Objectives: Liver retransplant is considered the only hope for patients with irreversible graft failure after primary transplant. In most Western centers, retrans­plant is done mainly from deceased donors; so far, only few published studies have reported on outcomes of liver retransplant with living donors. In this study, our aim was to analyze the outcomes of living-donor liver retransplant.

Materials and Methods: Patients who underwent liver retransplant between February 2011 and February 2019 were included in the study. Preoperative, operative, and postoperative data were analyzed. Results from 2 patient groups were compared: liver retransplant with living donors and liver retransplant with deceased donors.

Results: Thirty-two patients underwent liver retrans­plant (21 adult and 11 pediatric patients). The most common indications for liver retransplant were hepatic artery thrombosis (28.5%) and primary graft nonfunction (23.8%) in adults and hepatic artery thrombosis (45.5%) and chronic rejection (36.4%) in pediatric patients. Seventeen retransplant patients (53.1%) required early retransplant (within 1 mo), mainly due to hepatic artery thrombosis (52.9%) and primary graft nonfunction (35.3%). Late retransplant was mainly due to chronic rejection (40%) and recurrence of primary disease (26.7%). Seventeen patients (53.1%) underwent living-donor retransplant, and 5 donors underwent robotic right hepatectomy. Graft and patient survival rates at 1, 3, and 5 years were 81.3% for living-donor and 51.4% for deceased-donor liver retransplant recipients (P = .08). On multivariate analyses, we observed significant differences between both groups in pretransplant Model for End-Stage Liver Disease and Pediatric End-Stage Liver Disease scores (P = .05), preoperative international normalized ratio (P = .012), and cold ischemia time (P = .046).

Conclusions: The use of living donors for liver retransplant, despite its technical demand, was shown to be a safe and feasible option, especially when there is scarcity of deceased donors.

Key words : Deceased-donor liver retransplant, Graft dysfunction, Hepatic artery thrombosis, Living-donor liver retransplant

Introduction

Orthotopic liver transplant (LT) is the ideal treatment modality for end-stage liver disease, with continuous advancements in surgical technique and rising numbers of performed surgeries all over the world. Along with this increase, there has been an increased incidence of graft loss, especially after the use of organs from deceased donors after cardiac death and marginal donors.^1-4 Multiple controversies remain with regard to liver retransplant (re-LT). Although it has a lower outcome and more economic and ethical issues than primary transplant, it is the only life-saving therapeutic option for irreversible graft failure.^5,6

Most studies on re-LT have focussed on deceased-donor liver retransplant (re-DDLT). However, their use decreases the pool of available livers for patients waiting for a first LT, especially in countries with scarce numbers of organs from deceased donors. This fact has raised enthusiasm with regard to living-donor liver retransplant (re-LDLT), which has been considered an alternative to re-DDLT.^7,8

In this study, our aim was to determine the indications and risk factors that may affect survival of re-LT and to investigate outcomes of using re-LDLT versus re-DDLT and outcomes with early versus late re-LT.

Materials and Methods

We conducted a retrospective study on adult and pediatric (< 18 y) patients who underwent re-LDLT or re-DDLT between February 2011 and February 2019 at King Faisal Specialist Hospital and Research Center (KFSHRC), Riyadh, Kingdom of Saudi Arabia. Data were collected prospectively through a dedicated organ transplant database at KFSHRC. Our study was approved by the local institutional review board. Informed consent was obtained from all patients before surgery to use their related prospective data as needed for research work.

Patient characteristics and relevant preoperative, operative, and postoperative clinical and biochemical parameters were collected. These data included age of primary and re-LT donors, indications for primary transplant, causes of graft failure, time between first and second transplant (early was defined as retransplant within 30 days; late was defined as retransplant at more than 30 days after primary transplant), preoperative laboratory tests, and Model of End-Stage Liver Disease (MELD) and Pediatric End-Stage Liver Disease (PELD) scores. Results from 2 pati­ent groups (re-LDLT and re-DDLT) were compared.

There were no specific selection criteria of recipients for re-LDLT versus re-DDLT, with selection dependent on the availability of donors. In re-LDLT, donors were related donors. Preoperative assessments of living donors were per usual steps: laboratory investigations, imaging, and then liver biopsy. Robotic donor hepatectomy was started in our center in May 2018; therefore, re-LDLT procedures after this date were done by robotic surgery.

Surgical technique and posttransplant care
For early (emergency) re-LT (within 1 month), we observe few adhesions and dissections are usually not difficult. For late re-LT, especially with chronic rejection or recurrent viral hepatitis, there are usually heavy adhesions, and these new collaterals require difficult and sharp dissections. Re-LT was done by inferior vena cava (IVC) preservation. Venous anastomosis was done by classic piggyback technique or side-by-side cavoplasty of the recipient IVC at the level of hepatic veins and donor IVC. If the recipient hepatic artery or portal vein was short after hepatectomy, as with difficult dissections in late re-LT, we used vessel grafts from deceased donors for anastomosis. In re-LDLT with right lobe grafts, outflow of anteromedial sector with significant V5 and/or V8 was preserved by polytetrafluoroethylene synthetic grafts or vessels from deceased donors. Biliary anastomosis was done either by duct-to-duct anastomosis or by hepaticojejunostomy, according to the length and quality of the recipient bile duct.

Our postoperative immunosuppression protocol was to start with triple immunosuppression with steroids, mycophenolate mofetil, and calcineurin inhibitors (such as tacrolimus).

In our study, we defined primary graft nonfunction (PNF) as liver function inconsistent with need for retransplant up to day 10 or death, with absence of other causes of liver failure as vascular complications.⁹

All patients after discharge were followed up weekly at the outpatient clinic during the first 3 months and then every 1 to 3 months, depending on the patient’s clinical condition. All patients maintained their follow-up schedules. Median follow-up was 27 months (range, 4-82 mo). The Clavien-Dindo classification was used to grade early postoperative complications (grade I to V). Possible risk factors affecting survival were tabulated and studied.

Statistical analyses
We used SPSS software (version 21.0; SPSS, Inc., Chicago, IL, USA) for statistical analyses. Fisher exact or chi-square tests were used to compare categorical variables. Mann-Whitney and t tests were used for continuous nonparametric and parametric variables, respectively. Logistic regression test was used for multivariate analyses to compare results between re-LDLT and re-DDLT. Kaplan-Meier curves were used to estimate 1-, 3-, and 5-year patient survival rates, and log-rank test was used to compare survival between the groups. P < .05 was considered statistically significant.

Results

From February 2011 to February 2019, there were 1000 patients (630 adult and 370 pediatric patients) who underwent LT. Of these, 32 patients (3.2%) underwent re-LT, which included 21 adult (65.6%) and 11 pediatric patients (34.4%). Seventeen patients had re-LDLT, and 15 patients underwent re-DDLT.

Patient characteristics and preoperative, operative, and postoperative clinical and pathologic data are summarized in Tables 1, 2, and 3. For adults, the main indications for primary transplant were liver cirrhosis secondary to hepatitis B virus (33.3%) and hepatitis C virus (HCV) with hepatocellular carcinoma (23.8%). For pediatric patients, the main indication for primary transplant was biliary atresia (27.2%). For adults, 28.5% required re-LT for hepatic artery thrombosis (HAT) (with 3.2% incidence for all transplant patients) and 23.8% required re-LT for PNF (with 0.9% incidence for all transplant patients). For pediatric patients, the main indications for re-LT were HAT (45.5%) and chronic rejection (36.4%).

Seventeen patients (53.1%) had early re-LT (within 1 month after primary transplant). The main indications for early re-LT were HAT (52.9%) and PNF (35.3%). The main indications for late re-LT were chronic rejection (40%) and recurrence of the primary disease (26.7%). As shown in Table 4, univariate analyses revealed significant differences between early re-LT and late re-LT as regard to indications for re-LT (P = .01), preoperative MELD/PELD scores (P = .02), preoperative alanine aminotransferase level (P = .01), and preoperative international normalized ratio (INR) (P = .02). Our multivariate analyses revealed no significant differences.

Table 5 compares data between the re-LDLT and re-DDLT groups. Thirteen patients received both the primary LT and re-LT from living donors. Five donors underwent robotic right hepatectomy, and 1 donor underwent laparoscopic left lateral hepatectomy. In the re-LDLT group, 11 patients had living-donor right lobe grafts, 5 patients had left lateral grafts, and 1 patient had left lobe graft. Three patients in the re-LDLT group required deceased-donor iliac artery interposition grafts between the aorta and graft artery for arterial anastomosis, 2 patients required deceased-donor iliac vein grafts for portal vein reconstruction, and 1 patient had 2 interposition grafts for hepatic artery and portal vein anastomosis.

Our univariate analyses showed significant differences between groups as regard to indications for re-LT (P = .03), preoperative MELD/PELD scores (P = .01), MELD/PELD scores of > 23 (P = .02), preoperative platelet count (P = .03), preoperative INR (P = .02), donor age of re-LT (P = .03), cold ischemia time (P = .01), and total operative time (P = .05).

Our multivariate analyses showed a significant difference between both groups in pretransplant MELD/PELD score (P = .05) (hazard ratio [HR] = 2.1; 95% confidence interval [95% CI], 0.9-4.8), preo­perative INR (P = .012) (HR = 1.0; 95% CI, 1.0-1.0), and cold ischemia time (P = .046) (HR = 1.01; 95% CI, 1.0-1.03).

Three patients (17.6%) who had re-LDLT had early postoperative mortality, and 7 patients (46.7%) who had re-DDLT had early mortality, with no significant difference between groups (P = .08). The overall perioperative mortality was 31.3%. Graft and patient survival rates in adults at 1, 3, and 5 years were all 66.7%, whereas the survival rates in pediatric patients were 60% (P = .61). In the re-LDLT group, 1-, 3-, and 5-year patient and graft survival rates were 81.3%, whereas, in the re-DDLT group, these rates were 51.4% (Figure 1), with trend to significance shown between group (P = .08).

As shown in Table 6, risk factors that affected survival in adult patients on univariate analysis included high preoperative bilirubin level (P = .03), high preoperative creatinine level (P = .04), older age of donor for primary transplant (P = .02), patient on mechanical ventilation preoperatively (P = .01), and actual graft weight (P = .05). In multivariate analysis, only preoperative mechanical ventilation was a risk factor for survival (P = .03; HR = 9.7; 95% CI, 1.6-64.5).

Discussion

Liver retransplant is the most effective and feasible procedure for patients with primary graft failure, with acceptable short- and long-term outcomes. The incidence of re-LT varies in centers between 5% and 22%.^10-12 The lower incidence of 3.2% at our center may be due to good selection of donors and recipients for LT and low incidence of overall PNF and vascular complications in our patients. Another reason may be differences between centers as regard to selection parameters for re-LT.^5,8

The main indication for re-LT was HAT, which accounted for 11.5% to 36% of re-LT. Other studies have shown PNF to be the most common cause for re-LT, with incidence varying between 10% and 32.3%,^13,14 as seen in our study. Some studies have agreed that the indication of retransplant is not a risk factor for prognosis after re-LT,^14,15 others have shown that timing of re-LT can affect outcomes. Patients who had re-LT in the first month after primary transplant had lower survival rates than those with late re-LT after the first month. This was explained by higher incidence of a compromised general condition and postoperative sepsis during the early posttransplant period. In addition, patients who require early re-LT usually have acute liver failure, which contributes to lower survival rates versus chronic liver disease.^11,15 Although we found that timing of re-LT had no significant risk on patient survival, this may have been affected by the low number of study patients.

Different prognostic factors have been shown to predict outcomes after re-LT, including donor age, recipient age, high preoperative creatinine and bilirubin levels,16 timing of retransplant,^12,17 MELD score,^8,18 mechanical ventilation pretransplant,¹⁶ intraoperative blood transfusion, and the indication for re-LT.^7-15 In our study, we observed preoperative mechanical ventilation to be a risk factor for survival in adult patients.

Marudanayagam and colleagues3 reported 1-, 3-, and 5-year survival rates of 66%, 61%, and 57% after first retransplant in adult patients, which were comparable to our results.

Most studies on re-LT have analyzed outcomes with deceased donors, with few discussing results of re-LT with living donors. In a report of re-LDLT in 14 children seen at Kumamoto University Hospital, Miura and colleagues¹⁹ reported 3 patient deaths after re-LT. The 1- and 3-year survival rates were 85.7%, and the 5-year survival rate was 78.6%, which are comparable to our survival rate in re-LDLT of 84.6%. The study also reported that incidence of relaparotomy was higher and also affected survival,¹⁹ as seen in our study.

In one report of 126 re-LT procedures, Kim and associates reported that 16 patients (12.7%) underwent re-LDLT. The timing of retransplant, MELD score of more than 20, pretransplant portal vein thrombosis, pneumonia, and bacteremia were independent risk factors for survival after re-LDLT. Two patients in their study required interposition graft for portal vein anastomosis, and for hepatic artery anastomosis, the right gastroepiploic vein was used in 2 patients and the great saphenous vein was used in another 2 patients. Kim and associates reported comparable survival rates for re-LDLT versus re-DDLT.⁷

Dai and associates documented re-LDLT in 11 patients and re-DDLT in 27 patients; the main indications for re-LT were biliary and vascular complications. Significantly smaller graft sizes, younger donor age, shorter cold ischemia time, and longer operative time were shown in patients who had re-LDLT versus those who had re-DDLT, as also reported in our study. Dai and associates reported 1-, 3-, and 5-year patient survival rates of 90.9% in the re-LDLT group and 1-, 3-, and 5-year survival rates of 78.3%, 73.7%, and 63.8%, respectively, in the re-DDLT group, with no significant difference between the groups,20 as reported in our study.

In a study that compared 29 patients with re-DDLT and 15 patients with re-LDLT, Chok and colleagues6 also showed that the re-LDLT group had significantly smaller grafts and shorter cold ischemia time. In our study, we observed higher INR, lower serum albumin, higher MELD score, and more operative blood transfusion in re-DDLT, than the other group with re-LDLT.

Chok and colleagues also suggested that vascular grafts should be prepared for arterial or portal anastomosis, as the vessel from the primary transplant may be unfit; they required 3 long saphenous veins for vascular anastomosis in re-LT. Biliary enteric anastomosis was preferred than duct-to-duct biliary anastomosis as the old duct may be short or unhealthy during dissection of the native liver.⁸

As similarly reported in the previous study,⁸ we found no significant differences in early postoperative complications between the re-LDLT and re-DDLT groups (P = .27). Chok and associates had 2 patients in the re-LDLT group with hospital mortality. In our study there was 3 patients with hospital mortality in the group of re-LDLT with hospital mortality). Patient and graft survival rates were nearly similar between the study groups: 1-, 3-, and 5-year patient survival rates were 93%, 89.4%, and 89.4%, respectively, in the re-DDLT group and 93.3%, 93.3%, and 85.6%, respectively, in the re-LDLT group. They explained the favorable outcome to be due to their center having good experience with LDLT.8 In our study, patients in the re-DDLT group had a higher rate of hospital mortality (46.7% vs 17.6% in re-LDLT). This may be due to significantly higher MELD score and higher incidence of PNF in the re-DDLT group. Also we did not find a significant difference in patient and graft survival (P = .08) between the 2 groups, as reported in the previous study.

Our study had several limitations like small sample size -however to our knowledge it is the largest number of re-LDLT cases reported to date- the retrospective design may have resulted in selection bias, also our center has high patient volume and good experience with LDLT, which may have contributed to the favorable outcomes with re-LDLT.

Conclusions

Liver retransplant has acceptable short-term and long-term outcomes in both adult and pediatric patients, especially in high-volume specialized centers. Timing of re-LT as early or late had no significant risk on patient survival. Better outcomes with liver grafts from living donors encourage the use of re-LDLT as a justifiable alternative to re-DDLT. Despite its technical demand of expertise, the use of re-LDLT could be especially useful in countries with shortage of organs from deceased donors where need and supply are unbalanced.

References:

1. Zakaria HM, Sallam AN, Ayoub, II, et al. Predictors of outcome of living donor liver transplantation for hepatocellular carcinoma. Indian J Surg. 2017;79(4):299-307.
   CrossRef - PubMed
   
2. Aziz AM, Yassein T, Taha M, et al. Risk factors impacting mortality after living related liver transplantation for hepatocellular carcinoma: a retrospective cohort study. Egyptian J Surg. 2016;35(3):284.
   CrossRef
   
3. Marudanayagam R, Shanmugam V, Sandhu B, et al. Liver retransplantation in adults: a single-centre, 25-year experience. HPB (Oxford). 2010;12(3):217-224.
   CrossRef - PubMed
   
4. Zakaria HM, Taha M, Gad EH, et al. Living donor liver transplantation for patients with pre-existent portal vein thrombosis. J Liver. 2017;6(214):2167-2189.
   CrossRef
   
5. 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(10):1099-1104.
   CrossRef - PubMed
   
6. Chok KS, Fung JY, Chan AC, et al. Comparable short- and long-term outcomes in living donor and deceased donor liver transplantations for patients with Model for End-stage Liver Disease scores >/=35 in a hepatitis-B endemic area. Ann Surg. 2017;265(1):173-177.
   CrossRef - PubMed
   
7. Kim WJ, Song GW, Lee SG. Actual outcomes and technical feasibility of living donor-related liver retransplantation. HPB (Oxford). 2018;20(2):784-826.
   CrossRef
   
8. Chok KSH, Chan ACY, Fung JYY, Dai WC, Cheung TT, Lo CM. Comparable short- and long-term outcomes in deceased-donor and living-donor liver retransplantation. Hepatol Int. 2017;11(6):517-522.
   CrossRef - PubMed
   
9. Broering DC, Topp S, Schaefer U, et al. Split liver transplantation and risk to the adult recipient: analysis using matched pairs. J Am Coll Surg. 2002;195(5):648-657.
   CrossRef - PubMed
   
10. Pfitzmann R, Benscheidt B, Langrehr JM, Schumacher G, Neuhaus R, Neuhaus P. Trends and experiences in liver retransplantation over 15 years. Liver Transpl. 2007;13(2):248-257.
    CrossRef - PubMed
    
11. 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(5):1485-1487.
    CrossRef - PubMed
    
12. Lang H, Sotiropoulos GC, Beckebaum S, et al. Incidence of liver retransplantation and its effect on patient survival. Transplant Proc. 2008;40(9):3201-3203.
    CrossRef - PubMed
    
13. Jimenez M, Turrion VS, Alvira LG, Lucena JL, Ardaiz J. Indications and results of retransplantation after a series of 406 consecutive liver transplantations. Transplant Proc. 2002;34(1):262-263.
    CrossRef - PubMed
    
14. Deshpande RR, Rela M, Girlanda R, et al. Long-term outcome of liver retransplantation in children. Transplantation. 2002;74(8):1124-1130.
    CrossRef - PubMed
    
15. Facciuto M, Heidt D, Guarrera J, et al. Retransplantation for late liver graft failure: predictors of mortality. Liver Transpl. 2000;6(2):174-179.
    CrossRef - PubMed
    
16. Doyle HR, Morelli F, McMichael J, et al. Hepatic retransplantation--an analysis of risk factors associated with outcome. Transplantation. 1996;61(10):1499-1505.
    CrossRef - PubMed
    
17. Busuttil RW, Farmer DG, Yersiz H, et al. Analysis of long-term outcomes of 3200 liver transplantations over two decades: a single-center experience. Ann Surg. 2005;241(6):905-916; discussion 916-908.
    CrossRef - PubMed
    
18. Edwards E, Harper A. Does MELD work for relisted candidates? Liver Transpl. 2004;10(10 Suppl 2):S10-16.
    CrossRef - PubMed
    
19. Miura K, Sakamoto S, Shimata K, et al. The outcomes of pediatric liver retransplantation from a living donor: a 17-year single-center experience. Surg Today. 2017;47(11):1405-1414.
    CrossRef - PubMed
    
20. Dai WC, Chan SC, Chok KS, et al. Retransplantation using living-donor right-liver grafts. J Hepatobiliary Pancreat Sci. 2014;21(8):579-584.
    CrossRef - PubMed