Long-Term Outcomes After HCV Liver Transplant
"Delayed Onset of Severe Hepatitis C--Related Liver Damage Following Liver Transplantation: A Matter of Concern?"
Liver Transplantation, Vol 9, No 11 (November), 2003: pp 1152-1158
Marina Berenguer, * Victoria Aguilera, * Martin Prieto, * Domingo Carrasco, *Miguel Rayo´n, + Fernando San Juan, + Carmen Landaverde, * Jose ´ Mir, + and Joaquın Berenguer *
From the *Hepato-Gastroenterology Service, +Pathology Service, and
+Liver Transplantation Unit, Hospital Universitario La Fe, Valencia,
Although histological hepatitis occurs in the majority of hepatitis C virus (HCV)-infected liver transplant recipients, the natural history is highly variable. Whereas progression to cirrhosis occurs in up to 30% after 3 to 7 years, the disease remains stable in another third of patients, in whom protocol liver biopsies might be avoided. However, there is recent concern that with prolonged follow-up, some patients with initial benign recurrence may develop a late-onset aggressive course. Aims of the study are to determine the incidence and factors associated with this
Based on yearly protocol biopsies (median, five biopsies; range, three to seven biopsies), we evaluated the histological outcome of 57 HCV type 1b--infected trans-plant recipients with initial benign recurrence, defined as stable histological state (fibrosis stage F0 or F1) during the first 3 years posttransplantation. Severe late-onset liver damage is defined as progression to F3 or F4 in patients with previous benign recurrence.
Potential predictors of this event include demographics, donor-related factors,
liver enzyme levels at 1 and 3 (or baseline) years posttransplantation, activity grade and fibrosis stage at 1 and 3 years posttransplantation, nonalcoholic steatohepatitis-- related variables occurring within the first 3 years post-transplantation (diabetes, hyperlipidemia, obesity), use of some drugs (renin-angiotensin inhibitors, ursodeoxy-cholic acid), and the advent of any unusual event.
The incidence of severe late-onset liver damage was 35% (n=20). Twelve transplant recipients progressed to F3, whereas 8 transplant recipients progressed to F4. Sudden histological deterioration was observed on postoperative biopsy 5 in 12 patients; biopsy 6 or 7, in 7 patients; and biopsy 4, in 1 patient. Variables associated with this event in univariate analysis were fibrosis stage and activity grade (and its components) at baseline (P < .0001), recipient female gender (P = .04), alanine aminotransferase (ALT) level at 1 year posttransplantation (P =.02), and aspartate aminotransferase (AST) and ALT levels at baseline (P =.008 and P =.005, respectively). By multivariate analysis, only one variable was retained in the model: fibrosis stage at baseline (relative risk, 11; 95% confidence interval, 3 to 41; P =.0007), whereas AST level almost reached statistical significance (P = .07).
In conclusion, delayed HCV-related severe liver damage is not infrequent in transplant recipients with initial benign recurrence, occurring in approximately one third of them. The presence of some degree of fibrosis at baseline appears to predict this sudden change in the natural history of recurrent hepatitis C.
Based on these findings, we recommend continuing protocol biopsies and evaluating potential antiviral therapy in transplant recipients with evidence of some fibrosis (even if it is only portal).
Cirrhosis related to chronic infection with hepatitis C virus (HCV) has emerged as one of the leading indications for orthotopic liver transplantation world-wide,
accounting for approximately 50% of transplantations in the United States and Europe. Unfortunately, recurrence of HCV infection is universal after transplantation and is associated with a reduction in patient and graft survival in most studies. Overall, the course of hepatitis C is accelerated in these patients
compared with that observed in the immunocompetent population both before and after the development of cirrhosis. In one study, median duration from
transplantation to HCV-related graft cirrhosis was estimated to be approximately 9 to 12 years, with a 42% risk for clinical decompensation 1 year after the devel-opment of cirrhosis.
The natural history of posttransplantation hepatitis C is highly variable. Although some patients develop posttransplantation viremia with minimal or no liver
injury in the initial 3 to 5 years posttransplantation, others progress to severe forms of recurrence, including cirrhosis, in the same period. From available data, it appears that approximately one third of patients with recurrent hepatitis C can be classified as fast progressors; another third with benign outcome, at least during the first years posttransplantation, as slow progressors; and the remaining third, as intermediate progressors.4 It remains unclear whether all HCV-infected liver recipients eventually develop progressive fibrosis and whether the rate of fibrosis is constant or changes over time. There is some concern regarding the shape of fibrosis progression and, more specifically, whether unexpected acceleration may occur in transplant recipients with an initial benign outcome.
This information is extremely relevant for transplant centers for two main reasons: (1) to decide the need for protocol liver biopsies beyond postoperative year 5, par-ticularly in patients with previous biopsies showing minimal or mild damage and absence of progression, and (2) to allow early recognition of patients at high risk
for progression so that specific interventions can be implemented, particularly antiviral therapy, when tolerance and efficacy may be the greatest.
Specific aims of the study are to: (1) define whether a sudden histological deterioration or change in rate of HCV-related fibrosis progression exists in the post-transplantation setting; (2) determine the incidence of this event, i.e., late-onset aggressive course in patients with initial benign recurrence; and (3) determine the factors associated with this delayed onset of severe HCV-related liver damage.
Fifty-seven transplant recipients with initial benign HCV type 1b recurrence formed the study group. Initial benign HCV recurrence is defined as the presence of a stable histological state (no or portal fibrosis) during the first 3 years
posttransplantation, assessed through yearly protocol liver biopsies. This cohort was selected from a total of 194 HCV-infected liver transplant recipients who had undergone primary liver transplantation between 1991 and March 1998, after excluding those with comorbid conditions (hepatitis B or biliary complications; n=15), severe early HCV damage (fibrosis stage [F]3 or F4 in the first 3 years; n _ 58), earlier death or retransplantation (n =56), and no histological fol-low-
up after postoperative year 3 (n =8). Thus, only patients who had HCVinfection alone posttransplantation, defined as positivity of serum HCV RNA by reverse-transcription polymerase chain reaction; a minimum histological follow-up of 3
years (4 months) showing stable mild or no hepatitis; and absence of confounding causes of liver injury. The group included 57 Caucasians with a median age at
transplantation of 54 years (range, 30 to 65 years). Seventy-two percent of patients were men. Indications for transplan-tation were either end-stage cirrhosis (n =43) or cirrhosis with hepatocellular carcinoma (n =14). A history of alcohol
abuse before transplantation was present in 10 patients (17.5%). Median donor age was 27 years (range, 9 to 68 years). Patients were followed up until March 2003, death, and/or retransplantation, with a median follow-up of 93 months (range, 60 to 140 months). Ninety-five percent of patients were alive at the time of last follow-up.
Protocol liver biopsies were performed yearly. Additional biopsies were performed when clinically indicated. All biopsy specimens were reviewed by a single pathologist (M.R.) in a blinded fashion, and only those obtained before antiviral therapy was instituted were evaluated in this study. A total of 272 protocol liver biopsies were available for histological analysis, with a median of five specimens per patient (range, three to seven specimens). Sections were stained routinely with hematoxylin- eosin, reticulin, and Perls' and Orcein stains. When
findings were consistent with entities unrelated to HCV infection, such as biliary obstruction, acute or chronic rejection, or alcohol related, biopsy results were not included in the analysis.
Liver biopsies classified as "hepatitis" were scored evaluating both stage of fibrosis and degree of necroinflammatory activity, as previously reported.5 Grade was determined by combining the hepatic activity index scores for periportal
necrosis, lobular degeneration and necrosis, and portal inflammation, defined as follows: 1 to 2, minimal; 3 to 6, mild; 7 to 10, moderate; and 11 to 14, severe. Stage corre-sponded to the original hepatic activity index fibrosis score: 0,
none; 1, fibrous portal expansion; 3, bridging fibrosis; and 4, cirrhosis.
Prevalence of Late-Onset Severe HCV-Related Liver Damage
Twenty patients developed sudden histological deteri-oration
with an abrupt change in the rate of fibrosis progression, characterized by a significant worsening in fibrosis after previous benign histological evolution.
The prevalence of late-onset severe HCV-related liver damage was 35% (20 of 57 patients). Twelve transplant recipients progressed to F3, whereas 8 transplant recip-ients progressed to F4. The abrupt change in histological outcome was observed in postoperative biopsy 5 in 12 patients; postoperative biopsy 6 or 7, in 7 patients; and postoperative biopsy 4, in 1 patient.
Variables Associated With Late-Onset Severe HCV-Related Liver Damage
Univariate analysis. Of multiple variables evaluated by univariate analysis as possible predictors of late-onset severe liver damage, the following were found to be significantly (or close to) associated with outcome: history of acute histological hepatitis (P = .06); fibrosis stage and activity grade at baseline, i.e., in the third year posttransplantation liver biopsy (P=.0001 for both variables); recipient female gender (P = .04); warm ischemic time (P =.07); ursodeoxycholic acid treatment during the first years posttransplantation (P=.08); and liver enzyme levels at 1 and 3 years posttransplantation (P =.02 and =.07 for ALT and AST levels at 1 year; P=.005 and P=.008 for ALT and AST levels at baseline, respectively; Table 1).
Multivariate logistic regression analysis. The nine significant predictors of late-onset severe liver damage in univariate analyses were entered into a logistic regression model. Only one of the predictors identified in univariate associations remained significant in the multivariate model: fibrosis stage at baseline (relative risk, 11; 95% confidence interval, 3 to 41; P=.0007).
Thus, only 5 of 34 patients without fibrosis at baseline (15%) had subsequent onset of severe liver damage. Conversely, 63% of those with F1 seen during the initial
3-year liver biopsy (10 of 16 patients) developed this complication. This percentage increased to 71% in those who had progressed to F1 in postoperative year 3 (i.e., patients with F1 in earlier protocol biopsies; n =7). In addition, although not statistically significant,
there was a trend for AST level at 3 years (or baseline) to enter the model (P=.08). In that sense, only 1 of 16 patients without fibrosis and normal transaminase lev-els (6%) had delayed progression to severe fibrosis in the ollowing years.
HCV infection is an important indication for liver transplantation in most transplant centers. Unfortunately, recurrence of HCV infection is near universal in patients with pretransplantation viremia, and HCV-related disease occurs in the vast majority of patients. The natural history of this hepatitis is characterized by
its progression to cirrhosis in a percentage that ranges between 6% and 23% after a median of 3 years. This course of hepatitis C is clearly accelerated compared
with that observed in the nontransplantation population. However, and in a way similar to what we know from the immune competent population, the natural
history of recurrent hepatitis C is highly variable.
Although a third of patients progress to cirrhosis within the first years of transplantation, another third remain stable with near-normal livers during this period, and the remainder may be considered to have an inter-mediate
rate of progression. Whether patients with initial benign HCV recurrence
will follow the same course over the years is, at present, unknown. Recent data from the immune competent population suggest there is a change in the shape
of fibrosis progression with time of infection, so that progression is slower early after infection and becomes faster late after infection.10 Information regarding this
event in the liver transplantation population is extremely important given the lower efficacy and tolerance of antivirals 9 and the typically shorter time to
intervene compared with immune competent patients.
The main aims of this study thus are to assess whether acceleration of fibrosis progression occurs in transplant recipients with an initial benign outcome and define the profile of this type of patient.
The main findings of this study can be summarized as follows: (1) the rate of fibrosis is not constant over time in all patients; progression from F0 to F1 takes
longer than from F1 to F3 or F4; (2) this acceleration of fibrosis progression with sudden histological deterioration occurs in 35% of transplant recipients with initial
benign outcome; and (3) predictors of late-onset severe HCV-related liver damage include stage of fibrosis (portal versus none) observed in baseline protocol liver
Based on our findings, a patient at high risk for developing sudden or abrupt deterioration with late-onset HCV type 1b--related severe fibrosis would be a
transplant recipient with at least some degree of fibrosis in the third-year liver biopsy, regardless of baseline immunosuppression and age. Additional variables
potentially associated with this outcome include serum transaminase levels at 1 and 3 years posttransplantation, recipient female gender, presence of moderate to severe activity at baseline, and a history of acute histological hepatitis posttransplantation. However, because the latter associations did not reach statistical significance in multivariate analysis, they need to be confirmed
using larger databases. Thus, patients with high amino-transferase levels and more severe necroinflammatory changes on the initial biopsies appear to be more likely
to have worsening fibrosis than patients with lower transaminase levels and less hepatic inflammation.
These results are very similar to those previously reported in the nontransplantation population. The potential of liver histological characteristics for predicting outcome has been described previously in both immune competent patients and liver transplant recipients. We and others previously have shown that
progression to graft cirrhosis within the first 5 years is significantly more common in patients with moderate or severe activity grade in the first-year liver biopsy than in those with no or mild activity. In that sense, our findings confirm the importance of liver biopsies in predicting outcome, including prediction of a late
change in rate of fibrosis progression, from benign or stable course to accelerated progression.
Our findings also suggest that both donor selection and early posttransplantation management do not have a major role in late-onset disease progression, results at
odds with prediction of early disease progression. However, given the relatively small sample size, we believe these results need to be confirmed with larger
In conclusion, sudden worsening of disease with late-onset histological deterioration is not infrequent and may occur in up to one third of transplant recipients with initial benign HCV recurrence. Thus, progression to cirrhosis occurs at different rates in trans-plant recipients. Fibrosis appears to progress more
rapidly at some times, even in transplant recipients with an initial benign course. This event correlates best with the presence of at least some fibrosis, degree of necroin-flammatory activity on liver biopsy, and transaminase
Based on these findings, we recommend continuing protocol liver biopsies beyond the first years in patients with some degree of fibrosis and moderate to severe
necroinflammatory activity because these patients are at risk for developing late fibrosis. The lack of progression in the majority of patients with no fibrosis at 1 year, particularly those with transaminase levels within nor-mal range, could remove the need for continuing annual protocol biopsies in this group. Conversely, in those with elevated aminotransferase levels, the decision to continue performing liver biopsies should be on an individual basis because approximately a quarter of these patients may develop delayed-onset severe liver damage. In this subset of patients, protocol biopsies could potentially be performed with a lower frequency.
In addition, and because antiviral therapy appears to be better tolerated and more effective if started when the disease is at its early stages, therapy should be proposed to these patients.