Telephone: 44-207-472-6229; FAX: 44-207-472-6226
Immunosuppression and donor age with respect to severity of HCV recurrence after liver transplantation†
Article first published online: 17 MAR 2005
Copyright © 2005 American Association for the Study of Liver Diseases
Volume 11, Issue 4, pages 386–395, April 2005
How to Cite
Samonakis, D. N., Triantos, C. K., Thalheimer, U., Quaglia, A., Leandro, G., Teixeira, R., Papatheodoridis, G. V., Sabin, C. A., Rolando, N., Davies, S., Dhillon, A. P., Griffiths, P., Emery, V., Patch, D. W., Davidson, B. R., Rolles, K. and Burroughs, A. K. (2005), Immunosuppression and donor age with respect to severity of HCV recurrence after liver transplantation. Liver Transpl, 11: 386–395. doi: 10.1002/lt.20344
See Editorial on Page 384
- Issue published online: 21 MAR 2005
- Article first published online: 17 MAR 2005
- Hellenic Gastroenterology Society (2003)
In HCV cirrhotic patients after liver transplantation, survival and recurrence of HCV appears to be worsening in recent years. Donor age has been suggested as a cause. However, it is not clear if early and/or late mortality is affected and whether donor age is a key factor, as opposed to changes in immunosuppression. The aim of this study was to assess impact of donor age and other factors with respect to the severity of HCV recurrence posttransplant. A consecutive series of 193 HCV cirrhotic patients were transplanted with cadaveric donors, median age 41.5 years (13–73) and median follow-up of 38 months (1–155). Donor age and other factors were examined in a univariate/multivariate model for early/late survival, as well as fibrosis (grade 4 or more, Ishak score) with regular biopsies, 370 in total, from 1 year onwards. Results of the study indicated that donor age influenced only short-term (3 months) survival, with no significant effect on survival after 3 months. Known HCC independently adversely affected survival, as did the absence of maintenance azathioprine. Severe fibrosis (stage ≥ 4) in 51 patients was related to neither donor age nor year of transplantation, but it was independently associated with combined biochemical/histological hepatitis flare (OR 2.9, 95% CI 1.76-4.9) whereas maintenance steroids were protective (OR 0.4, 95% CI 0.23-0.83). In conclusion, in this cohort donor age did not influence late mortality in HCV transplanted cirrhotic patients or development of severe fibrosis, which was related to absence of maintenance steroids and a hepatitis flare. Maintenance azathioprine gave survival advantage. (Liver Transpl 2005;11:386–395.)
Hepatitis C virus (HCV) reinfection occurs in virtually all patients transplanted for HCV-related cirrhosis.1–4 The recurrent disease has a more rapid and aggressive course than in the nontransplant setting, adversely affecting patient and graft survival.5, 6 Histological evidence of chronic HCV infection develops in 50% to 90% of patients by 12 months after orthotopic liver transplantation (OLT),7 and cirrhosis occurs in about 20% of patients 4 to 5 years after transplant.8, 9 To date, the use of antivirals has had little impact on the course of posttransplant recurrent HCV because of a combination of poor efficacy and low tolerability.10, 11 Thus, the identification of recipients at greatest risk for more severe HCV recurrence after OLT has become important to plan new therapeutic strategies.
Many studies have evaluated host and viral factors that might predict progressive liver disease. In particular, immunosuppression,12–14 immune response,15–18 viral genotype,19–22 and levels of viraemia before and after transplant23–27 have been shown to be associated with more aggressive posttransplant HCV disease. Aging of donors has been associated with a worse outcome,28–30 and this fact has been used as an explanation for reports of a worse outcome in those transplanted more recently.5, 31 However, this is not confirmed in the most recent report of a large cohort,32 where the deleterious effect of older donor is confined to short-term graft survival.
Despite this fact, it is also the case that more varied immunosuppression has been used over time because of the availability of new drugs or new regimens. The precise relationships between several potential factors including immunosuppression affecting severe HCV recurrence remain uncertain. Methodological differences in study design, e.g., sample size, quantification methods, histopathological sampling, and immunosuppressive protocols, limit the ability to reach a consensus about which factors are associated with worsening fibrosis and which are not. However, evaluation and validation of the viral, host, and donor factors that may determine severity of HCV recurrence are essential to exploring new or different antiviral or immunosuppressive strategies.
Our aim was to evaluate the impact of donor age, immunosuppression, and other factors on short- and long-term survival, and on the evolution of significant fibrosis in a consecutive series of patients transplanted in our center for HCV cirrhosis.
Patients and Methods
From 1989 to June 2003, a total of 193 patients with end-stage HCV-related cirrhosis underwent 208 cadaveric orthotopic liver transplants in the Royal Free Hospital: 15 patients were retransplanted (hepatic artery thrombosis n = 6, primary graft failure n = 6, recurrent HCV cirrhosis n = 2 [1 with mild ductopenia], ductopenic rejection n = 1). We considered those who were retransplanted within a week as a single transplantation (n = 12) or censored at retransplantation (chronic rejection at 5 months, n = 1, recurrent HCV, n = 2). For the evaluation of fibrosis, the 2 patients who underwent a second transplant for recurrent HCV at 40 and 70 months were considered separately. Thus, 195 transplantations, 151 in males and 44 in females, were evaluated.
Hepatocellular carcinoma (HCC) was present in 59 cases; in 10 it was incidental and in 49 it was diagnosed before transplantation. Among the latter group, chemoembolization was used in 8, ethanol injection in 4, chemoembolization with ethanol injection in 2, chemoembolization plus radiofrequency ablation in 2, while 1 patient was treated with chemotherapy. There was no specific selection of donors for patients with known HCC. Concomitant alcoholic liver disease (based on history, usual clinical criteria, and compatible histological features) was present in 36: 3 patients returned to alcohol abuse posttransplant. There was coinfection with hepatitis B virus (HBV) in 11 patients, 2 of these with hepatitis D virus (HDV), (posttransplant HBV DNA remained undetectable—every patient had immunoglobulin B and/or lamivudine). As none of our patients coinfected with HBV became HBV-DNA positive, no active graft inflammation related to HBV was diagnosed posttransplant. No patient in this cohort has received antiviral therapy for recurrent HCV after liver transplantation up to the end of the study period. Follow-up continued to the end of June 2003 or death before this time point. Liver allocation was according to patient status, blood group compatibility, and donor-recipient size. Donor median age was 41.5 years (Table 1) and mean was 41.6 ± 14 (SD). We used previous cutoffs described in the literature, < 40 years (42%), 40–60 (48%), and > 60 (10%), or < 50 (70%) or > 50 (30%)31–33; 47 patients had a donor ≤30 years old. Donor and recipient gender were identical in 120 cases, male recipient with female donor in 55 cases, and female recipient with male donor in 20 cases.
|Number of patients||193|
|Number of transplants||208|
|Recipient age (yrs)|
|Causes of death|
|Pre OLT (n)||47|
|Post OLT (n)||61|
|Donor age (yrs)|
|Cold ischemia time|
|Warm ischemia time|
|Treatment for cellular rejection/patients:|
|Steroids used long-term (n)||128|
|Median (range) duration 4||(1–96)|
|Azathioprine used long-term (n)||118|
|Median (range) duration||12 (1–139)|
|Acute hepatitis HCV (n)||77|
|CMV hepatitis (n)||36|
|Ganciclovir (pre-emptive/therapeutic) (n)||43|
Triple therapy with cyclosporine or tacrolimus, azathioprine, and prednisolone was the initial immunosuppressive regimen in 119 patients (61%), whereas 47 patients (24%) received monotherapy with tacrolimus (n = 38) or cyclosporine (n = 9). In recent years tacrolimus alone or as part of triple therapy has been increasingly used either in the context of the TMC study34 or within a randomized trial of monotherapy vs. triple therapy.35 Only 8 patients received antithymocyte globulin (ATG), and 8 monoclonal antilymphocyte preparation (OKT3) (early on in the series) for resistant cellular rejection (more than 2 episodes). A combination of mycophenolate mofetil (MMF) with tacrolimus was used in 8 (7 with tacrolimus and 1 with cyclosporine); 7 received only azathioprine and prednisolone, while other combinations were used in 15 patients. The immunosuppressive regimen was not selected according to whether the recipient had concomitant HCC or alcoholic etiology. The regimens changed sequentially in time according to clinical trials ongoing in our center.34–36
Cyclosporine was started at a dose of 10 mg/kg/day in two divided doses. Doses were adjusted in the first year to maintain a plasma level of 100–150 ng/mL (by high performance liquid chromatography) or a whole blood level of 200–300 ng/mL (by enzyme immunoassay—Emi 2000 Cyclosporine Specific Assay, Behring Diagnostics Inc., Cupertino, CA). Tacrolimus was started at a dose of 0.1 mg/kg/day in two divided doses and for the first year doses were adjusted to maintain a whole blood level of 6–14 ng/mL (by microparticle enzyme immunoassay—IMX Tacrolimus II, Abbott Labs, IL). Azathioprine was used at 1 mg/kg/day over a median of 12 (range 1–139) months and was discontinued only if side effects, such as neutropenia, developed. In the early years, prednisolone (or equivalent doses of methylprednisolone) was started at a daily dose of 1 mg/kg in 128 patients (66%), gradually reduced (5 mg every week) or decreased more slowly if there was more than one episode of acute cellular rejection. Later on, prednisolone (or equivalent doses of methylprednisolone) was used, starting at 20 mg/day and tapered within 3 months, and completely withdrawn over a median of 4 (1–96) months. We analyzed whether azathioprine and/or prednisolone were started immediately posttransplant or not, and if started, if they then were discontinued. We used a cutoff of 3 months post-OLT to evaluate if early use of these drugs affected long-term recurrence of HCV. If there was significant renal impairment, MMF was given at 1 g twice daily. Intravenous 1 g methylprednisolone daily for 3 days was used to treat moderate or severe rejection in 126 patients (65%). Calcineurin inhibitors doses were reduced progressively if serum creatinine rose above the upper limit of the normal range (120 μmol men, 97 μmol women) with the addition of MMF; calcineurin inhibitors were occasionally stopped and substituted by MMF37 if the creatinine remained above 140 μmol.
The liver biopsies were reviewed by three pathologists (S. Davies moved to another center, substituted by A. Quaglia, so that A. P. Dhillon and A. Quaglia reviewed all biopsies), all having standardized their intra- and inter-observer reporting, and the final score was the result of their consensus. They did not have details of donor clinical data. Protocol liver biopsies immediately after or during the first months post OLT were evaluated for the diagnosis of acute cellular rejection or the presence of hepatitis. From 1 year onwards, 370 specimens were obtained from 136 patients at yearly intervals or when clinically indicated by derangements in the liver biochemistry test results: at 1 year (n = 126), 2 years (n = 66), 3 years (n = 61), 4 years (n = 36), 5 years (n = 41), 6 years (n = 16), and > 6 years (n = 24) after OLT. Liver biopsy samples were formalin-fixed, paraffin-embedded, and stained with haematoxylin and eosin. Gordon and Sweets staining for reticulin and chromotropic aniline blue or MSB staining was used for assessment of fibrosis. For each biopsy sample, there was histological evaluation for the stage of disease (fibrosis; 0–6) and the degree of necroinflammatory activity (grade: after combining the scores for piecemeal necrosis 0–4, confluent necrosis 0–6, focal necrosis 0–4, portal inflammation 0–4) according to Ishak et al.38 A stage score of 4 or more was considered as severe fibrosis and as an endpoint in our study; this score can be much more easily separated from stage 3 as a dividing point, as stage 3 is solely fibrosis confined to the portal tract.
Acute hepatitis C (AHC) was considered to develop in patients with increased alanine aminotransferase levels (> 2 times upper normal limit), together with histological changes consistent with hepatitis without diagnostic features of cellular rejection, duct loss, or any other cause of liver injury. Cytomegalovirus infection (CMV) hepatitis was suspected histologically in the presence of microabscess formation and/or viral inclusion and confirmed with immunohistochemistry, as appropriate. New-onset steatosis, and/or liver cell ballooning, and/or Mallory hyaline, and/or pericellular fibrosis, and/or lobular inflammation including a polymorphonuclear component, were considered as suggestive of alcoholic liver disease relapse, if supported by relevant clinical history.
Acute cellular rejection was graded according to the Royal Free Hospital (RFH) scoring system39 (which includes eosinophils) giving to a maximum a total score of 12. The global grading into mild, moderate, or severe cellular rejection correlates well with the published International Banff Criteria.40 A diagnosis of acute cellular rejection was made with a RFH score of 4 or more, and therapy for moderate and severe rejection was given when the score was 6 or more, unless there was concomitant or suspected bacterial infection or other clinical situations where adding steroids would not be advisable. Chronic ductopenic rejection was defined according to the recommendations of the International Panel.41
Serum samples drawn pretransplant and within the first three months after transplant were tested for determination of HCV genotype by reverse transcription PCR and reverse hybridization assay of the amplified sequence (InnoLipa HCV II, Innogenetics, Zwijnaarde, Belgium).42 The nomenclature system proposed by Simmonds et al. was used.43 Qualitative tests were performed by a reverse transcription PCR assay (Amplicor HCV, Roche Diagnostic Systems Inc., Branchburg, NJ—later Bayer TMA Component system, Berkeley, CA).44
We evaluated two sets of variables with respect to outcome, the first with pretransplant variables and the second including donor variables and others immediately after transplant. We evaluated in a time-dependent fashion the survival time overall, survival time from 3 months after transplantation, and time to reaching fibrosis Ishak stage 4 or more in the liver biopsy (with time to last biopsy for all patients who did not reach this stage).
Univariate data comparison among groups was performed by life tables (Kaplan-Meier estimate) using the log-rank test to assess statistical significance. Multivariable analysis was made by Cox proportional hazard regression model. The proportional assumption was verified graphically by plotting Log [-Log (survivor function)] against time to check parallelism. The maximum partial likelihood-ratio test was used for assessing the significance probability of each variable to be entered or removed. The Wald test was used for hypothesis testing. All analyses were 2-tailed and differences with P value < 0.05 were considered statistically significant. Default parameters of BioMedical Data Processing (BMDP dynamic version 7, Los Angeles, CA) were used for the analysis.
Univariate analysis was performed for overall survival, survival from 3 months, and severe fibrosis as defined by an Ishak stage score 4 or more, using pretransplant and posttransplant variables. Pretransplant variables were recipient age and gender, donor age and gender, gender mismatch, pretransplant HCC diagnosis, pretransplant concomitant alcoholic etiology, diabetes, year of transplant, and genotype of HCV. These variables were chosen because they either have been found to have prognostic significance for survival in the large European Liver Transplant Association Registry45 or they have been associated with HCV recurrence. Posttransplant variables were initial maintenance tacrolimus or cyclosporine, initial maintenance steroids and azathioprine, steroids and azathioprine still administered at 3 months, number of cellular rejection episodes, number of methylprednisolone boluses, use of ATG/OKT3, CMV infection and its treatment, and AHC (biopsy proven and more than 2-fold increase of transaminases).
Patient and Graft Survival
During the follow-up with a median period 38 months (1–155), 68 patients (35.2%) died; causes of death are shown in Table 1. Recurrent HCV cirrhosis caused 17% of the deaths (median time to death 40 months, range 12–114 months), compared to 19% due to sepsis (median 1 month, range 5 days–9 months), and 22% due to other transplant-related causes (median 0.25 month, range 1 day–3 months).
There was a higher mortality in those transplanted for HCC with a higher 3 months mortality, especially during the first years of our transplant program when current criteria were not used.46 Thus, cumulative survival for the non-HCC group was 80% at 1 year, 65% at 5 years, and 60% at 10 years; on the other hand, cumulative survival for the HCC group was 67% for 1 year, 44% at 5 years, and 35% at 10 years (Fig. 1). With regards to survival, we specifically focused on donor age. The mean donor age was significantly younger only in the first years, whereas it was fairly stable thereafter. There was a steady increase over time in 1-year graft survival from 74% to 80% (Table 2).
|Year of OLT||1989–1991||1992–1994||1995–1997||1998–2000||2001–6/2003|
|Number of patients||11||41||60||49||34|
|Mean donor age||29.5||40.1||42.6||42.1||44.6|
|First year graft survival||74%||76%||74%||80%||—|
|First year patient survival||81%||80%||80%||85%||—|
|3 years patient survival||55%||70%||70%||75%||—|
Factors Associated with Overall Survival
In the univariate analysis the following factors were associated with worse survival: presence of HCC (P = 0.0025), older donor age (P = 0.027), no maintenance steroids (P = 0.045), and no maintenance azathioprine (P = 0.004). With respect to donor age, only the comparison of the younger group (<40 years) with the oldest one (>60 years) was statistically significant (P = 0.007). In the multivariable analysis, variables independently associated with worse survival were donor age (OR 1.5, 95% CI 1.02–2.27) and presence of HCC (OR 2.1, 95% CI 1.12–3.87), whereas azathioprine still administered at 3 months (OR 0.3, 95% CI 0.18-0.64) was associated with better survival.
Factors Associated with Survival From 3 Months
In the univariate analysis donor age did not reach statistical significance (P = 0.184) and neither did the year of transplant (P = 0.147), while absence of maintenance azathioprine (P = 0.033), presence of histologically proven acute hepatitis HCV (P = 0.008), and no pretransplant concomitant alcohol etiology (P = 0.033) were associated with worse survival. In the multivariable analysis only initial maintenance azathioprine (OR 0.3, 95% CI 0.16-0.64) and pretransplant concomitant alcoholic etiology (OR 0.2, 95% CI 0.06-0.98) remained associated with better survival from 3 months.
Evaluation of Fibrosis
We had a median of 2 yearly follow-up biopsies per patient (range 1–7). The percentage biopsied among those alive at the yearly time points was as follows: 1 year 84% (126 of 150), 2 years 52% (66 of 129), 3 years 55% (61 of 111), 4 years 39% (36 of 94), 5 years 55% (41 of 74), 6 years 30% (16 of 53), and more than 7 years 60% (24 of 40).
A fibrosis score ≥ 4 was identified in 51 patients (37%) from the 136 individual patients with biopsies at 1 year or more, at a median of 36 months (range 12–84). In the first biopsy 15 (11%) had fibrosis stage 4 or more (11 at 12 months, 2 at 36 months, 1 at 72 months, 1 at 84 months). In the remaining 121, 40 developed stage 4 or more during follow-up. In the 85 who did not develop stage 4 fibrosis, 62 had 2 or more biopsies. These 62 had a median interval from last biopsy to last follow-up of 12 months. In the remaining 23 (who only had one biopsy at 1 year or more) the median interval from last biopsy to last follow-up was 11 months. Thus, there was no bias in the interval between last biopsy and last follow-up in the group who did not reach stage 4 fibrosis without a second biopsy (23 of 136, 17%) compared to those patients who did have a second biopsy or more. In the 51 patients with stage 4 fibrosis or more, 19 developed cirrhosis. Six patients alive at 1 year never had a biopsy as they refused it and are alive at 16, 66, 67, 71, 72, and 108 months without any evidence of progressive liver disease. Another 28 patients whose last biopsy did not show stage 4 fibrosis have a follow-up longer than 13 months from their last biopsy (mean 44 months, range 18–107). All except two are asymptomatic with no clinical or radiological evidence of progressive liver disease, with transaminase values less than twice normal (14 patients live abroad, 2 are anticoagulated, 2 refused biopsy, 8 biopsy still pending).
Fibrosing cholestatic HCV recurrence was diagnosed in only 2 patients—both stage 3 fibrosis—the first died 15 months after transplant from liver failure and the second was regrafted and died 6 months after the second transplant. Four patients had ductopenic rejection: one was regrafted, another with concomitant recurrent HCV (mild ductopenia) was also regrafted, and the 2 others died.
In the univariate analysis, the development of severe fibrosis was associated with the diagnosis of AHC (P = 0.0002), no maintenance steroids (P = 0.021), and no maintenance azathioprine (P = 0.029). As expected, maintenance or no maintenance of azathioprine was closely correlated with maintenance or no maintenance of steroids respectively (R = 0.753). The majority of patients who had taken maintenance azathioprine and/or steroids (even if subsequently discontinued) had a milder histological picture because most did not reach fibrosis ≥ 4 (Table 3). Neither donor age nor the year of transplant was significant for the development of severe fibrosis. Donor age was further examined in three fibrosis groups at last biopsy: mild, stage < 3 (mean 44 months), moderate, stage 3–5 (mean 54 months), and cirrhosis, score 6 (mean 45 months); there was no significant statistical association. In the multivariate model, the independent factors were AHC (OR 2.9, 95% CI 1.76–4.9) associated with worse fibrosis, and maintenance steroids (OR 0.4, 95% CI 0.23-0.83) associated with less fibrosis.
|Immunosuppressive (number of patients)||Fibrosis ≥4 median time to last biopsy (months)||Fibrosis <4 median time to last biopsy (months)|
|Initial azathioprine maintained (34)||23.5% (36)||70.5% (58)|
|Azathioprine discontinued (57)||35% (36)||65% (38)|
|Never azathioprine (46)||44% (36)||52% (24)|
|Initial maintenance steroids (then maintained or eventually discontinued) (97)||33% (36)||65% (44)|
|Never steroids (39)||41% (36)||54% (24)|
We repeated the same analysis for both fibrosis and survival excluding in turn patients who had retransplantation, pretransplant concomitant alcoholic etiology, and nonincidental HCC, as well as for patients with HCC who were outside the Milan criteria.46 The factors identified were the same.
Large cohort studies report worse survival for HCV cirrhotic patients compared to other groups who are transplanted,6, 47 and an accelerated rate of fibrosis has been described in those transplanted more recently.5, 31 An explanation put forward for this fact has been the increased use of older donors in recent years.28–31, 33, 48–52 Others have linked this to immunosuppression8, 14, 28, 31, 33, 53–55 or surgical variables,29, 51, 56 CMV infection,48, 57 HCV viral load and genotype,2, 5, 19, 22, 23, 25, 30, 52 and recipient characteristics.9, 28, 52, 56
The cumulative survival in our cohort was comparable with other cohorts in the literature over the same period. Survival has not decreased in more recent years (Table 2), reflecting better selection, greater technical expertise, and better management, in parallel with the improving results reported by the European Liver Transplant Registry.58 Similar experience was previously reported,30 and a recent study with the largest cohort also showed no decrease in survival over recent years.32 Moreover, data from the National Institute of Diabetes and Digestive and Kidney Diseases (NIDDK) data base showed that cirrhotic patients with HCV infection had a 10-year patient and graft survival similar to those of non-HCV cirrhotic patients following liver transplantation.59 Recurrent HCV cirrhosis caused 17% of the deaths, compared with 19% due to sepsis and 22% due to other transplant-related causes.
Donor age affected survival only during the early critical postoperative period58—not survival after 3 months, despite having in our cohort 30% of donors older than 50 years. This finding is similar to data reported by Russo et al.32 In another study, donor age older than 70 years increased the risk of mortality within the first month by 2.5 times.60 Of three large cohorts23, 33, 56 with a median donor age younger than 40 years, two23, 56 did not find donor age to be significantly associated with survival.
In our cohort, donor age was not a significant factor for the development of severe fibrosis, even when patients retransplanted or those with HCC or concomitant alcoholic etiology were excluded. This fact was also seen by the Paul Brousse group, whose donor cohort had a younger median donor age than did ours.61 The most recent report from the same group in liver transplant recipients who had liver biopsies at 10 years follow-up (without retransplantation) showed that donor age affected both cirrhotic patients with HCV (n = 123) and non-HCV (n = 142) in a similar manner, with increasing donor age giving rise to worsening fibrosis. The median donor age of that study was 30 years compared to our median of 41 years.62 It is interesting that Berenguer et al.,28 who first reported donor age to be associated with a worse outcome in HCV cirrhotic patients after transplantation, did not find donor age to be associated with a worse outcome in those HCV patients with a benign initial course who then developed severe disease.63 However, because our donor age was relatively stable over time and only 10% of the donors were older than 60 and 24% younger than 30 years old, this may have masked the association with donor age found in other series in which the proportion of older donors was higher.
Although pretransplant concomitant alcohol etiology was associated with better survival from 3 months, it was not associated with overall survival. We cannot find a plausible explanation for this fact, except that by chance most early deaths were associated in the HCC group, only 7 of whom had a concomitant alcoholic etiology.
AHC (biochemically and histologically proven) was an important and independent risk factor for survival from 3 months and for the development of severe fibrosis (OR 2.9), similar to Berenguer et al.,28 who found that it was associated with an increased risk of developing cirrhosis (OR 3.1).
Immunosuppression is one of the major factors that may account for the accelerated natural history of HCV infection posttransplant.54 Although less immunosuppression may seem to be the obvious strategy to minimize HCV progression, and some initial studies including our own (with only 13 monotherapy patients)14 suggested this, the situation is more complex. In many centers, steroids are either tapered very quickly or not used at all, because very high HCV RNA levels and increased frequency of acute hepatitis and earlier recurrence following steroid boluses was reported.2, 54 However, recently Brillanti et al.,53 Berenguer et al.,31 and Fasola et al.55 suggested that a more prolonged use of steroids, slowly tapered off over time, may prevent more aggressive forms of recurrent liver disease. The use of MMF has been correlated with worse outcome in recent reports.31, 48 However, only 8 of our patients received it, so no significant conclusions could be drawn.
Our cohort has the special characteristic that many patients received tacrolimus or cyclosporine with similar usage of prednisolone and azathioprine in combination. This fact allowed us to evaluate whether either combination with calcineurin inhibitor resulted in a better survival and/or less fibrosis. We found no difference, similar to a previous report by Zervos et al.64 but different from Ben-Ari et al., who had worse HCV recurrence with cyclosporine but similar survival rate to tacrolimus.65 This is interesting because recently cyclosporine has been found to have an antiviral effect with respect to HCV.66–68
In contrast, use of maintenance steroids (even if subsequently discontinued) was associated with less fibrosis (Table 3), supporting previous reports.31, 53, 55 More interestingly, azathioprine was associated with better survival and also less severe fibrosis. Thus, patients maintained on azathioprine did best and even those who took azathioprine and then discontinued it had a better course than those who had never taken it at all (Table 3). These results are similar to those by Berenguer et al., who found that induction without or with only a short course of azathioprine were related to increased risk for severe disease,28, 31 and also those of Hunt et al., who report that azathioprine-containing regimens reduce histological recurrence of hepatitis C posttransplant.69 These findings may be explained by recent experimental evidence of an antiviral effect of azathioprine on HCV.70 Thus, the rapid withdrawal of steroids (but also of azathioprine) prevalent in recent years may have allowed an early reconstitution of the immune system and its exposure to a large number of HCV-infected liver cells, leading to immune-mediated severe liver damage and thus to more frequent and severe forms of HCV recurrence. Other immunosuppressive treatments, such as methylprednisolone boluses, were not found to be significant in this analysis, similar to some studies31 and contrary to others.2, 51 Neither the occurrence nor the number of rejection episodes were statistically significant in our patient cohort. ATG and OKT3 were only used in 16 patients during the first years of the transplant program, so that the small number of patients did not allow us to study their impact.
Lastly, neither CMV nor its treatment were associated with worsening survival or fibrosis. We had a spread of HCV genotypes in our cohort but did not find a significant association with genotype 1b, contrary to some reports2, 6, 31 but similar to others,5, 22 nor with genotype 4, which was found to be important by others.30, 52
In our cohort of HCV cirrhotic patients who received liver transplants, donor age was only important for short-term survival. Donor age did not influence progression of recurrent HCV infection. The major determinants of the progression of HCV recurrence were the occurrence of AHC in the graft and the absence of maintenance steroids and azathioprine. The group with maintenance low-dose prednisolone and azathioprine had the smallest occurrence of severe fibrosis and survived the longest. These associations need to be tested in a prospective randomized fashion, preferably in association with the same calcineurin inhibitor, to remove the possibility of other factors related to changes in posttransplant care affecting the interpretation of the data on the course of HCV. Currently, we are testing tacrolimus monotherapy vs. tacrolimus with azathioprine and prednisolone in a randomized trial.35
- 16Defining the role of HCV specific T lymphocytes in hepatitis C recurrence after liver transplantation [abstract]. Gastroenterology 2002; 122: A632., , , , , , et al.
- 42Typing of hepatitis C virus isolates and characterization of new subtypes using a line probe assay. J Gen Virol 1993; 74(Pt 6): 1093–1102., , , , , , et al.
- 43Identification of genotypes of hepatitis C virus by sequence comparisons in the core, E1 and NS-5 regions. J Gen Virol 1994; 75(Pt 5): 1053–1061., , , , , , et al.
- 50Predictors of death or retransplantation in patients with hepatitis C following liver transplantation [abstract]. Hepatology 2003; 38(Suppl 1): 534A., , , , , , et al.
- 55A more severe hepatitis C recurrence (HCVR) post liver transplantation (OLT) observed in recent years may be explained by the use of lower dose corticosteroid (CS) maintenance protocols [abstract]. Hepatology 2003; 38(Suppl 1): 226A., , , , , , et al.