Modern Outcomes After Liver Retransplantation: A Single-center Experience : Transplantation

Journal Logo

Transplantation

Modern Outcomes After Liver Retransplantation: A Single-center Experience

Connor, Ashton A. MD, PhD1,2; Saharia, Ashish MD1,2,3; Mobley, Constance M. MD, PhD1,2,3; Hobeika, Mark J. MD1,2,3; Victor, David W. III MD1,2,3; Kodali, Sudha MD1,2,3; Brombosz, Elizabeth W. PhD1; Graviss, Edward A. PhD1,4; Nguyen, Duc T. MD, PhD4; Moore, Linda W. PhD1,3; Gaber, A. Osama MD1,2,3; Ghobrial, R. Mark MD, PhD1,2,3

Author Information
Transplantation ():10.1097/TP.0000000000004500, January 18, 2023. | DOI: 10.1097/TP.0000000000004500

Abstract

INTRODUCTION

Liver transplantation (LT) is a lifesaving procedure with excellent outcomes for patients with end-stage liver disease. Approximately 6% to 9%1,2 of LT recipients require liver retransplantation (reLT) at 1 y, which has decreased over time.3 However, the absolute number of reLTs is expected to rise because of the simultaneous increases in the number of LTs being performed, use of the extended criteria for donor liver allografts, prioritization of recipients with greater severity of liver dysfunction, and long-term survival of LT recipients.4 The allocation of scarce liver allografts to reLT may additionally strain organ networks and decrease waitlist survival, thus requiring careful scrutiny.3

Prior series of patients in the United States and elsewhere undergoing reLT have reported 5-y overall survival (OS) rates of 42% to 71%5–10 and graft survival (GS) rates of 36% to 60%,5,10,11 which are lower than recipients of single LT and decrease with each subsequent retransplant.5,6,8 Yet, some patients have had excellent outcomes after reLT, encouraging programs such as the University of Pittsburgh5 and University of California at Los Angeles12 to compile prognostic models that aim to identify the best candidates for reLT. Indications for reLT include primary nonfunction (PNF) of the original graft, hepatic artery thrombosis (HAT), cholangiopathies, and recurrent hepatitis after LT.3 Outcomes after reLT were associated with the time interval from first to second transplant, causes of graft failure, severity of liver dysfunction (Model for End-stage Liver Disease [MELD] score), recipient and donor ages, and other variables.5–7,11–16 Because many of these factors were not modifiable, reLT was performed infrequently.

More recent evidence suggests that reLT outcomes may be improving. Croome et al10 reported significantly improved GS following reLT in a 4-y period (2013–2017) compared with a historical 5-y (2002–2007) cohort. Takagi et al17 found that OS rates for reLT were higher in the 11-y period between 2007 and 2017 than in the 27-y period from 1979 to 2006 in a Dutch series. Similar trends were reported in Australia and New Zealand18 and France.19 An international, multicenter study by Abbassi et al20 reported encouraging outcomes after reLT in a select cohort. Thus, further studies of modern experience with reLT are needed.

The aim of this study was to examine contemporary outcomes of reLT utilizing granular retrospective data from a single center since 2008. These data provide insight into pretransplant variables associated with the risk of needing reLT, as well as changing etiologies and incidence over time, and outcomes.

MATERIALS AND METHODS

This report is a retrospective analysis of records of all adults undergoing LT or reLT at Houston Methodist Hospital (HMH) over a recent 13-y period. Procedures for accessing patient data and patient consent adhered to Houston Methodist IRB Protocol #Pro00000587. The research was conducted in accordance with the Declaration of Helsinki. Patients were grouped into 2 cohorts: the first undergoing a single LT and the second having undergone reLT at our center, regardless of where the patient received their primary LT. The decision to list patients for reLT was made by the multidisciplinary review board. The center’s guidelines for donor and recipient selection for reLT are presented in Table S1, SDC, https://links.lww.com/TP/C670. Selection criteria are similar for both LT and reLT, although with greater emphasis placed on prior treatment compliance in the latter.

Covariates were collected before the index hospitalization, perioperatively at each transplant, and in serial follow-up. Early acute rejection was defined as biopsy-proven acute cellular or antibody-mediated rejection after primary LT. For the LT cohort, OS was defined as the date of index LT to the date of last follow-up (censor) or death. GS was defined as the date of index LT to last follow-up (censor) or graft failure or death. For the reLT cohort, the first OS was defined as the date of index surgery to the date of last follow-up (censor) or death and the second OS (OS2) from second surgery to last follow-up (censor) or death. First GS (GS1) or “retransplant interval” was defined as the time interval from first to second transplant and second GS (GS2) from second transplant to last follow-up (censor) or graft failure or death.

Statistical Analysis

Patient characteristics were reported as frequencies and proportions for categorical variables and as median and interquartile range (IQR) for continuous variables. Differences between (1) single LT, (2) retransplant (reLT) and primary or index LT and (3) reLT and second LT were determined by chi-square or Fisher exact tests for categorical variables and Kruskal-Wallis tests for continuous variables as appropriate. Patient OS and GS at 5 y were depicted using the Kaplan-Meier survival curves. Differences in the outcomes between groups were determined by the log-rank test.

Multivariable logistic regression was used to determine factors associated with having reLT. Cox proportional hazard regression was used to determine factors associated with all-cause mortality and graft failure. Multivariable Cox regression was not performed for the reLT patients because of small sample size. Variables for the multivariable models were selected based on the clinical importance and also by the Stata Lasso technique with the cross-validation selection option.21,22 All the analyses were performed on Stata, version 17.0 (StataCorp LLC, College Station, TX). A P < 0.05 was considered statistically significant.

RESULTS

During the study period (January 1, 2008, to December 31, 2021), 1336 patients received an LT at HMH (Figure 1). The LT cohort included 1268 patients who had single LTs. The reLT cohort consisted of 68 patients who received a total of 128 LTs at our institution. This included 57 patients whose index and subsequent surgeries occurred at HMH and 11 patients whose initial LT occurred elsewhere, but the individual was followed at HMH, and the subsequent reLT was performed at HMH. The absolute number of patients having LT increased over each year (median of 3 reLT patients per year; IQR, 1.5–6.7), but the proportion of LT that were reLT remained relatively stable (3.9%, 2.6%–5.2%).

F1
FIGURE 1.:
Flowchart of the study population. HMH, Houston Methodist Hospital; LT, liver transplantation; reLT, liver retransplantation.

Patient, Donor, and Perioperative Characteristics

Patients undergoing a single LT at our center were comparable to patients in the reLT cohort at their index operation (Table 1; Table S2, SDC,https://links.lww.com/TP/C670), including time spent on the waitlist (P = 0.72; Table S2, SDC, https://links.lww.com/TP/C670). Patients needing reLT were more likely to have received their primary graft from an ABO incompatible donor (P < 0.001) and experience acute rejection (P < 0.001, Table 1). Donors were also similar between LT and reLT cohorts, including donor age (P = 0.44), donation after cardiac death (DCD; P = 0.27), and donor length of terminal hospitalization (P = 0.29; Table S2, SDC, https://links.lww.com/TP/C670). However, index donors in the reLT cohort had a higher Kidney Donor Risk Index (KDPI, P = 0.04, Table 1).

TABLE 1. - Recipient, donor, and perioperative factors stratified by single liver txp or index operation before subsequent retransplantation
Total Single LT ReLT (index LT) Bivariable Multivariable analysis
(N = 1325) (n = 1268) (n = 57) P OR (95% CI) P
Recipient age in years, median (IQR) 58.0 (49.0–64.0) 58.0 (49.0–64.0) 57.0 (45.0–64.0) 0.31 0.98 (0.96-1.01) 0.20
Recipient race and ethnicity, n (%) 0.25
 White 867 (65.4) 837 (66.0) 30 (52.6) (reference)
 Black 132 (10.0) 123 (9.7) 9 (15.8) 2.48 (1.05-5.86) 0.04
 Hispanic 276 (20.8) 260 (20.5) 16 (28.1) 1.53 (0.72-3.23) 0.27
 Asian 43 (3.2) 41 (3.2) 2 (3.5) 0.63 (0.07-5.34) 0.67
 Other 7 (0.5) 7 (0.6) 0 (0.0) - -
Recipient primary diagnosis at txp, n (%) 0.17
 Noncholestatic cirrhosis 770 (58.1) 741 (58.4) 29 (50.9) (reference)
 Cholestatic liver disease/cirrhosis 51 (3.8) 49 (3.9) 2 (3.5) 1.04 (0.21-5.17) 0.96
 Biliary atresia 1 (0.1) 1 (0.1) 0 (0.0) - -
 Acute hepatic necrosis 27 (2.0) 23 (1.8) 4 (7.0) 2.60 (0.60-11.29) 0.20
 Metabolic diseases 35 (2.6) 33 (2.6) 2 (3.5) 3.11 (0.67-14.34) 0.15
 Malignant neoplasms 388 (29.3) 369 (29.1) 19 (33.3) 1.82 (0.86-3.83) 0.12
 Others 53 (4.0) 52 (4.1) 1 (1.8) 0.58 (0.07-4.59) 0.60
Recipient/donor ABO incompatible, n (%) <0.001
 No 1297 (97.9) 1246 (98.3) 51 (89.5) (reference)
 Yes 28 (2.1) 22 (1.7) 6 (10.5) 7.13 (2.41-21.16) <0.001
Donor KDPI, median (IQR) 30.0 (12.0–55.0) 30.0 (12.0–55.0) 39.0 (19.0–77.0) 0.04 1.01 (1.00-1.02) 0.24
Txp procedure type, n (%) 0.22
 Whole 1319 (99.5) 1263 (99.6) 56 (98.2) (reference)
 Partial 1 (0.1) 1 (0.1) 0 (0.0) - -
 Split 5 (0.4) 4 (0.3) 1 (1.8) 9.58 (0.89-103.45) 0.06
Acute rejection episodes, n (%) <0.001
 No 1306 (98.6) 1253 (98.8) 53 (93.0) (reference)
 Yes 19 (1.4) 15 (1.2) 4 (7.0) 4.79 (1.10-20.86) 0.04
Recipient post-txp LOS (d), median (IQR) 16.0 (10.0–26.0) 15.0 (10.0–25.0) 23.5 (13.0–36.0) 0.002 1.01 (1.00-1.02) 0.01
Bold emphasis denotes P < 0.05.
CI, confidence interval; IQR, interquartile range; KDPI, Kidney Donor Profile Index; LOS, length of stay; LT, primary liver transplant; OR, odds ratio; reLT, liver retransplant; txp, transplant.

Several differences in perioperative variables were observed between the 2 cohorts. Patients receiving reLT required more packed red blood cells (PRBCs) during their index transplant operation (P = 0.04; Table S2, SDC,https://links.lww.com/TP/C670). Their index hospital length of stay was longer (P = 0.002), and they experienced acute rejection episodes in greater proportions (P < 0.001, Table 1).

Multivariable analysis identified several variables that were significantly associated with the need for reLT (Table 1). ABO incompatibility between the recipient and donor increased the risk of needing reLT (odds ratio [OR], 7.13; 95% confidence interval [CI], 2.41-21.16; P < 0.001). Experiencing acute rejection after LT also increased the odds of undergoing reLT (OR, 4.79; 95% CI, 1.10-20.86; P = 0.04). Black recipients were more likely to require reLT when adjusting for confounding variables (OR, 2.48; 95% CI, 1.05-5.86; P = 0.04; Table 1). Patients with a longer post-LT hospital length of stay were also at a greater risk of undergoing reLT (OR, 1.01; 95% CI, 1.00-1.02; P = 0.01, Table 1). When post-LT hospital length of stay was removed as a variable, ABO incompatibility (OR, 4.98, 95% CI, 1.74-14.20), acute rejection (OR, 4.71; 95% CI, 1.13-19.66), and Black recipient race (OR, 2.45; 95% CI, 1.10-5.47) all increased the chances of undergoing reLT (P < 0.05).

Cox proportional hazards analysis revealed several variables associated with primary graft failure in the reLT cohort (ie, GS1). Obese patients (body mass index [BMI] ≥ 30 kg/m2) had a higher risk of graft failure than patients with a normal BMI (hazard ratio [HR], 2.19; 95% CI, 1.11-4.35, P = 0.02). Patients on preoperative mechanical ventilation also had a greater risk of post-LT graft failure (HR, 1.94; 95% CI, 1.02-3.68, P = 0.04; Table 2). Factors associated with extended criteria donors, such as DCD (P = 0.26), donor length of hospital stay (P = 0.13), and donor age (P = 0.78; Table 2), did not increase the risk of primary graft failure. Primary graft failure risk was also not associated with era of transplant (2008–2014 versus 2015–2021; P = 0.14; Table 2).

TABLE 2. - Characteristics associated with primary graft failure in the retransplant cohort in univariable Cox proportional hazards analysis
Graft failure (n = 57) Univariable
HR (95% CI) P
Age at transplant (y), median (IQR) 57.0 (45.0–64.0) 1.01 (0.99-1.04) 0.25
Body mass index (kg/m2), median (IQR) 28.7 (25.3–32.9) 1.06 (1.01-1.11) 0.02
Body mass index (kg/m2), n (%)
 <18.5 1 (1.8) 2.09 (0.27-16.37) 0.48
 18.5–24.9 13 (22.8) (reference)
 25.0–29.9 17 (29.8) 1.60 (0.76-3.34) 0.22
 ≥30 26 (45.6) 2.19 (1.11-4.35) 0.02
Time on waitlist (d), median (IQR) 245.0 (10.0–393.0) 1.00 (1.00-1.00) 0.94
Laboratory MELD score at transplant, median (IQR) 28.0 (18.0–35.0) 1.01 (0.99-1.04) 0.36
MELD/status at transplant, n (%)
 6–15 6 (10.5) (reference)
 16–29 15 (26.3) 0.73 (0.28-1.94) 0.53
 30–40 33 (57.9) 1.20 (0.50-2.89) 0.68
 Status 1A 3 (5.3) 1.36 (0.33-5.54) 0.67
Preoperative ventilation, n (%)
 No 42 (73.7) (reference)
 Yes 15 (26.3) 1.94 (1.02-3.68) 0.04
Preoperative dialysis, n (%)
 No 36 (63.2) (reference)
 Yes 21 (36.8) 1.51 (0.86-2.65) 0.15
Recipient HCV serostatus, n (%)
 Negative 46 (80.7) (reference)
 Positive 11 (19.3) 0.86 (0.44-1.68) 0.67
 Unknown/not done 0 (0.0) - -
Donor age at transplant (y), median (IQR) 38.0 (26.0–46.0) 1.00 (0.98-1.02) 0.78
DCD, n (%)
 No 53 (93.0) (reference)
 Yes 4 (7.0) 0.55 (0.20-1.55) 0.26
Donor hospital length of stay (d), median (IQR) 4.0 (3.0–0.0) 1.07 (0.98-1.18) 0.13
Cold ischemic time (h), median (IQR) 6.2 (4.8–8.0) 1.00 (0.89-1.11) 0.95
Acute rejection episodes, n (%)
 No 53 (93.0) (reference)
 Yes 4 (7.0) 1.61 (0.57-4.53) 0.37
Intraoperative PRBC (units), median (IQR) 8.0 (6.0–14.0) 1.00 (0.97-1.04) 0.81
Era of retransplant, n (%)
 2008–2014 15 (26.3) (reference)
 2015–2021 42 (73.7) 1.71 (0.84-3.49) 0.14
Bold emphasis denotes P < 0.05.
CI, confidence interval; DCD, donation after circulatory death; HCV, hepatitis C virus; HR, hazard ratio; IQR, interquartile range; MELD, Model of End-stage Liver Disease; PRBC, packed red blood cells.

Interval From LT to reLT (Retransplant Interval)

The retransplant interval (ie, the interval from dates of LT to reLT or GS1) was not associated with patient OS (P = 0.86, Figure 2A). The reasons for initial graft failure, necessitating reLT, could be grouped by time intervals (0–7, 8–90, and >90 d). The most common indication for reLT within the first week was PNF. Patients with PNF (n = 17) had a median retransplant interval of 4.5 (IQR, 2.0–7.0) d. Of those patients, 14 (82.4%) had a retransplant interval of 0 to 7 d. Between 8 and 90 d, the most common indication for reLT was HAT. Patients with HAT underwent retransplant a median of 13 (IQR, 10.0–56.6) d after LT. Chronic rejection was the primary cause of graft failure resulting in reLT >90 d after LT. We observed only 1 case of ischemic cholangiopathy (IC) as an indication for reLT and no recurrences of primary hepatic disease. Patient survival was similar across all indications for reLT (Figure 2B).

F2
FIGURE 2.:
Patient overall survival up to 5 y stratified by (A) retransplant interval and (B) cause of index graft failure. HAT, hepatic artery thrombosis; PNF, primary nonfunction.

Patient Survival Outcomes for LT and reLT Cohorts

OSs at 1, 3, 5, and 10 y post-LT were 92%, 82%, 76%, and 59%, respectively. In comparison, the first OSs in the reLT cohort at the same intervals were 89%, 74%, 73%, and 68%. OS did not significantly differ between LT and reLT patients (P = 0.33, Figure 3A). GSs after single LT were 92%, 82%, 75%, and 59% at 1, 3, 5, and 10 y, respectively, compared with 86%, 73%, 70%, and 70% after reLT (GS2). LT and post-reLT (GS2) GSs were not significantly different (P = 0.10, Figure 3B).

F3
FIGURE 3.:
Survival after primary liver transplant and retransplant. Patient OS (A) and graft survival (B) in single LT and reLT cohorts, the latter stratified by index (OS1) and second transplant (OS2, GS2). LT, liver transplantation; OS, overall survival; reLT, liver retransplantation; txp, transplant.

Factors Associated With Graft Failure After reLT

Univariable Cox proportional hazards analysis identified several factors that predicted liver graft failure after reLT (GS2, Table 3). These variables were primarily associated with the reLT liver donor, such as receiving a graft from a donor who was of Black race (HR, 5.52; 95% CI, 1.78-17.12; P = 0.003). KDPI was also significantly associated with second graft failure, both as a continuous variable (HR, 1.02; 95% CI, 1.00-1.04, P = 0.04) and values with ≥80% (HR, 5.42; 95% CI, 1.19-24.59; P = 0.03). Receiving a liver graft from a national donor increased the risk of graft failure relative to a local donor (HR, 4.20; 95% CI, 1.16-15.27; P = 0.03). Neither length of donor terminal hospitalization, nor recipient waitlist time for reLT, nor cause of index graft failure, nor era in which reLT was performed was associated with second graft failure.

TABLE 3. - Characteristics associated with second graft survival after liver retransplant in univariable Cox proportional hazards analysis
Graft functioning (n = 48) Graft failure (n = 20) a Univariable
HR (95% CI) P
Recipient race and ethnicity, n (%)
 White 23 (47.9) 11 (55.0) (reference)
 Black 13 (27.1) 2 (10.0) 0.65 (0.14-3.04) 0.59
 Hispanic 11 (22.9) 4 (20.0) 1.10 (0.34-3.54) 0.87
 Asian 1 (2.1) 3 (15.0) 4.38 (1.18-16.16) 0.03
 Other - - - -
Laboratory MELD at txp, median (IQR) 33.0 (28.5–36.0) 36.0 (14.0–38.5) 0.98 (0.94-1.02) 0.36
MELD/status at txp, n (%)
 6–15 0 (0.0) 1 (5.0) (reference)
 16–29 - - - -
 30–40 38 (79.2) 15 (75.0) 0.25 (0.03-1.94) 0.19
 Status 1A 10 (20.8) 4 (20.0) 0.28 (0.03-2.58) 0.26
Indication for retransplant, n (%)
 Chronic rejection 14 (29.2) 4 (20.0) (reference)
 PNF 11 (22.9) 6 (30.0) 1.68 (0.47-5.99) 0.42
 HAT 17 (35.4) 7 (35.0) 1.19 (0.35-4.09) 0.78
 Other 6 (12.5) 3 (15.0) 0.87 (0.16-4.77) 0.87
Time on waitlist (d), median (IQR) 7.0 (2.5–40.5) 4.5 (2.5–24.0) 1.00 (1.00-1.00) 0.66
Donor race and ethnicity, n (%)
 White 30 (62.5) 8 (40.0) (reference)
 Black 4 (8.3) 6 (30.0) 5.52 (1.78-17.12) 0.003
 Hispanic 11 (22.9) 6 (30.0) 2.25 (0.76-6.70) 0.15
 Asian 3 (6.3) 0 (0.0) - -
DCD, n (%)
 No 48 (100.0) 20 (100.0) - -
 Yes 0 (0.0) 0 (0.0) - -
Allocation type, n (%)
 Local 21 (43.8) 8 (40.0) (reference)
 Regional 22 (45.8) 8 (40.0) 1.13 (0.41-3.11) 0.82
 National 5 (10.4) 4 (20.0) 4.20 (1.16-15.27) 0.03
KDPI, median (IQR) 14.0 (5.0–36.0) 26.0 (10.0–45.0) 1.02 (1.00-1.04) 0.04
KDPI, n (%)
 <80% 46 (97.9) 17 (89.5) (reference)
 ≥80% 1 (2.1) 2 (10.5) 5.42 (1.19-24.59) 0.03
Donor hospital length of stay (d), median (IQR) 3.0 (2.0–5.0) 3.0 (3.0–6.0) 1.00 (1.00-1.01) 0.12
Era of retransplant, n (%)
 2008–2014 8 (16.7) 5 (25.0) (reference)
 2015–2021 40 (83.3) 15 (75.0) 1.32 (0.42-4.10) 0.64
Bold emphasis denotes P < 0.05.
a17 of the 20 graft failures were due to patient death.
CI, confidence interval; DCD, donation after circulatory death; HAT, hepatic artery thrombosis; HR, hazard ratio; IQR, interquartile range; KDPI, Kidney Donor Profile Index; MELD, Model of End-stage Liver Disease; PNF, primary nonfunction; txp, transplant.

Factors Associated With Patient Survival After reLT

Several recipient variables were significantly associated with patient survival in reLT recipients, although this analysis was affected by low event rates and small sample size (Table 4). These factors included Asian recipient race at LT (HR, 7.75; 95% CI, 1.55-38.75) and reLT (HR, 5.56; 95% CI, 1.44-21.44), having a BMI <18.5 kg/m2 at LT (HR, 18.40; 95% CI, 1.63-208.26; P < 0.05; Tables 4 and 5), and having an etiology of liver failure as hepatitis B (HR, 27.50; 95% CI, 2.15-351.93) or acetaminophen exposure (HR, 20.47; CI, 1.70-246.16).

TABLE 4. - Characteristics associated with overall survival from date of first liver transplant in the retransplant cohort by univariable Cox proportional hazards analysis
Alive (n = 40) Dead (n = 17) Univariable
HR (95% CI) P
Recipient race and ethnicity, n (%)
 White 20 (50.0) 10 (58.8) (reference)
 Black 8 (20.0) 1 (5.9) 0.53 (0.07-4.29) 0.55
 Hispanic 12 (30.0) 4 (23.5) 1.11 (0.33-3.65) 0.87
 Asian 0 (0.0) 2 (11.8) 7.75 (1.55-38.75) 0.01
Body mass index (kg/m2), n (%)
 <18.5 0 (0.0) 1 (5.9) 18.40 (1.63-208.26) 0.02
 18.5–24.9 9 (22.5) 4 (23.5) (reference)
 25.0–29.9 13 (32.5) 4 (23.5) 0.71 (0.18-2.86) 0.63
 ≥30 18 (45.0) 8 (47.1) 1.22 (0.36-4.06) 0.75
Time on waitlist (d), median (IQR) 246.5 (12.5–385.5) 208.0 (6.0–557.0) 1.00 (1.00-1.00) 0.56
Laboratory MELD at transplant, median (IQR) 29.5 (18.5–37.0) 21.0 (15.0–31.0) 0.97 (0.93-1.01) 0.13
MELD/status at transplant, n (%)
 6–15 3 (7.5) 3 (17.6) (reference)
 16–29 9 (22.5) 6 (35.3) 0.73 (0.18-2.96) 0.66
 30–40 26 (65.0) 7 (41.2) 0.30 (0.07-1.19) 0.09
 Status 1A 2 (5.0) 1 (5.9) 0.16 (0.01-2.19) 0.17
Primary diagnosis at transplant, n (%)
 Acute hepatic necrosis 2 (5.0) 2 (11.8) 0.67 (0.07-6.48) 0.73
 Hepatitis B 0 (0.0) 1 (5.9) 27.50 (2.15-351.93) 0.01
 Hepatitis C 2 (5.0) 2 (11.8) 2.17 (0.36-13.04) 0.40
 Cirrhosis: acetaminophen 0 (0.0) 1 (5.9) 20.47 (1.70-246.16) 0.02
 Cirrhosis: other 10 (25.0) 3 (17.6) - -
 Alcoholic hepatitis/cirrhosis (reference)
 Nonalcoholic steatohepatitis 5 (12.5) 3 (17.6) 2.58 (0.51-13.12) 0.25
 Hepatocellular carcinoma 12 (30.0) 4 (23.5) 0.97 (0.22-4.40) 0.97
 Cholangiocarcinoma 2 (5.0) 1 (5.9) 1.59 (0.16-15.37) 0.69
 Benign tumor 1 (2.5) 0 (0.0) - -
Retransplant interval (d), median (IQR) 12.0 (7.0–215.0) 10.0 (5.0–64.0) 1.00 (0.99-1.00) 0.31
Retransplant interval, n (%)
 0–7 d 12 (30.0) 6 (35.3) (reference)
 8–90 d 16 (40.0) 8 (47.1) 0.81 (0.28-2.39) 0.70
 >90 d 12 (30.0) 3 (17.6) 0.48 (0.12-1.93) 0.30
Indication for retransplant, n (%)
 Chronic rejection 6 (15.0) 3 (17.6) (reference)
 PNF 12 (30.0) 5 (29.4) 1.26 (0.29-5.47) 0.76
 HAT 17 (42.5) 7 (41.2) 1.14 (0.28-4.55) 0.86
 Other graft failure 5 (12.5) 2 (11.8) 0.71 (0.11-4.40) 0.71
DCD, n (%)
 No 36 (90.0) 17 (100.0) - -
 Yes 4 (10.0) 0 (0.0) - -
Deceased donor cause of death, n (%)
 Anoxia 9 (22.5) 8 (47.1) 6.12 (1.29-28.97) 0.02
 Cerebrovascular 14 (35.0) 7 (41.2) 3.92 (0.80-19.25) 0.09
 Head trauma 17 (42.5) 2 (11.8) (reference)
Donor BUN (mg/dL), median (IQR) 17.0 (11.5–25.0) 22.0 (14.0–70.0) 1.02 (1.01-1.04) 0.002
Donor hospital LOS (d), median (IQR) 3.0 (2.0–4.5) 5.0 (4.0–6.0) 1.17 (1.04-1.31) 0.01
Era of retransplant, n (%)
 2008–2014 11 (27.5) 4 (23.5) (reference)
 2015–2021 29 (72.5) 13 (76.5) 2.45 (0.68-8.91) 0.17
Bold emphasis denotes P < 0.05.
BUN, blood urea nitrogen; CI, confidence interval; DCD, donation after circulatory death; HAT, hepatic artery thrombosis; HR, hazard ratio; IQR, interquartile range; LOS, length of stay; MELD, Model of End-stage Liver Disease; PNF, primary nonfunction.

TABLE 5. - Characteristics at liver retransplant associated with overall survival following second transplant by univariable Cox proportional hazards analysis
Alive (n = 50) Dead (n = 18) Univariable
HR (95% CI) P
Recipient race and ethnicity, n (%)
 White 25 (50.0) 9 (50.0) (reference)
 Black 13 (26.0) 2 (11.1) 0.82 (0.17-3.97) 0.81
 Hispanic 11 (22.0) 4 (22.2) 1.38 (0.41-4.65) 0.60
 Asian 1 (2.0) 3 (16.7) 5.56 (1.44-21.44) 0.01
 Other - -
Time on waitlist (d), median (IQR) 7.0 (2.0–43.0) 4.5 (3.0–9.0) 1.00 (1.00-1.00) 0.74
Laboratory MELD at transplant, median (IQR) 34.0 (29.0–37.0) 34.5 (11.0–37.0) 0.96 (0.92-1.01) 0.11
MELD/status at transplant, n (%)
 6–15 0 (0.0) 1 (5.6) (reference)
 16–29 - -
 30–40 40 (80.0) 13 (72.2) 0.22 (0.03-1.73) 0.15
 Status 1A 10 (20.0) 4 (22.2) 0.30 (0.03-2.70) 0.28
Indication for retransplant, n (%)
 Chronic rejection 15 (30.0) 3 (16.7) (reference)
 PNF 12 (24.0) 5 (27.8) 1.79 (0.43-7.53) 0.43
 HAT 17 (34.0) 7 (38.9) 1.65 (0.43-6.37) 0.47
 Other 6 (12.0) 3 (16.7) 1.21 (0.20-7.29) 0.83
Donor race and ethnicity, n (%)
 White 31 (62.0) 7 (38.9) (reference)
 Black 5 (10.0) 5 (27.8) 4.23 (1.28-13.96) 0.02
 Hispanic 11 (22.0) 6 (33.3) 2.65 (0.86-8.24) 0.09
 Asian 3 (6.0) 0 (0.0) - -
 Other - -
DCD, n (%)
 No 50 (100.0) 18 (100.0) - -
 Yes 0 (0.0) 0 (0.0) - -
Donor hospital length of stay (d), median (IQR) 3.0 (2.0–5.0) 4.0 (3.0–6.0) 1.00 (1.00-1.01) 0.10
Allocation type n (%)
 Local 23 (46.0) 6 (33.3) (reference)
 Regional 22 (44.0) 8 (44.4) 1.67 (0.55-5.10) 0.37
 National 5 (10.0) 4 (22.2) 5.98 (1.48-24.06) 0.01
KDPI, median (IQR) 14.0 (5.0, 32.0) 26.0 (10.0, 45.0) 1.02 (1.00-1.04) 0.03
KDPI, n (%)
 <80% 48 (98.0) 15 (88.2) (reference)
 ≥80% 1 (2.0) 2 (11.8) 6.06 (1.32-27.86) 0.02
Era of transplant, n (%)
 2008–2014 10 (20.0) 3 (16.7) (reference)
 2015–2021 40 (80.0) 15 (83.3) 2.90 (0.63-13.32) 0.17
Bold emphasis denotes P < 0.05.
CI, confidence interval; DCD, donation after circulatory death; HAT, hepatic artery thrombosis; HR, hazard ratio; IQR, interquartile range; KDPI, Kidney Donor Profile Index; MELD, Model of End-stage Liver Disease; PNF, primary nonfunction.

Mortality was not affected by recipient age at LT (HR, 1.03; 95% CI, 0.99-1.08; P = 0.12) or reLT (HR, 1.03; 95% CI, 0.99-1.07; P = 0.19). Additionally, recipient laboratory MELD score at LT (HR, 0.97; 95% CI, 0.93-1.01; P = 0.13) and reLT (HR, 0.96; 95% CI, 0.92-1.01; P = 0.11) did not affect the risk of dying. Receiving dialysis before LT (HR, 0.34; 95% CI, 0.10-1.20; P = 0.09) or reLT (HR, 0.67; 95% CI, 0.25-1.78; P = 0.42) did not affect mortality. Patients requiring mechanical ventilation before LT (HR, 0.45; 95% CI, 0.13-1.63; P = 0.23) or reLT (HR, 1.15; 95% CI, 0.43-3.05; P = 0.78) also did not experience a higher mortality hazard. The risk of dying was similar in the 2008 to 2014 era and the 2015 to 2021 era (HR, 2.45; 95% CI, 0.68-8.91; P = 0.17; Table 4; Figures S1 and S2, SDC,https://links.lww.com/TP/C670). Mortality risk was also similar regardless of the indication for retransplant (P > 0.05; Table 4).

Donor variables affecting mortality included receiving the LT graft from a donor who died from anoxia (HR, 6.12; 95% CI, 1.29-28.97; P = 0.02; Table 4), national reLT donor allocation (HR, 5.98; 95% CI, 1.48-24.06), Black reLT donor race (HR, 4.23; 95% CI, 1.28-13.96), and the KDPI of the reLT donor (HR, 1.02; 95% CI, 1.01-1.04; P = 0.002; Table 5). KDPI of the initial LT donor was not associated with survival (continuous, P = 0.78; KDPI ≥80%, P = 0.77). The donor’s terminal blood urea nitrogen (BUN) levels at initial LT also significantly affected the risk of death (HR, 1.02; 95% CI, 1.01-1.04; P = 0.002) and approached significance for reLT donors (P = 0.06). Donor terminal hospitalization at primary LT was associated with a greater risk of mortality (HR, 1.17; 95% CI, 1.04-1.31; P = 0.01; Table 4), although secondary donor hospitalization was not (HR, 1.00; 95% CI, 1.00-1.01; P = 0.10; Table 5). Donor age at initial LT (HR, 0.98; 95% CI, 0.95-1.01; P = 0.30) and reLT (HR, 1.04; 95% CI, 1.00-1.08; P = 0.07) was not associated with mortality.

No LT or reLT perioperative variables were significantly associated with mortality. For example, cold ischemia times at LT (HR, 0.90; 95% CI, 0.73-1.10; P = 0.30) and reLT (HR, 0.89; 95% CI, 0.68-1.18; P = 0.43) were not associated with an increased risk of death. The number of units of PRBCs administered during LT (HR, 0.94; 95% CI, 0.86-1.02; P = 0.13) and reLT (HR, 0.99; 95% CI, 0.92-1.06; P = 0.83) also did not significantly influence mortality.

DISCUSSION

The current retrospective cohort series from a single center has shown that reLT can have outcomes equivalent to those of single LT. Ethical management of scarce liver allografts requires liver allocation to waitlist patients who are most likely to benefit. The previous reports of poor reLT outcomes may have dissuaded centers from relisting LT recipients with graft failure to minimize waitlist and posttransplant mortality.4 As the absolute number of patients undergoing reLT has increased over time, 5-y reLT OS and GS have also improved relative to historical values6–8,23,24 to approximately 70% in this study, well above the 50% 5-y survival benchmark suggested as a cutoff by some authors.4 These results add to the growing body of literature demonstrating that currently reLT has similar outcomes as LT.10,17,18 This may be due in part to changes in transplant care universally. For example, Croome et al10 attributed similar LT and reLT outcomes to improvements in direct acting antiviral medications and reductions in hepatitis C recurrences.10,14 Our center provides advanced care in a transplant-specific intensive care unit, which has improved outcomes for our sickest patients, including patients requiring reLT.

This study identified multiple factors associated with requiring reLT. These factors included a high BMI at LT, pre-LT mechanical ventilation, donor–recipient ABO incompatibility, early acute rejection episodes, and a longer index hospital stay. These variables have also been identified as risk factors for reLT in other studies.25–27 However, they were not associated with survival following reLT, implying that patients with these risk factors still have good outcomes after reLT. For example, although pretransplant mechanical ventilation increased the risk of needing reLT in the current study, it was not associated with a poorer OS. These results are in contrast to examples in the literature, possibly because of differences in the critical care management of LT patients.5,11,12,28

Post-reLT graft failure was largely attributable to patient deaths with functioning grafts. Of 20 cases of post-reLT graft failure, 15 (75%) were due to patient death. Thus, factors associated with post-reLT GS are also associated with OS, including Asian recipient race, Black reLT donor race, reLT national organ allocation, and reLT KDPI. This highlights the importance of reLT donor quality.

Post-reLT survival was associated with several measures of donor quality, such as national organ sharing, LT donor death due to anoxia, LT donor terminal BUN, donor length of terminal hospitalization, and KDPI. Although KDPI was intended to predict deceased donor kidney allograft survival, it has been shown to accurately predict liver allograft survival.29 This is intuitive, as many of the variables included in the KDPI calculation have been shown to separately predict survival after reLT, including donor age,14,15,30 race,15 cause of death,15,31 and DCD donors.15,32 It is noteworthy that reLT KDPI and LT donor BUN significantly predicted mortality, but serum creatinine did not (HR, 1.18; 95% CI, 0.91-1.54; P = 0.22). Studies have suggested that BUN may predict outcomes in patients with liver disease more accurately than serum creatinine.33–35 These results highlight the importance of using high-quality donor livers for reLT because low-quality grafts are associated with reduced survival.15 Other variables associated with donor quality, such as donor age, donor BMI, and cardiac death, were not associated with negative outcomes.

Interestingly, many variables that predicted reLT outcomes in earlier studies were not significant in the present study. For example, indicators of intraoperative technical difficulty, such as PRBC utilization and cold ischemia time, were previously associated with poor outcomes after reLT.2,11,12,15,17 In the current study, the units of PRBCs administered were not associated with differences in OS. This is likely due to differences in techniques, including universal use of piggyback vena cava reconstruction at our institution. Piggyback reconstruction at the time of index transplant allows faster dissection during retransplant, reducing operative time relative to other reLT procedures. Recipient’s age, MELD score, medical condition at transplant, pretransplant mechanical ventilation, and dialysis were also not associated with mortality, in contrast with previous reports,13,36–40 suggesting that reLT may be safely offered to more severely ill patients with primary graft failure. These results are particularly interesting given the high median MELD of 28 in this cohort.

OS was independent of the time interval to reLT, also in contradiction to findings of earlier studies.6,27,41 The retransplant interval (ie, GS1) was associated with graft failure etiology but not with survival.24,26,40–42 The timing of reLT and its association with distinct etiologies are frequently discussed. In line with other studies,43,44 the most common indication for reLT within the first postoperative week of the index LT was PNF. A recent study by Abbassi et al20 reported worse outcomes for reLT patients with PNF. Yet, in our cohort, outcomes were not associated with the causes of index graft failure, which we attribute to rapid reLT for PNF, within the first few days after primary LT (median 4.5 d). ReLT within the next several months was generally performed for HAT and later reLT usually for chronic rejection. Many earlier reports associated late reLT with poor outcomes,2,6,7,13,28,45,46 but achieving excellent survival regardless of timing or etiology is in keeping with recent studies.47

Only 1 patient received reLT for IC. IC is associated with long ischemia times, organ donation following cardiac death, episodes of acute rejection, and CMV infection. Our center minimizes these risks by overlapping donor and recipient operations, maximizing piggyback caval reconstruction, highly selectively utilizing DCD donors, and maximizing inpatient and outpatient pharmacy oversight of immunosuppression and CMV prophylaxis.48–50 The combination of these measures may account for our low IC rates.

The improved outcomes after reLT in this study compared with historical cohorts are relevant to the debate on reLT futility. Although controversial,4,51 1 definition includes a 5-y GS rate and 1-y OS rate of <50%.4 The 5-y GS rate of 70% and a 1-y OS rate of 92% in this study are thus above this threshold. At the population level, we observe that reLT is not futile. Because of our small reLT sample size and few variables associated with reLT outcomes, we cannot address futility at the individual level.

The present study has several limitations. First, as a retrospective, single-center study, we cannot demonstrate external validity, although post-reLT outcomes may not be center-specific.52 Our academic practice includes recipients with high levels of liver dysfunction and advanced primary hepatic malignancies, which may worsen outcomes. However, patient risk factors did not differ significantly between our LT and reLT cohorts. Second, although HMH is a high-volume center, our sample size is nonetheless limited because the proportion of reLTs was lower than generally reported.2,17,18,32 This small sample size with low event rates impaired statistical power of many analyses. Third, although we endeavored to capture many covariates, it is possible that relevant features were missed. Without a replication cohort, we could not validate our multivariable model. Finally, it is likely that our reLT patient selection was informed by prior studies, potentially explaining the improved outcomes observed. Although we cannot rule out patient selection bias in our current dataset, the increased absolute number and stable frequency of reLTs over time argue against this bias.

In conclusion, reLT operations are justifiable, with both OS and GS that are statistically equivalent to single LT recipients. Many variables that were previously associated with reduced survival after reLT were not validated in our contemporary cohort, including the timing of retransplantation, preoperative mechanical ventilation, cold ischemia time, and PRBC administration. Transplant centers should closely monitor recipients for reLT risk factors, including high BMI, mechanical ventilation before LT, and donor–recipient ABO incompatibility. At reLT, attention should be paid to donor location and quality, which are most strongly associated with survival outcomes. Improved outcomes imply that more liberal use of reLT may be appropriate, especially given the lack of alternate treatment options for most of these patients.

REFERENCES

1. Kwong AJ, Ebel NH, Kim WR, et al. OPTN/SRTR 2020 annual data report: liver. Am J Transplant. 2022;22(Suppl 2):204–309.
2. Brüggenwirth IMA, Werner MJM, Adam R, et al.; all the other contributing centers (www.eltr.org), the European Liver, Intestine Transplant Association (ELITA). The Liver Retransplantation Risk Score: a prognostic model for survival after adult liver retransplantation. Transpl Int. 2021;34:1928–1937.
3. Kitchens WH, Yeh H, Markmann JF. Hepatic retransplant: what have we learned? Clin Liver Dis. 2014;18:731–751.
4. Biggins SW. Futility and rationing in liver retransplantation: when and how can we say no? J Hepatol. 2012;56:1404–1411.
5. Doyle HR, Morelli F, McMichael J, et al. Hepatic retransplantation--an analysis of risk factors associated with outcome. Transplantation. 1996;61:1499–1505.
6. Markmann JF, Markowitz JS, Yersiz H, et al. Long-term survival after retransplantation of the liver. Ann Surg. 1997;226:408–420.
7. Facciuto M, Heidt D, Guarrera J, et al. Retransplantation for late liver graft failure: predictors of mortality. Liver Transpl. 2000;6:174–179.
8. Watt KD, Lyden ER, McCashland TM. Poor survival after liver retransplantation: is hepatitis C to blame? Liver Transpl. 2003;9:1019–1024.
9. Yao FY, Saab S, Bass NM, et al. Prediction of survival after liver retransplantation for late graft failure based on preoperative prognostic scores. Hepatology. 2004;39:230–238.
10. Croome KP, Mathur AK, Pungpapong S, et al. Equivalent outcomes with retransplantation and primary liver transplantation in the direct-acting antiviral era. Transplantation. 2019;103:1168–1174.
11. Hong JC, Kaldas FM, Kositamongkol P, et al. Predictive index for long-term survival after retransplantation of the liver in adult recipients analysis of a 26-year experience in a single center. Ann Surg. 2011;254:444–449.
12. Markmann JF, Gornbein J, Markowitz JS, et al. A simple model to estimate survival after retransplantation of the liver. Transplantation. 1999;67:422–430.
13. Rosen HR, Prieto M, Casanovas-Taltavull T, et al. Validation and refinement of survival models for liver retransplantation. Hepatology. 2003;38:460–469.
14. Pelletier SJ, Schaubel DE, Punch JD, et al. Hepatitis C is a risk factor for death after liver retransplantation. Liver Transpl. 2005;11:434–440.
15. Northup PG, Pruett TL, Kashmer DM, et al. Donor factors predicting recipient survival after liver retransplantation: the retransplant donor risk index. Am J Transplant. 2007;7:1984–1988.
16. Bittermann T, Shaked A, Goldberg DS. When living donor liver allografts fail: exploring the outcomes of retransplantation using deceased donors. Am J Transplant. 2017;17:1097–1102.
17. Takagi K, Domagala P, Porte RJ, et al. Liver retransplantation in adult recipients: analysis of a 38-year experience in the Netherlands. J Hepatobiliary Pancreat Sci. 2020;27:26–33.
18. Jeffrey AW, Delriviere L, McCaughan G, et al. Excellent contemporary graft survival for adult liver retransplantation: an Australian and New Zealand registry analysis from 1986 to 2017. Transplant Direct. 2019;5:e472.
19. Schielke A, Scatton O, Boelle PY, et al. Ischemic-type biliary lesions: a leading indication of liver retransplantation with excellent results. Clin Res Hepatol Gastroenterol. 2019;43:131–139.
20. Abbassi F, Gero D, Muller X, et al. Novel benchmark values for redo liver transplantation. Does the outcome justify the effort? Ann Surg. 2022;276:860–867.
21. Hastie TR, Tibishirani R, Wainwright M. Statistical Learning with Sparsity: The Lasso and Generalizations. CRC Press2015.
22. StataCorp. Stata Lasso. Stata 16 Base Reference Manual. Stata Press2019.
23. Azoulay D, Linhares MM, Huguet E, et al. Decision for retransplantation of the liver: an experience- and cost-based analysis. Ann Surg. 2002;236:713–721.
24. Bilbao I, Figueras J, Grande L, et al. Risk factors for death following liver retransplantation. Transplant Proc. 2003;35:1871–1873.
25. Al-Freah MAB, Moran C, Foxton MR, et al. Impact of comorbidity on waiting list and post-transplant outcomes in patients undergoing liver retransplantation. World J Hepatol. 2017;9:884–895.
26. Enestvedt CK, Malik S, Reese PP, et al. Biliary complications adversely affect patient and graft survival after liver retransplantation. Liver Transpl. 2013;19:965–972.
27. Andres A, Gerstel E, Combescure C, et al. A score predicting survival after liver retransplantation for hepatitis C virus cirrhosis. Transplantation. 2012;93:717–722.
28. Zakaria H, Saleh Y, Zidan A, et al. Is it justified to use liver grafts from living donors for retransplant? A single-center experience. Exp Clin Transplant. 2020;18:188–195.
29. Adam R, Karam V, Delvart V, et al.; all contributing centers (www.eltr.org). Evolution of indications and results of liver transplantation in Europe. A report from the European Liver Transplant Registry (ELTR). J Hepatol. 2012;57:675–688.
30. Schmitt TM, Kumer SC, Pruett TL, et al. Advanced recipient age (>60 years) alone should not be a contraindication to liver retransplantation. Transpl Int. 2009;22:601–605.
31. Reyes JD, Dick AA, Hendele JB, et al. Adults transplanted as children as retransplant candidates: analysis of outcomes support optimism in a population mislabeled as high risk. Clin Transplant. 2020;34:e13880.
32. Montenovo MI, Hansen RN, Dick AA. Outcomes of adult liver re-transplant patients in the model for end-stage liver disease era: is it time to reconsider its indications? Clin Transplant. 2014;28:1099–1104.
33. Pose E, Ginès P. Improving outcome prediction in alcohol-associated hepatitis: blood urea nitrogen and albumin help. Mayo Clin Proc. 2022;97:436–438.
34. Yoneyama K, Taniguchi H, Kiuchi Y, et al. Prognostic index of liver cirrhosis with ascites with and without hepatocellular carcinoma. Scand J Gastroenterol. 2004;39:1272–1279.
35. Tenza E, Bernardo CG, Escudero D, et al. Liver transplantation complications in the intensive care unit and at 6 months. Transplant Proc. 2009;41:1050–1053.
36. Leithead JA, Smith MR, Materacki LB, et al. Intercurrent infection predicts mortality in patients with late hepatic artery thrombosis listed for liver retransplantation. Liver Transpl. 2012;18:1353–1360.
37. Aguero F, Rimola A, Stock P, et al. Liver retransplantation in patients with HIV-1 infection: an international multicenter cohort study. Am J Transplant. 2016;16:679–687.
38. Stankiewicz R, Kornasiewicz O, Grąt M, et al. Is elective status a predictor of poor survival in liver retransplantation? Transplant Proc. 2017;49:1364–1368.
39. Xu Q, McAlister VC, Leckie S, et al. Angiotensin II type I receptor agonistic autoantibodies are associated with poor allograft survival in liver retransplantation. Am J Transplant. 2020;20:282–288.
40. Mezochow AK, Abt PL, Bittermann T. Differences in early immunosuppressive therapy among liver retransplantation recipients in a national cohort. Transplantation 2021;105:1800–1807.
41. Marudanayagam R, Shanmugam V, Sandhu B, et al. Liver retransplantation in adults: a single-centre, 25-year experience. HPB (Oxford). 2010;12:217–224.
42. Ghabril M, Dickson RC, Machicao VI, et al. Liver retransplantation of patients with hepatitis C infection is associated with acceptable patient and graft survival. Liver Transpl. 2007;13:1717–1727.
43. Avolio AW, Franco A, Schlegel A, et al. Development and validation of a comprehensive model to estimate early allograft failure among patients requiring early liver retransplant. JAMA Surg. 2020;155:e204095.
44. Rana A, Petrowsky H, Kaplan B, et al. Early liver retransplantation in adults. Transpl Int. 2014;27:141–151.
45. Moon HH, Kim TS, Song S, et al. Early vs late liver retransplantation: different characteristics and prognostic factors. Transplant Proc. 2018;50:2668–2674.
46. Linhares MM, Azoulay D, Matos D, et al. Liver retransplantation: a model for determining long-term survival. Transplantation. 2006;81:1016–1021.
47. López MJC, Franco CC, Artacho GS, et al. Results of early liver retransplantation. Transplant Proc. 2020;52:1486–1488.
48. Goria O, Archambeaud I, Lemaitre C, et al. Ischemic cholangiopathy: an update. Clin Res Hepatol Gastroenterol. 2020;44:486–490.
49. Croome KP, Mathur AK, Aqel B, et al. Classification of distinct patterns of ischemic cholangiopathy following DCD liver transplantation: distinct clinical courses and long-term outcomes from a multicenter cohort. Transplantation. 2022;106:1206–1214.
50. Hobeika MJ, Saharia A, Mobley CM, et al. Donation after circulatory death liver transplantation: an in-depth analysis and propensity score-matched comparison. Clin Transplant. 2021;35:e14304.
51. Zimmerman MA, Ghobrial RM. When shouldn’t we retransplant? Liver Transpl. 2005;11(Suppl 2):S14–S20.
52. Reese PP, Yeh H, Thomasson AM, et al. Transplant center volume and outcomes after liver retransplantation. Am J Transplant. 2009;9:309–317.

Supplemental Digital Content

Copyright © 2023 Wolters Kluwer Health, Inc. All rights reserved.