SEARCH

SEARCH BY CITATION

Keywords:

  • Co-infection;
  • fibrosing cholestatic hepatitis;
  • hepatitis C virus;
  • human immunodeficiency virus;
  • liver transplantation

Abstract

  1. Top of page
  2. Abstract
  3. Introduction
  4. Patients and Methods
  5. Statistical Analysis
  6. Results
  7. Discussion
  8. Acknowledgments
  9. Disclosure
  10. References

We characterized fibrosing cholestatic hepatitis (FCH) in a large cohort of HIV/HCV co-infected patients. Between 1999 and 2008, 59 HIV infected patients were transplanted for end-stage liver disease due to HCV. Eleven patients (19%) developed FCH within a mean period of 7 months [2–27] after liver transplantation (LT). At Week 1 post-LT, the mean HCV viral load was higher in the FCH group: 6.13 log10 IU/mL ± 1.30 versus 4.9 log10 IU/mL ± 1.78 in the non-FCH group, p = 0.05. At the onset of acute hepatitis after LT, activity was moderate to severe in 8/11 HIV+/HCV+ patients with FCH (73%) versus 13/28 (46%) HIV+/HCV+ non-FCH (p = 0.007) patients. A complete virological response to anti-HCV therapy was observed in 2/11 (18%) patients. Survival differed significantly between the two groups (at 3 years, 67% in non-FCH patients versus 15% in FCH patients, p = 0.004). An early diagnosis of FCH may be suggested by the presence of marked disease activity when acute hepatitis is diagnosed and when a high viral load is present. The initiation of anti-HCV therapy should be considered at this point.


Abbreviations: 
3TC

lamivudine

ABC

abacavir

ALT

alanine aminotransferase

APV

amprenavir

AST

aspartate aminotransferase

AZT

zidovudine

d4T

stavudine

EFV

efavirenz

CyA

cyclosporine A

ENF

enfurtivide

fAPV

fosamprenavir

FCH

fibrosing cholestatic hepatitis

γ-GT

gamma glutamyl transferase

HAART

highly active antiretroviral therapy

HBV

hepatitis B virus

HCC

hepatocellular carcinoma

HCV

hepatitis C virus

HDV

hepatitis Delta virus

HIV

human immunodeficiency virus

IDV

indinavir

LPV/r

lopinavir

LT

liver transplantation

MELD

model for end-stage liver disease

MMF

mycophenolate mofetil

NFV

nelfinavir

NNRTI

nonnucleoside reverse transcriptase inhibitors

NVP

nevirapine

PEG-IFN

pegylated interferon

PI

protease inhibitors

RBV

ribavirin

RTV

ritonavir

SQV

saquinavir

Tac

Tacrolimus

TDF

tenofovir

VOD

veno-occlusive disease

Introduction

  1. Top of page
  2. Abstract
  3. Introduction
  4. Patients and Methods
  5. Statistical Analysis
  6. Results
  7. Discussion
  8. Acknowledgments
  9. Disclosure
  10. References

The introduction of antiretroviral therapies (HAART) has dramatically improved the survival of patients infected with human immunodeficiency virus (HIV) (1,2). End-stage liver disease (ESLD) has thus become the main cause of death among HIV patients co-infected with hepatitis C (HCV) or hepatitis B (HBV) viruses (1,3). Nevertheless, anti-HCV therapy with Pegylated Interferon and Ribavirin is efficacious in HIV/HCV co-infected patients and patients with a sustained virological response will have a low mortality rate, but in nonresponding or relapsing patients, HIV co-infection accelerates the course of liver disease and increases the mortality rate (3). The data available in the literature on HIV/HCV co-infection have suggested that co-infected patients are at risk from rapidly progressing liver disease and hepatocellular carcinoma, particularly in the context of a decreased CD4 cell count (4–6). For these reasons, in the past few years, liver transplantation (LT) has been made available to HIV-infected patients. Our group and others have demonstrated that LT is feasible in HIV-infected patients (7–10). Although the results of LT are excellent in HIV/HBV co-infected patients, survival is poorer in those with HIV/HCV co-infection because of the potential severity of HCV recurrence (7–10). We recently demonstrated that 2-year and 5-year survival rates reached 73% and 51% and 91% and 81% in co-infected and mono-infected patients, respectively (p = 0.004) (8).

The most important problem during the post-LT period consists in the severity of the recurrence of HCV infection on the liver graft, and in the same series, we observed that progression to fibrosis ≥F2 was significantly higher in the co-infected group (p < 0.0001) (8). Fibrosing cholestatic hepatitis (FCH) is a severe complication characterized by extensive dense portal fibrosis and cholestasis and a rapid deterioration of liver function (11). The aim of the present retrospective study was therefore to determine (1) the prevalence of FCH, (2) the clinical and pathological characteristics of FCH and (3) the predictive factors for FCH in a large cohort of HIV/HCV co-infected patients.

Patients and Methods

  1. Top of page
  2. Abstract
  3. Introduction
  4. Patients and Methods
  5. Statistical Analysis
  6. Results
  7. Discussion
  8. Acknowledgments
  9. Disclosure
  10. References

Between December 1999 and October 2008, 59 predominantly male (82%) HIV+ patients with a mean age of 44 years [29–63] and positive HCV RNA were transplanted in our center because of ESLD due to HCV (n = 40), HBV/HCV co-infection (n = 1), HBV/HCV/HDV co-infection (n = 1), HCV and alcohol (n = 6), or HCV and hepatocellular carcinoma (HCC) (n = 11). HCV carriage during the pretransplantation period was documented by positive results for hepatitis C antibody; HCV serology tests were performed using the Innotest HCV assay (Innogenetics, Ghent, Belgium) and HCV viremia was quantified using the Abbott Real Time HCV assay (Abbott Diagnostics) with a lower detection limit of 12 IU/mL. The HCV genotype was determined by direct sequencing of the NS5B region, as described elsewhere (12). In all patients, the HIV infection was under control, with a mean CD4 cell count of 294 cells/mm3[range: 60–1020]. The patients had not experienced any previous AIDS events or opportunistic infections, and displayed an undetectable HIV plasma viral load at the time of their registration on the LT waiting list (20 copies/mL, determined using the Cobas Taqman HIV assay, Roche Diagnostics, Meylan, France). Treatment regimens were not standardized. Antiretroviral therapy could include any of the 13 drugs available, namely lamivudine, zidovudine, enfuvirtide, abacavir; tenofovir, didanosine, atazanavir, darunavir, efavirenz, ritonavir, lopinavir and emtricitabine. The dosage of antiretroviral agents was adapted in each patient as a function of residual plasma concentrations of protease inhibitors (PI) or nonnucleoside reverse transcriptase inhibitors (NNRTI), in order to ensure that plasma concentrations remained at therapeutic levels and the plasma viral load remained undetectable.

A control group of 6 HCV+/HIV– patients transplanted between February 2003 and June 2008 (with a mean age of 54.3 years ± 11.5, range: 44–69) from our cohort of 223 HCV+ patients (2.6%) who had been transplanted in our center and who developed FCH within a mean period of 5.5 ± 1.64 [3–7] months were also studied.

Liver transplantation and immunosuppressive therapy

Liver grafts were obtained from deceased donor (n = 37) (63%) or living (n = 22) (37%) donors; in the latter case, grafts were obtained from patients transplanted for familial amyloidotic polyneuropathy (domino program) (n = 18, 30%), or a related living donor (n = 4, 7%). The median age of donors was 50 years [range: 23–78 years]. After transplantation, all patients received 100 mg Fluconazole as candida prophylaxis and 400/80 mg Cotrimoxazole/Trimethoprin as prophylaxis against pneumocystis carinii. Ganciclovir 10 mg/kg/day and Valganciclovir (900 mg/day) was administered for 3 months post-LT as prophylaxis against cytomegalovirus. Primary immunosuppression was based on Tacrolimus (n = 26, 44%), Tacrolimus and Mycophenolate Mofetil (MMF) (n = 10, 17%), Cyclosporine A (n = 7, 12%) and Cyclosporine A and MMF (n = 16, 27%). MMF was given to patients with renal insufficiency or as additional immunosuppression for those experiencing acute rejection. In all patients, corticosteroids were initiated at a dose of 5 mg/kg/day, then reduced to 0.3 mg/kg/day at the end of first week, tapered to 10–15 mg a day until the third month and then gradually decreased. With respect to HAART, the same regimen as that administered during the pre-LT period was reintroduced after LT when liver function had stabilized in HIV patients with or without FCH. The mean intervals before the reintroduction of HAART were 25 days [7–76] and 17 days [0–56], respectively (p = 0.12). Only one patient displayed a high level of HIV replication between LT and the reintroduction of HAART. Physical examinations and liver biochemistry tests were performed daily during the first week of the postoperative period, three times a week during the next month, then once a week for 3 months and every 2 months thereafter.

Virological monitoring

HCV RNA viremia was quantified every 3 months post-LT in the outpatient setting. Moreover, the HCV viral load was determined at 1 and 2 weeks post-LT; HIV RNA was determined once a week during the first month after LT, then every 3–4 months. For HBV/HDV co-infected patient, regular monitoring of anti-HBs antibody titers was ensured, and circulating HBV–DNA levels were quantified by PCR using the Cobas Taqman HBV assay with a lowest detection limit of 12 IU/mL (Roche Diagnostics). All patients had an undetectable HBV viral load at the time of LT. HDV replication was assessed using quantitative real time-PCR, as described elsewhere, with a lowest detection limit of 1000 copies/mL (13). HBV–DNA and HDV–RNA, were determined every 3–4 months after LT.

HLA typing

HLA typing for HLA-A, B and DR was performed by polymerase chain reaction (14).

Diagnostic criteria of FCH and pathological examination

Allograft liver biopsies were scheduled for all patients on day 0 before perfusion, at 6 months post-LT and then once a year post-LT. Additional liver biopsies were performed when liver function test results were abnormal. Liver biopsies were fixed in AFA, paraffin-embedded and stained with hematein eosin safran, picrosirius and Perls. Acute and chronic rejections were diagnosed according to the Banff classifications (15). Pathological features of veno-occlusive disease (VOD) were diagnosed if there was total or subtotal fibrous obliteration of the hepatic veins by connective tissue and centrilobular hemorrhagic necrosis. A diagnosis of ‘biliary obstruction’ was based on the combination of portal edematous fibrosis, ductular proliferation, portal mixed inflammatory infiltrate, cholangiolitis and cholestasis (16). Chronic hepatitis was scored according to the METAVIR scoring system (17).

FCH was defined according to the histological criteria described by Davies et al., that is, the presence of extensive, dense portal fibrosis with immature fibrous bands extending into the sinusoidal spaces, ductular proliferation, cholestasis and moderate mononuclear inflammation (11). Moreover, the severity and composition of the portal inflammatory infiltrate was assessed semi-quantitatively. In particular, the ratio of neutrophil polymorphs in the portal infiltrate was determined globally. Cholestasis and tissue changes due to cholestasis, that is, hepatocellular bile pigment, canalicular bile plugs, kupffer cell bile pigment and/or portal tract features such as bile ductular proliferation or inspissated bile in bile ductules, were also noted. In addition, cholestasis and sinusoidal fibrosis were also semi-quantified according to the scoring systems as previously described by Dixon et al. (18).

Liver biopsies performed at the time of acute hepatitis were also reviewed. Acute hepatitis was defined by the presence of necro-inflammatory activity on the liver biopsy performed because liver enzymes had started to rise again. The severity of activity was assessed using the METAVIR score (17).

Statistical Analysis

  1. Top of page
  2. Abstract
  3. Introduction
  4. Patients and Methods
  5. Statistical Analysis
  6. Results
  7. Discussion
  8. Acknowledgments
  9. Disclosure
  10. References

Continuous variables were expressed as the median with an interquartile range. Qualitative data were described using absolute values and percentages. For survival analysis, data were censored at the date of death or when the last information on life status was obtained, and no later than 31 October 2009. Survival times were calculated from the date of LT. A univariate analysis of survival was performed using the Kaplan–Meyer method. Curves were compared by means of the log-rank test. Statistical analyses were performed using the SPSS statistical software package, version 16.0 (SPSS, Chicago, IL, USA).

Results

  1. Top of page
  2. Abstract
  3. Introduction
  4. Patients and Methods
  5. Statistical Analysis
  6. Results
  7. Discussion
  8. Acknowledgments
  9. Disclosure
  10. References

Pathological characteristics at the time of FCH diagnosis

All 59 patients experienced HCV recurrence on the liver graft and among them, 11 patients (FCH group) out of the 59 patients in this cohort (19%) developed FCH according to the pathological features described in the Patients and Methods section. The kinetics of fibrosis and the distribution of fibrosis scores for the remaining 48 patients in the non-FCH group at the end of follow-up are detailed in Figure 1. At the end of follow-up, the proportions of patients qualified as F0, F1, F2, F3 and F4 were 26%, 22%, 35%, 9% and 9%, respectively.

image

Figure 1. Distribution of fibrosis scores in the non-FCH group at the end of follow-up.

Download figure to PowerPoint

FCH was diagnosed within a mean interval of 7 months [range: 2–27 months] post-LT. Different features suggestive of a histological diagnosis of FCH, and particularly sinusoidal fibrosis (Figure 2A and B) are summarized in Table 1(a). Table 1(b) gives the histological features of the six HIV–/HCV+ patients in the control group. Interestingly, sinusoidal fibrosis was more severe in this latter group. In addition, portal infiltrate was mostly mild in the co-infected group (8/11; 73%) when compared with the HIV–/HCV+ patients (2/6; 33%). Cholangiolitis was never observed (Figure 2C and D). Steatosis was rare and minimal in two of the 11 patients (18%). Acute rejection was not observed at the time of FCH.

image

Figure 2. (A) Sinusoidal fibrosis: Key features of FCH (picrosirius staining). (B) FCH at a later stage (picrosirius staining). Sinusoidal fibrosis graded 4 according to the fibrosis scoring system (18). CLV, centrilobular vein; PT, portal tract. (C) FCH (hematein–eosin safran staining). The portal tract is enlarged by edema, inflammation and bile duct proliferation, as in the case of biliary obstruction. (D) FCH (hematein–eosin safran staining). At a higher magnification, neutrophils are present around the cholangioles but not in the walls (no cholangiolitis present).

Download figure to PowerPoint

Table 1.  Pathological characteristics at the time of FCH diagnosis (A) in 11 HIV+/HCV+ patients and (B) in 6 HIV-/HCV+ patients
IDFibrosisInterface activityLobular activityCholestasis scoring system1Ductular proliferationSinusoidal fibrosisPortal oedemaSeverity of portal infiltrateNeutrophils ratio
  1. 1According to Dixon et al. (18).

(A) FCH diagnosis in 11 HIV+/HCV+ patients
 BPF2A1A21111Mild20%
 DMF2A1A21110Mild5–10% 
 HMF1A1A11120Mild10%
 IFF2A2A2/30110Moderate (nodular)5–10% 
 RDF2A1A30120Mild10%
 RNF2A1A23111Mild10%
 SFF2A1A24110Mild5–10% 
 TJF2AOA13111Minimal10%
 FLF2A1A23111Mild5%
 PHF2AOA11111Minimal5%
 DSF2A1A21111Mild5%
(B) FCH diagnosis in 6 HIV–/HCV+ patients
 LCF2A2A21330Moderate1%
 FMF2A1A13230Mild5%
 CAF2A1A12321Moderate10%
 FLF2A1A11331Moderate10%
 CFF2A2AO1230Moderate5%
 AAF2A1A21231Mild5%

Biological characteristics of the 11 HIV+/HCV+ patients at the time of FCH diagnosis

The biological and clinical characteristics of patients at the time of FCH diagnosis are described in Table 2. At this time, mean total bilirubin, γ-GT and ALT levels were respectively 175 μmol/L [range: 20–444], 637 IU/L [range: 175–1792] and 171 IU/L [range: 39–592]. The mean CD4 cell count was 159 cells/mm3[range: 100–360]. The mean HCV viral load was 7.2 log10 IU/mL ± 0.44 at 6 months post-LT and differed significantly from that of patients in the non-FCH group (6.35 log10IU/mL ± 0.9, p = 0.01) (Tables 2 and 3).

Table 2.  Biological characteristics of the 11 patients at the time of FCH diagnosis
IDDelay of FCH after LT (months)ALT (IU/L)gCT (IU/L)Total/Conj. Bilirubin (μmol/L)INRHCVRNA (log10 IU/mL)CD4 (cells count/mL)Delay of reintroduction (days)/and type of HAART
  1. Drugs: 3TC, lamivudine; AZT, zidovudine; T20, enfuvirtide; ABC, abacavir; TDF, tenofovir; DDI, didanosine; ATZ, atazanavir; TMC, darunavir; EFV, efavirenz; RTV, ritonavir; LPV/r, lopinavir; FTC, emtricitabine; fAPV, fosamprenavir.

  2. *= according to the METAVIR score,**= fibrosis not evaluated because of biliary obstruction.

BP563225131/761.156.610024/3TC-AZT-LPV/r
DM328088147/320.956.510020/ABC-ENF-TDF
HM11106397341/2001.06>814029/TDF-3/TC-LPV/r
IF459276596/611.266.826016/ENF-TDF-3TC-DDI
RD598175444/2181.227.212076/AZT-ABC-TDF
RN497382230/1231.727.65407/ENF-ABC-TDF-DRV-RTV
SF2198384250/1421.17.525028/ENF/3TC/TDF
TJ10941792160/911.397.113020/ABC/EFV/3TC
FL42171002104/640.847.136021/ABC/3TC/f APV/RTV
PH399724100/540.977.315014/ENF/TDF/FTC
DS273928020/50.987.110019/ABC/TDF/LPV/r
Table 3.  Pretransplant and transplant characteristics of HIV+/HCV+ co-infected patients who developed FCH after LT—comparison with patient in the non-FCH group and with HIV–/HCV+ patients.
 Patients with FCH HIV+/HCV+(1)Patients without FCH HIV+/HCV+ (2)Patients with FCH HIV-/HCV+ (3)P (1) versus (2)/(1) versus (3)
N11486 
Men9 (82%)39 (81%)5 (83%)0.6/0.9
Recipient Age (years)42.2 ± 2.344.4 ± 6.354 ± 11.50.2/0.004
MELD score17.5 ± 4.417.8 ± 9.320 ± 11.40.9/0.4
Genotype 18 (73%)25 (52%)4 (67%)0.5/0.79
HCV viral load at the time of LT (Log10 lU/mL)5.5 ± 1.25.5 ± 0.75.5 ± 0.60.9/0.9
CD4 cells count at the time of LT (Log10 lU/mL)162 ± 105138 ±  0.5/-
Donor Age (years)50.8 ± 12.749.6 ± 13.764.5 ± 3.80.8/0.02
Partial liver graft from living donar1 (9%)17 (35%)00.08/0.001
Matching HLA-A, B, DR ≥ 25/11(45%)10/45 (22%)00.1/<0.0001
Acute rejection episodes6/11 (54%)17 (35%)1(17%)0.3/< 0.0001
Steroid flashes episodes4/11 (36%)16/48 (33%)1(17%)0.9/<0.0001
Type of IS therapy:    
Cyclo + Steroids44100.53/0.01
Tac + Steroids71710.08/0.01
Cyclo or Tac + MMF43110.08/0.01
Cyclo + Steroids + MMF002NS/NS
Tac + Steroids + MMF001NS/NS
Tac + Steroids + Everolimus001NS/NS

Pre- and posttransplant characteristics of HIV/HCV co-infected patients who developed FCH after LT—Comparison with patients in the non-FCH group

The pretransplant clinical and laboratory features of co-infected patients who developed FCH are shown in Table 3. The mean age of these predominantly male patients (82%) at LT was 44 years [range: 29–63]; three patients had a previous history of alcohol abuse and one patient suffered from hepatopulmonary syndrome. The principal indication for LT was hepatic insufficiency (Child C = 10 patients) with a mean MELD score of 17.5 [range: 6–24] except for the patient with hepatopulmonary syndrome who was Child A and had a MELD score of nine at the time of LT. As shown in Table 3, there were no differences in terms of the severity of liver disease (MELD score), virological characteristics (genotype 1 frequency, HCV viral load), CD4 cell count and donor age. Interestingly, in the FCH group, we observed a higher level of HLA-A, B, DR matching (≥2) and acute rejection episodes (54% vs. 35%; p = 0.3) but no more requirement for steroid boluses (36% vs. 33%; p = 1). The types of immunosuppressive therapy did not differ significantly between the two groups. Moreover, the mean time until steroid suppression was 14.7 months in the non-FCH group, versus 9.3 months in the FCH group (p = 0.3).

Early quantification of HCV viral load after liver transplantation

At week 1 post-LT, the mean HCV viral load was higher in the FCH group than in the non-FCH group: 6.13 log10± 1.30 versus 4.9 log10± 1.78, p = 0.05. At Week 2 post-LT, the mean HCV viral load continued to be higher in the FCH group than in non-FCH patients: 6.14 log10± 1.95 versus with 5.77 log10± 1.17, p = 0.08 (Figure 3A and B).

image

Figure 3. (A) At week 1 post-LT, the mean HCV viral loads were 4.9 log10 IU/mL ± 1.78 and 6.13 log10 IU/mL ± 1.30 in the non-FCH and FCH groups, respectively, (p = 0.05). (B) At week 2 post-LT, the mean HCV viral loads were 5.77 log10 IU/mL ± 1.17 and 6.14 log10 IU/mL ± 1.95 in the non-FCH and FCH groups, respectively, (p = 0.08).

Download figure to PowerPoint

Characteristics of acute hepatitis after LT—Comparison between FCH and non-FCH groups (Table 4)

Table 4.  Characteristics of acute hepatitis after liver transplantation—comparison with patient in the non-FCH group and with HIV-/HCV+ patients
 (1) Patients with FCH HIV+/HCV+ (n = 11)(2) Patients without FCH HIV+/HCV+ (n = 48)(3) Patients with FCH HIV–/HCV+ (n = 6)P  (1) versus  (2)/  (1) versus  3) 
  1. 1Only 5/6 patients had a liver biopsy for confirmation of acute hepatitis.

  2. 21/6 patients had an episode of acute rejection.

Established diagnosis  n)112851 
Delay between diagnosis of acute hepatitis and LT  days)46.4 ± 21.330.5 ± 62.795 ± 96 0.4/0.03
GGT  IU/mL)531.2 ± 364.2435.4 ± 331.8819 ± 423 0.4/0.36
ALT  IU/mL)242.7 ± 192.4239.9 ± 197.6275 ± 177 0.9/0.83
Total blllrubln  μmol/L)77.8 ± 92.5 99.3 ± 131.560.4 ± 32  0.65/0.82
Moderate to severe activity8/11 73%)13/28 46%)2/5 40%)0.007/0.1 3
Degree of acute rejection  BANFF score)4.67 ± 2.8   4.5 ± 1.78620.9/-
HCV viral load  Log10 lU/mL)6.83 ± 0.755.5 ± 2 7.2 ± 0.70.08/0.43

At the time of diagnosis of acute hepatitis, the histological picture of HCV recurrence was similar in both groups: lobular hepatitis included focal hepatocyte necrosis and focal inflammation. Inflammatory infiltrate in both the portal tracts and lobules was essentially composed of lymphocytes. There was a significant difference between the groups regarding the severity of lobular activity: this was moderate to severe in 8/11 (73%) patients in the FCH group versus 13/28 (46%) in the non-FCH group (p = 0.007). γGT, ALT and total bilirubin levels were 531.2 ± 364.2 and 435.4 ± 331.8 IU/L (p = 0.4), 242.7± 192.4 and 239.9 ± 197.6 IU/L (p = 0.9), 77.8 ± 92.5 and 99.3 ± 131.5 μmo/L (p = 0.65), respectively, and no significant difference between the two groups (FCH and non-FCH) was noted regarding the time to onset of acute hepatitis: 46.4 ± 21.3 days and 30.5 ± 62.7 days (p = 0.4), respectively. At the time of acute hepatitis, the HCV viral load was higher in the FCH group: 6.83 log10IU/mL versus 5.5 log10 IU/mL in the non-FCH group (p = 0.08). As for other pathological features potentially associated with acute hepatitis, no differences were observed in the grading of acute rejection at the time of acute hepatitis; the Banff scores were 4.67 ± 2.8 and 4.5 ± 1.78, respectively (p = 0.9). No patients developed CMV reactivation or primary infection.

Other pathological features such as VOD and the pattern of biliary obstruction

Four of the 11 patients developed VOD, and in two of them, this preceded the development of FCH and was related to acute rejection. In the other two patients, VOD developed later and was not related to acute rejection. In the non-FCH group, only two out of 48 patients (4%) developed veno-occlusive lesions following acute rejection. Eleven patients in the non-FCH group experienced ‘biliary obstruction’, characterized by a combination of centrilobular cholestasis and portal changes including edema, mixed inflammatory infiltrate with neutrophil polymorphs, cholangiolar proliferation and cholangiolitis (19). This histological pattern was related to a true biliary complication (n = 4) or sepsis (n = 4), but remained unexplained in three cases. Interestingly, this pattern of VOD was not observed in the mono-infected group with FCH.

Survival

Overall cumulative patient survival reached 67%, 62% and 48% at 2, 3 and 5 years, respectively (Figure 4A). Twenty-five patients (42%) died within a mean period of 19 months (range: 2–68 months) post-LT. Thirteen patients died of HCV-related disease. Nine patients died of liver failure after developing FCH. There was a significant difference in survival between the FCH and non-FCH groups; the mean survival times were 69.8 ± 7.7 [range: 54.6–84.9] months and 26 ± 4.7 [16.8–35.2] months in the non-FCH and FCH groups, respectively (p = 0.004; Figure 4B). In the FCH group, where 9/11 (82%) patients died, the deaths of six patients occurred in a context of severe HCV recurrence (Table 4). Only one of the six HIV–/HCV+ patients died after LT.

image

Figure 4. (A) Overall cumulative survival of the 59 HIV/HCV co-infected patients transplanted between 1999 and 2008. (B) Survival of HIV/HCV co-infected patients after LT, with or without FCH.

Download figure to PowerPoint

Anti-HCV therapy and outcome in FCH patients

All patients with FCH were treated with anti-HCV therapy, Pegylated Interferon and Ribavirin within a mean period of 7.8 months [range: 3–19] post-LT. Anti-HCV therapy was introduced because of the presence of severe acute hepatitis in one patient and at the time of FCH diagnosis in 10 patients (Table 5). The regimen was initiated under very difficult conditions, in some cases even before discharge from hospital following LT; all patients had jaundice and three had ascites. Thus, except for one patient (BP) who received the full dose (for 2 months, then adjusted to anemia), treatment was started using a low-accelerating-dose regimen. The Pegylated Interferon (α2a or α2bPEG-IFN) doses administered were between 40 μg/week and 180 μg/week, while Ribavirin (RBV) was prescribed at between 400 and 800 mg/day, with the doses being adjusted to the adverse effects. The mean duration of treatment was 6 months (range: 0.5–11 months). None of the patients received complete full doses during treatment, and in all cases, the treatment was reduced or withdrawn before SVR was achieved. A complete biochemical response was observed in three patients (27%) (patients DM, IF, RD), with a normalization of bilirubin levels. A complete virological response was observed in two patients, but a relapse occurred when the drug dosage was reduced because of tolerability issues and adverse effects. Among the remaining patients, a partial response was observed in three patients (27%), while no response was seen in seven patients (64%). Adverse effects were reported in all patients. Anemia was present in 9/11 patients (82%), with a hemoglobin nadir of 7 g/dL. Marked psychiatric symptoms were observed in two patients.

Table 5.  Anti-HCV therapy modalities and outcomes of HIV+/HCV+ co-infected patients who developed FCH after LT HIV+/HCV+ co-infected patients who developed FCH after LT
IDReason for deciding introduction of anti-HCV therapy/Modality of anti-HCV therapyBiochemical response1Virological response2Secondary effects of anti HCV therapyOutcome / Delay of follow-up/Cause of death
  1. 1Complete biochemical response: normalization of liver tests; Partial biochemical response: improvement of liver tests.

  2. 2Partial response: Decrease of HCV viral load > 2 log10IU/mL; Complete virological response: HCV RNA < 12 IU/mL.

  3. NA = not available.

BPFCH (jaundice)/PEGIFNα2a (180 μg/W) + 800 mg/D Ribavirin  (M10)PartialPartialAnemia/agressivityDeath  (M24) Pulmonary hypertension
DMFCH (jaundice)/PEGIFNα2a (180 μg/W) + 800 mg/D Ribavirin (M4 to M21)CompletePartialAnemiaDeath (M44) Graft dysfunction due to HCV recurrence
HMFCH (jaundice)/PEGIFNα2b (70 μg/W) + 400 mg/D Ribavirin (M12 to M13)PartialNo responseAnemia/Depressive syndromeDeath (M13) Graft dysfunction due to HCV recurrence
IFFCH (jaundice, ascites)/PEGIFNα2b (70 μg/W) + 800 mg/D Ribavirine (M3 to M8) PEGIFNα2a (135 μg/W) + 600 mg/D Ribavirin (M11 to M49)CompletePartial/Relapse under therapyAnemiaDeath (M49) Graft dysfunction due HCV recurrence
RDFCH (jaundice,ascites)/PEGIFNa2b (125 μg/W) + 600 mg/D Ribavirin (M6 to M12)CompleteComplete Response/Relapse with diminution of dosageAnemia/trombopeniaDeath (M46) Graft dysfunction due to HCV recurrence
RNFCH (jaundice,ascites)/PEGIFNα2a (135 μg/W) + 600 mg/D Ribavirin (M2 to M5)PartialPartialNeutropenia/AnemiaDeath (M5) indeterminate cause
SFFCH (jaundice)/PEGIFNα2a (M3) (135 μg/W) + 600 mg/D Ribavirin (M3 to M4)PartialNo responseAnemia/neutropeniaDeath (M4) multiorgan failure
TJAcute hepatitis/PEGIFNα2b (65 μg/W) + 600 mg/D Ribavirin (M1 to M8)PartialNo responseNeutropeniaDeath (M26) multiorgan failure
FLFCH (jaundice)/PEGIFNα2b (50 μg/W) + 600 mg/D Ribavirin (M20 to M23)PartialNo responseAnemiaDeath (M26) Cardiac arrest
PHFCH (jaundice)/PEGIFNα2b (80 μg/W) + 800 mg/D Ribavirin (M4 to M9)PartialNo responseAnemia (M15) Well condition
DSFCH/PEGIFNα2b (40 μg/W) + 400 mg/D Ribavirin (M34 and in course)PartialPartialAnemia (M35) Well condition
 Modality of anti-HCV therapy
LCNAAggravationResponder (M4)AnemiaDeath (M26) Graft dysfunction due to HCV recurrence
FMPEGIFNα2b (55 μg/W) + 400 mg/D Ribavirin (M6 to M12)PartialNo responseAnemia/Right heart failure (M15) Well condition
CAPEGIFNα2b (100 μg/W) + 800 mg/D Ribavirin (M12 to M24)CompleteSustained ResponderAnemia/Fever (M28) Well condition
FLPEGIFNα2a (180 μg/W) + 1000 mg/D Ribavirin (M7-ongoing)CompleteNegative at M6 of treatmentAnemia (M14) Well condition
CFPEGIFNα2b (60 μg/W) + 400 mg/D Ribavirin (M6- ongoing)PartialNo responseAnemia/trombopenia (M14) Well condition
AAPEGIFNα2a (180μg/W) + 1000 mg/D Ribavirin (M15 to M27)CompleteResponderAnemia (M35) Well condition

Discussion

  1. Top of page
  2. Abstract
  3. Introduction
  4. Patients and Methods
  5. Statistical Analysis
  6. Results
  7. Discussion
  8. Acknowledgments
  9. Disclosure
  10. References

FCH is an uncommon but well-documented complication that affects transplant recipients infected by hepatitis B or C virus. FCH has been reported in patients with a recurrence of HBV infection without HBV prophylaxis and occurs within the first year of LT in 5% of HCV-infected patients. This severe complication is characterized histologically by extensive dense portal fibrosis with immature fibrous bands extending into sinusoidal spaces, ductular proliferation, marked canalicular and cellular cholestasis and mononuclear inflammation (11,18). However, a broad spectrum of FCH may exist with different degrees of mononuclear inflammation and ductular reaction and the possible coexistence of an expansion of portal fibrosis (20).

During this study, we demonstrated that FCH occurred in 11 out of 59 HIV/HCV co-infected patients (19%) shortly after LT, as it was diagnosed after a mean interval of 7 months. Furthermore, in eight patients (73%), FCH occurred within less than 6 months. Standard diagnostic criteria were applied, as described in Materials and Methods and as previously published by Davies et al. (11). All of our patients suffered from perisinusoidal fibrosis, predominating in the periportal area in nine of them (80%). This is the first description of fibrosis localization in a context of FCH in HIV/HCV co-infected liver transplant patients. Moderate to severe lobular activity was observed, in 8/11 (73%) patients, this reached ≥A2 using the METAVIR score. These results were consistent with those previously described, where activity was markedly stronger in patients with FCH (18). Moreover, the histological pattern is quite similar in comparison of a control group of HIV–/HCV+ patients transplanted in our center.

However, it may be difficult to distinguish FCH due to HCV recurrence and that resulting from a ‘bile duct obstruction’ pattern. Portal edema, portal neutrophilia (rather than periportal neutrophilia) and cholangiolar proliferation are common to both conditions; however, FCH portal edema was present in 6/11 (54%) of the patients, all six of whom were transplanted with a total liver from a deceased donor. No biliary obstruction occurred during the post-LT period; moreover, magnetic resonance cholangiography was performed in all these patients and did not reveal any bile duct abnormalities. Cholangiolar proliferation was observed in all cases; the neutrophil ratio was quite mild, at between 5% and 20%. Acute cholangiolitis was never present. VOD was observed in six HIV+ patients (four with and two without FCH). In four of these patients, VOD could be explained by previous episodes of acute rejection with predominantly endothelial involvement, but the cause remain unexplained in the other two.

Another interesting feature, which should be included in the diagnostic pattern is the HCV viral load. We recently demonstrated that the median HCV viral load rose from 5.83 log10/mL before transplantation to 6.58 log10/mL posttransplantation, without there being any changes to CD4 counts and T-cell subsets (20). If only one patient with FCH was included in this latter study, no HCV-specific responses were detectable in patients with severe hepatitis or a fibrosis score higher than F2 (20). The present report demonstrates that the HCV viral load could constitute an important feature because a significantly higher load was observed at the time of FCH diagnosis, which occurred within a mean delay of 7 months post-LT, when compared with the HCV viral load at 6 months post-LT in the non-FCH group. Moreover, and interestingly, the HCV viral load immediately after LT (weeks 1 and 2) was higher in the FCH group; moreover, in the present study the delay before the reintroduction of HAART was longer in the FCH group than in the non-FCH group (25 days versus 17 days, respectively; p = 0.12). The usefulness of instituting HAART therapy at an earlier stage in order to prevent a defective and early anti-HCV immune response needs to be demonstrated in further studies. Moreover, detection of a high HCV viral load in the short-term post-LT could indicate the development of a severe recurrence of HCV infection on the liver graft.

Predicting the onset of FCH and defining the patients at risk of developing this severe complication on the liver graft is a major clinical goal. At present, no reliable markers are available to identify recipients who will develop a severe recurrence on the liver graft. Nevertheless, two recent studies demonstrated that the activation of hepatic stellate cells, determined by smooth muscle actin (αSMA) immunostaining in liver allograft biopsies, could be predictive of the subsequent development of bridging fibrosis (21,22). However, immunohistochemistry techniques are a potential source of significant variability and error. Using univariate analysis, we could not determine any differences with respect to MELD score, donor age, HCV genotype, the number of acute rejection episodes and the number of steroid flashes. Nevertheless, at an early stage, that is, at the diagnosis of acute hepatitis, we were able to demonstrate that activity was more severe in the FCH group.

The prognosis of FCH in HIV/HCV co-infected patients is very serious (23). Nine of the 11 patients studied died within 26.3 months and 20.7 months of LT and the diagnosis of FCH, respectively. However, all these patients had been treated with pegylated IFN and RBV soon after the diagnosis of FCH, because anti-HCV therapy was instituted within a mean interval of 7.8 months after LT. None of the 11 patients displayed a sustained virological response. However, in three patients, a complete biochemical response was observed, with a normalization of bilirubin levels.

One of the main objectives when managing HIV/HCV co-infected patients is to avoid a severe recurrence of HCV infection on the liver graft. Presently, we lack predictive factors for a severe recurrence of HCV infection that might influence tailor-made management. It is therefore necessary to treat HCV infection in a context of HIV infection at the onset of acute hepatitis, particularly if the score is moderate to severe. Of course, the tolerance of pegIFN and RBV may be difficult at an early stage post-LT, particularly in terms of hematological disorders (24). Clearly, the introduction of EPO must always be discussed at the same time in order to avoid any reduction or discontinuation of the IFN and/or RBV dosage. Moreover, new anti-HCV drugs such as polymerase inhibitors need to be tested in the context of LT as soon as possible. This optimum therapeutic approach could provide great hope for HIV/HCV co-infected patients who experience a severe recurrence of HCV infection on the liver graft.

Acknowledgments

  1. Top of page
  2. Abstract
  3. Introduction
  4. Patients and Methods
  5. Statistical Analysis
  6. Results
  7. Discussion
  8. Acknowledgments
  9. Disclosure
  10. References

The authors would like to thank Pascaline Chaussenot, Claire Mony, Frédérique Blandin, Marine Monot and Victoria Hawken for their technical assistance.

Disclosure

  1. Top of page
  2. Abstract
  3. Introduction
  4. Patients and Methods
  5. Statistical Analysis
  6. Results
  7. Discussion
  8. Acknowledgments
  9. Disclosure
  10. References

The authors of this manuscript have no conflicts of interest to disclose as described by the American Journal of Transplantation.

References

  1. Top of page
  2. Abstract
  3. Introduction
  4. Patients and Methods
  5. Statistical Analysis
  6. Results
  7. Discussion
  8. Acknowledgments
  9. Disclosure
  10. References