Received: 13 November 2022 Accepted: 16 January 2023

DOI: 10.1111/ctr.14924

ORIGINAL ARTICLE

Variation in adult living donor liver transplantation in the

United States: Identifying opportunities for increased
utilization

Krista L. Lentine1 Tomohiro Tanaka2 Huiling Xiao1 Therese Bittermann3

Mary Amanda Dew4 Mark A. Schnitzler1 Kim M. Olthoff3 Jayme E. Locke5
Sukru Emre6 Heather F. Hunt7 AnnMarie Liapakis8 David A. Axelrod2

1
Saint Louis University Transplant Center,
SSM-Saint Louis University Hospital, St. Louis, Abstract
Missouri, USA
In the United States, living donor liver transplantation (LDLT) is limited to transplant
2
Organ Transplant Center, University of Iowa,
centers with specific experience. However, the impact of recipient characteristics on
Iowa City, Iowa, USA
3
University of Pennsylvania, Philadelphia,
procedure selection (LDLT vs. deceased donor liver transplant [DDLT]) within these
Pennsylvania, USA centers has not been described. Transplant registry data for centers that performed
≥1 LDLT in 2002–2019 were analyzed using hierarchal regression modeling to quan-
4
University of Pittsburgh, Pittsburgh,
Pennsylvania, USA
5
tify the impact of patient and center factors on the adjusted odds ratio (aOR) of LDLT
University of Alabama, Birmingham, Alabama,
USA (vs DDLT). Among 73,681 adult recipients, only 4% underwent LDLT, varying from <1%
6
Ege University School of Medicine, Izmir, to >60% of total liver transplants. After risk adjustment, the likelihood of receiving
Turkey
an LDLT rose by 73% in recent years (aOR 1.73 for 2014-2019 vs. 2002-2007) but
7
United Network for Organ Sharing,
remained lower for older adults, men, racial and ethnic minorities, and obese patients.
Richmond, Virginia, USA
8
Yale University, New Haven, Connecticut,
LDLT was less commonly used in patients with hepatocellular carcinoma or alcoholic
USA cirrhosis, and more frequently in those with hepatitis C and with lower severity of
illness (Model for End-Stage Liver Disease (MELD) score < 15). Patients with public
Correspondence
Krista L. Lentine, Saint Louis University insurance, lower educational achievement, and residence in the Northwest and South-
Transplant Center, 1201 S. Grand Blvd., St.
east had decreased access. While some differences in access to LDLT reflect clinical
Louis, MO, 63104, USA.
Email: [email protected] factors, further exploration into disparities in LDLT utilization based on center practice
and socioeconomic determinants of health is needed.
AnnMarie Liapakis and David A. Axelrod are
co-senior authors
KEYWORDS
access, disparities, liver transplantation, living donation, practice variation
Funding information
National Institute of Diabetes and Digestive
and Kidney Diseases, Grant/Award Number:
R01DK120518; Mid-America Transplant/Jane
A. Beckman Endowed Chair

1 INTRODUCTION in Asia where access to deceased donor organs remains limited.1 While
the technical complexity of both the donor and recipient LDLT proce-
Living donor liver transplant (LDLT) has evolved from a novel, infre- dures increases the risk of early bile duct and vascular complications
quently performed procedure to standard clinical practice, particularly when compared to deceased donor liver transplant (DDLT), access to

© 2023 John Wiley & Sons A/S. Published by John Wiley & Sons Ltd.

Clinical Transplantation. 2023;37:e14924. wileyonlinelibrary.com/journal/ctr 1 of 10

https://doi.org/10.1111/ctr.14924
 13990012, 2023, 7, Downloaded from https://onlinelibrary.wiley.com/doi/10.1111/ctr.14924 by Saint Louis University Pius XII, Wiley Online Library on [22/07/2023]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License
LENTINE ET AL . 2 of 10

earlier liver transplant (LT) results in decreased waitlist mortality com- 2 METHODS
pared to waiting for DDLT. Furthermore, as experience and technical
expertise has grown, recipients of LDLT have equivalent or supe- 2.1 Data source
rior outcomes to recipients of DDLT, as demonstrated in multicenter
retrospective and prospective series.2 This study used data from the Scientific Registry of Transplant
Although DDLT remains the dominant procedure in the United Recipients (SRTR). The SRTR data system includes data on all
States, the profound shortage of appropriate deceased donor allografts donors, waitlisted candidates, and transplant recipients in the
and resulting excessive rate of death on the waiting list have recently United States, submitted by the members of the OPTN, and has
led to expanded interest in LDLT. Following the first successful LDLT been described elsewhere.11 The Health Resources and Services
in 1989, there was rapid adoption of LDLT, which led to a peak of Administration (HRSA), US Department of Health and Human Ser-
524 cases in 2001. Subsequently, LDLT utilization declined to 200– vices, provides oversight to the activities of the OPTN and SRTR
300 cases per year in 2002 through 2014, following several reports of contractors.
donor and recipient complications, including a widely publicized donor
death.3 However, evidence of improved outcomes in international cen-
ters (mainly from Asia and Europe), growing clinical expertise, broader 2.2 Cohort definition
sharing of deceased donor allografts leading to increased severity of
illness at time of DDLT in many geographic areas have contributed We included all LT recipients at centers that performed ≥1 adult LDLT
to a rise in US LDLT utilization from 2014 through 2019.4 Although from 2002 through 2019. LT recipients at pediatric transplant cen-
the primary benefit of LDLT has historically been earlier access to LT, ters were excluded, irrespective of age. All multiorgan transplants and
LDLT also expands eligiblity for LT, including for patients with nonhep- retransplants were excluded.
atocellular malignancies (e.g., metastatic colorectal cancer5 ) who are
not allocated exception points under current Organ Procurement and
Transplantation Network (OPTN) policy.6 2.3 Exposures and covariates
Donor and recipient selection for LDLT is based on structured
protocols that maximize donor safety and minimize donor and recip- SRTR data were queried to assess patient demographic and clinical
ient complications. From a medical perspective, living liver donation characteristics (severity of illness, primary diagnosis, body mass index
is restricted to healthy individuals with minimal to no comorbidi- [BMI], sex) and social determinants of health (education, race and
ties and who are at low risk of operative complications or future ethnicity, employment status, insurance) (Table 1). Additional analytic
chronic liver disease. Surgically, inadequate donor remnant liver vol- variables included year of transplant and total DDLT volume over the
ume and donor arterial or biliary anatomic variations, which increase same period of analysis. Geographic areas are defined per UNOS as
donor risk, or complexity of recipient procedure may preclude safe follows (SDC Figure 1)12 : Northwest (WA, OR, ID, MT, AK, HI), South-
donation and LDLT. Recipient size and severity of illness must be con- west (CA, NV, UT, AZ, NM), North Midwest (ND, MN, SD, WY, NE, IA,
sidered, because recipients of partial allografts require an adequate CO, KS, MO), South Midwest (OK, TX), Great Lakes (WI, IL, IN, MI, OH),
graft-versus-recipient weight ratio (GRWR) to avoid postoperative Southeast (KY, AR, TN, NC, MS, AL, GA, SC, LA, FL, PR), Mid Atlantic
complications such as small-for-size syndrome.7 Center experience (WV, VA, PA, DC, MD, DE), and Northeast (NJ, NY, CT, RI, MA, VT,
remains a determinant of donor and recipient outcomes.8 Conse- NH, ME).
quently, current transplant regulations limit LDLT to selected centers
with sufficient volume and experience to safely perform these com-
plex operations.4,9 Additionally, lack of awareness of LDLT and financial 2.4 Primary outcome
barriers may affect the number of transplant candidates and potential
living liver donors who are even considered.10 The primary outcome was receipt of LDLT at a center performing both
Despite the promise and growth of LDLT, this procedure continues LDLT and DDLT.
to represent a small proportion of all LT in the United States. In October
2021, the American Society of Transplantation held a consensus con-
ference to identify important barriers to broader expansion of LDLT 2.5 Statistical analyses
in the United States, including data gaps, and to make recommenda-
tions for impactful and feasible mitigation strategies to overcome these 2.5.1 Unadjusted variation in LDLT
barriers. This analysis is a product of the pre-conference workgroup
that aimed to examine national transplant registry data to describe Univariate analyses were performed to identify patient and center
the epidemiology of US LDLT and to identify variation in utilization of characteristics that were correlated with the odds of undergoing an
LDLT among centers and patient groups to inform strategies to reduce LDLT rather than a DDLT. These univariate analyses were used to
disparities in access. define a population of “potentially eligible” LDLT recipients. This cohort
 13990012, 2023, 7, Downloaded from https://onlinelibrary.wiley.com/doi/10.1111/ctr.14924 by Saint Louis University Pius XII, Wiley Online Library on [22/07/2023]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License
3 of 10 LENTINE ET AL .

was defined as LT recipients with a weight, age, and allocation Model over time (Figure 1). Similarly, the proportion of all LTs performed with
for End-Stage Liver Disease (MELD) score at transplant ≤95th per- living donors has recently increased (2002-2007: 4.6%; 2008-2013:
centile of all LDLT recipients. This limited cohort was used to further 3.6%; 2014-2019: 4.8%). The proportion of recipients who underwent
assess variation in access by center. LDLT varied markedly across LT centers from <1% to >60% of all LTs
performed.

2.5.2 Combined center and case-level modeling

3.1 Recipient characteristics associated with
Bi-level hierarchical models were constructed to adjust rates of LDLT LDLT
among eligible candidates that were adjusted for clustering effects in
LDLT utilization. Level 1 comprised recipient factors, and level 2 repre- Patients undergoing LDLT were significantly different than recipi-
sented the transplant center. Empirical Bayes estimates provided the ents of DDLT (Table 1). LDLT recipients were younger; compared
adjusted proportion (with 95% confidence intervals [CIs]) of LDLT at a with recipients aged 18-30 years, patients 31-44 (aOR, .59 .70.83 ),
center, incorporating case-mix adjustment from the hierarchical model. 45-49 (aOR, .44 .52.61 ), and ≥60 years (aOR, .44 .51.61 ) were signifi-
A 95% CI for a given center’s LDLT proportion that does not include cantly less likely to receive an LDLT (Figure 2, SDC Table 1). While
the median national rate of use indicates a practice that is statistically women (aOR, 1.42 1.521.64 ) were more likely to undergo LDLT, patients
significantly different from expected considering clinical factors in the of non-White race or ethnicity (Black: aOR, .35 .42.51 ; Hispanic: aOR,
model. .76 .85.96 ; other race: aOR, .41 .49.59 ) were less likely to undergo LDLT.
Heterogeneity in LDLT use was quantified using median odds ratios LDLT use declined substantially with increases in BMI. For exam-
(MORs). The MOR provides the median of the odds that recipients ple, patients with a BMI > 40 (kg/m2 ) were 72% less likely (aOR,
with identical characteristics will undergo LDLT when two centers are .21 .28.38 ) to receive an LDLT compared to patients with a BMI of 18.5-
drawn at random (performed for all possible pairs of centers). For 24.9 kg/m2 . Compared to patients without functional limitations, LDLT
example, a MOR of 1.5 means that if centers are selected at random, was less commonly performed for patients with reduced functional
a recipient with a given set of reference characteristics is, on average, status (recipient requires some assistance: aOR, .72 .79.86 ; recipient
50% more likely to undergo LDLT at one of the randomly selected cen- requires total assistance: aOR, .52 .58.65 ).
ters than at the other.13 To account for clinical factors that may explain There was little change in the characteristics of the living liver donor
practice variation, the models also included age, BMI, severity of illness population across the study period (Table 2). Over time, donor age,
(i.e., MELD score), and cause of liver disease. Additional factors, such gender distribution, and racial composition were similar. There was a
as assessments of education, insurance, and race and ethnicity, were nonsignificant increase in mean BMI. Donor-to-recipient weight ratio
included when they were significant in the univariate analyses. remained at 1.
The adjusted odds ratio (95% LCL aOR95% UCL [LCL, lower confidence Likelihood of receiving an LDLT was inversely correlated with sever-
limit; UCL, upper confidence limit]) of receiving an LDLT was estimated ity of illness, being higher for patients with MELD score < 15 and
for patient factors, geography, and center volume, after account- decreasing with MELD score > 20. LDLT utilization varied markedly
ing for the effect of center differences using the hierarchical model. among patients with different causes of liver disease, even after
Data were analyzed using Stata 16, College Station, TX. Hierarchical controlling for severity of illness. Compared to patients with end-
logistic regression modeling was in Stata using the “xtmelogit” com- stage liver disease due to hepatitis C infection, patients with hepato-
mand with center as a random intercept. The MOR was calculated cellular carcinoma (HCC), alcohol-associated liver disease, or unknown
using the “xtmrho” (third-party suite) command. etiology of liver disease were less likely to undergo LDLT. Conversely,
patients with cholestatic liver disease were nearly twice as likely to
undergo LDLT (aOR, 1.70 1.912.13 ). Finally, after controlling for BMI,
2.6 Approval those with nonalcoholic steatohepatitis (NASH) were 50% more likely
(aOR, 1.32 1.521.76 ) to have an LDLT.
This analysis was approved by the institutional review board from Saint Social determinants appeared to have a significant association with
Louis University. access to LDLT. Patients who were working at the time of transplant
were 10% more likely to have an LDLT (aOR, 1.02 1.121.23 ). When com-
pared to LT recipients with at least a college education, patients with
3 RESULTS grade school/high school only were 22% less likely (aOR, .72 .78.84 )
and those with unknown education level were 17% less likely (aOR,
From 2002 through 2019, there were 4417 adult LDLTs performed at .74 .83.94 ) to receive an LDLT. Finally, patients with public insurance as
75 distinct US LT centers. LDLTs represented 4.5% of all 97,099 LTs their primary source of coverage were half as likely to receive an LDLT
performed during this period (6.0% of 73,681 LT at centers that per- (Medicaid: aOR, .45 .51.59 ; Medicare: aOR, .59 .65.71 ) than recipients with
formed at least one LDLT). Annual volume of LDLT varied significantly private insurance.
 13990012, 2023, 7, Downloaded from https://onlinelibrary.wiley.com/doi/10.1111/ctr.14924 by Saint Louis University Pius XII, Wiley Online Library on [22/07/2023]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License
LENTINE ET AL . 4 of 10

F I G U R E 1 Living donor liver transplants performed annually in the United States (2002–2019). Left axis: Annual LDLT volume (bars). Right
axis: LDLT as percentage of total LT in the year (orange line).

18 to 30 <18.5
Age

31 to 44
18.5 to 24.9
45 to 59
BMI

≥60 25 to 29.9
30 to 34.9
Sex

Female 35 to 39.9
Male
≥40
White
Race

Hispanic 6 to 9.9
MELD at Transplant

Black 10 to 14.9
Other 15 to 19.9
20 to 24.9
Education

College & Higher

Grade/High Schools 25 to 29.9
Unknown 30 to 35
>35
Private
Insurance

Medicaid Northwest
Medicare Southwest
Self/Other
Geography

North Midwest
HCC South Midwest
HCV Great Lakes
Cause of Disease

HBV Southeast
Metabolic Mid Atlanc
Alcoholic Northeast
Cholestasis
NASH 2002-2007
Year

Other
2008-2013
Unknown
2014-2019

Adjusted Odds Rao 0.5 1.0 1.5 2.0 2.5 0 0.5 1 2 3 4 5 6 7 8 20

Less LDLT More LDLT Less LDLT More LDLT

F I G U R E 2 Recipient characteristics associated with receipt of a living donor versus deceased donor liver transplant in multivariable adjusted
analysis. BMI, body mass index; HBV, hepatitis B, virus; HCC, hepatocellular carcinoma; HCV, hepatitis C virus; LDLT, living donor liver transplant;
MELD, Model for End-stage Liver Disease (score); NASH, nonalcoholic steatohepatitis; Q, quartile.
 13990012, 2023, 7, Downloaded from https://onlinelibrary.wiley.com/doi/10.1111/ctr.14924 by Saint Louis University Pius XII, Wiley Online Library on [22/07/2023]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License
5 of 10 LENTINE ET AL .

TA B L E 1 Characteristics of LDLT and DDLT recipients TA B L E 1 (Continued)

(N = 73681), among centers that performed LDLT in 2002–2019.
LDLT DDLT
LDLT DDLT (n = 4417) (n = 69264)
(n = 4417) (n = 69264) Characteristic n (col %) n (col %)
Characteristic n (col %) n (col %) Recipient employment status
Recipient age, years Working 1091 (24.7) 10256 (14.8)
18–30 360 (8.2) 2855 (4.1) Not working 2635 (59.7) 49709 (71.8)
31–44 707 (16) 8912 (12.9) Unknown 691 (15.6) 9299 (13.4)
45–59 2088 (47.3) 36859 (53.2) Recipient functional status
≥60 1262 (28.6) 20638 (29.8) Activities with no assistance 2474 (56) 25504 (36.8)
Recipient sex Activities with some assistance 1097 (24.8) 16504 (23.8)
Men 2456 (55.6) 46291 (66.8) Activities with total assistance 600 (13.6) 23237 (33.6)
Women 1961 (44.4) 22973 (33.2) Unknown 246 (5.6) 4019 (5.8)
Recipient race and ethnicity Recipient primary source of payment
Black 156 (3.5) 6442 (9.3) Private 3159 (71.5) 40352 (58.3)
Hispanic 487 (11) 9679 (14) Medicaid 350 (7.9) 10094 (14.6)
White 3618 (81.9) 48880 (70.6) Medicare 836 (18.9) 17961 (25.9)
Other 156 (3.5) 4263 (6.2) Self/Other 72 (1.6) 857 (1.2)
Recipient BMI Recipient geographic area
<18.5 97 (2.2) 1360 (2) Northwest 3 (.1) 1585 (2.3)
18.5–24.9 1645 (37.2) 19245 (27.8) Southwest 735 (16.6) 11911 (17.2)
25–29.9 1542 (34.9) 23519 (34) North Midwest 697 (15.8) 6563 (9.5)
30–34.9 749 (17) 14773 (21.3) South Midwest 197 (4.5) 4584 (6.6)
35–39.9 235 (5.3) 6381 (9.2) Great Lakes 639 (14.5) 10399 (15)
≥40 49 (1.1) 2737 (4) Southeast 80 (1.8) 14798 (21.4)
Unknown 100 (2.3) 1249 (1.8) Mid Atlantic 1002 (22.7) 10260 (14.8)
Recipient cause of ESLD Northeast 1064 (24.1) 9164 (13.2)
Hepatocellular carcinoma 662 (15) 19921 (28.8) Cohort
Hepatitis C virus 838 (19) 14715 (21.2) 2002–2007 1411 (31.9) 21530 (31.1)
Hepatitis B virus 49 (1.1) 1490 (2.2) 2008–2013 1142 (25.9) 21863 (31.6)
Metabolic liver disease 130 (2.9) 2018 (2.9) 2014–2019 1864 (42.2) 25871 (37.4)
Alcoholic liver disease 468 (10.6) 10120 (14.6) Center average annual DDLT volume
Cholestatic liver disease 1408 (31.9) 8215 (11.9) Q1 (0–30) 241 (5.5) 4312 (6.2)
NASH 436 (9.9) 4988 (7.2) Q2 (31–50) 336 (7.6) 11297 (16.3)
Other 331 (7.5) 5882 (8.5) Q3 (51–75) 1422 (32.2) 18905 (27.3)
Unknown 95 (2.2) 1915 (2.8) Q4 (>75) 2418 (54.7) 34750 (50.2)
Recipient lab MELD at transplant
Note: Analysis limited to centers that performed LDLT.
6–9.9 795 (18) 9280 (13.4) Abbreviations: DDLT, deceased donor liver transplant; ESLD, end-stage liver
10–14.9 1558 (35.3) 13180 (19) disease; LDLT, living donor liver transplant; MELD, Model for End-stage
Liver Disease (score); NASH, nonalcoholic steatohepatitis; Q, quartile.
15–19.9 1291 (29.2) 13318 (19.2)
20–24.9 571 (12.9) 10427 (15.1)
25–29.9 146 (3.3) 8416 (12.2) 3.2 Transplant center characteristics associated
30–35 43 (1) 7908 (11.4) with LDLT utilization
>35 13 (.3) 6735 (9.7)
The use of LDLT varied significantly by location of the transplant
Recipient education level
center. Compared with LT recipients in the upper Midwest, LDLT
College & Higher 2318 (52.5) 30954 (44.7)
rates were lowest among recipients in the Northwest (aOR, .01 .03.43 )
Grade/High School 1458 (33) 29441 (42.5)
and Southeast (aOR, .04 .13.41 ) and highest in the Northeast (aOR,
Unknown 641 (14.5) 8869 (12.8)
1.33 3.9311.63 ). Patients who underwent transplant at high-volume cen-
(Continues) ters (>75 DDLTs annually) were significantly more likely to receive an
 13990012, 2023, 7, Downloaded from https://onlinelibrary.wiley.com/doi/10.1111/ctr.14924 by Saint Louis University Pius XII, Wiley Online Library on [22/07/2023]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License
LENTINE ET AL . 6 of 10

TA B L E 2 Demographic and clinical characteristics of US living liver donors from 2002 to 2019.

Donor characteristic LDLT overall (n = 4417) 2002–2007 (n = 1411) 2008–2013 (n = 1142) 2014–2019 (n = 1864)
Age, years
5th percentile 21.0 21.0 21.0 22.0
Median 37.0 38.0 36.0 36.0
95th percentile 55.0 54.0 55.0 55.0
Sex (%)
Men 49.0 51.2 48.9 47.4
Women 51.0 48.8 51.1 52.6
Race and Ethnicity (%)
White 82.1 83.0 84.1 80.1
Black 3.6 3.3 4.1 3.5
Hispanic 11.0 11.1 8.3 12.6
Other 3.4 2.7 3.5 3.8
BMI
5th percentile 20.4 20.5 20.3 20.5
Median 26.0 25.9 25.7 26.4
95th percentile 32.6 32.9 32.4 32.6
Weight Rate (Donor/Recipient)
5th percentile .7 .7 .7 .7
Median 1.0 1.0 1.0 1.0
95th percentile 1.6 1.6 1.5 1.6

Abbreviation: BMI, body mass index.

LDLT (aOR, 1.31 3.137.50 ) than those in lower volume centers. However,
there was no correlation between the median allocation MELD or cal-
culated MELD score at transplant center and the use of LDLT (P = not
significant).

3.3 Risk-adjusted proportion of LDLTs among

active centers

After bi-level adjustment to account for differences in recipient popu-

lation and center characteristics, LDLT practice remained significantly
different across the United States (Figure 3, SDC Table 2). Among
75 LDLT centers assessed, 22 centers (29.3%) performed significantly
more LDLTs than would be expected based on recipients’ characteris- F I G U R E 3 Proportion of centers’ liver transplant volume
tics. Conversely, 22 centers (29.3%) were significantly below expected performed using living donors, adjusted for recipient characteristics.
rates. The MOR was 2.86, demonstrating a nearly three-fold variation LDLT, living donor liver transplant.
in the likelihood of receiving an LDLT for a patient with similar clinical
characteristics in centers performing LDLTs.

plants. Among these potentially eligible patients, 3089 received LDLTs

3.4 Variation in use of LDLT in “potentially (10.0%). While 40.0% of centers performed LDLT in <5% of all poten-
eligible” candidates tially eligible patients, 14 centers (18.7%) performed LDLT for at least
20% of potentially eligible patients (Figure 4). If all centers performed
During the study period, 30,818 LT candidates were below the 95th LDLT for 25% of eligible patients, an additional 4615 LDLTs could have
percentile for age, weight, and MELD score of all LDLT liver trans- been performed from 2002 through 2019, allowing deceased donor
 13990012, 2023, 7, Downloaded from https://onlinelibrary.wiley.com/doi/10.1111/ctr.14924 by Saint Louis University Pius XII, Wiley Online Library on [22/07/2023]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License
7 of 10 LENTINE ET AL .

F I G U R E 4 Proportion of liver transplants performed using living donors among “potentially eligible” recipients across centers. The
“potentially eligible” cohort was defined as liver transplant recipients with a weight, age, and allocation Model-for End-stage Liver Disease score at
transplant ≤95th percentile of all living donor liver transplants (LDLTs). DDLT, deceased donor liver transplant. Each bar represents a center.

organs to be redirected to waitlisted patients who do not meet current weight, reduces access to LDLT. Obese patients were 70% less likely
LDLT eligibility criteria. to receive an LDLT, controlling for other factors.16,17 Unfortunately,
given the obesity epidemic, and the increasing prevalence of NASH
as an indication for LT, the proportion of patients who are eligible
4 DISCUSSION for LDLT may decrease over time.18–20 The association of weight and
access to LDLT may also contribute to the improved access that women
LDLT is underutilized in the United States. In this epidemiologic anal- have for LDLT, which differs significantly from DDLT.21 Women tend to
ysis, LDLTs were performed at half of 151 LT programs and LDLTs have lower MELD scores (given impact of serum creatinine on MELD),
constitute only 4.4% of the total US LT volume, despite the ongo- which leads to lower MELD-with-sodium scores and less abdominal
ing shortage of transplantable organs and persistent death on the domain to accept large DDLT grafts, resulting in reduced access to
waiting list. Importantly, at some LT centers, the proportion of LDLTs DDLT and excess waitlist mortality compared with men with similar
performed is substantially higher than expected given recipient char- severity of illness.21,22 Hence, access to LDLT is a vitally important
acteristics, suggesting the potential to increase utilization nationally. option to improve access to LT for women, as demonstrated in a recent
Access to LDLT is not uniform, with significant differences in access retrospective analysis at the University of Toronto.23 Our analysis con-
attributed to both biological (e.g., weight, MELD score, cause of ESLD, firms that women are more likely to receive an LDLT in the United
age, race and ethnicity) and socioeconomic (e.g., insurance, education) States, after adjustment for age, weight, and severity of liver disease.
characteristics. Patients with conditions associated with significant This could reflect the difficulty of finding appropriate living donors for
morbidity but low MELD scores (e.g., cholestatic liver disease), were men due to the greater graft weight required for an acceptable GRWR,
more often recipients of LDLT. Even after restricting the analysis to as well as the decision to perform LDLT for women with cholestatic dis-
LT in patients who were comparable to potentially eligible recipi- eases who are inadequately prioritized for DDLT by the current MELD
ents (<95th percentile for age, BMI, and MELD score for all LDLTs), scoring formulation.
only 10% of recipients underwent LDLT. If all LDLT centers increased While all LT is more complex in older patients, LDLT may pose
the proportion of the potentially eligible patients undergoing trans- additional risk. The right hepatic artery needs to be healthy in LDLT
plant with LDLT to 25%, nearly 5000 additional LTs could have been recipients, and older age, unfortunately, leads to a greater burden of
performed over the study period. atherosclerotic disease precluding LDLT.24 Other age-related factors
By its nature, the use of a partial allograft in LDLT limits the size include concern about physiologic reserve, coronary artery disease,
of potential recipients given the size of the donor allograft. In gen- and frailty.25 This analysis confirms other multicenter data from
eral, GRWRs <0.8% are not widely used in the United States as they the A2ALL group documenting limited utilization of LDLT in older
have been associated with higher rates of graft failure due to small- patients.16 In the A2ALL analysis, lack of acceptable donors was raised
for-size syndrome.7,14,15 The requirement for sufficient hepatic mass as a possible issue, given restrictions on the age of eligible donors.26
likely explains the observation that higher BMI, or greater recipient However, recent data suggest that older patients can successfully
 13990012, 2023, 7, Downloaded from https://onlinelibrary.wiley.com/doi/10.1111/ctr.14924 by Saint Louis University Pius XII, Wiley Online Library on [22/07/2023]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License
LENTINE ET AL . 8 of 10

undergo LDLT, and increased utilization may reduce waitlist death by eral, demonstrating that Black patients are less likely to be referred,
allowing older patients to undergo transplant with lower severity of complete evaluation, and undergo LT than White patients.43
illness and decreased risk of sarcopenia.27 Recent innovations have addressed several of the perceived bar-
Etiology and severity of ESLD have been clearly associated with riers to LDLT, which may allow expanded access. Portal venous flow
reduced utilization of LDLT in the United States. Patients with high modulation, in combination with careful calculation of required allo-
MELD scores appear to have less physiologic reserve, less ability to sur- graft volume to avoid small-for-size syndrome while decreasing the
vive early allograft dysfunction from a partial graft, and a higher rate minimal GRWR, may help to address disparity based on weight.44 ABO-
of perioperative complications.28 Higher MELD scores also increase incompatible LDLT using rituximab and plasma exchange to expand
priority for DDLT, decreasing the need for LDLT. Accordingly, many the living donor pool allows transplantation from medically and sur-
LT centers avoid LDLT in candidates with high MELD scores, despite gically appropriate donors who were previously declined.45 Paired
a recent report demonstrating successful LDLTs in patients with donor exchanges have been performed to overcome ABO or human
decompensated cirrhosis.29–31 Our national analysis further demon- leukocyte antigen incompatibilities, similar to national kidney paired
strated that rates of LDLT were correlated with etiology of ESLD. donation programs.46 Finally, the development of minimally inva-
Cholestatic liver disease patients are highly represented among LDLT, sive approaches for donor hepatectomy may allow earlier return to
presumably due to their difficulty obtaining DDLT with the current work and responsibilities, to reduce financial barriers to LDLT.47
MELD-based allocation.32 Additionally, compared to patients with Among the most significant determinants of LDLT access in this
end-stage liver disease due to hepatitis C virus, fewer patients with analysis were not the patient characteristics noted above, but rather
HCC, hepatitis B virus-related liver disease, and alcoholic liver disease center practices and commitment to expanding LDLT access. Among
underwent LDLT.33,34 HCC, especially beyond conventional criteria centers with established programs (at least one LDLT performed),
such as Milan criteria, has become one of the leading diagnoses among less than 5% of LTs were from living donors. Furthermore, for candi-
recipients of LDLT in several countries outside of the United States.35 dates who met the potentially eligible criteria defined in this study
LDLT also has the potential to expand the oncologic indications (age <66 years, MELD score <22, weight <101 kg), only 10% of eligi-
for LT for patients with nonhepatocellular malignancies (“transplant ble candidates received an LDLT. Because many LDLT-eligible patients
oncology”),36 who usually have quite limited access to DDLT in the received DDLTs, expanded access to LDLT would allow deceased
United States. Our study indicates this may be an area for LDLT to donor allografts to be redirected to patients whose clinical charac-
expand in the United States, as centers establish protocols and care teristics are believed to preclude safe partial transplant (e.g., higher
pathways that will allow for responsible expansion in this domain of MELD patients). The observed variation in risk-adjusted use of LDLT
emerging indications supported by international data and initial North should allow the identification of high-performing centers. Best prac-
American experience. tices from high-performing centers (above risk-adjusted proportion of
While higher age, weight, and severity of illness have biologic ratio- LDLT) should be identified and communicated to the transplant com-
nales for lower utilization of LDLT, this analysis demonstrates the munity, which was, in part, the mission of the AST LDLT Consensus
impact of social determinants of health, including insurance status, Conference effort.
educational achievement, and employment, on medically risk-adjusted This analysis has several important limitations. First, these data
LDLT access. Our data suggested that the lack of employment, lower are limited to recipients of an LT and do not consider potential LT
educational level, and public insurance (compared to private) appeared candidates who were not offered LT. Some of these patients may
to reduce LDLT utilization among patients who receive LT. Prior stud- have “dropped out” while waitlisted due to disease progression. Sim-
ies in the United States37 and other countries38 have demonstrated ilarly, occasional patients may have had a living donor approved but
adverse financial and psychosocial outcomes for some living donors, received a DDLT prior to moving forward to LDLT. Second, the anal-
which could reduce access for recipients of lower socioeconomic ysis preceded recent changes in the liver allocation system, including
status with similar potential donors. The significant observed reduc- the reduction of priority for patients with HCC. This policy change
tion in access to LDLT for publicly-insured patients may reflect the may affect the observed geographic disparities in LDLT, as areas with
impact of socioeconomic factors among potential donors in patients’ high MELD scores are now able to draw more organs, decreasing
social networks who often face similar economic barriers. Despite the need for LDLT; However, MELD score at transplant is not well
access to the National Living Donor Assistance Center (NLDAC) in the correlated to LDLT utilization, diminishing the impact of this change.
United States and other donor assistance programs that provide some While historically patients with HCC were less likely to receive LDLTs,
support, many donors continue to report a significant financial burden it is possible that reduction in DDLT allocation priority may lead to
from donation.39 In addition, potential donors with lower socioeco- changes in this practice pattern. Additional outcomes data are needed,
nomic status have higher rates of medical comorbidities that preclude although outcomes for LDLT in patients with HCC appear to often be
donation (e.g., diabetes, obesity, coronary artery disease, NASH).40 quite good.48 Third, anatomic concerns affecting the candidacy of LDLT
Our present data also demonstrate that Black patients were nearly (e.g., poor transjugular intrahepatic portocaval shunt placement, por-
60% less likely to receive an LDLT, even after adjustment for age, tal vein thrombosis, etc) were not assessed in the analysis of LDLT
weight, severity of illness, and diagnosis, which is consistent with prior versus DDLT utilization in transplant candidates. Fourth, the recent
reports.41,42 These data are consistent with national data on LT in gen- increase in the proportion of candidates with NASH and the marked
 13990012, 2023, 7, Downloaded from https://onlinelibrary.wiley.com/doi/10.1111/ctr.14924 by Saint Louis University Pius XII, Wiley Online Library on [22/07/2023]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License
9 of 10 LENTINE ET AL .

reduction in patients with hepatitis C virus–related liver disease may An abstract describing portions of this work was shared as an oral
affect LDLT utilization rates in the future.18–20 However, in the anal- presentation at the 2022 American Transplant Congress, June 5, 2022,
ysis of potentially eligible candidates, we addressed patients whose Boston, MA.
weight and severity of disease would be expected to allow LDLT using
current approaches, and demonstrated persistent and significant dif- CONFLICTS OF INTEREST
ferences between centers. While we define characteristics of actual Krista L. Lentine and David A. Axelrod are senior scientists for the
donors, it is not possible to determine what portion of the potential SRTR. Dr. Lentine is Scientific Director of the SRTR Living Donor Col-
donors this represents. Efforts such as the SRTR Living Donor Collec- lective, chair of the AST Living Donor Community of Practice, member
tive may bring much needed information on US donor candidates,49 as of the American Society of Nephrology Policy and Advocacy Commit-
well as data on other factors such as anatomic complexity. We also do tee, member of the National Living Donor Assistance Center Advisory
not study donor outcomes, which have been reported in large US and Group, and member of the National Kidney Foundation Transplant
international studies.50 Advisory Committee.
In conclusion, LDLT utilization is increasing nationally in the United
States. We note with concern that patients with low socioeconomic DATA AVAILABILITY STATEMENT
status continue to have significantly reduced access to LDLT. Addition- SRTR data are publicly available.
ally, even after accounting for recipient characteristics, LDLT center
practice was still immensely variable, with some centers performing ORCID
almost no LDLTs and others having performing LDLT in up to 60% Krista L. Lentine https://orcid.org/0000-0002-9423-4849
of LT recipients. Despite improving LDLT outcomes and increasing Tomohiro Tanaka https://orcid.org/0000-0001-6139-8444
waitlist mortality risk, only a small minority of apparently eligible Therese Bittermann https://orcid.org/0000-0002-8576-0193
recipients seem to receive this option to undergo transplant sooner Mary Amanda Dew https://orcid.org/0000-0002-4666-1870
with a high-quality organ. Further outreach is needed to identify Jayme E. Locke https://orcid.org/0000-0002-0220-8716
and support donors, particularly for members of at-risk populations, AnnMarie Liapakis https://orcid.org/0000-0003-4484-6190
which may help to reduce disparity in access to LDLT for racial David A. Axelrod https://orcid.org/0000-0001-5684-0613
and ethnic minority populations and those with low socioeconomic
status. REFERENCES
1. Rela M, Reddy MS. Living donor liver transplant (LDLT) is the way
ACKNOWLEDGMENTS forward in Asia. Hepatol Int. 2017;11:148-151.
2. Abu-Gazala S, Olthoff KM. Status of adult living donor liver trans-
KLL is supported by a grant from National Institute of Diabetes
plantation in the United States: results from the adult-to-adult living
and Digestive and Kidney Diseases (R01DK120518) and by the Mid- donor liver transplantation cohort study. Gastroenterol Clin North Am.
America Transplant/Jane A. Beckman Endowed Chair in Transplanta- 2018;47:297-311.
tion. This work was conducted under the auspices of the Hennepin 3. Miller C, Florman S, Kim-Schluger L, et al. Fulminant and fatal gas
gangrene of the stomach in a healthy live liver donor. Liver Transpl.
Healthcare Research Institute (HHRI), contractor for the Scientific
2004;10:1315-1319.
Registry of Transplant Recipients (SRTR), as a deliverable under con- 4. Cotter TG, Minhem M, Wang J, et al. Living donor liver trans-
tract no. 75R60220C00011 (US Department of Health and Human plantation in the United States: evolution of frequency, outcomes,
Services, Health Resources and Services Administration, Healthcare center volumes, and factors associated with outcomes. Liver Transpl.
2021;27:1019-1031.
Systems Bureau, Division of Transplantation). The US government (and
5. Simoneau E, D’Angelica M, Halazun KJ. Liver transplantation for col-
others acting on its behalf) retains a paidup, nonexclusive, irrevocable,
orectal liver metastasis. Curr Opin Organ Transplant. 2019;24:175-181.
worldwide license for all works produced under the SRTR contract, and 6. Lunsford KE, Javle M, Heyne K, et al. Liver transplantation for
to reproduce them, prepare derivative works, distribute copies to the locally advanced intrahepatic cholangiocarcinoma treated with neoad-
public, and perform publicly and display publicly, by or on behalf of the juvant therapy: a prospective case-series. Lancet Gastroenterol Hepatol.
2018;3:337-348.
Government. The data reported here have been supplied by HHRI as
7. Kiuchi T, Kasahara M, Uryuhara K, et al. Impact of graft size mis-
the contractor for SRTR. The interpretation and reporting of these data matching on graft prognosis in liver transplantation from living donors.
are the responsibility of the author(s) and in no way should be seen as Transplantation. 1999;67:321-327.
an official policy of or interpretation by SRTR or the US Government. 8. Olthoff KM, Abecassis MM, Emond JC, et al. Outcomes of adult liv-
ing donor liver transplantation: comparison of the adult-to-adult living
The authors thank:
donor liver transplantation cohort study and the national experience.
Liver Transpl. 2011;17:789-797.
∙ The LDLT Consensus Conference co-chairs (Michelle Jesse, Vineeta 9. Li C, Mi K, Wen T, et al. A learning curve for living donor liver
Kumar, Elizabeth Verna, and Anjana Pillai, along with co-chair and transplantation. Dig Liver Dis. 2012;44:597-602.
10. Nephew LD, Serper M. Racial, gender, and socioeconomic disparities in
workgroup 1 member AnnMarie Liapakis) for review and feedback
liver transplantation. Liver Transpl. 2021;27:900-912.
∙ The American Society of Transplantation Education Committee for
11. Leppke S, Leighton T, Zaun D, et al. Scientific registry of transplant
review and feedback, and recipients: collecting, analyzing, and reporting data on transplantation
∙ SRTR colleague Anna Gillette for manuscript editing. in the United States. Transplant Rev (Orlando). 2013;27:50-56.
 13990012, 2023, 7, Downloaded from https://onlinelibrary.wiley.com/doi/10.1111/ctr.14924 by Saint Louis University Pius XII, Wiley Online Library on [22/07/2023]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License
LENTINE ET AL . 10 of 10

12. United Network for Organ Sharing (UNOS). COVID-19 and solid organ 34. Singhvi A, Welch AN, Levitsky J, Singhvi D, Gordon EJ. Ethical consider-
transplant. Accessed: 5/30/2021. https://unos.org/covid/ ations of transplantation and living donation for patients with alcoholic
13. Merlo J, Chaix B, Ohlsson H, et al. A brief conceptual tutorial of mul- liver diseases. AMA J Ethics. 2016;18:163-173.
tilevel analysis in social epidemiology: using measures of clustering in 35. Goldaracena N, Gorgen A, Doyle A, et al. Live donor liver transplan-
multilevel logistic regression to investigate contextual phenomena. J tation for patients with hepatocellular carcinoma offers increased
Epidemiol Community Health. 2006;60:290-297. survival vs. deceased donation. J Hepatol. 2019;70:666-6673.
14. Tuttle-Newhall JE, Collins BH, Desai DM, Kuo PC, Heneghan MA. The 36. Sapisochin G, Hibi T, Toso C, et al. Transplant oncology in primary and
current status of living donor liver transplantation. Curr Probl Surg. metastatic liver tumors: principles, evidence, and opportunities. Ann
2005;42:144-183. Surg. 2021;273:483-493.
15. Feng Y, Han Z, Wang X, Chen H, Li Y. Association of graft-to-recipient 37. DiMartini A, Dew MA, Liu Q, et al. Social and financial outcomes of
weight ratio with the prognosis following liver transplantation: a meta- living liver donation: a prospective investigation within the adult-to-
analysis. J Gastrointest Surg. 2020;24:1869-1879. adult living donor liver transplantation cohort study 2 (A2ALL-2). Am J
16. Trotter JF, Wisniewski KA, Terrault NA, et al. Outcomes of donor eval- Transplant. 2017;17:1081-1096.
uation in adult-to-adult living donor liver transplantation. Hepatology. 38. Levy GA, Selzner N, Grant DR. Fostering liver living donor liver
2007;46:1476-1484. transplantation. Curr Opin Organ Transplant. 2016;21:224-230.
17. Yamashiki N, Sugawara Y, Tamura S, et al. Selection of liver-transplant 39. Thuluvath AJ, Peipert J, Berkowitz R, et al. Donor quality of life after
candidates for adult-to-adult living donor liver transplantation as living donor liver transplantation: a review of the literature. Dig Med
the only surgical option for end-stage liver disease. Liver Transpl. Res. 2021;4.
2006;12:1077-1083. 40. Volaco A, Cavalcanti AM, Filho RP, Precoma DB. Socioeconomic sta-
18. Noureddin M, Vipani A, Bresee C, et al. Leading cause of liver trans- tus: the missing link between obesity and diabetes mellitus? Curr
plant in women: updated analysis of indications for liver transplant and Diabetes Rev. 2018;14:321-326.
ethnic and gender variances. Am J Gastroenterol. 2018;113:1649-1659. 41. Nobel YR, Forde KA, Wood L, et al. Racial and ethnic disparities in
19. Wong RJ, Aguilar M, Cheung R, et al. Nonalcoholic steatohepatitis is access to and utilization of living donor liver transplants. Liver Transpl.
the second leading etiology of liver disease among adults awaiting liver 2015;21:904-913.
transplantation in the United States. Gastroenterology. 2015;148:547- 42. Norman SP, Lu Y. Structural barriers to African American living donor
555. kidney transplant. JAMA Surg. 2021;156:1130.
20. Goldberg D, Ditah IC, Saeian K, et al. Changes in the prevalence of hep- 43. Mazumder NR, Simpson D, Atiemo K, et al. Black patients with cir-
atitis C virus infection, nonalcoholic steatohepatitis, and alcoholic liver rhosis have higher mortality and lower transplant rates: results from
disease among patients with cirrhosis or liver failure on the waitlist for a metropolitan cohort study. Hepatology. 2021;74:926-936.
liver transplantation. Gastroenterology. 2017;152:1090-1099 e1. 44. Masuda Y, Yoshizawa K, Ohno Y, Mita A, Shimizu A, Soejima Y. Small-
21. Kim WR, Mannalithara A, Heimbach JK, et al. MELD 3.0: the model for for-size syndrome in liver transplantation: definition, pathophysiology
end-stage liver disease updated for the modern era. Gastroenterology. and management. Hepatobiliary Pancreat Dis Int. 2020;19:334-341.
2021;161(6):1887-1895 e4. 45. Lee SD, Kim SH, Kong SY, Kim YK, Lee SA, Park SJ. ABO-incompatible
22. Wahid NA, Rosenblatt R, Brown RS. A review of the current state of living donor liver transplantation without graft local infusion and
liver transplantation disparities. Liver Transpl. 2021;27:434-443. splenectomy. HPB (Oxford). 2014;16:807-813.
23. Karnam RS, Chen S, Xu W, et al. Sex disparity in liver transplant and 46. Delmonico FL, Morrissey PE, Lipkowitz GS, et al. Donor kidney
access to living donation. JAMA Surg. 2021;156:1010-1017. exchanges. Am J Transplant. 2004;4:1628-1634.
24. Wang JC, Bennett M. Aging and atherosclerosis: mechanisms, func- 47. Broering D, Sturdevant ML, Zidan A. Robotic donor hepatectomy:
tional consequences, and potential therapeutics for cellular senes- a major breakthrough in living donor liver transplantation. Am J
cence. Circ Res. 2012;111:245-259. Transplant. 2022;22:14-23.
25. Durand F, Levitsky J, Cauchy F, Gilgenkrantz H, Soubrane O, Francoz 48. Limkemann AJP, Abreu P, Sapisochin G. How far can we go with hep-
C. Age and liver transplantation. J Hepatol. 2019;70:745-758. atocellular carcinoma in living donor liver transplantation? Curr Opin
26. Ryu S, Yoon SC, Hong KE, Kim JM. Psychosocial issues related to Organ Transplant. 2019;24:644-650.
donor’s decision-making in living donor liver transplantation. Ann 49. Kasiske BL, Lentine KL, Ahn Y, et al. OPTN/SRTR 2020 Annual
Transplant. 2019;24:576-583. Data Report: Living Donor Collective. Am J Transplant. 2022;22(Suppl.
27. Hakeem AR, Fathima R, Padmanaban H, et al. Propensity score- 2):553-586.
matched analysis of posttransplant outcomes in living donor liver 50. Ladner DP, Dew MA, Forney S, et al. Long-term quality of life after liver
transplantation for older adult recipients. Liver Transpl. 2021;27:1273- donation in the adult to adult living donor liver transplantation cohort
1282. study (A2ALL). J Hepatol. 2015;62:346-353.
28. Ikegami T, Imai D, Wang H, et al. D-MELD as a predictor of early
graft mortality in adult-to-adult living-donor liver transplantation.
Transplantation. 2014;97:457-462. SUPPORTING INFORMATION
29. Humar A, Ganesh S, Jorgensen D, et al. Adult living donor ver-
Additional supporting information can be found online in the Support-
sus deceased donor liver transplant (LDLT versus DDLT) at a single
center: time to change our paradigm for liver transplant. Ann Surg.
ing Information section at the end of this article.
2019;270:444-451.
30. Yadav SK, Saraf N, Saigal S, et al. High MELD score does not adversely
affect outcome of living donor liver transplantation: experience in
1000 recipients. Clin Transplant. 2017:31. How to cite this article: Lentine KL, Tanaka T, Xiao H, et al.
31. Selzner M, Kashfi A, Cattral MS, et al. Live donor liver transplantation Variation in adult living donor liver transplantation in the
in high MELD score recipients. Ann Surg. 2010;251:153-157.
United States: Identifying opportunities for increased
32. Khungar V, Goldberg DS. Liver transplantation for cholestatic liver
utilization. Clin Transplant. 2023;37:e14924.
diseases in adults. Clin Liver Dis. 2016;20:191-203.
33. Lucey MR. Liver transplantation for alcoholic liver disease. Nat Rev https://doi.org/10.1111/ctr.14924
Gastroenterol Hepatol. 2014;11:300-307.