Constructing an Intent-to-Treat Score Index to Predict Survival Outcomes in Pediatric Liver Transplant Recipients.

BACKGROUND: Waitlist and posttransplant outcomes have been widely reported for pediatric liver transplantation. Yet, analyzing these metrics individually fails to provide a holistic perspective for patients and their families. Intent-to-treat (ITT) analysis fills this gap by studying the associations between waitlist outcomes, organ availability, and posttransplant outcomes. Our study aimed to construct a predictive index utilizing ITT analysis for pediatric liver transplant recipients (Pedi-ITT). METHODS: We performed a retrospective analysis utilizing de-identified data provided by the United Network for Organ Sharing (UNOS) from March 1, 2002, to December 31, 2021. We analyzed data for 12 926 pediatric recipients (age <18). We conducted a univariate and multivariable logistic regression to find the significant predictive factors affecting ITT survival. A scoring index was constructed to stratify outcome risk on the basis of the significant factors identified by regression analysis. RESULTS: Multivariable analysis found the following factors to be significantly associated with death on the waitlist or after transplant: gender, diagnosis, UNOS region, ascites, diabetes mellitus, age at the time of listing, serum sodium at the time of listing, total bilirubin at the time of listing, serum creatinine at the time of listing, INR at the time of listing, history of ventilator use, and history of re-transplantation. Using receiver operator characteristic analysis, the Pedi-ITT index had a c-statistic of 0.79 (95% confidence interval [CI]: 0.76-0.82). The c-statistics of the Model for End-Stage Liver Disease/Pediatric for End-Stage Liver Disease and pediatric version of the Survival Outcomes Following Liver Transplantation score indices were 0.74 (CI: 0.71-0.76) and 0.69 (CI: 0.66-0.72), respectively. CONCLUSIONS: The Pedi-ITT index provides an additional prognostic model with moderate predictive power to assess outcomes associated with pediatric liver transplantation. Further analysis should focus on increasing the predictive power of the index.

Estimating risk of prolonged mechanical ventilation after liver transplantation in children: PROVE-ALT score.

BACKGROUND: Children at high risk for prolonged mechanical ventilation (PMV) after liver transplantation (LT) need to be identified early to optimize pulmonary support, allocate resources, and improve surgical outcomes. We aimed to develop and validate a metric that can estimate risk for Prolonged Ventilation After LT (PROVE-ALT). METHODS: We identified preoperative risk factors for PMV by univariable analysis in a retrospective cohort of pediatric LT recipients between 2011 and 2017 (n = 205; derivation cohort). We created the PROVE-ALT score by mapping multivariable logistic regression coefficients as integers, with cutoff values using the Youden Index. We validated the score by C-statistic in a retrospectively collected separate cohort of pediatric LT recipients between 2018 and 2021 (n = 133, validation cohort). RESULTS: Among total 338 patients, 21% (n = 72) were infants; 49% (n = 167) had cirrhosis; 8% (n = 27) required continuous renal replacement therapy (CRRT); and 32% (n = 111) required management in hospital (MIH) before LT. Incidence of PMV post-LT was 20% (n = 69) and 3% (n = 12) required tracheostomy. Independent risk factors (OR [95% CI]) for PMV were cirrhosis (3.8 [1-14], p = .04); age <1-year (8.2 [2-30], p = .001); need for preoperative CRRT (6.3 [1.2-32], p = .02); and MIH before LT (12.4 [2.1-71], p = .004). PROVE-ALT score ≥8 [Range = 0-21] accurately predicted PMV in the validation cohort with 73% sensitivity and 80% specificity (AUC: 0.81; 95% CI: 0.71-0.91). CONCLUSION: PROVE-ALT can predict PMV after pediatric LT with a high degree of sensitivity and specificity. Once externally validated in other centers, PROVE-ALT will empower clinicians to plan patient-specific ventilation strategies, provide parental anticipatory guidance, and optimize hospital resources.

Predicting wait time for pediatric kidney transplant: a novel index.

BACKGROUND: Over one thousand pediatric kidney transplant candidates are added to the waitlist annually, yet the prospective time spent waiting is unknown for many. Our study fills this gap by identifying variables that impact waitlist time and by creating an index to predict the likelihood of a pediatric candidate receiving a transplant within 1 year of listing. This index could be used to guide patient management by giving clinicians a potential timeline for each candidate's listing based on a unique combination of risk factors. METHODS: A retrospective analysis of 3757 pediatric kidney transplant candidates from the 2014 to 2020 OPTN/UNOS database was performed. The data was randomly divided into a training set, comprising two-thirds of the data, and a testing set, comprising one-third of the data. From the training set, univariable and multivariable logistic regressions were used to identify significant predictive factors affecting wait times. A predictive index was created using variables significant in the multivariable analysis. The index's ability to predict likelihood of transplantation within 1 year of listing was validated using ROC analysis on the training set. Validation of the index using ROC analysis was repeated on the testing set. RESULTS: A total of 10 variables were found to be significant. The five most significant variables include the following: blood group, B (OR 0.65); dialysis status (OR 3.67); kidney disease etiology, SLE (OR 0.38); and OPTN region, 5 (OR 0.54) and 6 (OR 0.46). ROC analysis of the index on the training set yielded a c-statistic of 0.71. ROC analysis of the index on the testing set yielded a c-statistic of 0.68. CONCLUSIONS: This index is a modest prognostic model to assess time to pediatric kidney transplantation. It is intended as a supplementary tool to guide patient management by providing clinicians with an individualized prospective timeline for each candidate. Early identification of candidates with potential for prolonged waiting times may help encourage more living donation including paired donation chains.

A modified Kidney Donor Risk Index for pediatric kidney transplant recipients.

BACKGROUND: The Kidney Donor Risk Index (KDRI) by Rao et al. was developed to measure the quality of kidney allografts. While Rao's KDRI has been found to be a robust measure of kidney allograft survival for adult kidney transplant recipients, many studies have indicated the need to create a distinct pediatric KDRI. METHODS: Our retrospective study utilized data from the United Network for Organ Sharing database. We examined 9295 deceased donor recipients' data for age < 18 years from 1990 to 2020. We performed a multivariate Cox regression to determine the significant recipient and transplant factors impacting pediatric kidney allograft survival. RESULTS: Multivariate analysis found 5 donor factors to be independently associated with graft failure or recipient death: age, female sex, anoxia as the cause of death, history of cigarette use, and cold ischemia time. Using receiver operator characteristic (ROC) curve analysis and analyzing the predictive value of each KDRI at 1, 5, and 10 years, the proposed pediatric KDRI had a statistically significant and higher predictive value for pediatric recipients at 5 (0.60 versus 0.57) and 10 years (0.61 versus 0.57) than the Rao KDRI. CONCLUSIONS: The proposed pediatric KDRI may provide a more accurate and simpler index to assess the quality of kidney allografts for pediatric recipients. However, due to the mild increase in predictive capabilities over the Rao index, the study serves as a proof of concept to develop a pediatric KDRI. Further studies should focus on increasing the index's predictive capabilities. A higher resolution version of the Graphical abstract is available as Supplementary information.

Marginal allografts in liver transplantation have a limited impact on length of stay.

The study of marginal liver transplant outcomes, including post-transplant length of stay (LOS), is necessary for determining the practicality of their use. 50 155 patients who received transplants from 2012 to 2020 were retrospectively analyzed with data from the Scientific Registry of Transplant Recipients database using Kaplan-Meier survival curves and multivariable Cox regression. Six different definitions were used to classify an allograft as being marginal: 90th percentile Donor Risk Index (DRI) allografts, donation after cardiac death (DCD) donors, national share donors, donors over 70, donors with > 30% macrovesicular steatosis, or 90th percentile Discard Risk Index donors. 24% (n = 12 124) of subjects received marginal allografts. Average LOS was 15.6 days among those who received standard allografts. Among those who received marginal allografts, LOS was found to be highest in those who received 90th percentile DRI allografts at 15.6 days, and lowest in those who received DCD allografts at 12.7 days. Apart from fatty livers (95% CI .86-.98), marginal allografts were not associated with a prolonged LOS. We conclude that accounting for experience and recipient matching, transplant centers may be more aggressive in their use of extended criteria donors with limited fear of increasing LOS and its associated costs.

A new chapter in an evolving pandemic: Successful pediatric liver transplantation with SARS-CoV-2+ donors.

BACKGROUND: Amid a viral pandemic with poorly understood transmissibility and pathogenicity in the pediatric patient, we report the first pediatric liver transplants utilizing allografts from SARS-CoV-2+ donors. METHODS: We describe the outcomes of two pediatric liver transplant recipients who received organs from SARS-CoV-2 nucleic acid test-positive (NAT+) donors. Data were obtained through the respective electronic medical record system and UNet DonorNet platform. RESULTS: The first donor was a 3-year-old boy succumbing to head trauma. One of four nasopharyngeal (NP) swabs and 1 of 3 bronchoalveolar lavage (BAL) NAT tests demonstrated SARS-CoV-2 infection before organ procurement. The second donor was a 16-month-old boy with cardiopulmonary arrest of unknown etiology. Three NAT tests (2 NP swab/1 BAL) prior to procurement failed to detect SARS-CoV-2. The diagnosis was made when the medical examiner repeated 2 NP swab NATs and an archive plasma NAT, all positive for SARS-CoV-2. Both 2-year-old recipients continue to do well 8 months post-transplant, with excellent graft function and no evidence of SARS-CoV-2 transmission. CONCLUSIONS: This is the first report to describe successful pediatric liver transplantation from SARS-CoV-2+ donors. These data reinforce the adult transplant experience and support the judicious use of SARS-CoV-2+ donors for liver transplantation in children. With SARS-CoV-2 becoming endemic, the concern for donor-derived viral transmission must now be weighed against the realized benefit of life-saving transplantation in the pediatric population as we continue to work toward donor pool maximization.

Survey of public attitudes towards imminent death donation in the United States.

Imminent death donation (IDD) is described as living organ donation prior to a planned withdrawal of life-sustaining care in an imminently dying patient. Although IDD was ethically justified by United Network for Organ Sharing, the concept remains controversial due to presumed lack of public support. The aim of this study was to evaluate the public's attitudes towards IDD. A cross-sectional survey was conducted of US adults age >18 years (n = 2644). The survey included a case scenario of a patient with a devastating brain injury. Responses were assessed on a 5-point Likert scale. Results showed that 68% - 74% of participants agreed or strongly agreed with IDD when posed as a general question and in relation to the case scenario. Participants were concerned about "recovery after a devastating brain injury" (34%), and that "doctors would not try as hard to save a patient's life" (33%). Only 9% of participants would be less likely to trust the organ donation process. In conclusion, our study demonstrates strong public support for IDD in the case of a patient with a devastating brain injury. Notably, participants were not largely concerned with losing trust in the organ donation process. These results justify policy change towards imminent death donation.

Factors associated with long-term graft survival in pediatric kidney transplant recipients.

Pediatric kidney transplant recipients generally have good outcomes post-transplantation. However, the younger age and longer life span after transplantation in the pediatric population make understanding the multifactorial nature of long-term graft survival critical. This investigation analyzes factors associated with 10-year survival to identify areas for improvement in patient care. Kaplan-Meier with log-rank test and univariable and multivariable logistic regression methods were used to retrospectively analyze 7785 kidney transplant recipients under the age of 18 years from January 1, 1998, until March 9, 2008, using United Network for Organ Sharing (UNOS) data. Our end-point was death-censored 10-year graft survival after excluding recipients whose grafts failed within one year of transplant. Recipients aged 5-18 years had lower 10-year graft survival, which worsened as age increased: 5-9 years (OR: 0.66; CI: 0.52-0.83), 10-14 years (OR: 0.43; CI: 0.33-0.55), and 15-18 years (OR: 0.34; CI: 0.26-0.44). Recipient African American ethnicity (OR: 0.67; CI: 0.58-0.78) and Hispanic donor ethnicity (OR: 0.82; CI: 0.72-0.94) had worse outcomes than other donor and recipient ethnicities, as did patients on dialysis at the time of transplant (OR: 0.82; CI: 0.73-0.91). Recipient private insurance status (OR: 1.35; CI: 1.22-1.50) was protective for 10-year graft survival. By establishing the role of age, race, and insurance status on long-term graft survival, we hope to guide clinicians in identifying patients at high risk for graft failure. This study highlights the need for increased allocation of resources and medical care to reduce the disparity in outcomes for certain patient populations.

Pediatric length-of-stay index following liver transplantation.

BACKGROUND: PELD scores are used to reduce waitlist mortality, but they do not accurately predict likelihood of prolonged length-of-stay or higher costs associated with it. This study aims to create a pediatric length-of-stay (LOS) index to predict increased risk of prolonged stay following liver transplantation. METHODS: The scoring system generated predicts length-of-stay following pediatric liver transplantation. With univariate and multivariate analyses on data from 5669 pediatric liver transplant recipients, independent recipient/donor risk factors for prolonged stay (>30 days) were identified. Multiple imputations accounted for missing variables. RESULTS: The most significant factors were ICU admission (OR 2.92, CI 2.27-3.75), recipient bilirubin >32 (OR 2.35, CI 1.70-3.25), and hemodialysis 1 week before transplantation (OR 2.27, CI 1.57-3.27). The LOS index assigns weighted scoring points to factors to predict prolonged stay (C-statistic of .72). The index demonstrated discrimination across the population after dividing it into quartiles for prolonged stay. CONCLUSIONS: The pediatric LOS index, utilizing 13 donor/recipient factors, can assess the risk for pediatric liver transplantation prolonged stay. Important predictive factors are hemodialysis, ICU admission, recipient weight and bilirubin, and recipient life support status.

Orthotopic liver transplantation for Sensenbrenner syndrome.

Sensenbrenner syndrome, or cranioectodermal dysplasia, is a rare heterogeneic autosomal recessive disorder, affecting ~1 of 1 000 000 live births. The syndrome usually manifests within the first year of life and can present with progressive liver and renal involvement. For all Sensenbrenner patients, renal and liver diseases are the main contributors of morbidity and mortality. In this report, we present the case of a 7-year-old boy with congenital liver disease progressing to liver failure secondary to Sensenbrenner syndrome. For this patient, evidence of liver dysfunction was evident from 2 months of age and progressed to frank cirrhosis and severe portal hypertension with multiple episodes of life-threatening variceal bleeding by age 6. This report illustrates the capability of orthotopic liver transplantation as a viable therapy for those pediatric patients suffering from severe liver failure secondary to a congenital ciliopathy, such as Sensenbrenner syndrome. In fact, early emphasis should be placed on the renal and liver involvement associated with Sensenbrenner syndrome with particular consideration for early referral for transplantation in cases with severe disease. Although the condition is rare, clinicians should be aware of it and its association with fatal liver disease to facilitate appropriate evaluation and referral.

Healthcare Worker Attitudes to Living Donation Prior to Planned Withdrawal of Care.

Background and Aims: This study assesses the attitudes of healthcare practitioners toward Living Donation Prior to Planned Withdrawal of Care (LD-PPW): the recovery of a living donor organ before withdrawal of life-sustaining measures in a patient who does not meet criteria for brain death, but for whom medical care toward meaningful recovery is deemed futile. Methods: An electronic survey was administered to 1735 members of the American Society of Transplant Surgeons mailing list with 187 responses (10.8%). Results: Data from this study revealed that 70% of responding practitioners agreed with LD-PPW due to principles of beneficence and autonomy. Also, 65% of participants felt confident in their ability to declare the futility of care and 70% felt that LD-PPW should be added as an option when registering to become an organ donor. Conclusion: Currently, nearly half of all donation after circulatory determination of death do not proceed to donation. LD-PPW has been proposed as an alternative procedure targeted at increasing the quality and quantity of transplantable organs while respecting the donor's right to donate, though its implementation has been hindered by concerns over public and provider perception. This study revealed support for LD-PPW among healthcare practitioners as an alternative procedure to increase the quality and quantity of transplantable organs while respecting the donor's right to donate.

Selecting DCD Recipients Using Predictive Indices.

Donation after circulatory death (DCD) allografts might represent one of the largest untapped sources of liver allografts. Our aim was to identify independent recipient risk factors that predict mortality in DCD allograft recipients to preselect optimal candidates for successful transplantation. Furthermore, we compared the application of our newly constructed DCD Recipient Selector Index (RSI) score to previously developed models to determine superiority in predicting recipient survival. Methods: Using the Organ Procurement and Transplantation Network database, we performed univariate and multivariate retrospective analyses on 4228 DCD liver allograft recipients. Results: We identified 8 significant factors and incorporated them into the weighted RSI to predict 3-mo survival following DCD liver transplantation with a C-statistic of 0.6971. The most significant recipient risk factors were recipient serum sodium levels >150 mEq/L at transplant, recipient albumin <2.0 g/dL at transplant, and a history of portal vein thrombosis. Because Model for End-Stage Liver Disease (MELD) score components were included as individual predictors, the DCD RSI predicts survival independently of MELD. Upon comparison with 3 previous recipient risk scores-Balance of Risk, Renal Risk Index, Patient-Survival Outcomes Following Liver Transplantation-the DCD RSI was determined to be superior at selecting optimal candidates pre-DCD transplantation, yielding a C-statistic of 0.6971. Conclusions: After verifying the performance of predictive indices for selection of DCD recipients, the DCD RSI is best used to preselect patients for optimized outcomes after DCD transplantation. This can increase utilization of DCD donors by improving outcomes.

Elevated bile acids are associated with left ventricular structural changes in biliary atresia.

BACKGROUND: In children with biliary atresia (BA), pathologic structural changes within the heart, which define cirrhotic cardiomyopathy, are associated with adverse perioperative outcomes. Despite their clinical relevance, little is known about the pathogenesis and triggers of pathologic remodeling. Bile acid excess causes cardiomyopathy in experimental cirrhosis, but its role in BA is poorly understood. METHODS: Echocardiographic parameters of left ventricular (LV) geometry [LV mass (LVM), LVM indexed to height, left atrial volume indexed to BSA (LAVI), and LV internal diameter (LVID)] were correlated with circulating serum bile acid concentrations in 40 children (52% female) with BA listed for transplantation. A receiver-operating characteristic curve was generated to determine optimal threshold values of bile acids to detect pathologic changes in LV geometry using Youden index. Paraffin-embedded human heart tissue was separately analyzed by immunohistochemistry for the presence of bile acid-sensing Takeda G-protein-coupled membrane receptor type 5. RESULTS: In the cohort, 52% (21/40) of children had abnormal LV geometry; the optimal bile acid concentration to detect this abnormality with 70% sensitivity and 64% specificity was 152 µmol/L (C-statistics=0.68). Children with bile acid concentrations >152 µmol/L had ∼8-fold increased odds of detecting abnormalities in LVM, LVM index, left atrial volume index, and LV internal diameter. Serum bile acids positively correlated with LVM, LVM index, and LV internal diameter. Separately, Takeda G-protein-coupled membrane receptor type 5 protein was detected in myocardial vasculature and cardiomyocytes on immunohistochemistry. CONCLUSION: This association highlights the unique role of bile acids as one of the targetable potential triggers for myocardial structural changes in BA.

Critical elements of pediatric liver cancer surgery.

The appropriate management of pediatric liver malignancies, primarily hepatoblastoma and hepatocellular carcinoma, requires an in depth understanding of contemporary preoperative risk stratification, experience with advanced hepatobiliary surgery, and a good relationship with one's local or regional liver transplant center. While chemotherapy regimens have become more effective, operative indications more well-defined, and overall survival improved, the complexity of liver surgery in small children provides ample opportunity for protocol violation, inadequate resection, and iatrogenic morbidity. These guidelines represent the distillation of contemporary literature and expert opinion as a means to provide a framework for preoperative planning and intraoperative decision-making for the pediatric surgeon.

Trends in Survival for Pediatric Transplantation.

OBJECTIVES: Progress in pediatric transplantation measured in the context of waitlist and posttransplant survival is well documented but falls short of providing a complete perspective for children and their families. An intent-to-treat analysis, in which we measure survival from listing to death regardless of whether a transplant is received, provides a more comprehensive perspective through which progress can be examined. METHODS: Univariable and multivariable Cox regression was used to analyze factors impacting intent-to-treat survival in 12 984 children listed for heart transplant, 17 519 children listed for liver transplant, and 16 699 children listed for kidney transplant. The Kaplan-Meier method and log-rank test were used to assess change in waitlist, posttransplant, and intent-to-treat survival. Wait times and transplant rates were compared by using χ2 tests. RESULTS: Intent-to-treat survival steadily improved from 1987 to 2017 in children listed for heart (hazard ratio [HR] 0.96, 95% confidence interval [CI] 0.96-0.97), liver (HR 0.95, 95% CI 0.94-0.97), and kidney (HR 0.97, 95% CI 0.95-0.99) transplant. Waitlist and posttransplant survival also improved steadily for all 3 organs. For heart transplant, the percentage of patients transplanted within 1 year significantly increased from 1987 to 2017 (60.8% vs 68.7%); however, no significant increase was observed in liver (68.9% vs 72.5%) or kidney (59.2% vs 62.7%) transplant. CONCLUSIONS: Intent-to-treat survival, which is more representative of the patient perspective than individual metrics alone, steadily improved for heart, liver, and kidney transplant over the study period. Further efforts to maximize the donor pool, improve posttransplant outcomes, and optimize patient care while on the waitlist may contribute to future progress.

Recipient Age Predicts 20-Year Survival in Pediatric Liver Transplant.

Introduction: Pediatric liver transplant recipients have demonstrated excellent long-term survival. The purpose of this analysis is to investigate factors associated with 20-year survival to identify areas for improvement in patient care. Methods: Kaplan-Meier with log-rank test as well as univariate and multivariate logistic regression methods were used to retrospectively analyze 4,312 liver transplant recipients under the age of 18 between September 30, 1987 and March 9, 1998. Our primary endpoint was 20-year survival among one-year survival. Results: Logistic regression analysis identified recipient age as a significant risk factor, with recipients below 5 years old having a higher 20-year survival rate (p < 0.001). A preoperative primary diagnosis of a metabolic dysfunction was found to be protective compared to other diagnoses (OR 1.64, CI 1.20-2.25). African-American ethnicity (OR 0.71, CI 0.58-0.87) was also found to be a risk factor for mortality. Technical variant allografts (neither living donor nor cadaveric) were not associated with increased or decreased rates of 20-year survival. Conclusions: Our analysis suggests that long-term survival is inversely correlated with recipient age following pediatric liver transplant. If validated with further studies, this conclusion may have profound implications on the timing of pediatric liver transplantation.

Trends in Outcomes for Marginal Allografts in Liver Transplant.

IMPORTANCE: Investigating outcomes after marginal allograft transplant is essential in determining appropriate and more aggressive use of these allografts. OBJECTIVE: To determine the time trends in the outcomes of marginal liver allografts as defined by 6 different sets of criteria. DESIGN, SETTING, AND PARTICIPANTS: In this case-control, multicenter study, 75 050 patients who received a liver transplant between March 1, 2002, and September 30, 2016, were retrospectively analyzed to last known follow-up (n = 55 395) or death (n = 19 655) using the United Network for Organ Sharing Database. The study period was divided into three 5-year eras: 2002-2006, 2007-2011, and 2012-2016. Kaplan-Meier survival analysis with log-rank test and Cox proportional hazards regression analysis were used to examine the allograft after transplant with marginal allografts, which were defined as 90th percentile Donor Risk Index allografts (calculated over the entire study period), donor after circulatory death allografts, national share allografts, old age (donors >70 years) allografts, fatty liver allografts, and 90th percentile Discard Risk Index allografts. Statistical analysis was performed from August to December 2019. MAIN OUTCOMES AND MEASURES: Allograft failure after transplant as defined by the Organ Procurement and Transplantation Network database. RESULTS: Among the 75 050 patients (44 394 men; mean [SD] age, 54.3 [9.9] years) in the study, Donor Risk Index, patient Model for End-stage Liver Disease scores, and balance of risk scores significantly increased over time. Multivariate Cox proportional hazards regression analysis indicated that 90th percentile Donor Risk Index allograft survival increased across the study period (2002-2006: hazard ratio, 1.41 [95% CI, 1.34-1.49]; 2007-2011: hazard ratio, 1.25 [95% CI, 1.17-1.34]; 2012-2016: hazard ratio, 1.10 [95% CI, 0.98-1.24]). Secondary definitions of marginal allografts (donor after circulatory death, national share, old age donors, fatty liver, and 90th percentile Discard Risk Index) showed similar improvements in allograft survival. CONCLUSIONS AND RELEVANCE: The study's findings encourage the aggressive use of liver allografts and may indicate a need for a redefinition of allograft marginality in liver transplantation.

Trends in Outcomes for Marginal Allografts in Liver Transplant.

IMPORTANCE: Investigating outcomes after marginal allograft transplant is essential in determining appropriate and more aggressive use of these allografts. OBJECTIVE: To determine the time trends in the outcomes of marginal liver allografts as defined by 6 different sets of criteria. DESIGN, SETTING, AND PARTICIPANTS: In this cohort, multicenter study, 75 050 patients who received a liver transplant between March 1, 2002, and September 30, 2016, were retrospectively analyzed to last known follow-up (n = 55 395) or death (n = 19 655) using the United Network for Organ Sharing Database. The study period was divided into three 5-year eras: 2002-2006, 2007-2011, and 2012-2016. Kaplan-Meier survival analysis with log-rank test and Cox proportional hazards regression analysis were used to examine the allograft after transplant with marginal allografts, which were defined as 90th percentile Donor Risk Index allografts (calculated over the entire study period), donor after circulatory death allografts, national share allografts, old age (donors >70 years) allografts, fatty liver allografts, and 90th percentile Discard Risk Index allografts. Statistical analysis was performed from August to December 2019. MAIN OUTCOMES AND MEASURES: Allograft failure after transplant as defined by the Organ Procurement and Transplantation Network database. RESULTS: Among the 75 050 patients (44 394 men; mean [SD] age, 54.3 [9.9] years) in the study, Donor Risk Index, patient Model for End-stage Liver Disease scores, and balance of risk scores significantly increased over time. Multivariate Cox proportional hazards regression analysis indicated that 90th percentile Donor Risk Index allograft survival increased across the study period (2002-2006: hazard ratio, 1.41 [95% CI, 1.34-1.49]; 2007-2011: hazard ratio, 1.25 [95% CI, 1.17-1.34]; 2012-2016: hazard ratio, 1.10 [95% CI, 0.98-1.24]). Secondary definitions of marginal allografts (donor after circulatory death, national share, old age donors, fatty liver, and 90th percentile Discard Risk Index) showed similar improvements in allograft survival. CONCLUSIONS AND RELEVANCE: The study's findings encourage the aggressive use of liver allografts and may indicate a need for a redefinition of allograft marginality in liver transplantation.