Retransplantation for graft failure in chronic hepatitis C infection: A good use of a scarce resource?

Abstract

AIM: To investigate the outcome of patients with hepatitis C virus (HCV) infection undergoing liver retransplantation.

METHODS: Using the UK National Registry, patients undergoing liver transplantation for HCV-related liver disease were identified. Data on patient and graft characteristics, as well as transplant and graft survival were collected to determine the outcome of HCV patients undergoing retransplantation and in order to identify factors associated with transplant survival.

RESULTS: Between March 1994 and December 2007, 944 adult patients were transplanted for HCV-related liver disease. At the end of follow-up, 617 of these patients were alive. In total, 194 (21%) patients had first graft failure and of these, 80 underwent liver retransplantation, including 34 patients where the first graft failed due to recurrent disease. For those transplanted for HCV-related disease, the 5-year graft survival in those retransplanted for recurrent HCV was 45% [95% confidence interval (CI): 24%-64%] compared with 80% (95% CI: 62%-90%) for those retransplanted for other indications (P = 0.01, log-rank test); the 5-year transplant survival after retransplantation was 43% (95% CI: 23%-62%) and 46% (95% CI: 31%-60%), respectively (P = 0.8, log-rank test). In univariate analysis of all patients retransplanted, no factor analyzed was significantly associated with transplant survival.

CONCLUSION: Outcomes for retransplantation in patients with HCV infection approach agreed criteria for minimum transplant benefit. These data support selective liver retransplantation in patients with HCV infection.

Key Words: Hepatitis C; Liver; Recurrence; Retransplantation; Outcome

INTRODUCTION

End-stage liver disease due to chronic hepatitis C virus (HCV) is the most common indication for liver transplantation in Europe and the United States[1,2]. Recurrent viral infection of the allograft is near universal. Ninety percent of all patients transplanted with HCV have evidence of HCV in the allograft within 6 mo[3] and at least 20% of patients transplanted for HCV infection will develop cirrhosis of their allograft[1,4,5]. Allograft cirrhosis is associated with recurrence of disease before 6 mo post-transplantation[6], severe necroinflammation and confluent necrosis at the time of first histological diagnosis, pre-transplant viral load and episodes of treated rejection[7]. Cirrhosis is associated with high rates of graft loss to recurrent disease[8]. Recurrent HCV infection in the setting of immunosuppression rarely follows the more aggressive course of recurrent cholestatic hepatitis which may cause early graft failure.

Current data show that the prognosis for patients with cirrhosis and graft failure from recurrent HCV in their grafts is poor, with only 41% surviving 1 year after the development of decompensation[5]. Antiviral therapy is largely ineffective in this setting[9] and retransplantation is the only therapeutic option. Consideration of retransplantation in the setting of graft failure due to recurrent HCV infection is therefore an increasing clinical problem and results in increasing pressure on an already limited resource. Retransplantation, however, is associated with a worse clinical outcome than for first grafts regardless of the indication[10]. There have been several studies addressing the issue of retransplantation in patients with HCV and results from these studies have yielded discordant results: these are summarized in Table 1[10-19]. Initial studies reported an increased mortality following retransplantation in patients with HCV infection, frequently in the first months after retransplantation[11,13,18]. These studies were all retrospective and largely database-dependent. Indeed the United Network for Organ Sharing (UNOS) data has been re-evaluated on 2 occasions with the predictors of poor outcomes changing with time: the most recent analysis showing no difference between outcomes for patients with and without HCV infection undergoing retransplantation[14]. That said, there have been particular factors [including model for end-stage liver disease (MELD) score and donor age] and subgroups (i.e. those retransplanted between 90 d and 1 year after first liver transplant) identified which increased the risk of retransplant failure. There may be differences amongst those transplanted for HCV infection whose grafts fail due to recurrent disease and those whose grafts fail for other reasons although this, too, is controversial[10,20]. Overall, there is no consensus on the role of retransplantation for recurrent HCV infection.

Table 1 Comparison of peer-reviewed studies on retransplantation outcome in patients with hepatitis C virus infection.

Author	Design	Number with HCV infection	Outcome	Predictors
Database studies
Rosen et al[11]	UNOS database	HCV total 357	54% 5-yr survival	HCV associated with mortality (HR 1.3)
Watt et al[12]	UNOS database	HCV total 899	45% 5-yr survival	MELD > 25
Petellier et al[13]	SRTR database	HCV total 464	NA	HCV associated with mortality (HR 1.3)
Ghabril et al[14]	UNOS database	HCV total 1034	rOLT 90 d-1 yr: 44% 3-yr survival	MELD > 25
			rOLT > 1 yr: 59% 3-yr survival	Donor > 60
				Donor black
				WIT > 75 min
Rowe et al, current study	UK Transplant database	34 recurrent HCV	Recurrent HCV rOLT 5-yr graft survival 45%	None
		46 HCV	HCV rOLT other reasons HCV 5-yr graft survival 80%
Center-led studies
Roayaie et al[15]	Retrospective single center	HCV total 42	52% alive at 6 mo	Donor < 60
				PT < 16 s
Neff et al[16]	Retrospective single center	HCV total 22	50% 1-yr survival	Normal physical activity
Carmiel-Haggai et al[17]	Retrospective single center	47 recurrent HCV	37% 1-yr mortality	Donor > 60
		62 HCV		Cirrhosis FCH
McCashland et al[18]	Prospective multicenter	HCV total 43	3-yr survival 49%	MELD not predictive
Bahra et al[19]	Retrospective single center	18 recurrent HCV	Recurrent HCV rOLT 5-yr survival 59%	MELD > 25
		11 HCV	HCV rOLT other reasons 5-yr survival 84%	Bilirubin > 15 mg/dL
Ghabril et al[10]	Retrospective single center	25 recurrent HCV	63% 3-yr survival	Donor age > 60
		48 HCV		Recipient > 60

UNOS: United Network for Organ Sharing; SRTR: Scientific registry of transplant recipients; HCV: Hepatitis C virus; rOLT: Retransplantation; MELD: Model for end-stage liver disease; PT: Prothrombin time; FCH: Fibrosing cholestatic hepatitis; WIT: Warm ischemia time.

In recent years in the United Kingdom (UK) the number of transplants performed has fallen whilst the number of patients waiting has risen by over 90%[20]. Because of the disparity between the numbers of donor organs and potential recipients, rationing must occur and selection criteria for transplant and retransplant are in place in transplant programs around the world. In the UK a meeting of health care professionals, patient representatives, ethicists and the public agreed an expected transplant survival probability of at least 50% at 5 years following liver transplantation as a benchmark of minimum expected survival following liver transplantation[21].

Liver retransplantation in patients with HCV infection is controversial and has been inconsistently associated with transplant survival rates of less than 50% at 5 years. The aim of this current study was therefore to determine overall outcomes after retransplantation in patients with HCV infection in the UK. In addition, we sought to determine whether failure of the first graft due to recurrent HCV-related disease had a significant effect on survival after retransplantation, and whether we could identify factors which predict a poor outcome.

MATERIALS AND METHODS

Data were obtained from the National Transplant database. This database is maintained by the Organ Donation and Transplantation Directorate of NHS Blood and Transplant on behalf of transplant services in the UK and the Republic of Ireland. All 944 patients aged 17 years and older undergoing first elective liver transplants for HCV-related liver disease in the UK between 1 March 1994 and 31 December 2007 were included. Patients who were recorded as HCV-positive at time of registration and/or time of transplantation but who did not have an indication for transplantation recorded as HCV were excluded from the analyses. Also excluded were patients receiving grafts from non-heart-beating donors and patients with intestinal failure-related liver disease. Details of the indication were categorized into 4 groups: HCV alone, HCV and alcohol-related liver disease (ALD), HCV and hepatocellular carcinoma (HCC), and HCV and ALD and HCC. In addition, patients undergoing liver retransplantation for reasons other than HCV-related liver disease were identified to act as a comparison group.

Patient, graft and transplant survival were obtained for the 4 HCV groups using the Kaplan Meier estimation method. Survival was compared using the log-rank test. Patient survival was measured from the time of first liver transplant to patient death. Graft survival was defined as time from liver transplant (or retransplant) to graft failure (patient deaths with a functioning graft were censored). Transplant survival was defined as time from transplant (or retransplant) to patient death or graft failure whichever was the earlier (deaths with a functioning graft were included as events).

Graft and transplant survival for the 80 retransplanted patients were obtained using the Kaplan-Meier method. Univariate Cox regression models were used to identify factors associated with liver transplant survival after retransplantation, where the first transplant failed due to disease recurrence. Factors considered were donor age, blood group, gender, height and weight, and recipient age, blood group, gender, height and weight, HCV group for first graft, MELD score[22] at time of second graft, year of first graft, year of second graft, cold ischemia time, hospitalization status, time from first graft to second graft and donor organ appearance. Missing values for donor height, donor weight, recipient height and weight, MELD score and cold ischemia time were replaced with the median value, and missing values for hospitalization status and donor organ appearance were replaced with the modal value.

RESULTS

Patient characteristics

Of the 944 patients transplanted for HCV-related liver disease during the study period, the majority (n = 521, 55%) were transplanted for HCV alone and for HCV with HCC (n = 261, 28%). In total, 698 (74%) patients were alive at the end of the study period. The 5-year patient survival estimate for patients transplanted for HCV alone was 70% [95% confidence interval (CI): 65%-74%] and for patients transplanted for HCV with HCC the 5-year patient survival was 63% (95% CI: 55%-69%).

Of those transplanted, 617 (65%) first grafts were still functioning at the end of the study period, 133 (14%) patients died with a functioning graft and 194 (21%) grafts failed. In total, the reason for graft failure was indicated as recurrent HCV in 77 patients, and other causes, most commonly acute vascular occlusion and chronic or ductopenic rejection, in 117 patients. Where the first graft failed, 89 patients (45.9% of those with graft failure) were relisted and 80 (41.2%) were retransplanted. Of the 80 patients retransplanted, the reason for failure of the first graft was recorded as recurrent disease in 26 cases, and other reasons in 54 cases. Further analysis of the other reasons for graft failure, as recorded in free text by transplant centers, indicated that graft failure of 8 additional patients was considered to be due to disease recurrence giving a total of 34 patients retransplanted for recurrent disease and 46 patients retransplanted for other reasons. There was no difference in the proportion of patients retransplanted where the first graft failed due to recurrent disease or due to other reasons (χ² test, P = 0.76).

Outcome after liver retransplantation

The clinical information of those patients who underwent retransplantation after first liver transplant for HCV-related liver disease is given in Table 2. The groups were similar in demographic data and severity of liver disease at the time of retransplantation. However, those patients retransplanted for recurrent HCV were less likely to be hospitalized at the time of retransplantation. Those retransplanted for reasons other than recurrent HCV also had a shorter interval between first and second liver transplant and the majority of those patients were retransplanted for acute vascular occlusion (n = 17, 37.0%) and primary graft non-function (n = 11, 23.9%). These 28 patients were transplanted as an emergency and these indications contributed largely to the shortened interval between first and second graft and the increased likelihood of hospitalization seen in this group. The remainder of the patients retransplanted for reasons other than recurrent disease were retransplanted for late vascular complications (6 patients), biliary complications (3 patients), chronic or ductopenic rejection (3 patients), or other reasons (4 patients).

Table 2 Characteristics of those patients undergoing retransplatation (mean ± SD) n (%).

Characteristic	First graft failed due to		P-value
	Recurrent disease	Other reasons
n	34	46
Recipient age (yr)	50.0 ± 8.16	53.1 ± 8.14	0.090
Recipient blood group
O	11 (32.3)	18 (39.1)	0.800
A	16 (47.1)	18 (39.1)
B	4 (11.8)	7 (15.2)
AB	3 (8.8)	3 (6.5)
Recipient gender, male	27 (79.4)	33 (71.7)	0.400
Recipient height (cm)	173.3 ± 9.72	169.6 ± 9.94	0.090
Recipient weight (kg)	81.0 ± 13.93	77.6 ± 17.48	0.400
HCV group for first graft
HCV alone	25 (73.5)	24 (52.2)
HCV + HCC	3 (8.8)	16 (34.8)	0.030
HCV + ALD	6 (17.6)	6 (13.0)
MELD score	19.9 ± 5.08	22.1 ± 10.79	0.280
Year of first graft
1994-2000	23 (67.6)	22 (47.8)	0.080
2001-2007	11 (32.4)	24 (52.2)
Year of second graft
1994-1998	7 (20.6)	14 (30.4)	0.060
1999-2003	13 (38.2)	24 (52.2)
2004-2007	14 (41.2)	8 (17.4)
Cold ischemia time (h)	11.9 ± 2.61	9.7 ± 3.29	0.020
Hospitalization status, Outpatient	19 (55.9)	4 (8.7)	< 0.001
Time from first graft to second graft (yr)	2.8 ± 2.17	0.38 ± 1.03	< 0.001
Donor organ appearance, normal	28 (82.4)	40 (87.0)	0.600

SD: Standard deviation; HCC: Hepatocellular carcinoma; ALD: Alcoholic liver disease.

Overall graft survival for patients first transplanted for HCV and subsequently retransplanted for recurrent HCV was lower than for those retransplanted for other reasons, as shown in Figure 1A. Five-year graft survival was estimated as 45% (95% CI: 24%-64%) for the recurrent disease group, compared with 80% (95% CI: 62%-90%) for those retransplanted for other reasons (P = 0.01). However, there was no difference in overall transplant survival between these 2 groups, as shown in Figure 1B. Five-year transplant survival in those with recurrent disease was 43% (95% CI: 23%-62%) and 46% (95% CI: 31%-60%) in those retransplanted for other reasons (P = 0.8). The difference between graft and transplant survival between these 2 groups was related to deaths occurring with a functioning graft: one death in the patients retransplanted for recurrent disease and 17 deaths in the patients retransplanted for other reasons. The cause of death in these patients was largely multiple organ failure and infective complications.

Figure 1 Graft survival is reduced in patients undergoing liver retransplantation for recurrent hepatitis C virus related disease. Kaplan-Meier survival curves were generated for A: Graft survival after second liver transplant in hepatitis C virus (HCV)-infected patients; B: Transplant survival after second liver transplant in HCV-infected patients. Graft or transplant survival is indicated for those retransplanted for recurrent HCV-related liver disease (dotted line) and for those retransplanted for reasons other than HCV-related liver disease.

Following retransplantation for recurrent HCV, the total number of grafts lost was 16 (47%). The most frequent causes of graft loss were acute vascular occlusion (n = 4), recurrent disease (n = 3) and chronic rejection (n = 2).

Factors predicting outcome after liver retransplantation

Several factors including pre-transplant disease severity and donor factors have previously been identified as predictors of poor outcome after liver retransplantation in patients with HCV infection. These factors, together with other factors considered to impact on retransplant outcome including the timing of liver retransplantation after first liver transplant, were considered in univariate Cox regression analyses in those patients retransplanted for HCV-related liver disease. None of the factors considered in these analyses, including donor age and MELD score at the time of retransplantation, was found to be associated with transplant survival after retransplantation in patients with HCV, as shown in Table 3.

Table 3 Predictive factors for transplant survival following liver retransplantation for patients with hepatitis C virus infection.

Variable	Category	n	Hazard ratio (95% CI)	P value
Recurrent disease	No	46	1
	Yes	34	0.92 (0.49-1.73)	0.8
Donor age (yr)	Linear	80	1.00 (0.98-1.03)	0.7
Donor blood group	O	43	1
	A, B, AB	37	0.91 (0.49-1.70)	0.8
Donor gender	Male	37	1
	Female	43	1.23 (0.66-2.29)	0.5
Donor height (cm)	Linear	80	0.98 (0.95-1.01)	0.3
Donor weight	Linear	80	0.98 (0.96-1.01)	0.2
Recipient age (yr)	Linear	80	0.99 (0.96-1.03)	0.6
Recipient blood group	O	29	1
	A, B, AB	51	1.00 (0.52-1.91)	0.99
Recipient gender	Male	60	1
	Female	20	1.38 (0.70-2.72)	0.4
Recipient height (cm)	Linear	80	0.98 (0.95-1.01)	0.1
Recipient weight (kg)	Linear	80	0.99 (0.98-1.02)	0.6
HCV group for first graft	HCV alone	49	1
	HCV + HCC	19	2.15 (1.07-4.30)	0.03
	HCV + ALD	12	1.13 (0.46-2.81)	0.8
MELD score	Linear	80	1.00 (0.96-1.05)	0.9
Year of first graft	Linear	80	1.04 (0.89-1.21)	0.6
Year of second graft	Linear	80	0.96 (0.87-1.06)	0.47
Cold ischemia time (h)	Linear	80	0.99 (0.92-1.06)	0.74
Hospitalization status	Outpatient	24	1
	Inpatient	56	1.36 (0.66-2.78)	0.4
Time from first to second graft (yr)	Linear	80	1.04 (0.89-1.21)	0.6
Donor organ appearance	Normal	68	1
	Abnormal	12	1.04 (0.44-2.47)	0.9

HCV: Hepatitis C virus; HCC: Hepatocellular carcinoma; MELD: Model for end-stage liver disease; ALD: Alcoholic liver disease.

DISCUSSION

For those patients developing graft failure, and in particular from recurrent HCV-related liver disease, the outlook is poor and there is no alternative to consideration of liver retransplantation. The outcome of patients retransplanted with chronic HCV infection is relatively poor, and, in the context of organ shortage, use of scarce resources for this indication is, therefore, controversial. As the number of patients being transplanted for HCV is increasing, so is the number of patients with graft failure due to recurrent HCV infection. Previous studies are conflicting, with some suggesting such a poor outcome in these patients that use of a liver graft cannot be justified[16,23]. More recent reports however have stated that patients should not be denied retransplantation merely based on a diagnosis of chronic HCV infection[10,14,18]. Furthermore, repeated analyses of the UNOS database initially demonstrated that HCV infection had a negative impact on survival after liver retransplantation but later studies did not support this finding[14] suggesting that, with increased experience, transplant centers may have improved outcomes for patients with graft failure and HCV infection. It should be noted that there is inconsistent reporting of confidence intervals of survival rates in these studies and these findings have contributed to the controversy regarding the utility of retransplantation in HCV infection.

The shortage of donor organs has focused attention on the optimal use of liver allografts. A consensus meeting of UK surgeons and physicians, health care professionals, ethicists, patients and the public agreed a minimum transplant survival of 50% at 5 years[21]. This is in keeping with the statement from the International Liver Transplantation Society Expert Panel who suggested a 1-year survival rate of at least 60% should be anticipated before listing a patient for retransplantation with recurrent HCV-related disease[24]. This guidance allows for the best use of a scarce resource, and also gives much needed transparency to transplant programs. Importantly, this will however disadvantage some patients who would otherwise have gained some benefit from transplantation.

In this study, representing a highly selected patient population, the 5-year transplant survival for patients with chronic HCV infection who were retransplanted was less than 50%, regardless of the cause of failure of the first graft, although the confidence intervals reported here were wide due to the relatively small numbers of patients included. However, in patients retransplanted for recurrent HCV, 5-year graft survival was 45%. In contrast, the 5-year graft survival was 80% in those transplanted for reasons other than recurrence. This discrepancy between graft and transplant survival in patients retransplanted for reasons other than recurrent HCV is likely to be explained in part by the short interval between first and second liver transplant. Many of these patients are, by definition, not fully recovered from the first transplant, and the proportion of patients dying from multiple organ failure and infective complications likely reflect this. This important observation indicates that, of the patients retransplanted for reasons other than HCV disease recurrence, those who survive the early weeks after transplantation have a relatively good long-term outlook.

In total, 5% of all patients retransplanted with HCV infection lost their second graft to recurrent disease, including 3 of 34 (8.8%) patients retransplanted for recurrent HCV-related disease. This is a clinically relevant proportion of these patients and this calls into question findings of previous studies addressing this. Since transplant failure following retransplantation is greater than after first transplant regardless of indication, factors that further impact on transplant survival after retransplantation are thus important. The timing of recurrent disease in the first liver transplant has been suggested to impact on the prevalence of recurrent disease in the second liver transplant in this group[25], although this too has been questioned[17]. In this study a large proportion of patients were retransplanted between 1 and 3 years after the first liver transplant. The small numbers of patients transplanted at times other than this has meant that we have been unable to reliably investigate this matter.

Several studies have investigated factors associated with graft loss in HCV infection following retransplantation. These have determined negative predictors of mortality in the recipient, including MELD score greater than 25, elevated serum creatinine and bilirubin, and prolonged prothrombin time[7,14]. In addition, donor age greater than 60 years has consistently been associated with negative outcomes in individuals retransplanted with HCV infection. Analyses of these factors in this study did not show any association with transplant survival. This is likely to be related, at least in part, to the relatively low numbers of patients retransplanted with HCV infection, the heterogeneity of this group as well as the skill of the surgeon in matching donor and recipient. At present, in the UK, there are national guidelines for selection of patients but the allocation of donor livers is done on a center-specific basis with the surgeon determining the most appropriate recipient for a liver donated from a heartbeating or non-heartbeating donor. Similarly, there are center-specific protocols regarding the management of recurrent HCV infection after liver transplantation. In this study, the details of these protocols and whether patients being considered for liver retransplantation had received antiviral therapy were not recorded. However, widespread use of combination antiviral therapy has only recently entered clinical practice and although we saw no clear effect of year of first or second liver transplant on transplant outcome, it is likely too soon to assess what effect this strategy may have.

In this national study of transplant outcome after retransplantation for HCV, we have shown that transplant survival rates fall below those normally accepted for survival after transplant, regardless of whether retransplantation was related to HCV recurrence or not. Furthermore, we have not been able to demonstrate any factors associated with increased, or indeed decreased, transplant failure. Although numbers are relatively small, this series is one of the largest conducted.

Whilst there is much work focused on the early identification of individuals with rapid development of fibrosis after transplantation[26], and antiviral treatment outcomes may be improving[27], over the coming years many of those transplanted with chronic HCV infection will develop graft failure and the question of retransplantation will be raised. Previously published studies are conflicting and this study adds to the evidence base supporting selective retransplantation in the setting of chronic HCV infection. Clear and careful consideration must therefore be given when assessing these individuals for retransplantation.

COMMENTS

Background

Hepatitis C virus (HCV) infection is common and is a leading indication for liver transplantation. Following transplantation, however, HCV infects the transplanted liver and disease recurrence is rapid and may lead to transplant failure. This inevitably leads to questions regarding the utility of repeat liver transplantation. It is not clear whether liver retransplantation is appropriate in patients with either recurrent HCV or with HCV who require liver retransplantation for other reasons.

Research frontiers

There is significant controversy as to whether patients with HCV infection should undergo liver retransplantation, and if so which patients would be most appropriate.

Innovations and breakthroughs

The authors demonstrate that those patients with recurrent HCV infection undergoing liver retransplantation fare worse, particularly with regard to graft survival, than those undergoing liver retransplantation for other reasons. However, at 5 years after liver retransplantation, transplant survival in both groups is at the limit of acceptability (50% survival at 5 years after transplant).

Applications

This study adds weight to the evidence supporting selective liver retransplantation in patients with HCV infection.

Terminology

Graft survival is defined as time from liver transplant (or retransplant) to graft failure. In this case a patient death with a functioning graft is not considered as an event. Transplant survival however is defined as time from transplant (or retransplant) to patient death or graft failure whichever was the earlier.

Peer review

This is a well-designed and structured manuscript addressing a controversial issue and is worth publication.