This decision analytic model study examines data for adult deceased liver transplant donors and waitlist candidates to determine the association between small stature and access to liver transplant and then creates a simulation of novel allocation policies to examine changes in access.
Key Points
Question
Is a small stature in waitlist candidates associated with decreased access to liver transplant, and what policy interventions can be used to overcome the disadvantage?
Findings
In this decision analytic model with 41 341 liver donors and 84 201 waitlist candidates, the smallest 25% of candidates by body surface area were disadvantaged on the waitlist based on size. A model that allocated the smallest 10% of donors to the smallest 15% of candidates overcame this disparity.
Meaning
Prioritizing allocation of smaller liver donors to smaller candidates may help overcome the disparity associated with candidate size.
Abstract
Importance
Small waitlist candidates are significantly less likely than larger candidates to receive a liver transplant.
Objective
To investigate the magnitude of the size disparity and test potential policy solutions.
Design, Setting, and Participants
A decision analytical model was generated to match liver transplant donors to waitlist candidates based on predefined body surface area (BSA) ratio limits (donor BSA divided by recipient BSA). Participants included adult deceased liver transplant donors and waitlist candidates in the Organ Procurement and Transplantation Network database from June 18, 2013, to March 20, 2020. Data were analyzed from January 2021 to September 2021.
Exposures
Candidates were categorized into 6 groups according to BSA from smallest (group 1) to largest (group 6). Waitlist outcomes were examined. A match run was created for each donor under the current acuity circle liver allocation policy, and the proportion of candidates eligible for a liver based on BSA ratio was calculated. Novel allocation models were then tested.
Main Outcomes and Measures
Time on the waitlist, assigned Model for End-Stage Liver Disease (MELD) score, and proportion of patients undergoing a transplant were compared by BSA group. Modeling under the current allocation policies was used to determine baseline access to transplant by group. Simulation of novel allocation policies was performed to examine change in access.
Results
There were 41 341 donors (24 842 [60.1%] male and 16 499 [39.9%] female) and 84 201 waitlist candidates (53 724 [63.8%] male and 30 477 [36.2%] female) in the study. The median age of the donors was 42 years (IQR, 28-55) and waitlist candidates, 57 years (IQR, 50-63). Females were overrepresented in the 2 smallest BSA groups (7100 [84.0%] and 7922 [61.1%] in groups 1 and 2, respectively). For each increase in group number, waitlist time decreased (234 days [IQR, 48-700] for group 1 vs 179 days [IQR, 26-503] for group 6; P < .001) and the proportion of the group undergoing transplant likewise improved (3890 [46%] in group 1 vs 4932 [57%] in group 6; P < .001). The smallest 2 groups of candidates were disadvantaged under the current acuity circle allocation model, with 37% and 7.4% fewer livers allocated relative to their proportional representation on the waitlist. Allocation of the smallest 10% of donors (by BSA) to the smallest 15% of candidates overcame this disparity, as did performing split liver transplants.
Conclusions and Relevance
In this study, liver waitlist candidates with the smallest BSAs had a disadvantage due to size. Prioritizing allocation of smaller liver donors to smaller candidates may help overcome this disparity.
Introduction
Transplant waitlist candidates with a small stature are significantly less likely than large-stature candidates to receive a liver transplant,1,2,3 and many but not all small candidates are female. This is one of many factors that has been identified as contributing to sex disparities in liver transplant waitlist mortality. Although the proposed revision to the Model for End-Stage Liver Disease (MELD), MELD 3.0,4 includes a covariate for female sex, which will likely mitigate much of the creatinine-based sex disparity for women,5,6 access to appropriately sized lifesaving liver grafts for short-statured individuals of any sex will not be rectified. Policy makers and stakeholders evaluating the MELD 3.0 acknowledged this and have proposed that body size be included as a separate component in the planned composite allocation score for liver transplant,7,8 similar to the recently approved changes to lung transplant allocation.9 Identifying an optimal metric to account for body size in a liver transplant allocation model will be a critical step in this process.
Although several authors studying this issue have used height as the anthropomorphic measurement for small stature, body surface area (BSA) may be a better metric. In healthy people, BSA is an accurate predictor of liver volume10,11 and a better measurement of metabolic mass than weight12 and as such is a more reliable estimate of liver volume. In the liver transplant population, too-small livers can result in small-for-size syndrome and too-big livers in large-for-size syndrome. Prior work by Fukazawa et al13 and Reyes et al,14 performed independently and using different methodology, showed that BSA ratio (donor BSA divided by recipient BSA) outside certain thresholds is predictive of graft loss.
In this study, we aim to first identify the effect of BSA on MELD score and waitlist outcomes. We then seek to describe the disparity in deceased donor liver allocation caused by size using BSA ratio and explore novel allocation rules to help overcome this disparity.
Methods
Data Source
We conducted a retrospective analysis of all adult candidates (age ≥18 years) on the Organ Procurement and Transplantation Network (OPTN) liver waiting list as well as donors of whole livers. The data for this analysis were OPTN data released April 1, 2020. The United Network for Organ Sharing (UNOS) as the contractor for the OPTN supplied these data. The University of Washington Human Subjects Division exempted this study from human participant review because the OPTN database is deidentified and publicly available.
Waitlist Study Population
We examined data for all liver waitlist candidates from June 18, 2013, to March 20, 2020. The start date of the study was the date the OPTN Rule of Share 35 became active.15 Candidates listed for multiorgan transplant were excluded. Candidate demographic factors recorded included age, sex, ABO blood type, assigned MELD score at transplant (as recorded in the database), weight, and height. The range of MELD scores is 6 to 40, with increasing numbers indicating higher risk of death on the waitlist and therefore greater urgency in needing a transplant. Body surface area was calculated using the Mosteller equation: BSA in m2 = √(height in cm × weight in kg)/3600.16 Linear regression was used to evaluate the trend in values across the ordered recipient groups through including group number as a continuous explanatory variable. Other candidate factors collected to construct the liver allocation match runs included the blinded code of the transplant center, UNOS region of the transplant center, and initial and end listing dates for each waitlist candidate.
Donor Population
For the donor data set, we conducted a review of all deceased liver donors whose livers were allocated and transplanted during the same period. Donors allocated to children were excluded. Donor characteristics recorded included age, sex, ABO blood type, distance in miles from donor to transplant center, weight, height, and calculated BSA. Other donor factors used in construction of the liver allocation match runs included blinded code for organ procurement organization (OPO) and the state of the donor location.
Baseline Model Generation
A match run was created for each donor for all active candidates on the liver waitlist at the date of the donor procurement. Three allocation rules, based primarily on the current UNOS acuity circle liver allocation model, were followed for all of our modeling scenarios. First, the donor could only be allocated to candidates according to UNOS policy 9.8.C, Allocation of Livers by Blood Type.17 In summary, donor livers were allocated to potential candidates of same blood type except blood type O donor livers could be allocated to blood type B candidates if the MELD score was 30 or higher. Second, donor livers were only allocated to candidates in a 500-mile circle from the donor. The 500-mile circles were approximated using a combination of 2 constraints: allocations were not allowed to occur when a transplant program’s UNOS region was always greater than 500 miles from the donor’s state or when the average distance organs had to travel from the donor’s OPO to the transplant center (after adjusting for distance variance17,18) was greater than 500 miles. Third, livers from donors could only be allocated to candidates with a donor-to-recipient BSA ratio (donor BSA divided by recipient BSA) between 0.68 and 1.25, according to Reyes et al.14 Body surface area ratio has been shown to be predictive of poor transplant outcomes outside these limits.
Liver waitlist candidates were ordered based on increasing BSA, and BSA groups were created with the following cutoffs, group 1: 1st to 10th percentile, group 2: 11th to 25th percentile, group 3: 26th to 50th percentile, group 4: 51st to 75th percentile, group 5: 76th to 90th percentile, group 6: 91st to 100th percentile. The allocation of specific donors to specific candidates was not determined, but for each donor, the proportion of candidates in each group that could be assigned that particular donor was recorded. Finally, we calculated the average probability of being assigned donors for each candidate group based on all donors in the study. We used Python version 3.8.5 and the Pandas 1.1.3 module for the match runs and calculations.
Novel Allocation Models
After running our baseline model using the 3 allocation rules noted above, and determining liver donor access based on BSA, we then created 3 additional models (models 1-3) that each included an additional novel allocation rule based on preferential allocation of smaller donors to smaller candidates (based on BSA percentiles) to determine if the disparity due to size mismatch could be overcome. We also explored the potential effect of a split liver transplant, by creating a split liver model that allowed splitting of livers deemed splitable by UNOS policy18 from the larger donor groups (donor BSA groups 3-5) into smaller recipients (recipient BSA groups 1 and 2). The split liver policy states that donor livers have the potential to be split if they meet all the following criteria: (1) age younger than 40 years, (2) receiving a single vasopressor or less, (3) values for transaminases no greater than 3 times the normal level, and (4) a body mass index of 28 or less (calculated as weight in kilograms divided by height in meters squared). Split livers were not considered as part of the allocation in models 1 through 3.
Continuous variables were reported as median and IQR and categorical variables as counts and percentages. This data set had an extremely small amount of missing data: 0.23% recipient height, 0.15% recipient weight, and 0.063% donor weight. As a result, only recipients and donors with complete data were used. χ2 analysis was used to compare categorical variables, and continuous variables were compared with a t test, analysis of variance, or Mann-Whitney test, depending on the distribution. All results were considered significant with a P value less than .05. The descriptive and comparison statistics were performed using JMP Pro version 15.2.0 (SAS Institute). Data were analyzed from January 2021 to September 2021.
Results
Study Population and Waitlist Outcomes
During the study period, there were 41 341 donors and 84 201 waitlist candidates. Donors tended to be younger (median age, 42 years [IQR, 28-55], vs 57 years [IQR, 50-63] for waitlist candidates), more likely to be female (16 499 [39.9%] vs 30 477 [36.2%] for waitlist candidates), and have a lower weight and BSA than waitlist candidates (Table 1). When examining the BSA distribution, the median donor BSA was 1.95 m2 (IQR, 1.78-2.14) compared with 1.82 m2 (IQR, 1.67-1.99) for female candidates and 2.08 m2 (IQR, 1.93-2.25) for male candidates (Figure).
Table 1. Donor and Waitlist Candidate Demographic Data.
| Characteristic | Median (IQR) | P value | |
|---|---|---|---|
| Donors (n = 41 341) | Candidates (n = 84 201) | ||
| Age, y | 42 (28-55) | 57 (50-63) | <.001 |
| Sex, No. (%) | <.001 | ||
| Male | 24 842 (60.1) | 53724 (63.8) | |
| Female | 16 499 (39.9) | 30 477 (36.2) | |
| Blood type, No. (%) | |||
| O | 21 374 (51.7) | 39 243 (46.6) | <.001 |
| A | 13 852 (33.5) | 31 632 (37.6) | <.001 |
| B | 4868 (11.8) | 10 214 (12.1) | .07 |
| AB | 1247 (2.0) | 3112 (3.7) | <.001 |
| Weight, kg | 80 (68-94.4) | 83.5 (71.2-98.0) | <.001 |
| Height, cm | 172 (165-178) | 172 (164-178) | .17 |
| BSA, m2 | 1.95 (1.78-2.14) | 2.0 (1.81-2.18) | <.001 |
| Distance from donor to center, miles | 75 (10-207) | ||
| Assigned MELD scorea | 25 (16-32) | ||
Abbreviations: BSA, body surface area; MELD, Model for End-Stage Liver Disease.
MELD score ranges from 6 to 40 with increasing numbers indicating higher risk of death on the waitlist.
Figure. Histogram of Donor, Male Candidate, and Female Candidate Body Surface Area.
The vertical line in each panel indicates the median body surface area for donors of 1.95 m2.
Characteristics of waitlist candidates in BSA group 1 (smallest) through group 6 (largest) are shown in Table 2. The proportion of female candidates was highest in the lowest BSA group (group 1: 7100 [84.0%]) and significantly decreased as BSA increased (from group 2: 7922 [61.1%] were female, through group 6: 679 [7.9%] female; P = .001). Assigned MELD score was associated with BSA group, was lowest in the lowest BSA group (group 1), and significantly increased with increasing BSA group. The median waiting time for transplant decreased with increasing BSA, such that the lowest BSA group (group 1) had a waiting time 55 days longer than the highest BSA group (group 6). The proportion of patients undergoing transplant was lowest in the lowest BSA group (n = 3890 [46%] in group 1) and significantly higher with increasing BSA group (n = 4932 [57%] in group 6).
Table 2. Waitlist Candidate Groups According to BSA Group From Smallest (1) to Largest (6)a.
| No. (%) | P value | ||||||
|---|---|---|---|---|---|---|---|
| Group 1 | Group 2 | Group 3 | Group 4 | Group 5 | Group 6 | ||
| No. of candidates | 8457 (10.0) | 12 560 (14.9) | 20 920 (24.8) | 20 940 (24.9) | 12 672 (15.1) | 8652 (10.3) | |
| Recipient BSA, range, m2 | 0.64-1.659 | 1.66-2.81 | 1.815-1.996 | 1.997-2.18 | 2.183-2.356 | 2.357-3.65 | <.001 |
| Height, median (IQR), cm | 157.6 (149.9-162.6) | 165 (160-170) | 170.2 (165-175.3) | 175.3 (170-180.3) | 177.8 (172.7-182.9) | 182.9 (177.8-188) | <.001 |
| Weight, median (IQR), kg | 56 (52-59) | 66.7 (63.8-69.4) | 77.4 (71.2-81) | 89 (85.8-93.9) | 103.2 (99.2-107.5) | 121.1 (114.8-130) | <.001 |
| Sex | .001 | ||||||
| Male | 1357 (16.0) | 4638 (36.9) | 12 758 (61.0) | 16 153 (77.1) | 10 845 (85.6) | 7973 (92.1) | |
| Female | 7100 (84.0) | 7922 (61.1) | 8162 (39.0) | 4787 (22.9) | 1827 (14.4) | 679 (7.9) | |
| Age, median (IQR), y | 56 (47-63) | 57 (49-63) | 58 (50-63) | 58 (51-63) | 57 (57-63) | 56 (49-61) | .38 |
| Blood type | |||||||
| O | 4002 (47.3) | 5817 (46.3) | 9803 (46.9) | 9722 (46.4) | 5918 (46.7) | 3981 (46.0) | .24 |
| A | 3002 (35.5) | 4612 (36.7) | 7762 (37.1) | 8072 (38.6) | 4837 (38.2) | 3347 (38.7) | .005 |
| B | 1104 (13.1) | 1633 (13.0) | 2608 (12.5) | 2432 (11.6) | 1429 (11.3) | 1008 (11.7) | .02 |
| AB | 349 (4.1) | 498 (4.0) | 747 (3.6) | 714 (3.4) | 488 (3.9) | 316 (3.7) | .31 |
| Assigned MELD score, median (IQR)b | 24 (13-32) | 24 (14-32) | 25 (16-32) | 25 (16-31) | 25 (17-32) | 26 (18-33) | .06 |
| Waiting time, median (IQR), d | 234 (48-700) | 229 (47-657) | 218 (45-602) | 211 (45-579) | 195 (36-533) | 179 (26-503) | <.001 |
| Outcomesc | |||||||
| Transplanted | 3890 (46.0) | 6029 (48.0) | 10 878 (52.0) | 11 308 (54.0) | 7096 (56.0) | 4932 (57.0) | <.001 |
| Died or too sick | 1903 (22.5) | 2801 (22.3) | 4330 (20.7) | 4188 (20.0) | 2484 (19.6) | 1713 (19.8) | .01 |
| Still on waitlist | 905 (10.7) | 1256 (10.0) | 2029 (9.7) | 1906 (9.1) | 1178 (9.3) | 822 (9.5) | .05 |
Abbreviations: BSA, body surface area; MELD, Model for End-Stage Liver Disease.
Linear regression was used to evaluate the trend in values across the ordered recipient groups through including group number as a continuous explanatory variable.
MELD score ranges from 6 to 40 with increasing numbers indicating higher risk of death on the waitlist.
Totals for each column do not add to 100% because other outcomes were also possible, such as becoming temporarily inactive or removal from the waitlist due to improvement in condition.
Baseline Model
The results of the baseline allocation model are shown in Table 3. Based on the allocation rules for the baseline model, which represents current acuity circle allocation policies, recipients with the lowest BSA (groups 1 and 2) and highest BSA (group 6) are disadvantaged, with 37.0%, 7.4%, and 6.8%, respectively, fewer livers allocated to each of these BSA groups relative to their proportional representation on the waitlist. Conversely, BSA groups 3 through 5 receive a disproportionately higher share of livers.
Table 3. Recipient Group and Disparity for the Baseline Allocation Model and Novel Allocation Modelsa.
| Groupb | BSA percentile | Proportion of waiting list, % | Proportion of group (disparity), % | ||||
|---|---|---|---|---|---|---|---|
| Baseline model | Model 1 (25/25)c | Model 2 (10/25)c | Model 3 (10/15)c | Split liver model | |||
| 1 | 1-10 | 10.00 | 6.3 (−37.0) | 12.7 (+21.3) | 8.6 (−16.3) | 10.0 (0) | 10.1 (+0.9) |
| 2 | 11-25 | 14.90 | 13.8 (−7.4) | 23.1 (+35.5) | 17.1 (+12.9) | 15.7 (+5.1) | 15.0 (+0.7) |
| 3 | 26-50 | 24.80 | 26.3 (+6.0) | 20.0 (−24.0) | 23.7 (−4.6) | 23.7 (−4.6) | 24.6 (−0.8) |
| 4 | 51-75 | 24.90 | 27.9 (+12.0) | 22.1 (−12.7) | 25.7 (+3.1) | 25.7 (+3.1) | 26.2 (+5.0) |
| 5 | 76-90 | 15.10 | 16.1 (+6.6) | 13.3 (−13.5) | 15.4 (+1.9) | 15.4 (+1.9) | 15.1 (0) |
| 6 | 91-100 | 10.30 | 9.6 (−6.8) | 8.8 (−17.0) | 9.5 (−8.4) | 9.5 (−8.4) | 8.9 (−15.7) |
Abbreviation: BSA, body surface area.
A negative value represents fewer allocated livers than the proportional representation on the waitlist.
Groups are numbered from smallest BSA (1) to largest (6).
Model 1 allocated the smallest 25% of donors by BSA to the smallest 25% of recipients, model 2 allocated the smallest 10% of donors to the smallest 25% of recipients, and model 3 allocated 10% of the smallest donors to 15% of the smallest recipients.
Novel Allocation Models
To explore potential allocation solutions that overcome this disparity for the lowest BSA groups, we first created a model that allocated the smallest 25% of donors by BSA to the smallest 25% of recipients (model 1, 25/25). Model 1 overallocated livers to recipients with the smallest BSA (groups 1 and 2) at the expense of all other recipient BSA groups (Table 3). In model 2 (10/25), we allocated the smallest 10% of donors by BSA to the smallest 25% of recipients, which improved the equity of the allocation but, compared with model 1, underallocated to recipients in the lowest BSA group (group 1) and overallocated to recipient BSA group 2. In model 3 (10/15), we allocated 10% of the smallest donors to 15% of the smallest recipients by BSA. This model appeared the most equitable (Table 3).
For the split liver model, we used the baseline allocation rules and allowed splitting of livers from the larger donors (donor BSA groups 3-5) into smaller recipients (recipient BSA groups 1 and 2). The split liver model overcame the disparity for the lower recipient BSA groups 1 and 2 but not excessively at the expense of all the other groups (Table 3).
Sex Disparities
Within the UNOS cohort of liver waitlist candidates used in our study, there were 30 477 female candidates, composing 36.2% of the waitlisted population. To explore the effect of our BSA-based allocation models on sex disparities in liver transplant, we compared the proportion of female waitlist candidates who were allocated livers under our modeling scenarios. Specifically, when comparing allocation sex disparities between the baseline model and model 3 (10/15), female candidates were allocated 33.9% of the livers in the baseline model (representing a disparity of −6.8%) compared with 36.8% in model 3 (representing a disparity of +1.6%). Female candidates were also well-served by the split liver model, with 36.9% of livers being allocated to them (disparity of +1.9%).
Discussion
In this study, we have shown that BSA is associated with waitlist outcomes, including lower assigned MELD score, longer time on the waitlist, and lower rate of transplant for low-BSA groups. In addition, we showed that the smallest 25% of waitlist candidates are disadvantaged because of their body size and that this disparity may be overcome by prioritizing the allocation of small liver donors to small candidates, while still maintaining priority to the sickest candidates on the waitlist (MELD score >35). In addition, broader utilization of split liver transplants could help lessen the size-based disparity. While this is an issue of size mismatch between donors and candidates, it is intertwined with that of sex, as the vast majority of small candidates are female.
Our work contributes to that of others showing that small size, measured in this study by BSA, affects liver allocation for both male and female candidates, but it disproportionately disadvantages female candidates, who are more likely to be small.19 In our BSA categorization, it is the smallest 25% who are disadvantaged based on BSA matching, representing 50.4% of females on the waitlist. Female waitlist candidates are disadvantaged by factors other than size. Our study demonstrated lower assigned MELD score for smaller waitlist groups, likely reflecting the weight of the creatinine value in the MELD score, and correspondingly lower rates of transplant, significantly longer wait times (>5 months when comparing smallest and largest BSA groups), and increased death on the waitlist. Further disadvantages for waitlisted female candidates include greater frailty at similar MELD scores and being more likely to be declined an organ offer.1,2,4,19,20
All allocation policy changes have consequences, and prioritizing 1 group of candidates means another group is deprioritized. UNOS began to allocate livers based on MELD score in 2002 in order to use a more objective and reproducible system.7 The introduction of MELD-based allocation decreased the waitlist mortality overall, and the mortality prediction of the score was further improved by adding sodium with MELD-Na in 2016.7,21 While MELD-based allocation was successful in improving waitlist mortality, studies have consistently demonstrated that female patients are disadvantaged at every step of the process, including access to listing22 and worse outcomes once on the liver transplant waiting list,3 with implicit bias likely contributing.23 Significant sex-based disparities have been identified since 2008.24 While a disparity in covariate-adjusted transplant rates for female patients existed prior to the initiation of using MELD for organ allocation, the disparity increased after the implementation of MELD, as waitlist mortality risk increased for females.25 A recent analysis of data from 2013 through 2018 shows that, compared with male candidates, female candidates are 8.6% more likely to die while on the liver transplant waiting list and are 14.4% less likely to receive a deceased donor liver transplant.26 While factors contributing to this disparity are broad and multifactorial, including disparities in care prior to listing for liver transplant, there are some factors that are dictated purely by the prioritization and allocation of deceased donor liver transplants.3 These postlisting factors are the result of allocation policies and could be addressed with targeted changes.3
The recent National Academies of Sciences, Engineering, and Medicine report specifically suggested to “modify the MELD scoring system for liver allocation and prioritization or establish an alternative overall prioritization scheme to include a modifier based on body size or muscle mass to overcome the demonstrated disparities observed for patients of smaller size.”27 Work is currently under way to move all organ allocation in the United States to continuous distribution models, and candidate size is 1 variable being considered for inclusion. A 5% weighting for candidate height is being incorporated into continuous distribution of lungs,28 and Ge and Lai29 have identified a height of less than 166 cm as the cutoff defining short stature at which waitlist mortality increases. While allocating points by height is 1 option, height alone may not fully represent the difficulty in finding an acceptable liver for a patient, as short-stature candidates with large ascites may have more options than a patient of the same size with no ascites. As such, BSA may be a better metric.
Another policy consideration is prioritization of split livers for small recipients. We showed that by splitting all livers acceptable by OPTN criteria, the size disparity could be overcome. However, this is likely an overestimate of the true impact because splitting a liver requires surgical expertise that not all centers have, as well as additional staffing for 2 recipient teams. Furthermore, there may not be enough acceptable pediatric recipients to allow splitting of livers to overcome the disparity. Splitting of livers likely represents only part of the solution.
Limitations
There were several limitations to our study. Our calculations are based on BSA ratio limits established by our prior work. These limits are similar to those independently established by Fukazawa et al13 (Reyes et al14 vs Fukazawa et al13: lower limit, 0.68 vs 0.78; upper limit, 1.25 vs 1.24), showing increased hazard for graft loss for transplants done outside this range, based on nationally available data. In clinical practice, our team uses this calculation to guide organ offers and finds it helpful, particularly when considering larger donors for smaller recipients. We chose to use BSA instead of height or weight alone because of the better ability to predict donor liver volume and the agreement of previous studies on BSA cutoffs. The presence of ascites in a candidate, which would increase a recipient’s BSA and decrease the donor-to-recipient BSA ratio, is not granularly recorded in the OPTN database and hence not included as a separate factor in the BSA models. Therefore, the effect of ascites on BSA ratio calculations is difficult to predict and could not be accounted for in this study. A recent single-center study questioned the validity of these cutoffs, showing no difference in short-term mortality.30 However, given the study numbers presented, that study was far underpowered to detect a difference, and no power analysis was done.
Conclusions
This study detected that a disparity in access to liver transplant was associated with size discrepancy between liver donors and waitlisted candidates, as measured by BSA ratio. Although the issues of size and sex are intertwined, our work along with that of others shows that the size disparity can be overcome with evidence-based allocation policy changes. Allocation policies that prioritize small donors for small recipients should be considered, in addition to advocating for continued expansion of split liver and living donor liver transplants.
Data sharing statement
References
- 1.Allen AM, Heimbach JK, Larson JJ, et al. Reduced access to liver transplantation in women: role of height, MELD exception scores, and renal function underestimation. Transplantation. 2018;102(10):1710-1716. doi: 10.1097/TP.0000000000002196 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 2.Lai JC, Terrault NA, Vittinghoff E, Biggins SW. Height contributes to the gender difference in wait-list mortality under the MELD-based liver allocation system. Am J Transplant. 2010;10(12):2658-2664. doi: 10.1111/j.1600-6143.2010.03326.x [DOI] [PMC free article] [PubMed] [Google Scholar]
- 3.Verna EC, Lai JC. Time for action to address the persistent sex-based disparity in liver transplant access. JAMA Surg. 2020;155(7):545-547. doi: 10.1001/jamasurg.2020.1126 [DOI] [PubMed] [Google Scholar]
- 4.Kim WR, Mannalithara A, Heimbach JK, et al. MELD 3.0: the model for end-stage liver disease updated for the modern era. Gastroenterology. 2021;161(6):1887-1895.e4. doi: 10.1053/j.gastro.2021.08.050 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 5.Myers RP, Shaheen AAM, Aspinall AI, Quinn RR, Burak KW. Gender, renal function, and outcomes on the liver transplant waiting list: assessment of revised MELD including estimated glomerular filtration rate. J Hepatol. 2011;54(3):462-470. doi: 10.1016/j.jhep.2010.07.015 [DOI] [PubMed] [Google Scholar]
- 6.Mindikoglu AL, Regev A, Seliger SL, Magder LS. Gender disparity in liver transplant waiting-list mortality: the importance of kidney function. Liver Transpl. 2010;16(10):1147-1157. doi: 10.1002/lt.22121 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 7.Kriss M, Biggins SW. Evaluation and selection of the liver transplant candidate: updates on a dynamic and evolving process. Curr Opin Organ Transplant. 2021;26(1):52-61. doi: 10.1097/MOT.0000000000000829 [DOI] [PubMed] [Google Scholar]
- 8.Organ Procurement and Transplantation Network . Continuous distribution. Accessed January 9, 2022. https://optn.transplant.hrsa.gov/policies-bylaws/a-closer-look/continuous-distribution/
- 9.United Network for Organ Sharing . New lung allocation policy approved. December 15, 2021. Accessed January 9, 2022. https://unos.org/news/new-lung-allocation-policy-approved/
- 10.Urata K, Kawasaki S, Matsunami H, et al. Calculation of child and adult standard liver volume for liver transplantation. Hepatology. 1995;21(5):1317-1321. doi: 10.1002/hep.1840210515 [DOI] [PubMed] [Google Scholar]
- 11.Vauthey JN, Abdalla EK, Doherty DA, et al. Body surface area and body weight predict total liver volume in Western adults. Liver Transpl. 2002;8(3):233-240. doi: 10.1053/jlts.2002.31654 [DOI] [PubMed] [Google Scholar]
- 12.Dubois E. Basal Metabolism in Health and Disease. Lea & Febiger; 1936. [Google Scholar]
- 13.Fukazawa K, Yamada Y, Nishida S, Hibi T, Arheart KL, Pretto EA Jr. Determination of the safe range of graft size mismatch using body surface area index in deceased liver transplantation. Transpl Int. 2013;26(7):724-733. doi: 10.1111/tri.12111 [DOI] [PubMed] [Google Scholar]
- 14.Reyes J, Perkins J, Kling C, Montenovo M. Size mismatch in deceased donor liver transplantation and its impact on graft survival. Clin Transplant. 2019;33(8):e13662. doi: 10.1111/ctr.13662 [DOI] [PubMed] [Google Scholar]
- 15.Massie AB, Chow EKH, Wickliffe CE, et al. Early changes in liver distribution following implementation of Share 35. Am J Transplant. 2015;15(3):659-667. doi: 10.1111/ajt.13099 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 16.Mosteller RD. Simplified calculation of body-surface area. N Engl J Med. 1987;317(17):1098. doi: 10.1056/NEJM198710223171717 [DOI] [PubMed] [Google Scholar]
- 17.Organ Procurement and Transplantation Network . Liver and Intestine distribution using distance from donor hospital. December 2018. Accessed March 2, 2021. https://optn.transplant.hrsa.gov/governance/public-comment/liver-and-intestine-distribution-using-distance-from-donor-hospital/
- 18.Organ Procurement and Transplantation Network . Policies. January 11, 2022. Accessed January 13, 2022. https://optn.transplant.hrsa.gov/media/eavh5bf3/optn_policies.pdf
- 19.Nephew LD, Goldberg DS, Lewis JD, Abt P, Bryan M, Forde KA. Exception points and body size contribute to gender disparity in liver transplantation. Clin Gastroenterol Hepatol. 2017;15(8):1286-1293.e2. doi: 10.1016/j.cgh.2017.02.033 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 20.Lai JC, Ganger DR, Volk ML, et al. Association of frailty and sex with wait list mortality in liver transplant candidates in the multicenter functional assessment in liver transplantation (FRAILT) study. JAMA Surg. 2021;156(3):256-262. doi: 10.1001/jamasurg.2020.5674 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 21.Nagai S, Chau LC, Schilke RE, et al. Effects of allocating livers for transplantation based on model for end-stage liver disease-sodium scores on patient outcomes. Gastroenterology. 2018;155(5):1451-1462.e3. doi: 10.1053/j.gastro.2018.07.025 [DOI] [PubMed] [Google Scholar]
- 22.Bryce CL, Angus DC, Arnold RM, et al. Sociodemographic differences in early access to liver transplantation services. Am J Transplant. 2009;9(9):2092-2101. doi: 10.1111/j.1600-6143.2009.02737.x [DOI] [PMC free article] [PubMed] [Google Scholar]
- 23.Lai JC, Pomfret EA, Verna EC. Implicit bias and the gender inequity in liver transplantation. Am J Transplant. 2022;22(6):1515-1518. doi: 10.1111/ajt.16986 [DOI] [PubMed] [Google Scholar]
- 24.Moylan CA, Brady CW, Johnson JL, Smith AD, Tuttle-Newhall JE, Muir AJ. Disparities in liver transplantation before and after introduction of the MELD score. JAMA. 2008;300(20):2371-2378. doi: 10.1001/jama.2008.720 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 25.Mathur AK, Schaubel DE, Gong Q, Guidinger MK, Merion RM. Sex-based disparities in liver transplant rates in the United States. Am J Transplant. 2011;11(7):1435-1443. doi: 10.1111/j.1600-6143.2011.03498.x [DOI] [PMC free article] [PubMed] [Google Scholar]
- 26.Locke JE, Shelton BA, Olthoff KM, et al. Quantifying sex-based disparities in liver allocation. JAMA Surg. 2020;155(7):e201129. doi: 10.1001/jamasurg.2020.1129 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 27.National Academies of Sciences, Engineering and Medicine . A fairer and more equitable, cost-effective, and transparent system of donor organ procurement, allocation, and distribution. Accessed May 23, 2022. https://www.nationalacademies.org/our-work/a-fairer-and-more-equitable-cost-effective-and-transparent-system-of-donor-organ-procurement-allocation-and-distribution
- 28.Organ Procurement and Transplantation Network . Notice of OPTN policy changes: establish continuous distribution of lungs. Accessed January 13, 2022. https://optn.transplant.hrsa.gov/media/b13dlep2/policy-notice_lung_continuous-distribution.pdf
- 29.Ge J, Lai JC. Identifying a clinically relevant cutoff for height that is associated with a higher risk of waitlist mortality in liver transplant candidates. Am J Transplant. 2020;20(3):852-854. doi: 10.1111/ajt.15644 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 30.Kubal CA, Mihaylov P, Fridell J, et al. Donor-recipient body size mismatch has no impact on outcomes after deceased donor whole liver transplantation: role of donor liver size measurement. Clin Transplant. 2021;35(6):e14299. doi: 10.1111/ctr.14299 [DOI] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.
Supplementary Materials
Data sharing statement