Abstract
Introduction
End-of-life care for liver transplant recipients is often characterized by high utilization of invasive procedures, prolonged hospital stays, and elevated health care costs. Despite evidence demonstrating that palliative care can reduce aggressive interventions, improve patient-centered outcomes, and lower costs, its integration into transplant care remains inconsistent.
Methods
A retrospective analysis was conducted using the National Inpatient Sample database (2016-2021). Hospitalizations ending in death for liver transplant recipients were compared to non-recipients regarding invasive procedures, health care costs, and the impact of palliative care consultations. Assessed procedures included: mechanical ventilation, tracheostomy, enteral and parenteral nutrition support, red blood cell transfusion, renal replacement therapy, central line placement, and cardiopulmonary resuscitation. Multivariable regression models adjusted for demographic and clinical covariates were utilized.
Results
Among 4,582,658 terminal hospitalizations, liver transplant recipients (n = 5995) were younger (mean age: 66.0 vs 70.9 years, P < 0.001), had higher comorbidity burdens, and were more likely to have undergone one or more of the assessed procedures (74.7% vs 58.4%, P < .001) compared to non-recipients. Hospitalization costs were increased in transplant recipients ($62,630 vs $46,930, P < .001). Palliative care consultations were associated with reduced procedure utilization (69.9% vs 83.7%, P < .001), shorter hospital stays, and lower costs ($46,930 vs $62,630, P < .001).
Discussion
Liver transplant recipients face unique end-of-life care challenges, including greater reliance on high-intensity interventions and associated costs. Palliative care is associated with less invasive procedures and lower costs, highlighting the need for its integration into transplant care pathways.
Keywords
Introduction
End-of-life care represents a cornerstone of compassionate health care, particularly for patients with complex conditions such as liver transplant recipients. These individuals often face unique challenges, including a high burden of chronic comorbidities, graft-related complications, and the intensive medical interventions that characterize terminal hospitalizations.1-3 Despite the clear need for tailored care strategies, the end-of-life phase for this population is frequently marked by disproportionate utilization of invasive procedures and elevated health care costs, which may not align with patient-centered goals.4,5
Palliative care (PC) provides a valuable framework for addressing the multidimensional needs of these patients by focusing on symptom management, emotional support, and goal-concordant care.6,7 Studies have demonstrated that PC can improve quality of life, reduce aggressive interventions, and lower health care costs, yet its integration within liver transplant programs remains limited and inconsistent.8,9 Misconceptions that PC is solely for terminal patients, rather than a supportive approach applicable at any stage of serious illness, further contribute to its underutilization. 10
The absence of standardized guidelines for integrating PC into transplant programs creates additional challenges. Health care providers often cite time constraints, prognostic uncertainty, and institutional priorities that emphasize survival over symptom management as key barriers. 11 Additionally, cultural stigma surrounding end-of-life discussions hinders effective communication between patients, families, and medical teams, delaying essential conversations about advance care planning and symptom control. 12 Addressing these systemic obstacles requires a multifaceted approach, including provider education, structured referral pathways, and greater interdisciplinary collaboration.13,14
To better characterize end-of-life outcomes in liver transplant recipients and the role of palliative care, we performed a retrospective analysis of terminal hospitalizations in patients with a liver transplantation. We theorized that liver transplant recipients would undergo more invasive procedures at the end-of-life than patients with no history of liver transplantation. Additionally, we theorized that PC consultations would be associated with reduced invasive procedures and cost in this population.
Methods
Study Population
We identified all adult (≥18 years) hospitalizations that ended in in-hospital death between 2016 and 2021 using the National Inpatient Sample (NIS). Liver transplant recipients were defined by the presence of the ICD-10 diagnosis code Z94.4 (Supplement Table S1), which indicates prior liver transplant. To ensure we captured only patients with established transplant history, we excluded cases where a liver transplant procedure was performed during the same hospitalization. The comparison group consisted of all other adult terminal hospitalizations with no indication of prior liver transplant. This allowed us to compare end-of-life care in patients with and without a history of liver transplantation. However, due to the structure of the NIS, we were unable to determine the time elapsed between liver transplantation and terminal hospitalization.
Study Design and Variables
This retrospective, cross-sectional study utilized the NIS, a publicly available, deidentified all-payer inpatient database designed to estimate utilization and outcomes for the U.S. inpatient population. The NIS contains data from approximately 20% of all hospitalizations in the United States, collected through a stratified systematic sampling method to allow for national estimates when standard weighting procedures are applied. Weighted analyses were performed to produce nationally representative results. Institutional Review Board approval was waived due to the deidentified nature of the database.
Characteristics of Terminal Hospitalizations in Patients With and Without a History of Liver Transplantation.
*P < .05, **P < .001.
Incidences of Inpatient Procedures During Terminal Hospitalizations in Patients With and Without a History of Liver Transplantation.
aAdjusted for the presence or absence of a palliative care encounter, age, race, income quartile, primary payer, Charlson Comorbidity Index, hospital region, and hospital bed size.
*P < .05, **P < .001.
Objectives
The primary objective was to determine whether liver transplant recipients experience higher rates of invasive procedures and incur higher end-of-life costs compared to patients without a history of liver transplantation. The secondary objective was to assess whether PC consults reduce invasive procedures and health care costs in liver transplant recipients at the end of life.
Statistical Analysis
Continuous variables are reported as means and compared using Student’s t-tests, while categorical variables are reported as proportions and compared using chi-square tests. Multivariable logistic regression was employed to assess the association between liver transplantation status and invasive procedures, with results expressed as adjusted odds ratios (aORs) and absolute risk differences (ARDs). Cost comparisons were conducted using a multivariable log-gamma model, with rate ratios (aRRs) representing the arithmetic mean ratios.
To improve comparability between groups, we adjusted for demographic and clinical predictors of end-of-life care, including age, race, income quartile, primary payer, Charlson Comorbidity Index, PC consults, hospital region, and hospital bed size. The Charlson Comorbidity Index, a validated measure of comorbidity burden, was included to account specifically for illness severity and baseline prognostic differences. 15 Hospitalizations with missing data were included in the overall analysis but excluded from analyses where the missing variable was an outcome. All analyses were performed using STATA, version 17.0.
Results
Cohort Characteristics
A total of 4,582,658 terminal hospitalizations were identified from 2016 to 2021, including 5995 (0.1%) for patients with a documented history of liver transplantation. The overall cohort had a mean age of 70.9 years, with 53.5% being male and 69.7% identifying as white. Compared to non-recipients, liver transplant recipients were younger (mean age: 66.0 vs 70.9 years, P < .001), more likely to be male (65.9% vs 53.5%, P < .001), and more likely to be white (72.9% vs 69.7%, P = .019). Liver transplant recipients had a significantly higher burden of chronic illness, with a mean Charlson Comorbidity Index of 5.3 compared to 3.6 in non-recipients (P < .001).
The leading primary admission diagnoses for terminal hospitalizations among liver transplant recipients are detailed in Supplement Table S3. Notably, common admission diagnoses in transplant recipients included sepsis, COVID-19, and respiratory failure, suggesting that many patients died of systemic illness or multiorgan failure rather than liver-specific causes. The frequency of PC consults was lower in liver transplant recipients (65.6% vs 69.2%, OR = 0.85, 95% CI: 0.75-0.96, P = .008), while the prevalence of do-not-resuscitate (DNR) orders did not differ significantly between groups (55.9% vs 58.3%, P = .092).
Invasive Procedure Utilization
Liver transplant recipients experienced a higher burden of invasive procedures compared to non-recipients, with 74.7% undergoing at least one procedure compared to 58.4% in the overall cohort (ARD = 163:1,000, aOR = 1.77, 95% CI: 1.54-2.05, P < .001) (Table 2). Among specific procedures, liver transplant recipients had significantly higher rates of mechanical ventilation (ARD = 127:1,000, aOR = 1.54, 95% CI: 1.35-1.75, P < .001), tracheostomy (ARD = 12:1,000, P < .05), enteral nutrition support (ARD = 15:1,000, P < .05), RBC transfusion (ARD = 50:1,000, P < .05), RRT (ARD = 163:1,000, P < .001), and central line placement (ARD = 31:1,000, P < .05). Notably, CPR rates were lower in liver transplant recipients compared to non-recipients (Table 2). Unadjusted odds ratios for procedure utilization are shown in Supplement Table S2.
Healthcare Costs and Length of Stay
The mean hospitalization cost was $32,507 across the cohort but was significantly higher in liver transplant recipients ($52,334 vs $32,480, aRR = 1.39, 95% CI: 1.28-1.51, P < .001). Liver transplant recipients also had longer hospital stays, with a mean of 11.9 days compared to 8.0 days for non-recipients (P < .001). While the mean cost per day was higher for liver transplant recipients, the difference was not statistically significant ($5739 vs $5,143, aRR = 1.07, 95% CI: 0.99-1.15, P = .102).
Impact of Palliative Care Consults
Incidences of Inpatient Procedures During Terminal Hospitalizations in Patients With a History of Liver Transplantation Stratified by the Presence of a Palliative Care Consult.
aAdjusted for the presence or absence of a race, income quartile, age, sex, race, primary payer, Charlson Comorbidity Index, hospital region and hospital bed size.
*P < .05, **P < .001.
Discussion
This study highlights the significant burden of high-intensity medical interventions among liver transplant recipients at the end of life. Compared to non-recipients, these patients underwent more frequent mechanical ventilation, renal replacement therapy, and enteral or parenteral nutrition support, resulting in longer hospital stays and increased health care costs.
Despite evidence demonstrating the benefits of PC in reducing aggressive medical interventions and improving patient-centered outcomes, its utilization remains low in this population.4,5 Findings from this study align with prior research indicating that early PC involvement is associated with lower health care costs, decreased ICU admissions, and improved symptom management, particularly in transplant populations. 8 However, systemic barriers, physician reluctance, and misconceptions regarding PC’s role continue to hinder its integration into liver transplant pathways.16,17
Numerous potential explanations exist for why liver transplant recipients underwent more end-of-life procedures. Prior studies, including in patients with malignancy and hematopoietic stem cell transplant populations, have shown that patients with high comorbidity burdens tend to undergo increased invasive procedures at the end-of-life.18-21 Similarly, liver transplant recipients in our study had an increased burden of comorbidities compared to non-transplant patients. This could result in greater rates of multiorgan failure prior to death and lead to interventions such as mechanical ventilation and RRT.
Although acute diagnoses such as sepsis and respiratory failure were common proximate causes of terminal hospitalization, these likely reflect a broader trajectory of chronic illness and cumulative physiological decline. The frequency of repeat hospitalizations and high comorbidity burden in this population may contribute to impaired quality of life and diminished functional reserve. These findings reinforce the need for earlier palliative care integration—not only for acute symptom management but also for longitudinal support in patients with chronic transplant-related vulnerability.
It is also possible that it is cultural attitudes towards transplant patients that is driving the differences in end-of-life outcomes. For example, a study by Murakami et al 22 showed that over 90% of transplant nephrologists are concerned that offering palliative care to patients with failing allografts will lead to fear and anxiety. Another critical cultural factor influencing high-intensity end-of-life care is the perception of transplant recipients as “investment patients,” which reinforces aggressive medical management even when prognosis is poor. 9 In hematopoietic stem cell transplantation, PC is often introduced late despite evidence from randomized trials showing that early intervention reduces symptom burden and improves psychological outcomes. 21 Our results show this pattern holds true in liver transplant recipients as well, despite evidence from other chronic disease populations showing that PC improves quality of life, facilitates advance care planning, and reduces unnecessary interventions. 23
Despite the well-documented benefits of PC, physicians in transplant programs face multiple challenges in referring patients. Studies indicate that barriers include lack of time, difficulty in prognostication, and institutional pressure to prioritize survival metrics over symptom management.8,24 The uncertainty surrounding prognosis in liver transplantation, particularly for patients with fluctuating clinical trajectories, may further contribute to physician hesitation in initiating PC discussions. 14 Addressing these challenges through automatic referral triggers, embedded PC specialists in transplant teams, and increased interdisciplinary collaboration could facilitate earlier PC integration without disrupting transplant evaluation pathways. 25
To address the underutilization of PC in liver transplantation, structured referral pathways must be implemented within transplant programs. One potential strategy is the automatic referral of high-risk transplant recipients, such as those with multiple hospitalizations, severe graft dysfunction, or declining functional status, to PC specialists as part of routine transplant evaluation. 26 This model has been successfully applied in other serious illnesses and has demonstrated improvements in symptom management, advance care planning discussions, and patient and family satisfaction. 27 Embedding PC specialists within multidisciplinary transplant teams would facilitate real-time consultation and collaboration between hepatologists, transplant surgeons, and PC providers, ensuring that symptom management and quality of life are considered alongside disease-modifying treatments. 28 Additionally, structured physician education programs focused on integrating PC with curative treatments could help overcome the misconception that PC is only for terminal patients, as demonstrated in other serious illnesses such as advanced cancer. 29 A study on automated triggers for PC consultations in hospitalized cirrhosis patients found that early intervention reduced ICU admissions and hospital costs, suggesting that similar models could be beneficial in transplant populations. 25
As clinical care with regard to patient autonomy continues to advance, including in liver transplantation, health care systems must prioritize not only survival but also focus on maximizing quality of life. The evidence presented in this study reinforces the urgent need for structured PC integration in transplant medicine. Future research should explore long-term patient-reported outcomes and cost-effectiveness of early PC integration, ensuring interventions align with patient-centered goals while reducing the burden of aggressive, non-beneficial interventions. 30
Strengths of this study include the large, generalizable inpatient sample size and the assessment of both invasive procedures and total costs. Limitations include the use of an administrative database reliant on ICD-10 codes and the lack of narrative assessments of hospitalizations. We were unable to determine the timing of when patients received their liver transplant, nor the timing of palliative care consultation during hospitalization, which limits our ability to assess whether PC was introduced early or late in the patient’s clinical course. Additionally, palliative care consultations were identified using ICD-10 code Z51.5. While this code is widely used in administrative research as a proxy for specialist PC involvement, it may also capture care provided by non-specialist providers and does not specify the nature, provider specialty, or timing of services delivered, which limits interpretation of the exact role and scope of palliative care involvement. 23
Liver transplant recipients face a higher burden of invasive procedures, longer hospital stays, and greater health care costs at the end of life, yet they receive fewer PC consultations than non-recipients. PC is associated with reduced high-intensity interventions, lower costs, and shorter hospitalizations, underscoring the need for its integration into transplant care pathways.
Supplemental Material
Supplemental Material - Terminal Hospitalizations in Liver Transplant Recipients: Reducing Costs and High-Intensity Care Through Palliative Care
Supplemental Material for Terminal Hospitalizations in Liver Transplant Recipients: Reducing Costs and High-Intensity Care Through Palliative Care by Stephanie Rodriguez, Spencer R. Goble, and Thomas M. Leventhal in American Journal of Hospice and Palliative Medicine®
Footnotes
Author Contributions
SR, SG, and TL contributed to the design of the study. SG performed data analysis. SR, SG, and TL contributed to interpretation of the data. SR drafted the initial version of the manuscript. SR, SG and TL provided critical review and editing of the manuscript. TL provided supervision for the project.
Declaration of Conflicting Interests
The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.
Funding
The author(s) received no financial support for the research, authorship, and/or publication of this article.
Ethical Statements
Data Availability Statement
The data supporting the findings of this study are publicly available.
Supplemental Material
Supplemental material for this article is available online.
References
Supplementary Material
Please find the following supplemental material available below.
For Open Access articles published under a Creative Commons License, all supplemental material carries the same license as the article it is associated with.
For non-Open Access articles published, all supplemental material carries a non-exclusive license, and permission requests for re-use of supplemental material or any part of supplemental material shall be sent directly to the copyright owner as specified in the copyright notice associated with the article.
