Journal Logo

Original Clinical Science—Liver

Regional Variation in Utilization and Outcomes of Liver Allografts From Donors With High Body Mass Index and Graft Macrosteatosis: A Role for Liver Biopsy

Steggerda, Justin A. MD1; Kim, Irene K. MD2; Malinoski, Darren MD3; Klein, Andrew S. MD2; Bloom, Matthew B. MD4

Author Information
doi: 10.1097/TP.0000000000002379

Abstract

The prevalence of obesity in the United States has reached epidemic proportions, with 36.5% of the population considered obese (body mass index [BMI], ≥30 kg/m2).1 Obesity carries with it significant health risks, including insulin resistance, cardiovascular disease, and hepatic steatosis.2

Considering the potential impact of these associated health risks on graft quality and function, high BMI (hBMI) donors (≥30 kg/m2) have been extensively evaluated for transplantation. In kidney transplantation alone, BMI has been included in the Kidney Donor Profile Index as a predictor of delayed graft function and worse outcomes.3 In liver transplantation, the role of donor BMI remains controversial. Yoo et al4 found no difference in rates of graft primary nonfunction (PNF) and early retransplantation across 4 donor BMI groups, including those with BMI of 30 kg/m2 or greater. In the Donor Risk Index, the Eurotransplant Donor Risk Index, and donor Model for End Stage Liver Disease (dMELD), multivariable regressions did not find donor BMI to be a significant predictor of outcomes when considering all transplant recipients.5-7 However, in an analysis of more contemporary populations, Bloom et al8 showed that, among recipients with a mean follow-up of 74 days, lower donor BMI was a predictor of early graft survival.

Liver steatosis is a hepatic manifestation of the metabolic syndrome and is strongly correlated with donor BMI. Steatosis has been identified in 30% of all deceased donors, and up to 76% of the potential living donors with BMI greater than 28 kg/m2.9,10 While grafts with hepatic macrosteatosis (MaS) less than 30% are routinely transplanted, association of moderate steatosis (MaS ≥30%) with increased risks of PNF, early allograft dysfunction (EAD), and overall worse survival are well documented.9,11-13 Grafts with severe steatosis (MaS >60%) are generally considered unsuitable for transplant. Still, individual reports have shown good outcomes with donor livers exhibiting 40% MaS or more.14,15 Additionally, consensus on definitions and evaluation of MaS is lacking, particularly distinctions between small-droplet MaS and microsteatosis,16 as it has been shown that MaS is a risk factor for graft loss while microsteatosis alone, is not.17

As waitlist growth outpaces transplantation rates, efforts to expand the donor pool and limit waitlist mortality are ongoing. Continued evaluation of the available donor pool and selection criteria is paramount to increase beneficial utilization. Using a nationally derived database, we evaluate the prevalence of obesity among potential liver donors as well as trends in biopsy performance, organ utilization, and associated outcomes over the past decade across donation regions.

MATERIALS AND METHODS

Donor and Recipient Populations

The United Network for Organ Sharing prospectively collects data on potential donors and recipients which is reported in the Standard Transplant Analysis and Research file. We performed a retrospective review of this database, identifying all potential adult donors between January 1, 2006, and June 30, 2016. After excluding pediatric donors (<18 years old), donation after circulatory death (DCD) donors, and those without BMI information, 67 154 potential donors were available for analysis. Donor-recipient pairs, matched by unique United Network for Organ Sharing identification numbers, and excluding all pediatric recipients (<18 years old), multiorgan recipients, and recipients of DCD organs within this same time period, left 51 917 transplants for evaluation. Donors and recipients with missing data were further excluded from multivariable analyses, leaving 64 816 potential donors evaluated for organ utilization and 50 742 transplants evaluated for graft survival.

For donors, data were collected for demographic characteristics, including age, gender, ethnicity, BMI, blood type, and DCD status as well as medical and social histories for presence of hypertension, diabetes, prior myocardial infarction, hepatitis C viral infection, hepatitis B viral infection, history of alcohol, cigarette, or drug use, and Centers for Disease Control classification as high-risk for blood borne disease. When liver biopsy was performed, reported levels of MaS were included for analysis. Donor-recipient pairs were also evaluated for demographic factors, MELD at transplant, health status at time of transplantation (ie, on life support, on ventilator, recent dialysis) and medical histories including history of diabetes, portal vein thrombosis, or prior transjugular intrahepatic portovenous shunt. In the case of split liver transplantation, donors were counted once while each recipient was individually evaluated.

Outcomes

Eleven Organ Procurement and Transplant Network (OPTN) regions were identified and evaluated. Two primary outcomes of interest were utilization rates of allografts from hBMI donors and grafts with MaS on biopsy for transplantation, as well as graft survival after transplantation. Secondary outcomes included rates of liver biopsy performance in potential donors and utilization of biopsied organs. Graft failure was defined as the period between transplantation and graft loss necessitating retransplantation or recipient death, whichever occurred first.6,18 Graft survival was assessed for the duration of follow-up or 3 years after transplantation, whichever came first.

Statistical Analysis

Statistical analysis was performed in JMP Pro 13 (SAS Institute Inc., Cary, NC). Single variable analyses were performed using Student t tests and analysis of variance where appropriate. χ2 analysis was performed with multicategorical variables. In the donor population, univariate and multivariate logistic regression models were used to assess likelihood of organ utilization and performance of a predonation biopsy across regions. In the transplant population, univariate and multivariable analyses were performed using Cox proportional hazards regression modeling. For both models, a baseline multivariate model was determined and then factors of interest (ie, donor BMI, graft MaS, and region of transplant) were included to assess their effects on the outcome. The baseline multivariate model was identified using univariate logistic regression. Factors with P value less than 0.10 were then included for multivariate analysis. Multivariable model was constructed using a stepwise manual regression, with factors with P value <0.01 remaining in the final baseline model. All factors remained in the model when the additional factors of interest were included, even if P value increased above 0.01. Kaplan-Meier survival analyses were performed and were censored for patients with a functioning graft at last known follow-up or 3 years. Median survival was not reached in all subgroups analyzed, therefore 75% survival rates are reported for comparison; however, P values are reported according to Wilcoxon-Rank Sum and assess the entire duration of survival. Findings were considered significant with P value less than 0.05; however, when comparing variables within a multivariable regression, Bonferoni corrections were calculated and more stringent P values were applied accordingly.

RESULTS

Organ Utilization in hBMI Donors

Organ utilization was assessed by donor BMI and found to vary significantly across groups (Table 1, P < 0.0001). Donors with normal BMI (18.5 to <25 kg/m2) were used most frequently and therefore were considered the reference group. Univariate logistic regression to assess likelihood of organ utilization was performed and significant factors were included in a multivariable model which was used to adjust for all further donor analyses (Table S1, SDC,http://links.lww.com/TP/B610). Compared to donors with normal BMI, likelihood of organ utilization decreased significantly as donor BMI increased and showed a significant drop as donor BMI increased above 30 kg/m2, even after applying a Bonferoni correction (P < 0.008 for significance). When all donors with BMI of 30 kg/m2 or greater were combined, the likelihood of organ utilization was odds ratio (OR) of 0.509 (95% confidence interval [CI], 0.487-0.533; P < 0.0001).

TABLE 1
TABLE 1:
Likelihood of organ utilization by donor BMI

Differences between potential donors with low BMI and hBMI donors (≥30 kg/m2) were assessed (Table S2, SDC,http://links.lww.com/TP/B610). Briefly, hBMI donors were older and a greater proportion were black or female. They were significantly more likely to have anoxia and CVA or stroke as cause of death and much less likely to have death resulting from Trauma. Predonation biopsy was performed in 57.7% of hBMI donors compared with 35.5% of lower BMI donors (P < 0.0001). Nearly 54.8% of hBMI donors had hypertension and 13.6% had diabetes compared with 33.3% and 9.7% of lower BMI donors, respectively (P < 0.0001 for both). Overall, 84.3% of available lower BMI donors were used for transplantation, whereas only 73.7% of hBMI donors were used (P < 0.0001). Furthermore, only 56.0% of hBMI donors were multiorgan donors, whereas 69.8% of lower BMI donors provided a multiorgan donation (P < 0.0001).

The rate of donor biopsy increased as BMI increased (Figure 1A, P < 0.0001 across BMI groups). Allograft MaS of 30% or greater has been associated with increased rates of PNF and EAD.9,12,13 The incidence of MaS of 30% or greater on predonation biopsy also increased as donor BMI increased (Figure 1B, P < 0.0001 across BMI groups). The rates of organ utilization when a predonation biopsy was performed were lower than the overall utilization rate for all donors except those with BMI of 40 kg/m2 or greater. However, organ utilization was increased for donors with BMI of 30 kg/m2 or greater when MaS less than 30% was identified on predonation biopsy (Figure 1C, P < 0.0001 across groups for all BMI).

FIGURE 1
FIGURE 1:
Allograft biopsy and utilization across donor BMI groups. A, Percent of available donors within each BMI group who had predonation liver biopsy. B, Percent of biopsies showing MaS of 30% or greater from donors within each BMI group. C, Organ utilization is reported as percent of available donors within each BMI group. Subgroups examined overall donor utilization, utilization of donors who had pretransplant biopsy, and utilization of biopsied organs with MaS of 30% or greater.

Logistic regression was used to evaluate organ utilization by biopsy status (Table 2). In both lower BMI and hBMI donors, donors with predonation biopsy were less likely to be used for transplantation. Among donors who had pretransplant biopsy, the likelihood of utilization with findings of MaS of 30% or greater carried an OR of 0.122 (95% CI, 0.109-0.137; P < 0.0001) in low BMI donors and was similarly low in hBMI donors (OR, 0.135; 95% CI, 0.121-0.150; P < 0.0001). When compared with donors who did not undergo biopsy, lower BMI donors were less likely to be transplanted regardless of the findings of MaS. However, hBMI donors with MaS less than 30% were more likely to be utilized (OR, 1.343; 95% CI, 1.230-1.467; P < 0.0001), whereas the likelihood of organ utilization among those with MaS of 30% or greater remained extremely low.

TABLE 2
TABLE 2:
Likelihood of organ utilization by donor BMI

Regional Variation in Biopsy Performance and Donor Utilization

Regional differences in transplantation practices, with disparities in median MELD at transplant and ratios of deaths of eligible waitlisted candidates to transplants performed have been previously reported.19 Despite these differences, specific donation practices across regions are poorly characterized, so regional differences in the utilization of livers from hBMI donors and those with a predonation biopsy were examined.

The proportion of potential hBMI donors varied significantly across all regions, with the highest proportions in regions 2, 7, 8, and 10 and the lowest proportions in regions 5 and 6 (Figure 2A, P < 0.0001). Rate of predonation biopsy performance, both overall and in hBMI donors alone, also varied significantly across all regions (Figure 2B, P < 0.0001). Donor utilization rate significantly varied across all regions, with the highest utilization rates in regions 1, 3, and 8—all using more than 84% of potential donors (Figure 2C, P < 0.0001). Region 5 had the lowest utilization rate, transplanting livers from only 73.2% of available donors. The utilization of hBMI donors with and without biopsy varied across regions. Only regions 1, 6, and 7 showed higher rates of hBMI donor utilization with biopsy performance regardless of biopsy results. The rates of MaS of 30% or greater varied across regions, as did utilization of donor organs with macrosteastosis of 30% or greater (Figure 2D, P < 0.0001). Regions 1, 3, and 11 had the highest utilization rate of organs with MaS of 30% or greater, whereas region 6 had the lowest utilization rate. Rates of donor biopsy and organ utilization were used to evaluate between regions; however, given different sizes of population served, actual numbers of donors biopsied and organs transplanted are presented in Table S3, SDC (http://links.lww.com/TP/B610).

FIGURE 2
FIGURE 2:
Regional variation in donor biopsy performance and organ utilization. A, Percent of all total donors utilized for liver transplantation per region (P < 0.0001 across regions). B, Percent of potential donors undergoing predonation biopsy in the whole population as well as hBMI donors (P < 0.0001 across regions for both). C, Percent of potential donors utilized for transplantation among all available donors as well as hBMI donors with and without predonation biopsy. D, Percent of predonation biopsies with MaS of 30% or greater and utilization of organs with MaS of 30% or greater (P < 0.0001 across regions).

Differences in hBMI donor biopsy and organ utilization were further assessed within regions using multivariable logistic regression (Table 3A, B). Compared with donors without biopsy, region 1 had the highest likelihood of utilization when predonation biopsy showed MaS less than 30% with OR of 1.837 (95% CI, 1.124-3.000; P = 0.0153) for low BMI donors and OR of 5.445 (95% CI, 2.736-10.836; P < 0.0001) in hBMI donors. Lower BMI donors in region 9 were significantly less likely to be used, even with biopsy showing less than 30% MaS (OR, 0.674; 95% CI, 0.495-0.917; P = 0.0119). Among hBMI donors, however, biopsy with less than 30% MaS increased likelihood of utilization in regions 1, 3, 5, and 6. For all donors, regardless of BMI, biopsy with 30% or greater significantly decreased the likelihood of organ utilization.

TABLE 3
TABLE 3:
Regional variation in organ utilization with predonation biopsy

Multivariable logistic regression was used to compare donor utilization across regions. Region 9 had the median utilization rate for hBMI donors (n = 771, 75.3%) and therefore was used as reference region; Bonferoni correction yielded P less than 0.0045 as significant. Region 3 had the highest likelihood of hBMI donor utilization (OR, 1.178; 95% CI, 1.052-1.318; P = 0.0045) (Figure 3A). Region 6 had the lowest likelihood of hBMI donor utilization (OR, 0.399; 95% CI, 0.346-0.462; P < 0.0001). The likelihood of predonation biopsy in hBMI donors was most likely in region 6 (OR, 2.668; 95% CI, 2.668-3.366; P < 0.0001) and was also high in regions 11, 1, 8, and 7, all with OR greater than 1.5000 (Figure 3B). Region 3 was least likely to perform a predonation biopsy with an OR of 0.754 (95% CI, 0.647-0.878; P = 0.0003). The utilization of organs with MaS of 30% or greater varied across regions (Figure 3C). However, no regions were significantly more likely than region 9 to use these donors and only region 2 and region 6 were significantly less likely than region 9 to use an allograft with 30% or greater MaS. Finally, the likelihood of utilization of an hBMI donor with predonation biopsy showing less than 30% MaS was evaluated. Compared with region 9, utilization was significantly more likely in regions 1 and 3, whereas organ utilization was significantly less likely in region 6 (Figure 3D).

FIGURE 3
FIGURE 3:
Regional variation in donor utilization. Multivariable logistic regression was performed to compare OPTN regions for the likelihood of (A) hBMI donor utilization, (B) performance of predonation biopsies in hBMI donors, (C) utilization of organs with MaS less than 30% on predonation biopsy, and (D) utilization of organs with MaS ≥30% on predonation biopsy. Region 9 was considered the reference region for all comparisons. ORs and 95% CIs are reported.

Graft Survival

Graft survival was assessed at 3 years posttransplant. Overall, there was no difference in graft survival between recipients with lower BMI and hBMI donors (Figure 4A; P = 0.4428). Grafts with MaS had a significantly lower survival at 3 years (Figure 4B, P = 0.0043).

FIGURE 4
FIGURE 4:
Kaplan-Meier analysis for liver allograft survival. Unadjusted 3 year graft survival was assessed between (A) hBMI donors and lower BMI donors; (B) recipients with grafts showing MaS less than 30% versus 30% or greater. Recipient 3-year graft survival was assessed by performance of biopsy and findings of MaS for low BMI donors (C, unadjusted; E, adjusted) and hBMI donors (D, unadjusted; F, adjusted). Adjusted 3-year graft survival was compared across regions when (G) hBMI donors were used or (H) grafts with MaS of 30% or greater were used.

Regional differences in recipient MELD score and other characteristics are presented in Table S3, SDC,http://links.lww.com/TP/B610. A multivariable Cox proportional hazards model for allograft survival was created from factors found to significantly influence organ utilization as well as transplant and recipient-specific characteristics (Table S4, SDC,http://links.lww.com/TP/B610). In lower BMI donors, unadjusted graft survival was significantly different by biopsy results; however, this difference disappeared when adjusting for covariates (Figures 4C, E). In hBMI donors, there was no difference between grafts without biopsy and those with biopsy showing less than 30% MaS; however, both were higher than grafts with biopsy showing MaS ≥30% (Figure 4D, P = 0.0017). This difference persisted when adjusting for covariates (Figure 4F).

Unadjusted graft survival was assessed across all OPTN regions at 90 days and 3 years from donors with and without biopsy (Figure S1, SDC, http://links.lww.com/TP/B610). Adjusted graft survival from hBMI donors varied significantly across regions (Figure 4G, P < 0.0001). Regions 6, 8, and 4 exhibited the best graft survival with 3-year estimated survival of 85.2% (95% CI, 80.9%-86.6%), 81.0% (78.4%-83.3%), and 80.7% (78.2%-82.9%), respectively. The worst graft survival from hBMI donors was in regions 7, 9, and 2. Differences in outcomes among allografts with MaS of 30% or greater were evaluated and was not significantly different across regions (Figure 4H).

DISCUSSION

Allograft availability is the primary limiting factor to overcoming the growing waitlist for liver transplantation. Efforts to expand the donor pool with use of marginal donors, that is, DCD donors, older aged donors, has been undertaken with relatively good success.20,21 As the proportion of obese individuals continues to rise in the United States, we examined variations in the performance of liver biopsy in donor evaluation, and utilization of hBMI donors and those with MaS on liver biopsy.

Through this study, we found that donor BMI significantly influences the likelihood of organ utilization and becomes nearly half as likely as BMI increases above 30 kg/m2. Previous studies have not found BMI alone to be a risk factor for graft loss in multivariable analyses.6,22 Therefore, this discrepancy in utilization may be explained, in part, by our data showing that hBMI donors constitute a relatively sicker population with higher rates of diabetes and hypertension and overall, likely reflects a level of uncertainty when it comes to hBMI donors.

Overall, performance of liver biopsy was associated with decreased likelihood of organ utilization in lower and hBMI groups. However, the results of the biopsy did alter utilization in hBMI donors, where allografts had an OR 1.343 (95% CI, 1.230-1.467) when less than 30% MaS was identified. In both lower BMI donors and hBMI donors, findings of MaS of 30% or greater reduced organ utilization by 85% to 90%. Macrosteatosis of 30% or greater is a known risk factor for EAD and PNF,9 and utilization rates were low across all regions when MaS is 30% or greater was identified on biopsy. Furthermore, analysis of organs recovered but not ultimately transplanted showed biopsy results were the reported reason for refusal in nearly 50% of discarded organs (data not shown).

In our regional analysis, we found that compared with hBMI donors without a biopsy, findings of MaS less than 30% increased organ utilization in 4 of 11 OPTN regions and had no effect in the other 7. In lower BMI donors, only region 1 showed the same effect. Notably, region 6 had one of the highest biopsy rates in hBMI donors and one of the highest utilization rates of hBMI donors when biopsy showed MaS less than 30%. Conversely, region 6 also had one of the lowest utilization rates when biopsy showed MaS of 30% or greater in both lower BMI and hBMI donors. This selectivity translated into excellent outcomes as they exhibited the highest estimated graft survival rate at 3 years. This likely reflects the significant role biopsy played in donor selection, as region 6 had the lowest utilization rate for hBMI donors overall.

Region 8 had the second highest biopsy rate overall and third highest biopsy rate in hBMI donors. Similar to region 6, region 8 had the second lowest organ utilization when biopsy showed MaS of 30% or greater on biopsy in hBMI donors. Ultimately, this resulted in the third highest rate of hBMI donor utilization and the second best 3-year graft survival from hBMI donors. Conversely, region 9 was seventh of the 11 regions in biopsy rate in hBMI donors and had the second worst graft survival from hBMI donors. Furthermore, in region 9, hBMI donors trended toward a decreased likelihood of utilization, even with MaS less than 30% on biopsy. Region 10 also had poor graft survival from hBMI donors (eighth of 11) and had the lowest biopsy rates overall and in hBMI donors.

Although pretransplant biopsy is time-consuming and subject to variability in interpretation between pathologists, it is our recommendation that pretransplant biopsy be used in hBMI donors, and especially in those with additional metabolic comorbidities, including diabetes and hypertension. Ruling out high levels of MaS in potential allografts may increase confidence in utilization of hBMI donors for transplantation. Further study into the clinical practice of specific centers within regions showing success with hBMI donors and subsequent widespread application should be carried out. Ultimately, this may increase the acceptable donor pool, reduce the assumed risk in transplanting organs from hBMI donors, and improve outcomes for recipients of hBMI donor allografts.

This study is not without its own limitations. First, this a retrospective study of prospectively collected data from across the entire nation. Large databases are subject to error and patients are lost to follow-up, which may diminish the ability to estimate long-term outcomes. Furthermore, although we have tried to account for many factors to reduce known regional differences in our multivariable analysis (ie, MELD score at transplant, recipient and donor ages), and used adjusted Kaplan-Meier estimates to account for differences in volume, there are inherent differences across regions (including size and population demographics) that cannot be fully overcome. Therefore, we recognize the need for further studies at the organ procurement organization and center level to fully ascertain differences in practice that lead to higher utilization rates with improved organ selection when considering hBMI donors. Lastly, it is important to recognize that there is a selection bias when evaluating graft survival outcomes in transplantation. Many potential donor livers are rejected before implantation, and we are unable to predict outcomes from those livers due to the variability in recipients to which they are available. Prospective studies with the liberal use of both predonation biopsy and hBMI donors should be undertaken to evaluate their influence on graft survival.

Expansion of the donor pool is paramount to overcoming a growing need for liver transplantation in the United States. Despite efforts to increase organ utilization, careful organ selection remains a necessary part of maintaining and improving outcomes for transplant recipients. Here we have identified variations across 11 OPTN regions with regards to donor evaluation by pretransplant biopsy, utilization of hBMI donors, and most importantly outcomes of recipients with allografts from those hBMI donors. Our results indicate that regions, which have high rates of biopsy performance may increase their utilization of organs from hBMI donors, improve their organ selection, all of which results in better allograft survival.

REFERENCES

1. Ogden CL, Carroll MD, Flegal KM. Prevalence of obesity in the United States. JAMA. 2014;312:189–190.
2. Clinical Guidelines on the Identification, Evaluation, and Treatment of Overweight and Obesity in Adults–The Evidence Report. National Institutes of Health. Obes Res. 1998;6(Suppl 2):51S–209S.
3. Meier-Kriesche HU, Arndorfer JA, Kaplan B. The impact of body mass index on renal transplant outcomes: a significant independent risk factor for graft failure and patient death. Transplantation. 2002;73:70–74.
4. Yoo HY, Molmenti E, Thuluvath PJ. The effect of donor body mass index on primary graft nonfunction, retransplantation rate, and early graft and patient survival after liver transplantation. Liver Transpl. 2003;9:72–78.
5. Braat AE, Blok JJ, Putter H, et al. The Eurotransplant donor risk index in liver transplantation: ET-DRI. Am J Transplant. 2012;12:2789–2796.
6. 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.
7. Halldorson JB, Bakthavatsalam R, Fix O, et al. D-MELD, a simple predictor of post liver transplant mortality for optimization of donor/recipient matching. Am J Transplant. 2009;9:318–326.
8. Bloom MB, Raza S, Bhakta A, et al. Impact of deceased organ donor demographics and critical care end points on liver transplantation and graft survival rates. J Am Coll Surg. 2015;220:38–47.
9. Verran D, Kusyk T, Painter D, et al. Clinical experience gained from the use of 120 steatotic donor livers for orthotopic liver transplantation. Liver Transpl. 2003;9:500–505.
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;7:409–414.
11. Trevisani F, Sica G, Bernardi M. Autonomic neuropathy in advanced liver disease. Hepatology. 1996;24:1549.
12. Adam R, Reynes M, Johann M, et al. The outcome of steatotic grafts in liver transplantation. Transplant Proc. 1991;23(1 Pt 2):1538–1540.
13. Selzner M, Clavien PA. Fatty liver in liver transplantation and surgery. Semin Liver Dis. 2001;21:105–113.
14. Wong TC, Fung JY, Chok KS, et al. Excellent outcomes of liver transplantation using severely steatotic grafts from brain-dead donors. Liver Transpl. 2016;22:226–236.
15. Westerkamp AC, de Boer MT, van den Berg AP, et al. Similar outcome after transplantation of moderate macrovesicular steatotic and nonsteatotic livers when the cold ischemia time is kept very short. Transpl Int. 2015;28:319–329.
16. Choi WT, Jen KY, Wang D, et al. Donor liver small droplet macrovesicular steatosis is associated with increased risk for recipient allograft rejection. Am J Surg Pathol. 2017;41:365–373.
17. Spitzer AL, Lao OB, Dick AA, et al. The biopsied donor liver: incorporating macrosteatosis into high-risk donor assessment. Liver Transpl. 2010;16:874–884.
18. Kwong AJ, Goel A, Mannalithara A, et al. Improved posttransplant mortality after share 35 for liver transplantation. Hepatology. 2018;67:273–281.
19. Sheehy E, O'Connor KJ, Luskin RS, et al. Investigating geographic variation in mortality in the context of organ donation. Am J Transplant. 2012;12:1598–1602.
20. Gastaca M. Extended criteria donors in liver transplantation: adapting donor quality and recipient. Transplant Proc. 2009;41:975–979.
21. Vodkin I, Kuo A. Extended criteria donors in liver transplantation. Clin Liver Dis. 2017;21:289–301.
22. Rana A, Hardy MA, Halazun KJ, et al. Survival outcomes following liver transplantation (SOFT) score: a novel method to predict patient survival following liver transplantation. Am J Transplant. 2008;8:2537–2546.

Supplemental Digital Content

Copyright © 2018 Wolters Kluwer Health, Inc. All rights reserved.