Objectives: Previous studies of liver transplant recipients have reported discrepancies with regard to gender and/or sex differences but have focused on pretransplant outcomes. Female candidates are less likely to receive liver transplant and more likely to die or be delisted than their male counterparts. Here, we examined differences in men versus women with alcoholic liver disease before liver transplant and the effects of these differences on posttransplant survival.
Materials and Methods: We analyzed the Scientific Registry of Transplant Recipients records of adult, deceased-donor, whole liver transplant recipients with decompensated alcoholic liver disease from 2002 to 2017 to evaluate the effects of gender on survival in 2 alcoholic liver disease cohorts: (a) including and (b) excluding recipients with additional diagnoses. Pretransplant characteristics were compared using chi-square or t tests. Kaplan-Meier and multivariable proportional hazards regression models were used to evaluate the main and covariable-adjusted effects of gender on survival.
Results: Of 13 781 transplant recipients with decompensated end-stage liver disease, as defined by Model for End-Stage Liver Disease score ≥ 15, 10 924 (79%) were men and 2857 (21%) were women. Women had higher Model for End-Stage Liver Disease scores, higher rates of stage 4 and 5 chronic kidney disease, and were more likely to be on dialysis or ventilator support at time of transplant (all P < .05). Among all recipients, and after adjusting for risk factors, men were approximately 9% more likely than women to experience long-term graft loss (hazard ratio = 1.093; 95% confidence interval, 1.00-1.19; P = .043). However, sex difference was not associated with risk of graft loss among those without additional diagnoses (hazard ratio = 1.09; 95% confidence interval, 0.99-1.21; P = .095).
Conclusions: Although women with alcoholic liver disease who undergo liver transplant have higher severity of illness than their male counterparts, long-term outcomes are comparable.
Key words : Alcoholism, Female, Graft survival, Scientific Registry of Transplant Recipients
The Model for End-Stage Liver Disease (MELD) score was established in 2002 as a validated metric for use in the allocation of liver allografts.1 The goal of the MELD score was to use an objective, transparent method in the allocation of liver allografts that would thus reduce disparities. However, registry database studies examining wait list outcomes (among candidates having various primary causes of end-stage liver disease) before and after implementation of the MELD liver allocation system reported discrepancies based on gender and/or sex. Female candidates were reported to have lower rates of liver transplant (LT) after the introduction of the MELD allocation system and to be more likely to die on the wait list compared with male candidates.2,3 These gender- and sex-related discrepancies have been shown to persist in recent studies.4-7 Lai and colleagues used multivariable competing risk methods and national registry data to examine effects of gender and other covariables on wait list mortality.7 In a sample of 34 711 LT candidates that included 7002 with alcoholic liver disease (ALD), Lai and colleagues found women, who comprised 24% of those with ALD, to have significantly greater risk of wait list mortality than men and that height was an important contributor to this disparity. Specifically, they reported that “Adjustment for MELD, age, African-American race, etiology of cirrhosis, region risk group, ABO group, and significant interactions between these covariates explained only a small part of the association between gender and wait list mortality risk.”7 Women have higher mortality rates on the wait list compared with men and continue to be less likely to receive LT. Furthermore, women are more likely than men to be delisted for being “too sick,” despite having comparable survival after delisting.6
Few studies, however, have focused on sex- and gender-related differences in post-LT outcomes. Mathur and colleagues examined national registry data from 2002 to 2008 and reported that female recipients were more likely to receive liver allografts of lower quality compared with male recipients, with no difference in graft survival after controlling for graft quality.8 They did not, however, evaluate the effect of etiology of liver disease on graft survival. Schoening and colleagues, in a more recent single-center study of 2144 LT recipients that included 509 people with ALD-induced liver failure (150 women, 29% of those with ALD), reported significantly better unadjusted long-term graft survival in women compared with men.9 The multivariable analysis, which did not evaluate whether a sex- or gender-related effect differed by etiology of liver disease, indicated that donor risk index and impaired kidney function were associated with reduced long-term graft survival. Neither of these studies evaluated the effect of sex or gender on posttransplant outcomes in LT recipients transplanted for ALD or whether there was an interaction effect on posttransplant outcomes with regard to etiology and sex- and gender-related differences.
The proportion of patients undergoing LT for ALD has increased over the past 15 years, making ALD the most common indication for LT in the United States.10-12 The number of wait list additions for young and middle-aged female candidates have doubled, and the proportion of female candidates undergoing LT for ALD has also increased over time.10,11 Therefore, in this study, we used national transplant registry data to evaluate whether outcomes differed between male and female LT recipients with ALD, with the null hypothesis that female and male LT recipients experience statistically comparable posttransplant graft and patient survival. Specifically, the aims of this research were to (1) characterize and compare pretransplant clinical and demographic characteristics of female and male ALD LT recipients and (2) examine the effects of sex differences on unadjusted graft and patient survival and the risk factor-adjusted likelihood of long-term graft loss after LT for ALD.
Materials and Methods
Database, inclusion criteria, and data encoding
National transplant registry data, the June 2017 release Standard Analysis Files of the Scientific Registry of Transplant Recipients (SRTR), were used in this study. The SRTR data system includes data on all donors, wait list candidates, and transplant recipients in the United States, submitted by members of the Organ Procurement and Transplantation Network (OPTN). The Health Resources and Services Administration, US Department of Health and Human Services provides oversight to the activities of the OPTN and SRTR contractors. After Institutional Review Board approval, relevant transplant recipient records were linked with candidate, donor, and follow-up data elements. Records were identified for adult (age ≥ 18 years) deceased-donor whole LT recipients between March 2002 and June 2017 based on a diagnosis of ALD. Additional inclusion criteria were laboratory MELD score at transplant greater than or equal to 15 (as an indicator of decompensated cirrhosis) and no prior transplant.
The effect of sex differences was evaluated for demographic and clinical characteristics known to be related to survival after LT or likely to differ between women and men. Waiting time (months) was computed from transplant and listing dates, and the liver donor risk index (LDRI) was computed using previously reported methods.13 Recipients’ body mass index values (BMI; kg/m2) were stratified based on standards for clinical interpretation: underweight (reference: < 18.5), normal/healthy weight (18.5-24.9), overweight (25.0-29.9), obese class 1 (30.0-34.9), obese class 2 (35.0-39.9), and obese class 3 (≥ 40.0). Estimated glomerular filtration rate (eGFR; mL/min/1.73 m2) was computed among those who were not on dialysis from the serum creatinine values used to compute the last laboratory MELD score, race, and sex using the Chronic Kidney Disease Epidemiology Collaboration (CKD-EPI) equations.14 Renal function categories were stratified based on eGFR as follows: normal (≥ 90), CKD stages 2 (60-89), 3A (45-59), 3B (30-44), 4 (15-29), and 5 (< 15) or on dialysis. Listing and transplant laboratory MELD scores were stratified in 5-point increments from < 15, 15 to 19 through 35 to 39, and ≥ 40 in selected analyses.
Statistical methods and secondary cohort (exclusive ALD)
The effect of sex differences on demographic and clinical characteristics was evaluated using chi-square tests with z tests of column proportions or t tests, as appropriate. Analysis of covariance was used to test the associations of patients’ sex and listing laboratory MELD on waiting time. We used the Kaplan-Meier method with log-rank tests to evaluate the unadjusted effect of sex differences on graft and patient survival. We used multivariate Cox proportional hazards regression to test the effect of patients’ sex on the likelihood of graft loss after adjusting for age at transplant (years), waiting time (months), laboratory MELD at transplant, LDRI, and whether the following recipient characteristics applied (reference: not applicable): (1) hepatitis C virus (HCV) infection, (2) approved for hepatocellular carcinoma (HCC) exception points, (3) mechanical ventilator support, (4) diabetes, (5) history of portal vein thrombosis (PVT), and (6) dialysis support. A secondary multivariable model, which necessarily excluded the effects of HCV and HCC exception, was conducted among ALD recipients who did not have any co-diagnoses such as HCV or HCC. Cases having missing data for one or more covariables were excluded from the multivariable models. All analyses were conducted using SPSS statistical software (version 25.0; IBM, Armonk, NY, USA), and statistical significance was defined as P < .05.
A total of 13 781 patients with decompensated end-stage liver disease, as defined by a laboratory MELD score ≥ 15, were transplanted during the study period, of whom 10 924 (79%) were men and 2857 (21%) were women. Demographic and clinical characteristics of the cohort are summarized in Table 1. Female patients who received LT for ALD were younger, with lower BMI, higher LDRI, higher listing and transplant MELD scores, and shorter waiting times (all P < .001). Consistent with the analyses of MELD as continuous variables, higher proportions of women than men comprised the 3 highest transplant MELD categories (30-34, 35-39, ≥ 40), and lower proportions of women had MELD scores in the lower 2 MELD categories (15-19, 20-24) (all P < .05). A similar pattern was observed for listing laboratory MELD categories. After adjusting for the significant effect of listing laboratory MELD on waiting time (P < .001), there was no effect of gender on waiting time (P = .458).
Women were significantly more likely to be on ventilator support at the time of transplant (P < .001), but men had higher rates of diabetes, HCV, HCC, and PVT (all P < .01). Women had lower serum creatinine levels (P < .001) at the time of transplant but had higher rates of stage 4 and 5 CKD and dialysis (all P < .05), as well as lower rates of normal renal function or stage 2 or 3A CKD at the time of transplant.
Unadjusted Kaplan-Meier analyses indicated that there was no statistically significant difference in graft (P = .066) or patient survival (P = .197) between male and female patients. Median graft survival was 157 and 144 months and median patient survival was 158 and 151 months for female and male patients, respectively. The Cox multivariable model summarized in Table 2, which was determined among the 13 167 recipients (96% of the cohort) having complete data on all covariables, indicates that, after adjusting for recipient characteristics (HCV diagnosis, HCC exception points, ventilator support, diabetes, age, BMI class, history of PVT, laboratory MELD score, dialysis, and waiting time and LDRI), male patients had an overall 9% increased risk of long-term graft loss versus female patients (hazard ratio = 1.093; 95% confidence interval, 1.003-1.192; P = .043) (Figure 1A). Individual covariable parameters indicated that HCV infection, ventilator support, diabetes, increased age, history of PVT, increased laboratory MELD at transplant, waiting time and LDRI were independently associated with long-term graft failure (all P ≤ .01).
The secondary Cox model demonstrated that, among those ALD LT recipients without co-diagnoses and after adjusting for covariables, the effect of gender on graft survival was not statistically significant (P = .095) (Table 3, Figure 1B).
The present study is the first to evaluate gender-related disparities between male and female LT recipients with underlying decompensated ALD and adds to the growing body of knowledge regarding disparities in LT between men and women. In patients undergoing LT for ALD, men had higher rates of HCC and HCV than women, which was expected and consistent with previously reported data. However, women had significantly higher laboratory MELD scores at listing and transplant and were more likely to be on ventilator support at the time of transplant than men. When laboratory MELD at transplant was stratified in 5-point increments between 15 and ≥ 40, women had higher rates of being in the MELD categories above 30 at time of listing and transplant. Women with ALD are listed with higher MELD scores, which is likely related to their having shorter waiting times compared with men. Women undergo LT with higher MELD scores and require more extensive medical support at the time of LT. Despite these differences, women in our study have slightly better long-term graft survival than their male counterparts.
Our data demonstrated that, while the female patients had lower serum creatinine levels, the rate of advanced kidney disease based on GFR as estimated using the CKD-EPI approach, was in fact higher among females. Prior data suggest that the severity of renal failure, assessed by creatinine level-dependent measurements, is underestimated in women because women are generally smaller and have less metabolically active lean body mass than males. A recent study by Allen and associates examined OPTN data from 2002 to 2013 and found that LT candidates who were 165 cm or shorter (over 50% of whom were women) were 10% to 15% less likely to undergo LT compared with LT candidates taller than 175 cm.5 Additionally, their center-specific data indicated that women had lower creatinine and MELD scores that were up to 2 points lower compared with men with similar renal function as measured by iothalamate clearance GFR. It is conceivable that, for women to achieve MELD scores sufficiently high to compete with men, their degree of renal or hepatic function must be worse than men with “similar” renal function per creatinine-based assessment. In other words, women with ALD need to be more ill to compete with men for LT. Our findings of higher MELD scores at listing and transplant and increased use of ventilatory support in women, which are reflective of the magnitude of their illness, support this theory.
Being more ill at the time of LT may also reflect delayed presentation or accelerated progression of ALD in female LT candidates. A recent study revealed that women with alcohol use disorder are less likely than men to seek alcohol addiction treatment.15 The study showed that a significant barrier to seeking treatment in women is the perception that alcohol addiction can be self-limiting. Another plausible barrier for women is the fear, in some instances, of losing child custody when the caring parent (the mother in this case) is deemed an alcoholic. The delay in disclosing alcohol use and seeking alcohol addiction treatment can result in delayed presentation in these female patients until the severity of hepatic illness prompts them to overcome these barriers and seek medical attention. Moreover, sex-related differences in alcohol dehydrogenase activity can have accelerated progression of ALD in women compared with men who consume the same amount of alcohol.16 Therefore, even if women and men with ALD present for medical care after the same duration and extent of alcohol use, women’s livers may be “sicker” at that time. Although this metabolic disparity may not be overcome, intensive community-based counseling programs targeting women, as an audience, about the natural course of ALD in women and the importance of seeking early medical attention are certainly warranted.
Limitations of our study are related to the use of a national registry database, including incompletely populated variables and lack of granularity of the data, such as standardized metrics related to renal function. For instance, we used the CKD-EPI equations to calculate eGFR from demographic and clinical data, including serum creatinine values recorded in association with the last reported laboratory MELD score, to determine pretransplant renal function. Additionally, detailed data relating to alcohol use history and social support, which are not captured in this database, may be important factors when evaluating sex- and gender-related discrepancies in graft survival.
Our study suggests that women with ALD who receive LT have a higher severity of pretransplant illness compared with their male counterparts, as illustrated by higher listing and transplant MELD scores, higher rates of ventilator support, and higher rates of advanced chronic kidney disease. Women undergoing LT for ALD may be disadvantaged in 2 ways: women need to be more ill to achieve competitive MELD scores (with lower muscle mass that underestimates the degree of renal insufficiency) and once listed, their shorter stature (in general) makes it less likely they will receive LT. However, women in our study had improved post-LT outcomes compared with their male counterparts, or at least equivalent outcomes, when examining the cohort with only ALD diagnosis. As the proportion of women needing LT for ALD increases, addressing issues related to sex and gender will become increasingly critical. Our findings call for further studies evaluating ways to ensure equality for women and men in access to LT for patients with ALD.
DOI : 10.6002/ect.2020.0063
From the 1Division of Hepatobiliary Surgery and Liver Transplantation, Section
of Surgical Sciences, Vanderbilt University Medical Center, Nashville, Tennessee
USA; the 2Vanderbilt Transplant Center, Nashville, Tennessee USA; the
of Gastroenterology, Hepatology, and Nutrition, Department of Medicine,
Vanderbilt University Medical Center, Nashville, Tennessee USA; and the
4Departments of Surgery and Biostatistics, Vanderbilt University Medical Center,
Nashville, Tennessee, USA
Acknowledgements: The authors have no sources of funding for this study and have no conflicts of interest to declare.
As a condition of the SRTR data use agreement for Standard Analysis Files, we note the following: The data reported here have been supplied by the Minneapolis Medical Research Foundation as the contractor for the SRTR. The interpretation and reporting of these data are the responsibility of the authors and in no way should be seen as an official policy of or interpretation by the SRTR or the US Government.
Corresponding author: Lea Matsuoka, 801 Oxford House, 1313 21st Avenue South, Nashville, TN 37232 USA
Table 1. Characteristics by Sex in Patients Undergoing Liver Transplant for Alcoholic Liver Disease
Table 2. Multivariable Model of the Likelihood of Graft Failure in Patients Undergoing Liver Transplant for Alcoholic Liver Disease
Table 3. Multivariable Model of the Likelihood of Graft Failure in Patients Undergoing Liver Transplant for Alcoholic Liver Disease after Excluding those with Multiple Diagnoses