Hong Kong, China

Survival outcomes of right-lobe living donor liver transplantation for patients with high Model for End-stage Liver Disease scores

Kenneth SH Chok, See Ching Chan, James YY Fung, Tan To Cheung, Albert CY Chan, Sheung Tat Fan and Chung Mau Lo

Hong Kong, China

BACKGROUND:Controversy exists over whether living donor liver transplantation (LDLT) should be offered to patients with high Model for End-stage Liver Disease (MELD) scores. This study tried to determine whether a high MELD score would result in inferior outcomes of right-lobe LDLT.

METHODS:Among 411 consecutive patients who received right-lobe LDLT at our center, 143 were included in this study. The patients were divided into two groups according to their MELD scores: a high-score group (MELD score ≥25;n=75) and a low-score group (MELD score <25;n=68). Their demographic data and perioperative conditions were compared. Univariable and multivariable analyses were performed to identify risk factors affecting patient survival.

RESULTS:In the high-score group, more patients required preoperative intensive care unit admission (49.3% vs 2.9%;P<0.001), mechanical ventilation (21.3% vs 0%;P<0.001), or hemodialysis (13.3% vs 0%;P=0.005); the waiting time before LDLT was shorter (4 vs 66 days;P<0.001); more blood was transfused during operation (7 vs 2 units;P<0.001); patients stayed longer in the intensive care unit (6 vs 3 days;P<0.001) and hospital (21 vs 15 days;P=0.015) after transplantation;more patients developed early postoperative complications (69.3% vs 50.0%;P=0.018); and values of postoperative peak blood parameters were higher. However, the two groups had comparable hospital mortality. Graft survival and patient overall survival at one year (94.7% vs 95.6%; 95.9% vs 96.9%), three years (91.9% vs 92.6%; 93.2% vs 95.3%), and fi ve years (90.2% vs 90.2%; 93.2% vs 95.3%) were also similar between the groups.

CONCLUSIONS:Although the high-score group had signif icantly more early postoperative complications, the two groups had comparable hospital mortality and similar satisfactory rates of graft survival and patient overall survival. Therefore, a high MELD score should not be a contraindication to rightlobe LDLT if donor risk and recipient benef i t are taken into full account.

(Hepatobiliary Pancreat Dis Int 2013;12:256-262)

Model for End-stage Liver Disease;living donor liver transplantation; survival; right-lobe

Introduction

Fig. 1.Median waiting time for deceased donor liver transplantation in Hong Kong, China.

Liver transplantation is the best treatment option for patients with end-stage liver diseases. In places, Asian regions in particular, where organs donated by the deceased are scarce, living donor liver transplantation (LDLT) plays an important role in meeting the huge demand. However, there were reports suggesting that outcomes of LDLT may be inferior, especially in patients with high Model for End-stage Liver Disease (MELD) scores. Thus, the New York State Committee on Quality Improvement in Living Liver Donation has recommended that LDLT should not be provided for patients with MELD scores >25.[1]On the other hand, a relatively recent report concluded that post-LDLT graft function and survival rates in patients with high MELD scores were excellent and that a high MELD score should not be a contraindication to LDLT.[2]At our center, all patients requiring liver transplantation are encouraged to consider LDLT irrespective of their MELD scores because of the very long time it takes awaiting a deceased-donor graft (Fig. 1). The aim of this study is to determine whether MELD score per se has any impact on the short- and long-term outcomes of right-lobe LDLT. The learning curve effect and cases of hepatocellular carcinoma or fulminant hepatic failure were excluded from analysis so as to avoid any inf l uence these factors might have on graft survival and patient survival.[3-5]

Methods

This is a retrospective study covering the period from May 1996 to March 2011. Prospectively collected data of 411 consecutive patients who received right-lobe LDLT at our center were reviewed. Before transplantation, written informed consent to transplantation and to use of their data for research was obtained from all donors and recipients (if mentally fi t) and their family members. Use of such data for research is approved by the institution and the local authority.

To avoid the learning curve effect, the fi rst 50 cases were excluded from the study. Patients with hepatocellular carcinoma, fulminant hepatic failure or acute fl are of hepatitis B were also excluded to avoid heterogeneity of disease. The remaining patients were divided into a low-score group (MELD score <25) and a high-score group (MELD score ≥25). The dividing MELD score was set at 25 according to the literatures.[1,2]The study's primary endpoint was hospital mortality and the secondary endpoints were graft survival and patient overall survival at one year, three years and fi ve years after transplantation.

We adopted the standard perioperative management protocol and surgical techniques for donor and recipient operations.[3,6]In brief, donor work-up was in a stepwise fashion. Apart from physical well-being and blood group compatibility, psychological readiness as assessed by experienced clinical psychologists was also a criterion for donation. Liver volumetry was done by 2.5-mm-cut computed tomography and liver volumes were calculated along the proposed transection line according to the principle of Heymsf i eld.[7]Donor right hepatectomy[8,9]and right-liver graft implantation have been described elsewhere.[8]Duct-to-duct anastomosis was the preferred method for bile duct reconstruction if the condition was favorable. All grafts included the middle hepatic vein except in three cases.

Recipients were followed up at our outpatient clinic weekly within the fi rst three months after transplantation, and every one to three months afterwards depending on their clinical conditions. In general, as told before informed consent was acquired, donor mortality is 0.5% with a morbidity of 20%, recipient mortality is 5%-10% with a morbidity of 30%, and recipient fi ve-year survival is around 90%.[9]

Graft cold ischemic time was the time between cross-clamping of donor vessels and portal reperfusion in the recipient. Recipient warm ischemic time was the time between the fi rst venous anastomosis and portal reperfusion in the recipient. Hospital mortality was def i ned as death occurring during the hospital stay during which the primary operation took place. Early postoperative complication was def i ned as complication occurring within three months of transplantation. Complications of Clavien-Dindo[10]Grade 3 or above were regarded as major complications.

Immunosuppression and prophylaxis regimens

Before January 2001, immunosuppression regimen featured the administration of tacrolimus and corticosteroids, including intraoperative intravenous injection of 1 g of hydrocortisone and a postoperative tapering dose of methylprednisolone starting at 200 mg/d to 40 mg/d by day 7 after transplantation. Steroid treatment was maintained with oral prednisolone at 20 mg/d and was gradually tapered off six months after transplantation. Oral tacrolimus was given at a dosage of 0.15 mg/kg body weight/d within 12 hours of transplantation. The dose was titrated to achieve a trough level of 10-15 ng/mL. Since January 2001, 20 mg of basiliximab (Novartis, Basel, Switzerland) was given intravenously within 6 hours of graft reperfusionand on postoperative day 4. Steroid injection was given intraoperatively with 1 g of hydrocortisone and on postoperative day 1 with 500 mg of hydrocortisone. Postoperative immunosuppression was maintained with oral tacrolimus given within 12 hours of transplantation at a dosage of 0.15 mg/kg body weight/d, and the dose was titrated to achieve a trough level of 5-10 ng/mL. Mycophenolate mofetil (CellCept?, Roche, Basel, Switzerland) at a dosage of 1.5 g/d was started within 48 hours of transplantation and was gradually tapered off at three months. Maintenance steroid was not given routinely. All patients also took 200 mg of oral fl uconazole and 480 mg of co-trimoxazole daily and 400 mg of acyclovir three times a day after transplantation for three months as prophylaxis. For patients with renal dysfunction (e.g. arbitrarily twice of the level before tacrolimus was started), the trough tacrolimus level was kept at 3-5 ng/mL and the addition of prednisolone at 20 mg/d was necessary. Pentamidine inhalation (300 mg) was given monthly for three months to patients with renal impairment or glucose-6-phosphate dehydrogenase def i ciency because the use of acyclovir or co-trimoxazole on them was prohibited.

Statistical analysis

All statistical analyses were performed with the SPSS software, version 12.0. Comparison of categorical variables was performed with Pearson's Chi-squared test or Fisher's exact test where appropriate. Nonparametric continuous variables were compared by the Mann-WhitneyUtest and presented as medians with range. The cut-off values for univariable analysis were medians in continuous variables. Univariable analysis was performed using the logistic regression model and signif i cant variables were put into the Cox proportional hazards model to determine the risk factors associated with overall survival. Survival was estimated with the Kaplan-Meier estimator and survival comparison was done with the log-rank test.Pvalues of less than 0.05 were regarded as statistically signif i cant and allPvalues were two-tailed.

Results

After exclusion as stated before, 143 patients were grouped into the low-score group (MELD score <25,n=68) and the high-score group (MELD score ≥25,n=75). The two groups had comparable donor characteristics (Table 1). In the high-score group, there were more hepatitis B carriers (62 vs 44;P=0.014); more patients required preoperative intensive care unit (ICU) admission (37 vs 2;P<0.001), preoperative mechanical ventilation (16 vs 0;P<0.001), and preoperative hemodialysis (10 vs 0;P=0.005); the waiting time before LDLT was shorter (4 vs 66 days;P<0.001) (Table 1); more blood was transfused during operation (7 vs 2 units;P<0.001); the graft cold ischemic time was longer (112 vs 101.5 min;P=0.023) (Table 2); patients stayed longer in the ICU (6 vs 3 days;P<0.001) and hospital (21 vs 15 days;P=0.015) after operation and had higher postoperative peakserum total bilirubin level (Fig. 2A); and more patients developed early major postoperative complications (69.3% vs 50.0%;P=0.018) (Tables 3, 4). However, the two groups had comparable hospital mortality and postoperative levels of serum aspartate transaminase, alanine transaminase and creatinine (Fig. 2B-D). Graft survival and patient overall survival at one year (94.7% vs 95.6%; 95.9% vs 96.9%), three years (91.9% vs 92.6%; 93.2% vs 95.3%) and fi ve years (90.2% vs 90.2%; 93.2% vs 95.3%) were also similar between the groups (Table 3; Figs. 3, 4).

Table 1.Patient demographics and donor characteristics

Table 2.Intraoperative parameters

Fig. 2.Peak serum levels of (A) total bilirubin, (B) aspartate transaminase and (C) alanine transaminase in the fi rst week after transplantation and (D) half-yearly median serum levels of creatinine in 5 years after transplantation.

Table 3.Postoperative outcomes

Table 4.Types of early major postoperative complications

Fig. 3.Comparison of graft survival.

Fig. 4.Comparison of patient overall survival. Four patients with MELD score <25 and 2 patients with MELD score ≥25 had retransplantation and were excluded from patient survival analysis.

Analysis was done in the Cox proportional hazards model to investigate the impact of MELD score as a continuous and categorical variable (high vs low scores) on overall survival but no signif i cant differences were found. MELD score as a continuous variable was associated with a hazard ratio of 1.057 (conf i dence interval 0.994-1.125;P=0.075), while MELD score >25 as a categorical variable was associated with a hazard ratio of 1.196 (conf i dence interval 0.320-4.465;P=0.790). Univariable analysis showed that MELD score >35, preoperative hemodialysis and poor renal function were signif i cant factors affecting overall survival (Table 5). These factors were then put into the Cox proportional hazard model (a stepwise forward conditional method), and it was found that preoperative hemodialysis was the only factor associated with overall survival (hazard ratio: 9.345; 95% CI: 2.228-39.186;P=0.002).

Table 5.Univariable analysis of risk factors affecting overall survival

Discussion

Since the introduction of MELD score in the allocation of liver grafts in 2002,[11]the number of LDLTs performed each year has increased, especially in places with severe shortage of deceased-donor livers.[12]However, the New York State Committee on Quality Improvement in Living Liver Donation recommended that LDLT for patients with MELD scores >25 should be prohibited because of anticipated inferior outcomes in these patients in addition to potential risks to donors.[1]The recommendation was based on the speculative theory that a high MELD score brings about bad outcomes and has only gained support from small studies in theWest,[13,14]while some high-volume transplant centers have produced opposing evidence.[2,3]Some studies showed that MELD score was predictive of preoperative mortality but not postoperative outcomes,[15-18]whereas some showed that high MELD score was associated with prolonged postoperative ICU stay and increased hospital costs but not with postoperative mortality.[17,19]A study that had the largest LDLT cohort concluded that LDLT can provide excellent graft function and survival rates in patients with high MELD scores,[2]and its results are echoed by our present study.

Our present study found that patients with high MELD scores, compared with their low-score counterparts, did have a more diff i cult recovery as manifested by longer ICU and hospital stay, higher peak values of blood parameters, and a higher incidence of early complication after transplantation. Nonetheless, high MELD score did not bring about any signif i cant worsening of hospital mortality, patient survival, or graft survival. When patient overall survival was considered, MELD score ≥35 was signif i cant on univariable analysis (Table 5) but insignif i cant on multivariable analysis. As the sample size was small and the factor of hemodialysis was also put into the analyses, the signif i cance of MELD score still needs more investigation. Moreover, in view of the relatively high rates of mortality (0.5%) and morbidity (20%) of right-liver donors at our center,[20]whether patients with a MELD score ≥35 should be routinely offered right-liver LDLT needs more consideration.

In the present series, the median ratio of graft weight to recipient estimated standard liver volume was around 50% (Table 2), and almost every graft included the middle hepatic vein. So the results derived can be quite different from those obtained by using smaller grafts or using grafts without the middle hepatic vein.

Patients in the high-score group had more preoperative ICU admissions and more cases of organ failure requiring organ support. Moreover, 75% (56/75) of the patients in this group had acute-on-chronic liver failure. Their prognosis would be dismal if they did not receive liver transplantation in a few days' time. A previous study from our center also reported the same fi nding.[21]Possible factors contributing to the acute decompensation included hepatorenal syndrome, hepatic coma, gastrointestinal tract bleeding, and infection.

Center experience is a factor in the success of a rightlobe LDLT, as with other operations. Studies have shown that right-lobe LDLT is extremely technically demanding and the learning curve effect must be taken into account when its success rate is considered.[3-5]It was found that a center having performed more than 20 cases of LDLT would see a signif i cantly lowered risk of graft failure in later LDLTs.[4]A previous study by our center also found that signif i cantly lowered hospital mortality came after a 50-case experience of right-lobe LDLT.[3]In the present study, more blood was transfused during operation and the graft cold ischemic time was longer in the high-score group, possibly ref l ecting more diff i cult operations, but the operation time did not differ much between the two groups.

The study also recorded that more high-score patients developed early postoperative complications, and this may be a reason for the longer postoperative ICU and hospital stay of the group. With longer ICU and hospital stay, probably more hospital expenses were incurred, but this study did not delve into the issue of resource utilization. The relationship between MELD score and healthcare cost has been studied and it was reported that high MELD score cost more.[22,23]We believe high-score patients have the right to best medical treatments as other patients do and should not be discriminated simply because their high MELD scores would mean consuming more resources. However, more research is desirable to work out how to improve the cost-effectiveness of treating high-score patients.

The present study has several limitations. First, it is a single-center study and the fi ndings may not be reproducible at low-volume centers. Second, the sample size was relatively small, which increases the risk of type-II error. Third, the study reports experience of the East, which is different from the West. In Western regions, patients of liver diseases consist of more hepatitis C carriers and there are more cases of alcoholic cirrhosis. Furthermore, the study worked on right-lobe LDLT only; it is not clear whether the outcomes would be the same with deceased donor liver transplantation.

In conclusion, the study has challenged the recommendation from the New York State Committee on Quality Improvement in Living Liver Donation. In right-lobe LDLT, patients with high MELD scores have satisfactory survival outcomes, similar to those in patients with low MELD scores, and a high MELD score should not be a contraindication to right-lobe LDLT. In the face of the grave shortage of liver grafts from the deceased, LDLT is a justif i able and wise alternative to deceased donor liver transplantation as long as center experience is considered and donor volunteerism and informed consent are guaranteed. Of course, further research including prospective study and subgroup analysis is needed for a more profound understanding of the issue.

Contributors:CKSH designed the study, collected and analyzed the data, and drafted the manuscript. All other authors contributed to the critical revision of the manuscript. CSC, FST and LCM alsoapproved the manuscript. CSC is the guarantor.

Funding:None.

Ethical approval:Not needed.

Competing interest:No benef i ts in any form have been received or will be received from a commercial party related directly or indirectly to the subject of this article.

1 New York State Department of Health NYSCoQliLLD. Accessed in 2002. Available from: http://www.health.state.ny.us.

2 Selzner M, Kashf i A, Cattral MS, Selzner N, McGilvray ID, Greig PD, et al. Live donor liver transplantation in high MELD score recipients. Ann Surg 2010;251:153-157.

3 Chan SC, Lo CM, Ng KK, Fan ST. Alleviating the burden of small-for-size graft in right liver living donor liver transplantation through accumulation of experience. Am J Transplant 2010;10:859-867.

4 Olthoff KM, Merion RM, Ghobrial RM, Abecassis MM, Fair JH, Fisher RA, et al. Outcomes of 385 adult-to-adult living donor liver transplant recipients: a report from the A2ALL Consortium. Ann Surg 2005;242:314-325.

5 Olthoff KM, Abecassis MM, Emond JC, Kam I, Merion RM, Gillespie BW, et al. Outcomes of adult living donor liver transplantation: comparison of the Adult-to-adult Living Donor Liver Transplantation Cohort Study and the national experience. Liver Transpl 2011;17:789-797.

6 Fan ST, Lo CM, Liu CL. Technical ref i nement in adult-toadult living donor liver transplantation using right lobe graft. Ann Surg 2000;231:126-131.

7 Heymsf i eld SB, Fulenwider T, Nordlinger B, Barlow R, Sones P, Kutner M. Accurate measurement of liver, kidney, and spleen volume and mass by computerized axial tomography. Ann Intern Med 1979;90:185-187.

8 Chan SC, Lo CM, Liu CL, Wong Y, Fan ST, Wong J. Tailoring donor hepatectomy per segment 4 venous drainage in right lobe live donor liver transplantation. Liver Transpl 2004;10: 755-762.

9 Fan ST. Living donor liver transplantation, 2nd ed. Singapore: World Scientif i c Publishing Company;2011.

10 Dindo D, Demartines N, Clavien PA. Classif i cation of surgical complications: a new proposal with evaluation in a cohort of 6336 patients and results of a survey. Ann Surg 2004;240:205-213.

11 Freeman RB Jr, Wiesner RH, Harper A, McDiarmid SV, Lake J, Edwards E, et al. The new liver allocation system: moving toward evidence-based transplantation policy. Liver Transpl 2002;8:851-858.

12 Shimada M, Fujii M, Morine Y, Imura S, Ikemoto T, Ishibashi H. Living-donor liver transplantation: present status and future perspective. J Med Invest 2005;52:22-32.

13 Marubashi S, Dono K, Asaoka T, Hama N, Gotoh K, Miyamoto A, et al. Risk factors for graft dysfunction after adult-to-adult living donor liver transplantation. Transplant Proc 2006;38:1407-1410.

14 Wang ZX, Yan LN, Wang WT, Xu MQ, Yang JY. Impact of pretransplant MELD score on posttransplant outcome in orthotopic liver transplantation for patients with acute-onchronic hepatitis B liver failure. Transplant Proc 2007;39:1501-1504.

15 Hayashi PH, Forman L, Steinberg T, Bak T, Wachs M, Kugelmas M, et al. Model for End-Stage Liver Disease score does not predict patient or graft survival in living donor liver transplant recipients. Liver Transpl 2003;9:737-740.

16 Li C, Wen T, Yan L, Li B, Wang W, Xu M, et al. Does model for end-stage liver disease score predict the short-term outcome of living donor liver transplantation? Transplant Proc 2010;42:3620-3623.

17 Oberkof l er CE, Dutkowski P, Stocker R, Schuepbach RA, Stover JF, Clavien PA, et al. Model of end stage liver disease (MELD) score greater than 23 predicts length of stay in the ICU but not mortality in liver transplant recipients. Crit Care 2010;14:R117.

18 Bernardi M, Gitto S, Biselli M. The MELD score in patients awaiting liver transplant: strengths and weaknesses. J Hepatol 2011;54:1297-1306.

19 Foxton MR, Al-Freah MA, Portal AJ, Sizer E, Bernal W, Auzinger G, et al. Increased model for end-stage liver disease score at the time of liver transplant results in prolonged hospitalization and overall intensive care unit costs. Liver Transpl 2010;16:668-677.

20 Chan SC, Fan ST, Lo CM, Liu CL, Wong J. Toward current standards of donor right hepatectomy for adult-to-adult live donor liver transplantation through the experience of 200 cases. Ann Surg 2007;245:110-117.

21 Chan AC, Fan ST, Lo CM, Liu CL, Chan SC, Ng KK, et al. Liver transplantation for acute-on-chronic liver failure. Hepatol Int 2009;3:571-581.

22 Axelrod DA, Koffron AJ, Baker T, Al-Saden P, Dixler I, McNatt G, et al. The economic impact of MELD on liver transplant centers. Am J Transplant 2005;5:2297-2301.

23 Washburn WK, Pollock BH, Nichols L, Speeg KV, Halff G. Impact of recipient MELD score on resource utilization. Am J Transplant 2006;6:2449-2454.

Received December 18, 2012

Accepted after revision April 8, 2013

AuthorAff i liations:Department of Surgery (Chok KSH, Chan SC, Cheung TT, Chan ACY, Fan ST and Lo CM); State Key Laboratory for Liver Research (Chan SC, Fung JYY, Fan ST and Lo CM); and Department of Medicine (Fung JYY), The University of Hong Kong, 102 Pokfulam Road, Hong Kong, China

Prof. See Ching Chan, Department of Surgery, The University of Hong Kong, 102 Pokfulam Road, Hong Kong, China (Tel: 852-22553025; Fax: 852-28175475; Email: seechingchan@gmail.com)

Part of the material in the paper was presented at the Vanguard Debate Session of the 17th Annual International Congress of the International Liver Transplantation Society held in Valencia, Spain, on June 22-25, 2011. ? 2013, Hepatobiliary Pancreat Dis Int. All rights reserved.

10.1016/S1499-3872(13)60042-9