Post–liver transplant cholestatic hepatitis C: A systematic review of clinical and pathological findings and application of consensus criteria

Authors


Abstract

Liver transplantation is currently the only definitive modality for the treatment of end-stage liver disease due to chronic hepatitis C. However, recurrent hepatitis C after liver transplantation is nearly universal. Cirrhosis may develop in 20% of recipients within 5 years, and recurrent hepatitis C may lead to graft failure, retransplantation, and even death. A subset of recipients may develop post–liver transplant cholestatic hepatitis C (PLTCHC), which is characterized by cholestasis, hepatocyte ballooning, and rapid progression to graft failure. We present a systematic review of PLTCHC that is focused on hepatitis C–infected liver transplant recipients. We compare the pathological definitions of PLTCHC, clinical factors, management strategies, and outcomes reported in studies. We found differences among studies in the types of histological criteria used to diagnose PLTCHC during liver biopsy and in the types of clinical information provided. Three of the 12 studies published after 2003 used the definition of PLTCHC published by the first International Liver Transplantation Society expert panel consensus conference on liver transplantation and hepatitis C. We propose that studies on PLTCHC use the consensus criteria for diagnosis and suggest clinical information that should be provided in future studies with the goal of improving our understanding and management of this deadly disease. Liver Transpl 16:1228-1235, 2010. © 2010 AASLD.

End-stage liver disease caused by hepatitis C virus (HCV) is the leading indication for liver transplantation in the United States. HCV recurrence after liver transplantation is one of the leading complications faced by hepatologists, and in recent years, the progression of fibrosis has seemed to be occurring more rapidly, with up to 20% of recipients developing cirrhosis within 5 years of transplantation.1–12 Hepatitis C after liver transplantation includes a spectrum of disease presentations ranging from recurrent hepatitis C with inflammation with or without fibrosis to post–liver transplant cholestatic hepatitis C (PLTCHC). PLTCHC is an uncommon and aggressive form of recurrent hepatitis C that is characterized by the rapid development of inflammatory infiltrates on liver biopsy, cholestasis, hepatocyte ballooning, and eventually advanced liver fibrosis leading to graft failure and death.13–16 Although PLTCHC may fall along a spectrum of recurrent hepatitis C presentations, it presents as a condition distinctly different from that of recipients who develop recurrent hepatitis C with advanced fibrosis or cirrhosis over the course of several years. A profoundly distinct difference between PLTCHC and recurrent hepatitis C is the rapidity with which graft failure can develop with cholestasis and a poor response to therapy. PLTCHC has been reported to affect 2% to 9% of hepatitis C–infected liver transplant recipients and usually leads to liver failure and graft loss within the first 1 to 2 years after transplantation.13, 17–21

Studies on risk factors for recurrent hepatitis C after liver transplantation have typically included recipients who developed advanced fibrosis or cirrhosis within 3 to 5 years after liver transplantation and have not focused on PLTCHC. Patients with PLTCHC may be included in reports of recurrent hepatitis C, but the emphasis of these studies has not been PLTCHC. Studies that have focused on PLTCHC are typically case reports or case series that vary in the defining criteria or types of clinical information provided. Previous reviews of recurrent hepatitis C after liver transplantation with or without PLTCHC have focused on findings on liver biopsy and have not systemically reviewed clinical and histological features of liver transplant recipients alone.15, 16, 22–26 We conducted a systematic review of PLTCHC in hepatitis C–infected liver transplant recipients with the goal of identifying the defining criteria, reported risk factors, and effective treatment strategies in studies of PLTCHC. In addition, we evaluated whether the reviewed studies used the 2003 consensus criteria for defining and diagnosing this condition.14 On the basis of our review, we report the most commonly used histological criteria for PLTCHC and the most commonly used treatment strategies, and we suggest clinical information that should be routinely provided in studies of PLTCHC.

Abbreviations:

HCV, hepatitis C virus; NR, not reported; PLTCHC, post–liver transplant cholestatic hepatitis C; Th2, T helper 2.

MATERIALS AND METHODS

A literature search of the Ovid MEDLINE database and the Embase online database was conducted with the keywords fibrosing cholestatic hepatitis C, cholestasis and hepatitis C, and cholestatic hepatitis C for the period of January 1990 to October 2009. The references of articles were also reviewed for additional studies. Two authors independently extracted data from the articles (T.K.N. and M.W.R.). Because fibrosing cholestatic hepatitis C and cholestatic hepatitis C are believed to be diseases along the same spectrum, they were treated as 1 category (PLTCHC). Our analysis included only PLTCHC and excluded cholestatic hepatitis in nonliver patients, nontransplant patients, and non–hepatitis C studies. The articles were separated into case reports and series, clinical trials, and review articles. Data on the clinical, laboratory, and histological criteria used to define PLTCHC, time of diagnosis, immunosuppression, treatment, survival, and cause of death were extracted. Individual patient data are presented when they are available; otherwise, summary data are presented in the tables and supporting tables. We specifically sought to determine if studies on PLTCHC published after 2003 used the definition proposed by the consensus conference in 2003.14

RESULTS

The literature search identified 46 articles: 39 clinical studies or case series and reports (22 retrospective studies and 17 case reports and series) and 7 review articles. One hundred seventy-five subjects were reported to have cholestatic hepatitis C. Thirty-one of the 39 articles (clinical studies and case reports and series) included liver transplant recipients with PLTCHC.2, 9, 13, 17–19, 27–51 Two articles appeared to report the same case, and only 1 was included for further review.49, 51 The following articles were not included in the analysis: 4 articles about kidney transplant recipients with cholestatic hepatitis C,52–55 2 reports about a heart transplant recipient with cholestatic hepatitis C,56, 57 and 2 reports that described patients with cholestatic hepatitis C without a history of solid organ transplantation.58, 59 Henceforth, we report the results of the 30 articles related only to PLTCHC (Supporting Table 1).

Table 1. Laboratory Features of PLTCHC Patients in the Literature
Authors (Year)Bilirubin (mg/dL)Alkaline Phosphatase (U/L)Alanine Aminotransferase (U/L)
  1. NOTE: Some authors2, 17–19, 27–29, 31, 34, 35, 46, 47 did not report these features for PLTCHC subjects.

Schluger et al.13 (1996)3.4303243
Dickson et al.33 (1996)18.3, 31, 14.7, 4.7224, 632, 435, 419NR, 405, 162, 339
Rosen et al.9 (1998)3.7 ± 1.4281.7 ± 99.2111.1 ± 28.1
Taga et al.37 (1998)24NRNR
Pessoa et al.32 (1999)29 (range = 12.5-72)NR192 (92-933)
Cotler et al.36 (2000)4.4, 8.2, 3.8, 3.1, 3.5, 3.7, 5.1218, 77, 113, 234, 110, 157, 429NR
Tolan et al.40 (2001)84 μmol/LNRNR
Ong et al.41 (2001)50.8, 26.4, 62.7, 24.3, 41NR490, 351, 969, 104
Deshpande et al.38 (2001)42.5, 52.7, 23.2, 24.5, 24.7, 22.8, 27.7, 31.8NRNR
Gopal et al.34 (2001)6.4 (median)NR261 (median)
Takahashi et al.49 (2003)7.9115375
Gopal and Rosen45 (2003)10.7 ± 5.4NR279 ± 203
Neff et al.39 (2005)15.7NR480
Castells et al.50 (2006)16, 21, 18185, 256, 294290, 440, 296
Taniguchi et al.30 (2006)NRNR110
Sohara et al.42 (2006)7.9115375
Bolkhir et al.43 (2007)12.3500700
Carrion et al.48 (2007)3.4 (0.6-34)NR183 (44-497)
Miyake et al.44 (2009)35475200

Definition of PLTCHC

Twenty-eight articles provided a definition for PLTCHC, although the definitions and diagnostic criteria varied among the articles (Supporting Table 2). Fifteen used histology on liver biopsy alone for the diagnosis of PLTCHC,2, 9, 17, 28, 32, 33, 36, 37, 41–44, 46, 48, 49 and 13 articles used a combination of clinical criteria and histology.13, 18, 19, 27, 29–31, 38–40, 45, 47, 50 The most common defining criterion on liver biopsy was cholestasis, which was reported in 20 articles.2, 9, 13, 19, 27–29, 33, 36, 38–41, 44–50 The other common histological features that were reported included hepatocyte changes such as ballooning degeneration on liver biopsy (15 articles),2, 9, 17, 19, 27, 29, 31, 33, 36, 37, 39, 40, 43, 44, 46 fibrosis (periportal, portal, and bridging; 12 articles),2, 18, 28, 33, 36, 37, 41–44, 48, 49 and ductal/cholangiolar proliferation (7 articles).27, 28, 31, 36, 37, 42, 49 Figure 1 shows typical histological features of PLTCHC on liver biopsy as well as features described by the consensus criteria.14 Other less common liver biopsy findings that were reported included centrilobular necrosis29 and cirrhosis.40, 41, 48 Schluger et al.'s article13 was most frequently cited.2, 17, 19, 29, 30, 32, 35–37, 39, 44, 45, 47, 49, 50 Three of the 12 studies published after 200329, 31, 46 used the definition of PLTCHC published by the first International Liver Transplantation Society expert panel consensus conference on liver transplantation and hepatitis C.14

Figure 1.

(A) Early PLTCHC: pericentral hepatocyte ballooning degeneration with canalicular cholestasis. Scattered apoptotic hepatocytes are indicated by arrows (original magnification, ×40; hematoxylin and eosin). (B) Progressive PLTCHC: portal expansion by a mixed inflammatory cell infiltrate. Abundant interfacial inflammatory activity with hepatocyte ballooning degeneration and acidophil necrosis is indicated by arrows (original magnification, ×20; hematoxylin and eosin). (C) Progressive PLTCHC: portal fibrosis with periportal/perisinusoidal extension into hepatic lobules (original magnification, ×20; Masson's trichrome). (D) Late PLTCHC: portal expansion with a ductular reaction and prominent hepatocyte ballooning/reactive changes (original magnification, ×40; hematoxylin and eosin).

Table 2. Proposed Diagnostic and Reporting Criteria for PLTCHC
Proposed Consensus Diagnostic Criteria*
  • *

    Adapted from Wiesner et al.14

• Longer than 1 month after transplantation (usually <6 months)
• Serum bilirubin level greater than 6 mg/dL
• Serum alkaline phosphatase and gamma-glutamyl transpeptidase levels greater than 5 times the upper limits of normal
• Characteristic histological state with predominant ballooning of hepatocytes in the perivenular zone (not necrosis or fallout), paucity of inflammation, and variable degrees of cholangiolar proliferation without bile duct loss
• Very high serum HCV RNA levels
• Absence of surgical biliary complications (normal cholangiogram) and absence of evidence of hepatic artery thrombosis
Reporting Criteria Proposed by the Authors
• Serum HCV RNA (IU/mL) by polymerase chain reaction at the time of diagnosis
• Hepatitis C genotype
• Duration from liver transplantation to the diagnosis of PLTCHC
• Type of immunosuppression prescribed and drug level goals
• Total and direct bilirubin, aspartate aminotransferase, alanine aminotransferase, alkaline phosphatase, and gamma-glutamyl transpeptidase levels at diagnosis
• Histological findings on liver biopsy
• Treatment of PLTCHC (including the dosing and duration of therapy)
• Response to therapy (including the HCV viral load and liver tests)
• Outcome and survival after the diagnosis of PLTCHC
• In the case of death, the interval between the diagnosis of PLTCHC and death and the cause of death
• In the case of retransplantation, the interval between the diagnosis of PLTCHC and retransplantation
• Time from retransplantation to the last follow-up visit
• Autopsy findings if they are available and suggested cause of death

Clinical Data and Risk Factors for PLTCHC (30 Studies in All)

Thirteen articles reported serum HCV RNA levels at the time of diagnosis; in 17 articles, HCV RNA was not reported (NR), or it was not clear if the reported viral load was determined at the time of diagnosis (Supporting Table 1). Fourteen articles reported the HCV genotype, and genotype 1 was the most commonly reported. The interval between transplantation and the development of PLTCHC was reported in 19 articles2, 13, 17, 18, 30, 33, 36, 38–45, 47–50 and ranged from 42 to 1281 days30, 39 (Supporting Table 1). Immunosuppression was reported in 26 articles2, 9, 13, 17, 18, 27, 29, 31–36, 38–50: tacrolimus-based immunosuppression was reported in 9 articles, cyclosporine-based immunosuppression was reported in 6 articles, and tacrolimus-based or cyclosporine-based immunosuppression was reported in 11 articles (Supporting Table 3).

Management of PLTCHC

Twenty-one articles clearly reported the management of patients with PLTCHC.13, 27, 29, 30, 33–45, 47–50 Interferon-based therapy was the most widely used treatment strategy (18 articles),13, 27, 29, 30, 33, 34, 36, 37, 39–44, 47–50 whereas retransplantation was reported for 23 patients in 6 articles13, 33, 37–39, 47 (Supporting Table 4). Among the 18 articles that reported treatment with antiviral therapy for PLTCHC (interferon alone or in combination with ribavirin), 16 articles reported the response to therapy. Forty-two patients received antiviral therapy: 13 had a biochemical or virological response, 3 patients underwent retransplantation, 19 patients died, and outcomes were not provided for 7 patients (Supporting Table 4). Less frequent approaches included interferon with double-filtration plasmapheresis (2 articles)30, 44 and surgical revision of the biliary anastomoses (2 articles).27, 33

Outcomes of PLTCHC

Twenty-four articles reported data on survival,2, 13, 18, 19, 27, 29, 30, 32–45, 47, 49, 50 and 17 of these provided the time from diagnosis to death or the last follow-up visit. Six of the 23 patients who underwent retransplantation (6 articles)13, 33, 37–39, 47 were reported to have died, whereas 7 patients were alive (3 had recurrent cirrhosis, and 1 had graft failure due to hepatic artery thrombosis and was relisted for transplantation). For the remaining 10 recipients, the outcome after retransplantation was unclear. According to the 24 articles that reported survival (Supporting Table 4), 47 of 94 patients with PLTCHC died, and the time between transplantation and death ranged from 2 months to 312 days. The cause of death was reported in 11 articles,13, 18, 27, 33, 35, 38, 39, 41, 45, 47, 49 and the most common causes of death were graft failure (16 patients)27, 33, 38, 45, 47, 49 and sepsis (7 patients)39, 41 (Supporting Table 4).

DISCUSSION

The presentation of HCV recurrence after liver transplantation varies greatly and can range from mild hepatitis to aggressive and potentially fatal PLTCHC.2, 60, 61 Most studies of recurrent hepatitis C have been retrospective studies or clinical trials of recipients who developed advanced fibrosis over several months to years after transplantation. Although patients with PLTCHC have been included in some of these studies, PLTCHC has not been their focus. Studies of PLTCHC alone have typically been case reports or case series.13, 30, 33, 34, 40–45, 49

We focused our analysis on PLTCHC after liver transplantation because PLTCHC may behave differently in nontransplant patients, in patients with cholestatic hepatitis from hepatitis B, and in patients after kidney or heart transplantation for whom there may be more substantial differences in immunosuppression in comparison with liver transplant recipients. Fibrosing cholestatic hepatitis was initially reported in patients with hepatitis B after liver transplantation62 and has been regularly reported in patients with hepatitis C after transplantation.

Initial descriptions of PLTCHC on liver biopsy included progressive fibrosis and cholestasis with graft failure. According to our review, histological features on liver biopsy were not uniform among the articles (Supporting Table 2). The most common biopsy features for the diagnosis of PLTCHC included cholestasis with portal/periportal fibrosis and structural hepatocyte changes marked by ballooning degeneration. Some articles did not report specific histological criteria for PLTCHC.34, 35 We propose that studies of PLTCHC should not only report the histological features used to make the diagnosis but also explicitly state the defining criteria. Only 3 studies29, 31, 46 cited the consensus criteria developed in 2003.14 We propose that the histological criteria developed by the consensus conference, which include the hallmark liver biopsy features of cholestasis and ballooning hepatocytes, be used as the defining histological criteria for PLTCHC (Table 2).

Distinguishing PLTCHC from other forms of chronic hepatitis in liver allografts requires correlations with clinical, biochemical, and serological profiles.16 In PLTCHC, the earliest features are generally lobular without significant portal changes or changes typical of acute cellular rejection (endothelialitis, cholangitis, and mixed inflammatory cellular infiltrates). This includes centrilobular hepatocyte ballooning degeneration with mild architectural disarray and mild inflammatory activity. The degree of cholestasis can vary with canalicular bile plugging and hepatocellular and/or Kupffer cell bile pigment deposition. Pericellular/perisinusoidal fibrosis is usually minimal in this early stage as well. The progression of PLTCHC typically leads to more prominent portal changes with bile ductular proliferation and portal expansion due to mixed or even neutrophil-predominant inflammation. The degree of fibrosis progresses from early pericellular/perisinusoidal fibrosis to bridging fibrosis and in some cases to fully developed cirrhosis. A significant percentage of cases will exhibit marked parenchymal collapse with confluent necrosis and rapid progression to fibrosis (sometimes with nodular regeneration).15

Most articles that we reviewed included patients with recurrent hepatitis C, which is a less aggressive entity than PLTCHC. Histologically, recurrent hepatitis C may be associated with chronic portal and lobular inflammation and acidophil bodies with or without fibrosis. Chronic graft rejection may also lead to severe cholestasis. However, other histological features of chronic graft rejection, including a paucity of large-caliber bile ducts and foam cell changes in the hepatic arterioles, are not typical features of PLTCHC. In addition, the observation of prominent pericellular fibrosis should be helpful for excluding chronic rejection. Adverse drug reactions are common, and clinical information is critical for excluding a drug reaction as a significant contributing factor to graft dysfunction. A large bile duct obstruction or hepatic artery thrombosis can induce a histological picture similar to that of PLTCHC. Lobular changes, including hepatocellular ballooning, favor PLTCHC over duct obstruction. Duct obstruction and acute cholangitis are more likely to show portal edema and portal neutrophilia.

The risk factors reported to be associated with PLTCHC included the genotype, immunosuppression, and HCV viral load. Surprisingly, seemingly important clinical information was not regularly reported in the studies. For example, HCV genotyping was reported in only half of the articles. Furthermore, the type of immunosuppression and details of clinical outcomes, such as the intervals between transplantation and the development of PLTCHC, the time from transplantation to death, and the time from the diagnosis of PLTCHC to death, were not routinely reported. Data on survival and long-term follow-up were not routinely reported, and findings from this review may not accurately reflect the actual mortality due to PLTCHC. We have proposed clinical information that should be routinely provided in articles on PLTCHC (Table 2). The consistent reporting of significant clinical information in PLTCHC studies may be important for identifying risk factors for the development of PLTCHC and for formulating effective treatment strategies.

The pathogenesis of PLTCHC is poorly understood, but a proposed mechanism involves hepatocyte repopulation and regeneration after liver transplantation.63 Hepatocellular injury from PLTCHC is believed to be a direct cytopathic effect of HCV with a relative absence of inflammation, a high serum viral load, and a large number of HCV virions in hepatocytes.64 The concept of increased vulnerability of the dividing hepatocytes to HCV infection has also been suggested by in vitro studies. HCV replication has been noted to be more efficient in actively proliferating cells in the S and G2/M phases of the cell cycle. It has also been shown that the activity of the internal ribosome entry site that controls HCV translation is significantly increased in regenerating cells versus quiescent cells.63 During transplantation, the HCV-infected liver is removed and replaced by a graft. Consequently, there is a significant decrease in the serum HCV RNA concentration, and this continues after the implantation of the new graft. Despite this initial decrease in the viral load immediately after transplantation, there may be a rapid increase in the hepatitis C viral load within hours of graft reperfusion, and the load may reach pretransplantation levels in only a few days. However, the virus does not follow the same kinetics in all patients, and the clinical outcome depends on factors such as the recipient health status, rejection episodes, viremia levels, viral genotype, and immunosuppression.4, 64–67

The T helper 2 (Th2) immune response and quasispecies (a population of closely related yet unique genomic RNA viral sequences produced over time in an individual) have been implicated in the pathogenesis of PLTCHC.68–71 Studies have demonstrated that recipients with mild recurrent hepatitis C have greater genetic diversification (quasispecies) than recipients with PLTCHC; quasispecies become more homogeneous in patients with severe recurrence, although this association is controversial.71–73 Recipients with PLTCHC express up-regulation of Th2 cytokines, interleukin-10, and interleukin-4 in comparison with patients with recurrent hepatitis C and have higher levels of intrahepatic hepatitis C.74 These data suggest that the Th2 immune response is associated with increased viral replication and graft injury.

We hope that a better understanding of the pathogenesis of PLTCHC will lead to more successful treatment strategies because, in most cases, treatment is unsuccessful. The most common therapeutic approach is antiviral therapy. Although a definitive recommendation for antiviral therapy cannot be made because of limitations of the available data, some patients have a biochemical or virological response, and a trial of antiviral therapy should be considered. Retransplantation seemed to be unsuccessful in most studies; however, follow-up after retransplantation was not routinely provided.

Ideally, it is important to implement measures that may potentially prevent PLTCHC because PLTCHC responds so poorly to current treatment once it has developed. Modifiable risk factors include the therapy for rejection, the donor age, and the degree of immunosuppression. Cyclosporine may have antiviral properties and inhibit HCV replication through inhibition of the nonstructural protein 5B serine protease,75 but an obvious beneficial effect was not evident in our review. A study comparing recipients with recurrent hepatitis to those with PLTCHC did not find an association between the type of immunosuppression or cumulative steroid dose and PLTCHC, although the cumulative steroid dose was associated with the stage of fibrosis.74 Studies have suggested that the use of OKT3, the cumulative steroid dose, and the donor age are associated with more severe recurrent hepatitis C,5, 74, 76, 77 although data specifically for PLTCHC are lacking. Data on the role of immunosuppression and the risk of PLTCHC are lacking, although OKT3 has been associated with an increased risk of severe recurrent hepatitis C.14 Recommendations from the consensus conference include carefully weighing the risk of treating mild rejection with steroid boluses and antilymphocyte therapy against the risk of severe recurrent hepatitis C.

In conclusion, although PLTCHC occurs uncommonly, it poses a challenging therapeutic dilemma. There are no clearly identified risk factors that predict who may develop PLTCHC. Since 2004, the year after the publication of the consensus conference statement, 12 articles have been published on PLTCHC, but little new information has been provided. However, as suggested by the consensus conference statement, a multicenter study will be needed to better understand this disease and develop effective treatment strategies because of the relative rarity of this condition. On the basis of our systematic review, we propose using the consensus criteria14 for diagnosing PLTCHC, and we also propose clinical information that should be routinely reported in future studies of PLTCHC with the hope of improved management of this devastating disease.

Ancillary