- Email: [email protected]
Early Vs Late Liver Retransplantation: Different Characteristics and Prognostic Factors
Early Vs Late Liver Retransplantation: Different Characteristics and Prognostic Factors H.H. Moona, T.-S. Kimb, S. Songc, M. Shind, Y.J. Chunge, S. Leee, G.S. Choie, J.M. Kime, C.H.D. Kwone, S.-K. Leee, and J. Johe,* a Department of Surgery, Kosin University Gospel Hospital, Kosin University School of Medicine, Busan, Korea; bDepartment of Surgery, Keimyung University Dongsan Hospital, Keimyung University School of Medicine, Daegu, Korea; cDepartment of Surgery, Dankuk University Hospital, Dankuk University School of Medicine, Daejeon, Korea; dDepartment of Surgery, Konkuk University Medical Center, Konkuk University School of Medicine, Seoul, Korea; and eDepartment of Surgery, Samsung Medical Center, Sungkyunkwan University School of Medicine, Seoul, Korea
ABSTRACT Background. East Asia is a known endemic area for hepatitis B, and living donor liver transplantation is mainly performed. Liver retransplantation (ReLT) is expected to become an increasing problem because of a shortage of organs. This study aimed to compare early and late ReLT with consideration of specific circumstances and disease background of East Asians. Methods. Between October 1996 and January 2015, 51 patients underwent ReLT; we performed a retrospective analysis of data obtained from medical records of the patients. Clinical characteristics, indication, causes of death, survival rate, and prognostic factors were investigated. Result. The survival rate for early ReLT (n ¼ 18) was 51.5% and that for late ReLT (n ¼ 33) was 50.1% at 1 year postoperatively. Continuous venovenous hemodialysis and the use of mechanical ventilators were more frequent, and pre-retransplant intensive care unit stay and prothrombin time was longer in early ReLT than in late ReLT. Operation time was longer and the amount of intraoperative blood loss was greater in late ReLT than in early ReLT. Multivariate analysis showed that a higher C-reactive protein level increased mortality in early ReLT (P ¼ .045), whereas a higher total bilirubin level increased the risk of death in late ReLT (P ¼ .03). Conclusion. Patients with early ReLT are likely to be sicker pre-retransplantation and require adequate treatment of the pretransplant infectious disease. On the other hand, late ReLT is likely to be technically more difficult and should be decided before the total bilirubin level increases substantially.
T
HE INCREASING number of liver transplants is expected to result in a growing number of graft losses. The most common indications of early graft failure after primary liver transplantation (LT) are primary nonfunction, vascular complications, and uncontrollable acute cellular rejection, whereas the most frequent causes for late liver retransplantation (ReLT) are disease recurrence and biliary tract complications [1,2]. It is known that survival rates following ReLT are 20% to 50% lower than those after the first transplantation.
0041-1345/18 https://doi.org/10.1016/j.transproceed.2018.03.040
2668
ReLT will become an increasing problem in the next decade because of a shortage of organs [3]. In Asia, considering the background that the majority of primary LT
*Address correspondence to Jae-Won Joh, MD, PhD, Department of Surgery, Samsung Medical Center, Sungkyunkwan University School of Medicine, 81 Irwon-Ro, Gangnam-Gu, Seoul, Republic of Korea 06351. Tel: þ82-2-3410-3466, Fax: þ82-2-3410-0040. E-mail: [email protected] ª 2018 Elsevier Inc. All rights reserved. 230 Park Avenue, New York, NY 10169
Transplantation Proceedings, 50, 2668e2674 (2018)
EARLY VS LATE LIVER RETRANSPLANTATION
involves living donor LT (LDLT), it is difficult to find another living donor for ReLT, and the wait for a deceased donor LT is long. The aim of this study was to investigate the different characteristics and prognostic factors between early and late ReLT by comparing indications, characteristics, and outcomes. PATIENTS AND METHODS Study Population From October 1996 through January 2015, a total of 1581 patients underwent LT at Samsung Medical Center, Korea. ReLT was carried out in 58 patients within this period. The rate of ReLT was 3.7%. Among them, we excluded 7 cases of pediatric ReLT but included recipients who were 15 years or older at the time of their ReLT procedure. Two patients who received a third graft were included, but in these cases, combined or multivisceral transplant was performed. We retrospectively evaluated these patients using their medical records. Clinical characteristics, operative variables, postoperative complications, survival rates, and causes of death were investigated.
2669 drugs were methylprednisolone, mycophenolic acid, and FK506 or cyclosporine. The starting time and dose of maintenance immunosuppressive drugs were tailored to each patient’s condition. Most of the recipients who were at high risk of infection received a reduced dose of methylprednisolone.
Statistics We used the Mann-Whitney test for continuous variables and Pearson’s c2 test and Fisher exact test for categorical variables. The quantitative data are presented as median and range. The Cox proportional hazard model was used to identify independent predictive factors. Hazard ratio (HR) and corresponding 95% CIs were calculated. Variables associated with survival in the univariate analyses with a P value <.20 were entered into the multivariate logistic regression analysis, and adjusted risk ratios with 95% CI were calculated. P value <.05 was considered significant, but P value <.10 was regarded as borderline significant. The receiver operating characteristic curve was used to evaluate the predictive cutoff value of C-reactive protein (CRP) and total bilirubin (TB) for prognosis. The Kaplan-Meier method was used for survival analysis of patients. All statistical analyses were performed using SPSS 18.0 for Windows (IBM, Armonk, NY, United States).
Definition Patients were classified according to the interval between ReLT and their previous LT. Patients undergoing ReLT within 60 days were classified under early ReLT, and those undergoing ReLT at 60 days or more after their LT were classified under late ReLT. Primary nonfunction (PNF) is defined as graft failure requiring retransplantation within 1 week of prior transplant without identifiable causes. Delayed nonfunction is defined as graft failure requiring retransplantation after 1 week without any other definable causes [4]. We estimated intraoperative blood loss (IBL) using the following formula: IBL ¼ patient’s estimated blood volume (mL) (preoperative hematocrit in % postoperative hematocrit in %) þ (transfused leukocyte-depleted red blood cell in units 213 70%) þ (transfused Cell Saver [Haemonetics, Braintree, Mass, United States] blood in mL 55%).
Indications Indications for ReLT included chronic rejection (CR), PNF, biliary cirrhosis, hepatic artery thrombosis, hepatic artery stenosis, portal vein thrombosis, hepatitis recurrence, recurrence of primary sclerosing cholangitis, and other causes of hepatic failure after primary LT. We did not perform ReLT in patients with uncontrolled infection, noncompliance, substance abuse, or aggressive recurrent hepatitis C (HCV) or B (HBV) virus infection.
Surgical Technique The modified piggyback technique was used in deceased donor ReLT, and modified right lobe graft was used in living donor ReLT. As a bile duct reconstruction method, duct-to-duct anastomosis was carried out, but Roux-en-Y hepaticojejunostomy, as the second best option, was performed in some cases.
Immunosuppression Basiliximab was usually used as an induction immunosuppressive drug in cases in which the interval between primary transplantation and ReLT was more than 2 years. Maintenance immunosuppressive
RESULTS Indications for ReLT
Table 1 shows the causes of early and late ReLT. The most common cause for early ReLT was PNF (61.1%), and the next was hepatic artery thrombosis (22.2%). In contrast, CR (36.4%) was the most frequent cause of late ReLT, followed by biliary complications (21.2%), primary sclerosing cholangitis, HBV, and HCV. There was a significant difference Table 1. Causes of Early and Late Liver Retransplantation Early ReLT (N ¼ 18) Causes
Chronic Rejection Primary Nonfunction Delayed Nonfunction Biliary Problem* Hepatic Vascular Problem HAT HVS PVT Disease Recurrence HBV HCV PSC AIH Others† Total
Late ReLT (N ¼ 33)
No. (%)
P Value
11 (61.1) 2 (11.1) -
12 (36.4) 7 (21.2)
4 (22.2) 1 (5.6)
1 (3.0) 1 (3.0) -
0 18 (100)
3 2 3 1 2 33
(9.1) (6.1) (9.1) (3.0) (6.1) (100)
<.001
AIH, autoimmune hepatitis; HAT, hepatic artery thrombosis; HBV, hepatitis B virus; HCV, hepatitis C virus; HVS, hepatic vein stenosis; PSC, primary sclerosing cholangitis; PVT, portal vein thrombosis; ReLT, liver retransplantation. *Biliary problem included biliary cirrhosis and biliary complication due to hepatic artery stenosis. † Others were posttransplant lymphoproliferative disorder, autoimmune hepatitis after liver transplantation, and small-for-size syndrome.
2670
MOON, KIM, SONG ET AL Table 2. Characteristics of Early and Late Liver Retransplantation Early ReLT (N ¼ 18) Categorical Variable
No. (%)
Sex (M:F) Living donor:deceased donor HCC:non-HCC Child A:B:C Status before ReLT Home:Ward:ICU Variceal bleeding history SBP HRS CVVH Mechanical ventilator Bacteremia within 30 days before ReLT Mortality < POD 60
P Value
10:8 3:15 5:13 1:1:16
24:9 5:28 14:29 1:5:27
0:1:17 0 (0) 2 (11.1) 8 (44.4) 9 (50.0) 10 (55.5) 5 (31.3) 7 (38.9)
4:10:19 5 (14.7) 6 (17.6) 9 (26.4) 7 (20.6) 9 (26.5) 9 (29.0) 7 (21.6)
Continuous Variable Recipient age, y Donor age, y Interval, d Length of hospital stay before ReLT, d Length of ICU stay before ReLT, d Albumin, g/dL PT INR Total bilirubin, mg/dL CRP, mg/dL CTP score MELD Operation time, min CIT, min IBL,* mL Before 2005 After 2005 GRWR Length of ICU stay after ReLT, d Length of hospital stay after ReLT, d
Late ReLT (N ¼ 33)
.21 > .99 .37 .53 .008 .15 .70 .36 .03 .046 .88 .18
Median (range) 48 43 12 19 9 3.1 2.36 19.6 1.10 12 36 405 273 2164 4315 1088 2.05 13 35
(15e64) (10e63) (3e47) (4e57) (0e17) (2.3e4.1) (1.17e6.15) (1.1e47.6) (0.08e6.01) (6e15) (11e53) (270e720) (37e518) (266e10702) (903e10702) (266e2476) (0.68e2.98) (2e78) (2e92)
47 42 1266 56 8 3.04 1.95 26.8 3.04 11 30 580 283 3572 5723 3095 2.11 19 52
(15e64) (17e68) (64e5241) (0e251) (0e64) (2.1e4.4) (1.05e3.81) (0.8e53.1) (0.12e13.57) (6e15) (6e44) (281e945) (22e495) (647e19445) (2161e16381) (647e19445) (0.95e3.40) (1e72) (1e167)
.83 .86 <.001 .009 .03 .66 .04 .28 .28 .09 .11 <.001 .85 .02
.35 .84 .58
CIT, cold ischemic time; CRP, C-reactive protein; CTP, Child-Turcotte-Pugh score; CVVH, continuous venovenous hemodialysis; GRWR, graft-recipient weight ratio; HCC, hepatocellular carcinoma; HRS, hepatorenal syndrome; IBL, intraoperative blood loss; ICU, intensive care unit; MELD, Model for End-Stage Liver Disease score; POD, postoperative day; PT INR, international normalized ratio of prothrombin time; ReLT, liver retransplantation; SBP, spontaneous bacterial peritonitis. *IBL ¼ Patient’s estimated blood volume (mL) (preoperative hematocrit in % postoperative hematocrit in %) þ (transfused leukocyte-depleted red blood cell in units 213 70%) þ (transfused Cell Saver blood in mL 55%), IBL was calculated in only adult ReLT.
between the causes of early ReLT and those of late ReLT (P < .001). Characteristics of Early and Late ReLT
The baseline clinical, demographic, and laboratory characteristics of early and late ReLT patients are summarized in Table 2. Most of the early ReLT patients were in the intensive care unit (ICU) before the ReLT procedure, with half of the patients on continuous venovenous hemodilution and a mechanical ventilator at the time of ReLT. Accordingly, ICU stay before ReLT was longer in cases of early ReLT than in cases of late ReLT. In addition, the international normalized ratio of prothrombin time and Child-Turcotte-Pugh and model for end-stage liver disease scores of early ReLT patients were relatively higher than those of late ReLT patients (Table 2). The operation time for late ReLT was longer than that for early ReLT (median,
398 minutes vs 540 minutes; P < .001). Besides, IBL of late ReLT was more than that of early ReLT (median, 1273 mL vs 2523 mL; P ¼ .002). Comparing IBL before and after the year 2006, median IBL was 1170 mL in early ReLT and 1325 mL in late ReLT after 2006, which was significantly decreased from median 2557 mL in early ReLT and 3293 mL in late ReLT before 2006 (early ReLT, P ¼ .02; late ReLT, P ¼ .02). Causes of Death
The mean and median follow-up times were 35.9 and 8.5 months, respectively. Postoperative mortality in early and late ReLT at 60 days was 38.9% and 21.6%, respectively. The survival rate for early ReLT and late ReLT was 51.5% and 50.1% at 1 year, respectively. The overall survival rate for all ReLT was 51.5% and 43.7% at 1 year and 3 years, respectively. In early ReLT cases, 55.5% of deaths were
EARLY VS LATE LIVER RETRANSPLANTATION
2671
Table 3. Causes of Death After Early and Late Liver Retransplantation Early ReLT Cause
Late ReLT No. (%)
Sepsis Bleeding ICH PNF Rejection GVHD HAT HCC recurrence PSC recurrence total
7 1 2 2
(58.3) (8.3) (16.6) (16.6) 12 (100)
P Value
11 4 1 1 2 1 1 1 1 23
(47.8) (17.3) (4.3) (4.3) (8.7) (4.3) (4.3) (4.3) (4.3) (100)
.41
GVHD, graft-vs-host disease; HAT, hepatic artery thrombosis; HCC, hepatocellular carcinoma; ICH, intracranial hemorrhage; PNF, primary nonfunction; PSC, primary sclerosing cholangitis; ReLT, liver retransplantation.
sepsis related, but in late ReLT cases, 42.1% of deaths were sepsis related (Table 3). The second most common cause of death in early ReLT cases was PNF. Both cases of PNF developed after living donor ReLT. Risk Factors for Mortality After Retransplantation
In the present study, risk factors for mortality were evaluated in early ReLT and late ReLT (Table 4). In the multivariate analysis of risk factors for mortality in early ReLT, higher CRP level was a significant factor (P ¼ .045; HR, 1.87). In multivariate analysis of late ReLT, only TB was a significant factor related to mortality in late ReLT (P ¼ .03; HR, 1.05). We calculated cutoff values of CRP for prognosis in early ReLT and of TB in late ReLT by using receiver operating characteristic curve (Fig 1). For the prognosis of early ReLT, 3.0 mg/dL of CRP was a statistically significant value (Fig 2A). However, 24.5 mg/dL of TB was not significant (Fig 2B). Infectious Sites and Organisms
Sepsis was the main cause of mortality in both early and late ReLT. We analyzed pre- and post-retransplant infectious Table 4. Multivariate Analysis of Risk Factor for Mortality in Early and Late Liver Retransplantation Early ReLT (N ¼ 18) Variables
P Value
CRP Total bilirubin IBL HE CTP score GRWR OP time ReLT type Ventilator
.045 .97 .45 .80 -
Hazard Ratio (95% CI)
Late ReLT (N ¼ 33) P Value
1.87 (0.97e3.60)
.37 .03 .41
sites and organisms (Table 5). Thirty-three patients experienced infectious episodes within 30 days before ReLT (64.7%), and post-retransplant infectious complications were observed in 43 patients (84.3%). The sites of pre-retransplant infection were evenly distributed, and intra-abdomen was the most frequent infectious site before retransplant. The most frequent organism of infection was Enterococcus faecium both pre- and post- retransplantation. Interestingly, 61.5% of pre-retransplant infectious sites were correlated with post-retransplant infectious sites. In addition, 82.1% of pre-retransplant isolated bacteria arose again after ReLT.
Hazard Ratio (95% CI)
1.05 (1.03e1.09)
.13 .21 .61 .32 .92
CRP, C-reactive protein; CTP, Child-Turcotte-Pugh; GRWR, graft recipient weight ratio; HE, hepatic encephalopathy; IBL, intraoperative blood loss; OP time, operation time; ReLT, liver retransplantation.
DISCUSSION
In our series, the rate of ReLT was 3.7% (58/1581) at our center; this is lower than the rates reported in Western countries, which are usually in the range of 10% to 20%. The reasons for this low incidence of ReLT at our center may be the specific circumstances and disease background of the area. Referring to Bittermann et al, PNF is the most common cause of ReLT and represents the primary causative factor in over half of cases [5]. However, LDLT has been performed mainly as primary LT in 1581 cases at our center. The incidence of PNF after LDLT is low compared with that of PNF after deceased donor LT. The other reason is that Korea has a low prevalence of HCV and a high prevalence of HBV. This is because East Asia is an HBV endemic area, while HCV is more prevalent in Western countries. Because HCV has a high recurrence rate after LT, Western countries may have a higher rate of ReLT [6]. The causes of early ReLT versus late ReLT were significantly different. PNF and vascular complications were the main causes for early ReLT in our series. In the reports based on deceased donors, approximately 40% of the cases of ReLT are caused by PNF [7]. In this study, PNF accounted for only 20% of indications of early ReLT. Among them, the only one was PNF after LDLT. In contrast, the most common causes of late ReLT were CR and biliary complications. Generally, advances in immunosuppressive therapy and diagnostic means have contributed to a decrease in CR [8]. However, in the present study, a decrease in CR was not shown. Biliary complications were managed with combined endoscopic retrograde cholangiopancreatography and radiological intervention. Four cases of biliary complications were presumed secondary to hepatic artery stenosis or thrombosis. In these cases, the biliary complications were not resolved by interventions, and ReLT was ultimately the only effective therapy [9]. In a comparison of the clinical and laboratory demographics, the pre-retransplant location status (ie, at home, ward, or ICU) of early ReLT candidates was poorer than that of late ReLT. A poorer preoperative general condition may lead to high 30-day postoperative mortality rates in early ReLT. IBL may be used as an indicator of
2672
MOON, KIM, SONG ET AL
Fig 1. (A) The receiver-operating characteristic (ROC) curve of CRP level in early ReLT. (B) The receiver-operating characteristic (ROC) curve of total bilirubin level in late ReLT.
difficulty of operation [10]. The average IBL of late ReLT was more than that of early ReLT. This suggests that late ReLT is technically more difficult than early ReLT. Previous adhesions and fibrosis are a huge obstacle of explant of primary liver graft [11,12]. In addition, after removal of the primary graft, a sufficient length of the vena cava, hepatic artery, and distal bile duct should be preserved. Thus, experienced surgical skills and fine teamwork are crucial for success during ReLT [13,14]. When we compared IBL before 2005 and IBL after 2005 in early and late ReLT, the amount was significantly reduced. This observation suggests the improvement of surgical skill for ReLT at our center. The overall survival rate for ReLT was 60.2%, 45.5%, and 38.6% at 3 months, 1 year, and 3 years, respectively, in our series. These rates are 10% to 20% lower than those in other recent reports [11,15]. In this study, 81% of ReLT recipients had waited for a deceased donor. In such cases, it is difficult to be selected as a recipient because of the small number of deceased donors, and it is often a long wait. A long waiting time can result in a high rate of pretransplant infection, longer length of ICU stay, deterioration of pretransplant condition, and poor survival [16,17].
We have not found any differences with regard to survival rates and perioperative mortality between early and late ReLT. This finding is in accordance with the results of other reports [9,18]. However, it disagrees with other results that suggest that survival rates associated with elective ReLT were superior to those associated with urgent ReLT [15,19]. In the present study, 6 (54.5%) among the 11 deaths of early ReLT and 9 (42.8%) among the 21 deaths after late ReLT were due to sepsis. Except for 1 patient, all sepsisrelated deaths occurred within 9 months postoperatively. The patients who undergo ReLT are typically sicker than those who undergo primary LT. The higher mortality rate of immediate posttransplant is related to immediate postoperative bleeding or infectious problems [12,20,21]. When we investigated pre- and post-retransplant infectious complications, the presence of pre-retransplant infection did not contribute to the occurrence of post-retransplant infectious complications. However, we did identify a phenomenon that pre-retransplant infectious type and organism reappeared in post-retransplant infectious complications, even though the infectious syndrome had been considered as completely treated before retransplant. This observation supports the
Fig 2. (A) Kaplan-Meier Survival Function According to Pre-transplant CRP level of less or more than 3 mg/dL in early retransplantation. (B) Kaplan-Meier Survival Function According to Pre-transplant total bilirubin level of less or more than 24.5 mg/dL in late retransplantation.
EARLY VS LATE LIVER RETRANSPLANTATION
2673
Table 5. Comparison of the Site of Pre- and Posttransplant Infectious Syndrome Pretransplant Infectious Syndrome Within 30 days Before ReLT (N ¼ 34)
Noninfectious syndrome Type of infection Bacteremia Respiratory Intra-abdominal Biliary Urinary Wound
Posttransplant Infectious Complications (N ¼ 47)
20
7
14 9 9 9 2 -
32 6 34 2 10 4
idea that the treated organism may boost again under induction of immunosuppression at ReLT. Generally, reduced immunosuppression is suggested in ReLT, and as such, physicians should be aware of the propensity for infectious complications after ReLT. In addition, it is necessary to tailor immunosuppression according to the patient’s status and to initiate effective antimicrobial prophylaxis for patients while being aware of the preoperative infectious site and organism [1,16,20]. There have been many studies to identify risk factors and make strategies that predict the outcomes of ReLT. Previous studies have reported that factors associated with poor prognosis of ReLT include advanced age, renal failure, high serum bilirubin levels, United Network for Organ Sharing status 1, Model for End-Stage Liver Disease score, interval between primary transplantation and ReLT, emergency transplant, mechanical ventilation before ReLT, and recurrent HCV infection [21e23]. These can be summarized as the quality of the new graft, timing of the procedure, and preoperative patient status [8]. In the present study, the risk factors, which were related with survival, were significantly different between early ReLT and late ReLT. Although it may have a complex effect on various factors, through a multivariate analysis, a high CRP level was related to mortality after early ReLT. This suggests that the control of infection is likely to be important for early ReLT as part of the management of preoperative patient status. In contrast, a high TB level increases the mortality after late ReLT. For late ReLT, the severity of hyperbilirubinemia was a significant factor. Bilirubin level is likely to be helpful in assessing how urgent ReLT is and in determining the timing of late ReLT. The present study has several limitations. First, it has the limitations typical of a retrospective data analysis. Second, the number of ReLT cases was small for accurately analyzing prognostic factors from which to draw conclusions and may be slightly underestimated because we lost followup data of some primary LT patients. Third, the different types of donor and heterogeneous causes of ReLT could have an impact on different prognoses. Fourth, we obtained prognostic CRP values of early ReLT through statistics, but it seems unlikely to be applicable in clinical practice. The
prognostic factors are influenced by the complexity of several factors, and it is difficult to quantify and organize these. Lastly, the most common cause of death was sepsis, and immunosuppressants are related to sepsis. However, because the use of immunosuppressive drugs varies greatly depending on the individual patient’s status, it is not easy to consider the impact on prognosis.
CONCLUSIONS
Despite the limitations described above, our study reveals that patients undergoing early ReLT are likely to be sicker before retransplant and that the mortality after early ReLT may be associated with infectious disease. Late ReLT is likely to be technically more difficult, and the increase in TB level could be a prognostic factor of late ReLT. Therefore, early ReLT is likely to require sufficient treatment of the pretransplant infectious disease, whereas the appropriate decision of ReLT before TB level increases substantially, and delicate surgery is needed in late ReLT. Further research is needed to organize and quantify the risk factors associated with ReLT, with consideration of specific circumstances and disease background of the area. ACKNOWLEDGMENTS The authors have no relevant affiliations or financial involvement with any organization or entity with a financial interest in or financial conflict with the subject matter or materials discussed in this manuscript. We obtained writing assistance for preparation of this manuscript.
REFERENCES [1] Adani GL, Baccarani U, Risaliti A, et al. A single-center experience of late retransplantation of the liver. Transplant Proc 2005;37:2599e600. [2] Uemura T, Randall HB, Sanchez EQ, et al. Liver retransplantation for primary nonfunction: analysis of a 20-year singlecenter experience. Liver Transpl 2007;13:227e33. [3] Chen GH, Fu BS, Cai CJ, et al. A single-center experience of retransplantation for liver transplant recipients with a failing graft. Transplant Proc 2008;40:1485e7. [4] Pfitzmann R, Nussler NC, Hippler-Benscheidt M, et al. Longterm results after liver transplantation. Transpl Int 2008;21:234e46. [5] Bittermann T, Shaked A, Goldberg DS. When living donor liver allografts fail: exploring the outcomes of retransplantation using deceased donors. Am J Transplant 2017;17:1097e102. [6] Bahra M, Neumann UP, Jacob D, et al. Outcome after liver re-transplantation in patients with recurrent chronic hepatitis C. Transpl Int 2007;20:771e8. [7] Biggins SW, Beldecos A, Rabkin JM, et al. Retransplantation for hepatic allograft failure: prognostic modeling and ethical considerations. Liver Transpl 2002;8:313e22. [8] Martí J, Charco R, Ferrer J, et al. Optimization of liver grafts in liver retransplantation: a European single-center experience. Surgery 2008;144:762e9. [9] Chen GH, Fu BS, Yang Y, et al. Early liver retransplantation versus late liver retransplantation: analysis of a single-center experience. Chin Med J (Engl) 2008;121:1992e6. [10] Bang SR, Ahn HJ, Kim GS, et al. Predictors of high intraoperative blood loss derived by simple and objective method
2674 in adult living donor liver transplantation. Transplant Proc 2010;42:4148e50. [11] Yan JQ, Peng CH, Li HW, et al. Preliminary clinical experience in liver retransplantation. Hepatobiliary Pancreat Dis Int 2007;6:152e6. [12] Onaca N, Levy MF, Ueno T, et al. An outcome comparison between primary liver transplantation and retransplantation based on the pretransplant MELD score. Transpl Int 2006;19:282e7. [13] Reese PP, Yeh H, Thomasson AM, et al. Transplant center volume and outcomes after liver retransplantation. Am J Transplant 2009;9:309e17. [14] Feier F, da Fonseca EA, Candido HL, et al. Outcomes and technical aspects of liver retransplantation with living donors in children. Pediatr Transplant 2016;20:813e8. [15] Bellido CB, Martinez JM, Gomez LM, et al. Indications for and survival after liver retransplantation. Transplant Proc 2010;42:637e40. [16] Markmann JF, Markowitz JS, Yersiz H, et al. Long-term survival after retransplantation of the liver. Ann Surg 1997;226:408e20. [17] Rosen HR, Madden JP, Martin P. A model to predict survival following liver retransplantation. Hepatology 1999;29:365e70.
MOON, KIM, SONG ET AL [18] Kamei H, Al-Basheer M, Shum J, et al. Comparison of short- and long-term outcomes after early versus late liver retransplantation: a single-center experience. J Surg Res 2013;185: 877e82. [19] Linhares MM, Azoulay D, Matos D, et al. Liver retransplantation: a model for determining long-term survival. Transplantation 2006;81:1016e21. [20] Dreyzin A, Lunz J, Venkat V, et al. Long-term outcomes and predictors in pediatric liver retransplantation. Pediatr Transplant 2015;19:866e74. [21] Memeo R, Laurenzi A, Pittau G, et al. Repeat liver retransplantation: rationale and outcomes. Clin Transplant 2016;30: 312e9. [22] 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;39:230e8. [23] Torosian JC, Leiby BE, Faenkel JM, et al. Liver retransplantation for recurrence of HCV-related cirrhosis using hepatitis C-positive allografts: a 19-year OPTN analysis. Ann Transplant 2016;21:262e9.
ABSTRACT Background. East Asia is a known endemic area for hepatitis B, and living donor liver transplantation is mainly performed. Liver retransplantation (ReLT) is expected to become an increasing problem because of a shortage of organs. This study aimed to compare early and late ReLT with consideration of specific circumstances and disease background of East Asians. Methods. Between October 1996 and January 2015, 51 patients underwent ReLT; we performed a retrospective analysis of data obtained from medical records of the patients. Clinical characteristics, indication, causes of death, survival rate, and prognostic factors were investigated. Result. The survival rate for early ReLT (n ¼ 18) was 51.5% and that for late ReLT (n ¼ 33) was 50.1% at 1 year postoperatively. Continuous venovenous hemodialysis and the use of mechanical ventilators were more frequent, and pre-retransplant intensive care unit stay and prothrombin time was longer in early ReLT than in late ReLT. Operation time was longer and the amount of intraoperative blood loss was greater in late ReLT than in early ReLT. Multivariate analysis showed that a higher C-reactive protein level increased mortality in early ReLT (P ¼ .045), whereas a higher total bilirubin level increased the risk of death in late ReLT (P ¼ .03). Conclusion. Patients with early ReLT are likely to be sicker pre-retransplantation and require adequate treatment of the pretransplant infectious disease. On the other hand, late ReLT is likely to be technically more difficult and should be decided before the total bilirubin level increases substantially.
T
HE INCREASING number of liver transplants is expected to result in a growing number of graft losses. The most common indications of early graft failure after primary liver transplantation (LT) are primary nonfunction, vascular complications, and uncontrollable acute cellular rejection, whereas the most frequent causes for late liver retransplantation (ReLT) are disease recurrence and biliary tract complications [1,2]. It is known that survival rates following ReLT are 20% to 50% lower than those after the first transplantation.
0041-1345/18 https://doi.org/10.1016/j.transproceed.2018.03.040
2668
ReLT will become an increasing problem in the next decade because of a shortage of organs [3]. In Asia, considering the background that the majority of primary LT
*Address correspondence to Jae-Won Joh, MD, PhD, Department of Surgery, Samsung Medical Center, Sungkyunkwan University School of Medicine, 81 Irwon-Ro, Gangnam-Gu, Seoul, Republic of Korea 06351. Tel: þ82-2-3410-3466, Fax: þ82-2-3410-0040. E-mail: [email protected] ª 2018 Elsevier Inc. All rights reserved. 230 Park Avenue, New York, NY 10169
Transplantation Proceedings, 50, 2668e2674 (2018)
EARLY VS LATE LIVER RETRANSPLANTATION
involves living donor LT (LDLT), it is difficult to find another living donor for ReLT, and the wait for a deceased donor LT is long. The aim of this study was to investigate the different characteristics and prognostic factors between early and late ReLT by comparing indications, characteristics, and outcomes. PATIENTS AND METHODS Study Population From October 1996 through January 2015, a total of 1581 patients underwent LT at Samsung Medical Center, Korea. ReLT was carried out in 58 patients within this period. The rate of ReLT was 3.7%. Among them, we excluded 7 cases of pediatric ReLT but included recipients who were 15 years or older at the time of their ReLT procedure. Two patients who received a third graft were included, but in these cases, combined or multivisceral transplant was performed. We retrospectively evaluated these patients using their medical records. Clinical characteristics, operative variables, postoperative complications, survival rates, and causes of death were investigated.
2669 drugs were methylprednisolone, mycophenolic acid, and FK506 or cyclosporine. The starting time and dose of maintenance immunosuppressive drugs were tailored to each patient’s condition. Most of the recipients who were at high risk of infection received a reduced dose of methylprednisolone.
Statistics We used the Mann-Whitney test for continuous variables and Pearson’s c2 test and Fisher exact test for categorical variables. The quantitative data are presented as median and range. The Cox proportional hazard model was used to identify independent predictive factors. Hazard ratio (HR) and corresponding 95% CIs were calculated. Variables associated with survival in the univariate analyses with a P value <.20 were entered into the multivariate logistic regression analysis, and adjusted risk ratios with 95% CI were calculated. P value <.05 was considered significant, but P value <.10 was regarded as borderline significant. The receiver operating characteristic curve was used to evaluate the predictive cutoff value of C-reactive protein (CRP) and total bilirubin (TB) for prognosis. The Kaplan-Meier method was used for survival analysis of patients. All statistical analyses were performed using SPSS 18.0 for Windows (IBM, Armonk, NY, United States).
Definition Patients were classified according to the interval between ReLT and their previous LT. Patients undergoing ReLT within 60 days were classified under early ReLT, and those undergoing ReLT at 60 days or more after their LT were classified under late ReLT. Primary nonfunction (PNF) is defined as graft failure requiring retransplantation within 1 week of prior transplant without identifiable causes. Delayed nonfunction is defined as graft failure requiring retransplantation after 1 week without any other definable causes [4]. We estimated intraoperative blood loss (IBL) using the following formula: IBL ¼ patient’s estimated blood volume (mL) (preoperative hematocrit in % postoperative hematocrit in %) þ (transfused leukocyte-depleted red blood cell in units 213 70%) þ (transfused Cell Saver [Haemonetics, Braintree, Mass, United States] blood in mL 55%).
Indications Indications for ReLT included chronic rejection (CR), PNF, biliary cirrhosis, hepatic artery thrombosis, hepatic artery stenosis, portal vein thrombosis, hepatitis recurrence, recurrence of primary sclerosing cholangitis, and other causes of hepatic failure after primary LT. We did not perform ReLT in patients with uncontrolled infection, noncompliance, substance abuse, or aggressive recurrent hepatitis C (HCV) or B (HBV) virus infection.
Surgical Technique The modified piggyback technique was used in deceased donor ReLT, and modified right lobe graft was used in living donor ReLT. As a bile duct reconstruction method, duct-to-duct anastomosis was carried out, but Roux-en-Y hepaticojejunostomy, as the second best option, was performed in some cases.
Immunosuppression Basiliximab was usually used as an induction immunosuppressive drug in cases in which the interval between primary transplantation and ReLT was more than 2 years. Maintenance immunosuppressive
RESULTS Indications for ReLT
Table 1 shows the causes of early and late ReLT. The most common cause for early ReLT was PNF (61.1%), and the next was hepatic artery thrombosis (22.2%). In contrast, CR (36.4%) was the most frequent cause of late ReLT, followed by biliary complications (21.2%), primary sclerosing cholangitis, HBV, and HCV. There was a significant difference Table 1. Causes of Early and Late Liver Retransplantation Early ReLT (N ¼ 18) Causes
Chronic Rejection Primary Nonfunction Delayed Nonfunction Biliary Problem* Hepatic Vascular Problem HAT HVS PVT Disease Recurrence HBV HCV PSC AIH Others† Total
Late ReLT (N ¼ 33)
No. (%)
P Value
11 (61.1) 2 (11.1) -
12 (36.4) 7 (21.2)
4 (22.2) 1 (5.6)
1 (3.0) 1 (3.0) -
0 18 (100)
3 2 3 1 2 33
(9.1) (6.1) (9.1) (3.0) (6.1) (100)
<.001
AIH, autoimmune hepatitis; HAT, hepatic artery thrombosis; HBV, hepatitis B virus; HCV, hepatitis C virus; HVS, hepatic vein stenosis; PSC, primary sclerosing cholangitis; PVT, portal vein thrombosis; ReLT, liver retransplantation. *Biliary problem included biliary cirrhosis and biliary complication due to hepatic artery stenosis. † Others were posttransplant lymphoproliferative disorder, autoimmune hepatitis after liver transplantation, and small-for-size syndrome.
2670
MOON, KIM, SONG ET AL Table 2. Characteristics of Early and Late Liver Retransplantation Early ReLT (N ¼ 18) Categorical Variable
No. (%)
Sex (M:F) Living donor:deceased donor HCC:non-HCC Child A:B:C Status before ReLT Home:Ward:ICU Variceal bleeding history SBP HRS CVVH Mechanical ventilator Bacteremia within 30 days before ReLT Mortality < POD 60
P Value
10:8 3:15 5:13 1:1:16
24:9 5:28 14:29 1:5:27
0:1:17 0 (0) 2 (11.1) 8 (44.4) 9 (50.0) 10 (55.5) 5 (31.3) 7 (38.9)
4:10:19 5 (14.7) 6 (17.6) 9 (26.4) 7 (20.6) 9 (26.5) 9 (29.0) 7 (21.6)
Continuous Variable Recipient age, y Donor age, y Interval, d Length of hospital stay before ReLT, d Length of ICU stay before ReLT, d Albumin, g/dL PT INR Total bilirubin, mg/dL CRP, mg/dL CTP score MELD Operation time, min CIT, min IBL,* mL Before 2005 After 2005 GRWR Length of ICU stay after ReLT, d Length of hospital stay after ReLT, d
Late ReLT (N ¼ 33)
.21 > .99 .37 .53 .008 .15 .70 .36 .03 .046 .88 .18
Median (range) 48 43 12 19 9 3.1 2.36 19.6 1.10 12 36 405 273 2164 4315 1088 2.05 13 35
(15e64) (10e63) (3e47) (4e57) (0e17) (2.3e4.1) (1.17e6.15) (1.1e47.6) (0.08e6.01) (6e15) (11e53) (270e720) (37e518) (266e10702) (903e10702) (266e2476) (0.68e2.98) (2e78) (2e92)
47 42 1266 56 8 3.04 1.95 26.8 3.04 11 30 580 283 3572 5723 3095 2.11 19 52
(15e64) (17e68) (64e5241) (0e251) (0e64) (2.1e4.4) (1.05e3.81) (0.8e53.1) (0.12e13.57) (6e15) (6e44) (281e945) (22e495) (647e19445) (2161e16381) (647e19445) (0.95e3.40) (1e72) (1e167)
.83 .86 <.001 .009 .03 .66 .04 .28 .28 .09 .11 <.001 .85 .02
.35 .84 .58
CIT, cold ischemic time; CRP, C-reactive protein; CTP, Child-Turcotte-Pugh score; CVVH, continuous venovenous hemodialysis; GRWR, graft-recipient weight ratio; HCC, hepatocellular carcinoma; HRS, hepatorenal syndrome; IBL, intraoperative blood loss; ICU, intensive care unit; MELD, Model for End-Stage Liver Disease score; POD, postoperative day; PT INR, international normalized ratio of prothrombin time; ReLT, liver retransplantation; SBP, spontaneous bacterial peritonitis. *IBL ¼ Patient’s estimated blood volume (mL) (preoperative hematocrit in % postoperative hematocrit in %) þ (transfused leukocyte-depleted red blood cell in units 213 70%) þ (transfused Cell Saver blood in mL 55%), IBL was calculated in only adult ReLT.
between the causes of early ReLT and those of late ReLT (P < .001). Characteristics of Early and Late ReLT
The baseline clinical, demographic, and laboratory characteristics of early and late ReLT patients are summarized in Table 2. Most of the early ReLT patients were in the intensive care unit (ICU) before the ReLT procedure, with half of the patients on continuous venovenous hemodilution and a mechanical ventilator at the time of ReLT. Accordingly, ICU stay before ReLT was longer in cases of early ReLT than in cases of late ReLT. In addition, the international normalized ratio of prothrombin time and Child-Turcotte-Pugh and model for end-stage liver disease scores of early ReLT patients were relatively higher than those of late ReLT patients (Table 2). The operation time for late ReLT was longer than that for early ReLT (median,
398 minutes vs 540 minutes; P < .001). Besides, IBL of late ReLT was more than that of early ReLT (median, 1273 mL vs 2523 mL; P ¼ .002). Comparing IBL before and after the year 2006, median IBL was 1170 mL in early ReLT and 1325 mL in late ReLT after 2006, which was significantly decreased from median 2557 mL in early ReLT and 3293 mL in late ReLT before 2006 (early ReLT, P ¼ .02; late ReLT, P ¼ .02). Causes of Death
The mean and median follow-up times were 35.9 and 8.5 months, respectively. Postoperative mortality in early and late ReLT at 60 days was 38.9% and 21.6%, respectively. The survival rate for early ReLT and late ReLT was 51.5% and 50.1% at 1 year, respectively. The overall survival rate for all ReLT was 51.5% and 43.7% at 1 year and 3 years, respectively. In early ReLT cases, 55.5% of deaths were
EARLY VS LATE LIVER RETRANSPLANTATION
2671
Table 3. Causes of Death After Early and Late Liver Retransplantation Early ReLT Cause
Late ReLT No. (%)
Sepsis Bleeding ICH PNF Rejection GVHD HAT HCC recurrence PSC recurrence total
7 1 2 2
(58.3) (8.3) (16.6) (16.6) 12 (100)
P Value
11 4 1 1 2 1 1 1 1 23
(47.8) (17.3) (4.3) (4.3) (8.7) (4.3) (4.3) (4.3) (4.3) (100)
.41
GVHD, graft-vs-host disease; HAT, hepatic artery thrombosis; HCC, hepatocellular carcinoma; ICH, intracranial hemorrhage; PNF, primary nonfunction; PSC, primary sclerosing cholangitis; ReLT, liver retransplantation.
sepsis related, but in late ReLT cases, 42.1% of deaths were sepsis related (Table 3). The second most common cause of death in early ReLT cases was PNF. Both cases of PNF developed after living donor ReLT. Risk Factors for Mortality After Retransplantation
In the present study, risk factors for mortality were evaluated in early ReLT and late ReLT (Table 4). In the multivariate analysis of risk factors for mortality in early ReLT, higher CRP level was a significant factor (P ¼ .045; HR, 1.87). In multivariate analysis of late ReLT, only TB was a significant factor related to mortality in late ReLT (P ¼ .03; HR, 1.05). We calculated cutoff values of CRP for prognosis in early ReLT and of TB in late ReLT by using receiver operating characteristic curve (Fig 1). For the prognosis of early ReLT, 3.0 mg/dL of CRP was a statistically significant value (Fig 2A). However, 24.5 mg/dL of TB was not significant (Fig 2B). Infectious Sites and Organisms
Sepsis was the main cause of mortality in both early and late ReLT. We analyzed pre- and post-retransplant infectious Table 4. Multivariate Analysis of Risk Factor for Mortality in Early and Late Liver Retransplantation Early ReLT (N ¼ 18) Variables
P Value
CRP Total bilirubin IBL HE CTP score GRWR OP time ReLT type Ventilator
.045 .97 .45 .80 -
Hazard Ratio (95% CI)
Late ReLT (N ¼ 33) P Value
1.87 (0.97e3.60)
.37 .03 .41
sites and organisms (Table 5). Thirty-three patients experienced infectious episodes within 30 days before ReLT (64.7%), and post-retransplant infectious complications were observed in 43 patients (84.3%). The sites of pre-retransplant infection were evenly distributed, and intra-abdomen was the most frequent infectious site before retransplant. The most frequent organism of infection was Enterococcus faecium both pre- and post- retransplantation. Interestingly, 61.5% of pre-retransplant infectious sites were correlated with post-retransplant infectious sites. In addition, 82.1% of pre-retransplant isolated bacteria arose again after ReLT.
Hazard Ratio (95% CI)
1.05 (1.03e1.09)
.13 .21 .61 .32 .92
CRP, C-reactive protein; CTP, Child-Turcotte-Pugh; GRWR, graft recipient weight ratio; HE, hepatic encephalopathy; IBL, intraoperative blood loss; OP time, operation time; ReLT, liver retransplantation.
DISCUSSION
In our series, the rate of ReLT was 3.7% (58/1581) at our center; this is lower than the rates reported in Western countries, which are usually in the range of 10% to 20%. The reasons for this low incidence of ReLT at our center may be the specific circumstances and disease background of the area. Referring to Bittermann et al, PNF is the most common cause of ReLT and represents the primary causative factor in over half of cases [5]. However, LDLT has been performed mainly as primary LT in 1581 cases at our center. The incidence of PNF after LDLT is low compared with that of PNF after deceased donor LT. The other reason is that Korea has a low prevalence of HCV and a high prevalence of HBV. This is because East Asia is an HBV endemic area, while HCV is more prevalent in Western countries. Because HCV has a high recurrence rate after LT, Western countries may have a higher rate of ReLT [6]. The causes of early ReLT versus late ReLT were significantly different. PNF and vascular complications were the main causes for early ReLT in our series. In the reports based on deceased donors, approximately 40% of the cases of ReLT are caused by PNF [7]. In this study, PNF accounted for only 20% of indications of early ReLT. Among them, the only one was PNF after LDLT. In contrast, the most common causes of late ReLT were CR and biliary complications. Generally, advances in immunosuppressive therapy and diagnostic means have contributed to a decrease in CR [8]. However, in the present study, a decrease in CR was not shown. Biliary complications were managed with combined endoscopic retrograde cholangiopancreatography and radiological intervention. Four cases of biliary complications were presumed secondary to hepatic artery stenosis or thrombosis. In these cases, the biliary complications were not resolved by interventions, and ReLT was ultimately the only effective therapy [9]. In a comparison of the clinical and laboratory demographics, the pre-retransplant location status (ie, at home, ward, or ICU) of early ReLT candidates was poorer than that of late ReLT. A poorer preoperative general condition may lead to high 30-day postoperative mortality rates in early ReLT. IBL may be used as an indicator of
2672
MOON, KIM, SONG ET AL
Fig 1. (A) The receiver-operating characteristic (ROC) curve of CRP level in early ReLT. (B) The receiver-operating characteristic (ROC) curve of total bilirubin level in late ReLT.
difficulty of operation [10]. The average IBL of late ReLT was more than that of early ReLT. This suggests that late ReLT is technically more difficult than early ReLT. Previous adhesions and fibrosis are a huge obstacle of explant of primary liver graft [11,12]. In addition, after removal of the primary graft, a sufficient length of the vena cava, hepatic artery, and distal bile duct should be preserved. Thus, experienced surgical skills and fine teamwork are crucial for success during ReLT [13,14]. When we compared IBL before 2005 and IBL after 2005 in early and late ReLT, the amount was significantly reduced. This observation suggests the improvement of surgical skill for ReLT at our center. The overall survival rate for ReLT was 60.2%, 45.5%, and 38.6% at 3 months, 1 year, and 3 years, respectively, in our series. These rates are 10% to 20% lower than those in other recent reports [11,15]. In this study, 81% of ReLT recipients had waited for a deceased donor. In such cases, it is difficult to be selected as a recipient because of the small number of deceased donors, and it is often a long wait. A long waiting time can result in a high rate of pretransplant infection, longer length of ICU stay, deterioration of pretransplant condition, and poor survival [16,17].
We have not found any differences with regard to survival rates and perioperative mortality between early and late ReLT. This finding is in accordance with the results of other reports [9,18]. However, it disagrees with other results that suggest that survival rates associated with elective ReLT were superior to those associated with urgent ReLT [15,19]. In the present study, 6 (54.5%) among the 11 deaths of early ReLT and 9 (42.8%) among the 21 deaths after late ReLT were due to sepsis. Except for 1 patient, all sepsisrelated deaths occurred within 9 months postoperatively. The patients who undergo ReLT are typically sicker than those who undergo primary LT. The higher mortality rate of immediate posttransplant is related to immediate postoperative bleeding or infectious problems [12,20,21]. When we investigated pre- and post-retransplant infectious complications, the presence of pre-retransplant infection did not contribute to the occurrence of post-retransplant infectious complications. However, we did identify a phenomenon that pre-retransplant infectious type and organism reappeared in post-retransplant infectious complications, even though the infectious syndrome had been considered as completely treated before retransplant. This observation supports the
Fig 2. (A) Kaplan-Meier Survival Function According to Pre-transplant CRP level of less or more than 3 mg/dL in early retransplantation. (B) Kaplan-Meier Survival Function According to Pre-transplant total bilirubin level of less or more than 24.5 mg/dL in late retransplantation.
EARLY VS LATE LIVER RETRANSPLANTATION
2673
Table 5. Comparison of the Site of Pre- and Posttransplant Infectious Syndrome Pretransplant Infectious Syndrome Within 30 days Before ReLT (N ¼ 34)
Noninfectious syndrome Type of infection Bacteremia Respiratory Intra-abdominal Biliary Urinary Wound
Posttransplant Infectious Complications (N ¼ 47)
20
7
14 9 9 9 2 -
32 6 34 2 10 4
idea that the treated organism may boost again under induction of immunosuppression at ReLT. Generally, reduced immunosuppression is suggested in ReLT, and as such, physicians should be aware of the propensity for infectious complications after ReLT. In addition, it is necessary to tailor immunosuppression according to the patient’s status and to initiate effective antimicrobial prophylaxis for patients while being aware of the preoperative infectious site and organism [1,16,20]. There have been many studies to identify risk factors and make strategies that predict the outcomes of ReLT. Previous studies have reported that factors associated with poor prognosis of ReLT include advanced age, renal failure, high serum bilirubin levels, United Network for Organ Sharing status 1, Model for End-Stage Liver Disease score, interval between primary transplantation and ReLT, emergency transplant, mechanical ventilation before ReLT, and recurrent HCV infection [21e23]. These can be summarized as the quality of the new graft, timing of the procedure, and preoperative patient status [8]. In the present study, the risk factors, which were related with survival, were significantly different between early ReLT and late ReLT. Although it may have a complex effect on various factors, through a multivariate analysis, a high CRP level was related to mortality after early ReLT. This suggests that the control of infection is likely to be important for early ReLT as part of the management of preoperative patient status. In contrast, a high TB level increases the mortality after late ReLT. For late ReLT, the severity of hyperbilirubinemia was a significant factor. Bilirubin level is likely to be helpful in assessing how urgent ReLT is and in determining the timing of late ReLT. The present study has several limitations. First, it has the limitations typical of a retrospective data analysis. Second, the number of ReLT cases was small for accurately analyzing prognostic factors from which to draw conclusions and may be slightly underestimated because we lost followup data of some primary LT patients. Third, the different types of donor and heterogeneous causes of ReLT could have an impact on different prognoses. Fourth, we obtained prognostic CRP values of early ReLT through statistics, but it seems unlikely to be applicable in clinical practice. The
prognostic factors are influenced by the complexity of several factors, and it is difficult to quantify and organize these. Lastly, the most common cause of death was sepsis, and immunosuppressants are related to sepsis. However, because the use of immunosuppressive drugs varies greatly depending on the individual patient’s status, it is not easy to consider the impact on prognosis.
CONCLUSIONS
Despite the limitations described above, our study reveals that patients undergoing early ReLT are likely to be sicker before retransplant and that the mortality after early ReLT may be associated with infectious disease. Late ReLT is likely to be technically more difficult, and the increase in TB level could be a prognostic factor of late ReLT. Therefore, early ReLT is likely to require sufficient treatment of the pretransplant infectious disease, whereas the appropriate decision of ReLT before TB level increases substantially, and delicate surgery is needed in late ReLT. Further research is needed to organize and quantify the risk factors associated with ReLT, with consideration of specific circumstances and disease background of the area. ACKNOWLEDGMENTS The authors have no relevant affiliations or financial involvement with any organization or entity with a financial interest in or financial conflict with the subject matter or materials discussed in this manuscript. We obtained writing assistance for preparation of this manuscript.
REFERENCES [1] Adani GL, Baccarani U, Risaliti A, et al. A single-center experience of late retransplantation of the liver. Transplant Proc 2005;37:2599e600. [2] Uemura T, Randall HB, Sanchez EQ, et al. Liver retransplantation for primary nonfunction: analysis of a 20-year singlecenter experience. Liver Transpl 2007;13:227e33. [3] Chen GH, Fu BS, Cai CJ, et al. A single-center experience of retransplantation for liver transplant recipients with a failing graft. Transplant Proc 2008;40:1485e7. [4] Pfitzmann R, Nussler NC, Hippler-Benscheidt M, et al. Longterm results after liver transplantation. Transpl Int 2008;21:234e46. [5] Bittermann T, Shaked A, Goldberg DS. When living donor liver allografts fail: exploring the outcomes of retransplantation using deceased donors. Am J Transplant 2017;17:1097e102. [6] Bahra M, Neumann UP, Jacob D, et al. Outcome after liver re-transplantation in patients with recurrent chronic hepatitis C. Transpl Int 2007;20:771e8. [7] Biggins SW, Beldecos A, Rabkin JM, et al. Retransplantation for hepatic allograft failure: prognostic modeling and ethical considerations. Liver Transpl 2002;8:313e22. [8] Martí J, Charco R, Ferrer J, et al. Optimization of liver grafts in liver retransplantation: a European single-center experience. Surgery 2008;144:762e9. [9] Chen GH, Fu BS, Yang Y, et al. Early liver retransplantation versus late liver retransplantation: analysis of a single-center experience. Chin Med J (Engl) 2008;121:1992e6. [10] Bang SR, Ahn HJ, Kim GS, et al. Predictors of high intraoperative blood loss derived by simple and objective method
2674 in adult living donor liver transplantation. Transplant Proc 2010;42:4148e50. [11] Yan JQ, Peng CH, Li HW, et al. Preliminary clinical experience in liver retransplantation. Hepatobiliary Pancreat Dis Int 2007;6:152e6. [12] Onaca N, Levy MF, Ueno T, et al. An outcome comparison between primary liver transplantation and retransplantation based on the pretransplant MELD score. Transpl Int 2006;19:282e7. [13] Reese PP, Yeh H, Thomasson AM, et al. Transplant center volume and outcomes after liver retransplantation. Am J Transplant 2009;9:309e17. [14] Feier F, da Fonseca EA, Candido HL, et al. Outcomes and technical aspects of liver retransplantation with living donors in children. Pediatr Transplant 2016;20:813e8. [15] Bellido CB, Martinez JM, Gomez LM, et al. Indications for and survival after liver retransplantation. Transplant Proc 2010;42:637e40. [16] Markmann JF, Markowitz JS, Yersiz H, et al. Long-term survival after retransplantation of the liver. Ann Surg 1997;226:408e20. [17] Rosen HR, Madden JP, Martin P. A model to predict survival following liver retransplantation. Hepatology 1999;29:365e70.
MOON, KIM, SONG ET AL [18] Kamei H, Al-Basheer M, Shum J, et al. Comparison of short- and long-term outcomes after early versus late liver retransplantation: a single-center experience. J Surg Res 2013;185: 877e82. [19] Linhares MM, Azoulay D, Matos D, et al. Liver retransplantation: a model for determining long-term survival. Transplantation 2006;81:1016e21. [20] Dreyzin A, Lunz J, Venkat V, et al. Long-term outcomes and predictors in pediatric liver retransplantation. Pediatr Transplant 2015;19:866e74. [21] Memeo R, Laurenzi A, Pittau G, et al. Repeat liver retransplantation: rationale and outcomes. Clin Transplant 2016;30: 312e9. [22] 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;39:230e8. [23] Torosian JC, Leiby BE, Faenkel JM, et al. Liver retransplantation for recurrence of HCV-related cirrhosis using hepatitis C-positive allografts: a 19-year OPTN analysis. Ann Transplant 2016;21:262e9.