Since the first study to document recurrent hepatitis C virus (HCV*) infection after liver transplantation (1), it has become recognized that viremia is universal and that the majority of HCV-seropositive liver transplant recipients will eventually develop histologic evidence of HCV recurrence (2-4). The spectrum of allograft injury ranges from mild hepatitis to graft failure requiring repeat transplantation (5). The classic initial histologic appearance of recurrent HCV includes portal and parenchymal mononuclear infiltrates of varying intensity, acidophilic necrosis, bile duct damage, and steatosis (6). Atypical changes, which may be difficult to distinguish from rejection, obstruction, or ischemia, include extensive hepatocyte swelling or acidophilic necrosis with minimal inflammation(7). In addition, similar to what has previously been described with hepatitis B recurrence, HCV is occasionally manifested by a severe and progressive form of cholestatic hepatitis with atypical histologic features (8, 9). Therefore, it may be appropriate to include cholestasis grading when evaluating allograft specimens with HCV as an adjunct to the standard classifications of histologic grading (i.e., Knodell and Scheuer scores). The aim of this study was to semiquantitatively define histologic and biochemical features at the time of initial HCV recurrence that predict the course of patients with long-term follow-up.
PATIENTS AND METHODS
The study group comprised 34 consecutive patients (beginning in 1988) who underwent orthotopic liver transplantation (OLT) with cyclosporine-based immunosuppression for end stage liver disease secondary to chronic hepatitis C and who developed histologic evidence of HCV recurrence, with a minimal of 1 year of histologic follow-up. HCV infection was confirmed before and after OLT by polymerase chain reaction in all patients, as described previously(14). The 34 study patients were selected from a larger group of 50 consecutive HCV-positive liver transplant recipients; the remaining 16 were excluded because they either lacked evidence of histologic HCV recurrence or failed to survive for longer than 1 year. Informed consent was obtained from all patients, and the study protocol was approved by our institutional review board for human studies.
Liver biopsies were obtained by protocol at 7, 14, and 365 days, as indicated by an elevation from the baseline liver function biochemistries, and after treatment for acute cellular rejection. All liver transplant recipients at our program follow a protocol schedule to check their serum chemistries(including liver enzymes): twice weekly laboratory tests for the first 3 months, weekly tests for the next 3 months, and tests every other week between 6 and 12 months after transplantation. A 15% increase in any of the liver chemistry values usually warrants a liver biopsy to evaluate for rejection, cytomegalovirus, or recurrent HCV disease. Two hundred and seventy-eight serial allograft biopsies (mean: 6.85±0.62 per patient, range: 4-21), including 176 nonprotocol biopsies, were analyzed. Histologic HCV recurrence was diagnosed based on the following: portal and lobular infiltration by mononuclear cells, piecemeal necrosis, and/or acidophil bodies in the absence of any other specific causes. Disease severity was determined by the hepatic activity index (HAI) system of Knodell et al. (10), applied to each biopsy in a blinded fashion by two pathologists, and a consensus score was assigned. The HAI was used to quantitate four major elements of the biopsy, with a total score (0-22) and separate scores for amount of piecemeal necrosis (0-10), intralobular hepatocyte degeneration(0-4), portal inflammation (0-4), and hepatic fibrosis (0-4). The Scheuer score (11) for grading of necroinflammatory activity and staging of fibrosis was also applied to each allograft specimen, with excellent correlation between the HAI and the Scheuer score(rs=0.930, P<0.0001). For the purposes of this study, only the HAI data are presented. Hepatocyte ballooning degeneration and cholestasis were coded: 0 (none) or 1 (ballooning and/or canalicular plugging in zones 2 or 3 or panacinar). If these changes were noted in the first month after OLT and, therefore, potentially attributable to preservation injury, they were not considered a manifestation of HCV recurrence and were excluded from analysis. The final allograft specimen was the most recent biopsy, at the time of last follow-up, subsequent transplantation, or death.
HCV viremia (serum specimens frozen at -70 °C) was assessed by reverse transcription-polymerase chain reaction using conserved 5′-noncoding region primers and a 32P-labeled internal oligonucleotide probe as described previously (14). The reverse transcription-polymerase chain reaction assay sensitivity is less than 100 copies HCV RNA/ml of patient serum. HCV genotype was by determined by restriction fragment length polymorphism of the 5′-noncoding region, as described by Davidson et al. (12), and HCV genotypes were classified according to the Simmonds (13) nomenclature.
Using uni- and multivariate analyses, we studied the following variables in an attempt to predict long-term histologic scores: time to the onset of histologic HCV recurrence, biochemical values at the onset of recurrence and peak values in the ensuing month (i.e., aspartate aminotransferase [AST; in U/L] and peak AST; total serum bilirubin [in mg/dl] and peak total bilirubin[pT.Bili]), and total and subitem histologic scores. Findings not attributable solely to HCV recurrence (e.g., concurrent rejection, cytomegalovirus infection, and biliary obstruction) were excluded from analysis.
One-way analysis of variance and the Mantel-Haenzel chi-square test were used for statistical comparison of means (±SEM) and proportions between groups, respectively. The Kaplan-Meier product-limit estimate was used for the univariate analysis of time-dependent events (time to development of moderate-to-severe allograft hepatitis, i.e., piecemeal necrosis and/or lobular inflammation subscores ≥3) with comparison between groups performed via the log-rank test. The correlations among biochemical values, interval to histologic recurrence, and histologic severity scores were assessed by Spearman rank correlation (rs) coefficients. All factors found to be at least marginally associated with progression to allograft cirrhosis (P<0.15) were tested by multiple logistic regression analysis with allograft cirrhosis coded as a dichotomous outcome variable. AP- value of less than 0.05 was considered significant. The JMP (SAS Institute Inc, Cary, NC) statistical package was used.
The range of follow-up from date of OLT to last biopsy was 365-2269 days. The mean time to histologically documented recurrence was 204.9±42.6 days. The proportion of patients with mild, moderate, and severe histologic scores on final biopsy were 38% (n=13), 38% (n=13), and 24% (n=8), respectively. Allograft cirrhosis developed at a mean of 638.3±113.5 days. Table 1 outlines the characteristics of patients who developed allograft cirrhosis (group 1) versus patients without cirrhosis(group 2), with statistically equivalent mean follow-up periods in both groups(P=0.14). Moreover, the mean number of allograft biopsies per patient was comparable in both groups (8.2±1.26 vs. 6.4±0.70;P=0.21). Seven of the 8 group 1 patients had a history of ethanolism before OLT, as did 20 of the 26 group 2 patients (P=NS). HCV genotypes for the entire study group were as follows: 17 (50%) genotype 1a, 12(35%) genotype 1b, 2 (6%) genotype 2a, 1 (3%) genotype 2b, 1 (3%) genotype 3a, and 1 (3%) mixed 1b/2a. There was no statistically significant difference in the distribution of genotypes between groups 1 and 2 (P=0.87). Thirteen (38.3%) of the 34 patients received treatment for acute allograft rejection, and there was no difference in the rate of rejection(P=0.12) or the cumulative steroid dose (P=0.16) between both groups.
The mean and median intervals to the onset of histologic recurrence were comparable between patients in both groups. A slightly higher proportion of patients who eventually developed allograft cirrhosis had demonstrated initial histologic recurrence within 90 days after OLT, but the difference was not statistically significant (5 of 8 [62.5%] vs. 11 of 26 [42.3%];P=0.42) Moreover, linear regression did not demonstrate any significant relationship between interval to the onset of recurrence and either initial or final HAI (including subitem scores). At the time of recurrence, patients in group 1 had higher mean total serum bilirubin and peak total bilirubin (pT.Bili) than patients in group 2. Marked, transient hyperbilirubinemia after recurrence was not uncommon, and only in half of the patients did the bilirubin value at the onset of recurrence represent the peak value in the first month. Six of the eight patients who developed cirrhosis eventually normalized their serum bilirubin (mean: 247.6 days). Moreover, actuarial rates of moderate-to-severe hepatitis were significantly greater in patients with pT.Bili ≥3.5 mg/dl (at 3 years, 86% vs. 43% in patients with pT.Bili <3.5, P=0.004, log-rank test; Fig. 1).
There was no difference between groups 1 and 2 with regard to total HAI of the initial biopsy at the time of recurrence. The initial fibrosis subscore was higher in group 1 patients (P=0.045); the proportion of patients with any evidence of fibrous portal expansion at onset of recurrence was higher for group 1 patients, although the difference was not significant (50% vs. 23%, P=0.19; Fisher's exact test). The difference in mean piecemeal necrosis subitem score approached statistical significance(P=0.06). The necroinflammatory grade, a summated score of portal/periportal and lobular necroinflammation (14), at the time of initial histologic recurrence was equivalent in both groups(P=0.25). Moreover, the presence of early ballooning degeneration and/or cholestasis was detectable in 7 of 8 (88%) group 1 patients versus 12 of 26 group 2 patients (46%) (P=0.04, Fisher's exact test; seeFig. 2).
The aspartate aminotransferase/alanine aminotransferase (AST/ALT) ratio at the time of initial HCV reinfection of the allograft did not correlate with initial histologic scores, but there was a statistically significant linear relationship with ultimate HAI (Fig. 3). Because the variables studied were mutually correlated, statistically significant variables were studied by multivariate regression analysis. The pT.Bili (i.e., the peak serum bilirubin in the first month after recurrence entered as a continuous covariate) was the only independent predictor of progression to allograft cirrhosis (P=0.012).
Four patients in the first group developed graft loss from severe recurrent HCV: two underwent repeat transplantation, and two died of graft failure(including one patient who was precluded from consideration for repeat transplantation because of recidivant injection drug use. There were two deaths in group 2, both unrelated to recurrent HCV. Therefore, the rate of graft loss in group 1 was 50% (4 of 8) vs. 7.7% (2 of 26)(P=0.018).
Despite our current understanding of HCV-related allograft injury, including its high rate of occurrence and potential progression to cirrhosis, the pathological evolution and underlying mechanisms of this disease are incompletely understood. No study to date has attempted to define the features of initial HCV recurrence that are predictive of long-term allograft injury. In this study, we evaluated numerical scoring systems for grading histologic lesions (including necroinflammation and ballooning degeneration/cholestasis) as well as biochemical features and time to initial histologic recurrence to define predictive parameters. Potential confounding variables including HCV genotype, follow-up after OLT, the number of allograft biopsy specimens per patient, basal immunosuppression, prior ethanol use, and incidence of rejection (and cumulative steroid dose for rejection) were carefully evaluated. Univariate analysis identified the total serum bilirubin and the peak value in the ensuing month after HCV recurrence, the AST/ALT ratio, the fibrosis subitem score at the onset of recurrence, and the presence on histology of ballooning degeneration and/or cholestasis as being prognostically significant. All the allograft specimens were assigned a score according to the criteria devised by Knodell. In terms of disease activity, the HAI is a useful means of assessing the degree of histologic abnormality, but its true role is in comparative studies of disease progression or regression, particularly in response to specific treatment regimens(15). Having been devised for the assessment of chronic active hepatitis, the HAI is of relatively less value in milder degrees of inflammation (16). Moreover, the histologic appearance of HCV infection of the allograft may be modified by immunosuppression, as a result of its potential to enhance viral replication as well as ameliorate the inflammatory response.
Conventional scoring systems do not take into account an array of histologic features seen frequently in HCV-infected allograft specimens(cholestasis, ballooning degeneration, bile duct injury, and steatosis). In our series, hepatocyte ballooning degeneration (i.e., cellular swelling with cytoplasmic material clumped around the nucleus with lucent peripheral areas, particularly in centrilobular zones) and/or bile stasis were present in the majority (88%) of patients who went on to develop cirrhosis and in just under half of the patients who did not. As these features are not part of the classic scoring systems for chronic hepatitis, they should be considered an adjunct for classification of postOLT histologic severity. A number of factors have been postulated to cause cholestasis including inflammatory cytokines such as tumor necrosis factor-α and interleukin-6, which act on the basolateral membrane of hepatocytes and interfere with bile salt transport(17). A recent study demonstrated high circulating levels of these cytokines in liver transplant recipients with recurrent HCV(18). The exact role of these cytokines in the development of allograft injury, including development of cirrhosis, merits further investigation. In the current study, marked and transient hyperbilirubinemia commonly occurred within the first month after acute HCV reinfection of the allograft. Furthermore, the mean initial total serum bilirubin and peak bilirubin in the ensuing month were significantly higher in patients who ultimately developed cirrhosis. Peak bilirubin (pT.Bili) ≥3.5 mg/dl was associated with a greater likelihood of, and shorter interval to, the development of moderate-to-severe histologic hepatitis. Moreover, by multivariate logistic regression analysis, pT.Bili had prognostic significance for the development of cirrhosis. Recently, Schluger et al.(9) reported a severe cholestatic variant of recurrent HCV, but in all patients, once the bilirubin exceeded 3 mg/dl, it increased relentlessly. In our analysis, six of the eight (75%) patients who developed cirrhosis had eventually normalized their serum bilirubin (mean: 247.6 days).
Several other results of this study merit special mention. Although it seems intuitively logical that earlier recurrence would eventually lead to more severe histology, we were unable to demonstrate any correlation between the temporal development of recurrence and ultimate histologic severity (with time to recurrence coded both as continuous and grouped variables). A statistical difference might have been missed because of our sample size, and a larger study is under way to further address this issue. In addition, there appeared to be no significant relationship between liver enzymes (AST, ALT, alkaline phosphatase) and HCV recurrence. Indeed, a previous study by Schiffman et al. (19) showed that over half of patients with recurrent histologic HCV had prolonged periods during which ALT remained within the normal limits. The ratio of AST to ALT in non-transplant patients with viral hepatitis is typically less than 1.0 but rises, often to values greater than 1.0, as cirrhosis develops (20). Although we observed a linear relationship between the AST/ALT ratio at the onset of histologic recurrence and degree of ultimate hepatic injury, this relationship(rs) was weak and probably of little clinical value.
In summary, histology continues to be the gold standard to appropriately assess the degree of allograft injury; however, in the HCV-infected transplant recipient, atypical features need to be considered. Based on the results of this study, recognition of cholestasis and transient hyperbilirubinemia is prognostically significant. Defining early parameters predictive of histologic progression will, it is hoped, identify appropriate candidates for future antiviral therapy.
1. Martin P, Munoz SJ, Di Bisceglie AM, et al. Recurrence of hepatitis C virus infection after orthotopic liver transplantation. Hepatology 1991; 13: 719.
2. Gretch DR, Bacchi CE, Corey L, et al. Persistent hepatitis C virus infection after liver transplantation: clinical and virological features. Hepatology 1995; 22: 1.
3. Chazouilleres O, Kim M, Combs C, et al. Quantitation of hepatitis C virus RNA in liver transplant recipients. Gastroenterology 1994; 106: 994.
4. Feray C, Gigou M, Samuel D, et al. The course of hepatitis C virus infection after liver transplantation. Hepatology 1994; 20: 1137.
5. Rosen HR, O'Reilly PM, Shackleton CR, et al. Graft loss following liver transplantation in patients with chronic hepatitis C. Transplantation 1996; 62: 1724.
6. Araya V, Rakela J, Wright TL. Hepatitis C after orthotopic liver transplantation. Gastroenterology 1997; 112: 575.
7. Ferrell LD, Wright TL, Roberts J, Ascher N, Lake J. Hepatitis C viral infection in liver transplant recipients. Hepatology 1992; 16: 865.
8. Lim HL, Lau GK, Davis GL, Dolson DJ, Lau JYN. Cholestatic hepatitis leading to hepatic failure in a patient with organ-transmitted hepatitis C virus infection. Gastroenterology 1994; 106: 248.
9. Schluger LK, Sheiner PA, Thung SN, et al. Severe recurrent cholestatic hepatitis C following orthotopic liver transplantation. Hepatology 1996; 23: 971.
10. Knodell RG, Ishak KG, Black WC, et al. Formulation and application of a numerical scoring system for assessing histologic activity in asymptomatic chronic active hepatitis. Hepatology 1981; 5: 431.
11. Desmet VJ, Gerber M, Hoofnagle JH, Manns M, Scheuer PJ. Classification of chronic hepatitis: diagnosis, grading and staging. Hepatology 1994; 19: 1513.
12. Davidson F, Simmonds P, Ferguson JC, et al. Survey of major genotypes and subtypes of hepatitis C using RFLP of sequences amplified from the 5′ noncoding region. J Gen Virol 1995; 76: 1197.
13. Simmonds P, Holmes EC, Cha TA, et al. Classification of hepatitis C into six major genotypes and a series of subtypes by phylogenetic analysis of the NS-5 region. J Gen Virol 1993; 74: 2391.
14. Yano M, Kumada H, Kage M, et al. The long-term pathological evolution of chronic hepatitis C. Hepatology 1996; 23: 1334.
15. Hytirolglou P, Thung SN, Gerber MA. Histologic classification and quantitation of the severity of chronic hepatitis: keep it simple! Semin Liver Dis 1995; 15: 414.
16. McMahon RFT, Yates AJ, McLindon J, Babbs C, Love EM, Warnes TW. The histopathological features of asymptomatic hepatitis C virus-antibody positive blood donors. Histopathology 1994; 24: 517.
17. Green RM, Crawford JM. Hepatocellular cholestasis: pathobiology and histologic outcome. Semin Liver Dis 1995; 15: 372.
18. Fried MW, Seeff LC, Lam L, Gutekunst K, Hodge T, Boyer TD. Serum cytokine levels in patients with recurrent hepatitis C after liver transplantation [Abstract]. Hepatology 1995; 22: 107A.
19. Schiffman ML, Contos MJ, Luketic VA, et al. Biochemical and histologic evaluation of recurrent hepatitis C following orthotopic liver transplantation. Transplantation 1994; 57: 526.
20. Friedman LS, Martin P, Munoz SJ. Liver function tests and the objective evaluation of the patient with liver disease. In: Zakim D, Boyer TD, eds. Hepatology: a textbook of liver disease. Philadelphia: WB Saunders, 1996: 791.