hayashi PH, Axelrod DA, Galanko J, et al. Regional differences in deceased donor liver transplantation and their implications for organ utilization and allocation. Clin Transplant. 2011;25:156–163.
23. Kinkhabwala M, Lindower J, Reinus JF, et al. Expedited liver allocation in the United States: a critical analysis. Liver Transpl. 2013;19:1159–1165.
Copyright © 2018 Wolters Kluwer Health, Inc. All rights reserved.