Natural History of Clinically Compensated Hepatitis C Virus Related Graft Cirrhosis After Liver Transplantation

Transcription

1 Natural History of Clinically Compensated Hepatitis C Virus Related Graft Cirrhosis After Liver Transplantation MARINA BERENGUER, 1 MARTÍN PRIETO, 1 JOSÉ M. RAYÓN, 2 JULIO MORA, 1 MIGUEL PASTOR, 1 VICENTE ORTIZ, 1 DOMINGO CARRASCO, 1 FERNANDO SAN JUAN, 3 MANUEL-DE-JUAN BURGUEÑO, 3 JOSÉ MIR, 3 AND JOAQUÍN BERENGUER 1 The natural history of clinically compensated hepatitis C virus (HCV) cirrhosis after liver transplantation is unknown. This information is relevant to transplant centers to improve the management of these patients and decide the optimal timing for retransplantation. The aims of the study were (1) to describe the natural history of patients with HCV-cirrhosis transplants in a center with annual liver biopsies, and (2) to determine predictors for clinical decompensation, retransplantation, and mortality rates. A total of 49 patients with HCV-graft cirrhosis, 39 clinically compensated at histologic diagnosis of cirrhosis (post liver transplantation cirrhosis) were included and followed up for 1 year (15 days-3.5 years). All patients tested were infected with genotype 1b. Predictive variables included histologic activity index (HAI) at post liver transplantation cirrhosis, liver function tests, age, sex, and maintenance immunosuppression. Eighteen of 39 patients developed at least 1 episode of decompensation after a median of 7.8 months (4 days-2.6 years; 93% ascites). The cumulative probability of decompensation was 8%, 17%, and 42% at 1, 6, and 12 months, respectively. Graft and patient survival rates were 100%, 85%, and 71% and 100%, 92%, and 74% at 1, 6, and 12 months, respectively. Patient survival rates dropped significantly once decompensation developed (93%, 61%, and 41% at 1, 6, and 12 months, respectively). Variables associated with decompensation, retransplantation, and mortality rate included a high Child-Pugh score (>A), low levels of albumin at post liver transplantation cirrhosis, and a short interval between liver transplantation and post liver transplantation cirrhosis. The natural history of clinically compensated HCV-graft cirrhosis is shortened when compared with immunocompetent patients. If retransplantation is considered, it should be performed promptly once decompensation develops. (HEPATOLOGY 2000;32: ) Hepatitis C virus (HCV)-associated end-stage liver disease is a leading diagnosis in patients undergoing liver transplantation, accounting for half of transplantations in many centers. 1 Despite universal viral recurrence, 2 early posttransplantation infection generally results in indolent disease with good graft and patient survival rates, at least for the first 5 to 7 years. 3,4 The full consequences of HCV recurrence with longer follow-up are not as benign, 5-7 and may ultimately result in reduced graft and patient survival rates compared with patients transplanted for nonviral causes. Although the natural history of HCV infection in liver transplant recipients has been investigated by several centers, to date these studies have only evaluated the evolution of hepatitis C until the development of graft cirrhosis. Based on these studies, there is a consensus emerging that recurrent HCV infection results in liver failure in a significant proportion of patients and that the time course over which this progression occurs is shorter than in the immunocompetent population. 7 Data are lacking, however, on the outcome of liver transplant recipients who have reached the stage of fibrosis 4 or cirrhosis 5,8 in their graft. In the immunocompetent population, data from several large studies suggest that the outcome of compensated patients with HCV cirrhosis is sufficiently good, at least during 5 years, to favor observation over transplantation This information is extremely relevant to transplant centers because liver transplantation should only be considered a therapeutic option when the course of the disease is sufficiently advanced that medium-term survival is unlikely without this intervention. The same statement applies to liver transplant recipients who have developed HCV-related graft cirrhosis and who may be given the option of retransplantation. Hence, the purpose of this study was to assess the outcome of clinically compensated HCV-related graft cirrhosis in a center in which protocol annual liver biopsies are performed routinely among HCVinfected recipients. Abbreviations: HCV, hepatitis C virus; HBV, hepatitis B virus; HCC, hepatocellular carcinoma; ALT, alanine transaminase; AST, aspartate transaminase; HAI, histologic activity index. From the 1 HepatoGastroenterology Service, 2 Pathology Service, 3 Liver Transplantation and Surgery Unit, Hospital Universitari La Fe, Valencia, Spain. Received May 2, 2000; accepted July 19, Supported in part by grant FIS 97/0569 from the Fondo de Investigaciones Sanitarias, Ministerio de Sanidad, Spain. Address reprint requests to: Marina Berenguer, M.D., Servicio de HepatoGastroenterología, Hospital Universitario La Fe, Avenida Campanar, 21, Valencia, Spain. jbl01v@nacom.es; fax: (34) Copyright 2000 by the American Association for the Study of Liver Diseases /00/ $3.00/0 doi: /jhep PATIENTS AND METHODS Patients. Between January 5, 1991, and December 29, 1999, 557 patients underwent 598 orthotopic liver transplants at our institution. Two hundred and eighty-five of these patients underwent transplantation for cirrhosis secondary to chronic HCV infection, defined by the presence of anti-hcv antibody by EIA-II and confirmed by a RIBA-II assay. The term posttransplantation cirrhosis was defined as the time of diagnosis of fibrosis 4 5,8 in 1 of the liver biopsies performed regularly in HCV-infected recipients. Only patients who had HCV infection posttransplantation defined as positivity of serum HCV RNA by reverse transcription polymerase chain reaction and absence of biliary complications were included in this retrospective study. In 51 of the patients, a stage of fibrosis 4 was detected during

2 HEPATOLOGY Vol. 32, No. 4, 2000 BERENGUER ET AL. 853 the histologic follow-up. Two patients were excluded from the analysis because of hepatitis B virus (HBV) coinfection before and/or after the diagnosis of HCV-related graft cirrhosis. One of the excluded patients died, the other underwent retransplantation. The causes of death and retransplantation were cirrhosis of the graft related to both HBV and HCV infection. Therefore, 49 patients (27 men, 22 women; median age at transplantation 56 years, range years) with fibrosis 4 in at least 1 liver biopsy were included in this study and formed the study group. The indications for the original transplantation were either end-stage HCV-related cirrhosis (n 34) or HCV-related cirrhosis with hepatocellular carcinoma (HCC) (n 15). In 2 patients, alcohol contributed to liver injury before transplantation. Induction immunosuppression consisted in triple therapy in 39 patients, whereas 10 recipients had been treated with tacrolimus and prednisone. Ten of these patients had already developed clinical decompensation at the time of histologic diagnosis of cirrhosis, including the presence of ascites and/or edema (n 8), or hydrothorax (n 2). One of the patients who had presented with ascites as the first sign of decompensated graft cirrhosis was diagnosed with spontaneous bacterial peritonitis at the same time. Although these patients do not fall into the category of compensated cirrhosis and, hence, are not the strict focus of this study, their outcome will be reviewed. Follow-up. All patients were examined at least once every 1.5 months. The liver function tests were assessed on each of these visits. Abdominal ultrasonography and alpha-fetoprotein serum levels were performed every 4-6 months or when clinically indicated. The follow-up was terminated at the time of either the patient s death, retransplantation, or at the end of the observation period (April 2000). Definitions. The end-points chosen included: (1) clinical decompensation; (2) inclusion on the waiting list for retransplantation; (3) retransplantation; and (4) death. In patients treated with antiviral therapy, these end-points were assessed both before and after treatment was initiated. Clinical decompensation was defined as development of ascites with or without peripheral edema, hydrothorax, variceal bleeding, spontaneous bacterial peritonitis, or encephalopathy during the follow-up period. Criteria used for relisting is similar to that used for original listing. In addition, patients with a duration between the original transplantation and the need for retransplantation of less than 1 year are not considered suitable for relisting. The cause of death was categorized as causes related to HCV-cirrhosis including perioperative death after retransplantation for HCV-related graft cirrhosis, or liver disease unrelated. Liver retransplantation was considered as death for the calculation of graft survival rates. Predictive Factors for Morbidity and Mortality. Clinical, biochemical, and histologic variables were analyzed as predictors of decompensation, retransplantation, and/or death. These included sex, age at diagnosis of cirrhosis, age at transplantation, presence of HCC at transplantation, histologic activity at diagnosis of cirrhosis, Child-Pugh score at the time of diagnosis of graft cirrhosis and at the time of first clinical decompensation, liver biochemistry (serum values of alanine and aspartate transaminase [ALT, AST], gamma glutamyl transpeptidase, albumin, bilirubin, prothrombin time, cholesterol) at the time of diagnosis of graft cirrhosis and at the time of first decompensation, and type of initial decompensation. Genotype was not considered a potential predictive factor because all tested patients (n 39) were infected with genotype 1b. Because antiviral therapy was only administered after retransplantation, this variable was not included in the list of potential risk factors. Serologic and Virologic Assays. Anti-HCV was determined by using a commercial second-generation enzyme-linked immunosorbent assay (ELISA 2.0; Ortho Diagnostic System, Raritan, NJ). Positive serum samples were confirmed by using a second-generation recombinant immunoblotting assay (RIBA 2.0; Chiron Corporation, Emeryville, CA). Qualitative HCV RNA was determined by a nested reverse transcription polymerase chain reaction by using primers from the 5 untranslated region of the HCV genome, as previously described. 5 Viral genotype was assessed in serum samples obtained after transplantation by using a line probe assay (Inno-Lipa HCV; Innogenetics, Zwijndrecht, Belgium), as previously described. 5 Immunosuppression. Induction immunosuppression consisted of (1) standard triple therapy with cyclosporine, azathioprine (1-2 mg/ kg/d), and methylprednisolone given intravenously with tapering of the dose from 200 mg to 20 mg at day 6, at which time 20 mg/d of prednisone was administered orally; or (2) dual therapy with tacrolimus and methylprednisolone. At the time of diagnosis of graft cirrhosis, 8 patients were still receiving prednisone (median dose, 5 mg; range, mg), and none was receiving azathioprine. Histologic Assessment. Protocol liver biopsies were performed yearly for the first 5 years posttransplantation. Additional liver biopsies were performed for unexplained abnormalities in liver tests (serum transaminase values at least 1.5 times the upper limit of normal) and when clinically indicated. Diagnosis of HCV-related graft cirrhosis was made in all 39 compensated patients by means of a protocol liver biopsy. All biopsy specimens were reviewed by a single pathologist (J.M.R.) in a blinded fashion. The specimens were scored according to a slight modification of the histologic activity index (HAI) proposed by Knodell and their histologic grade (activity) and stage (fibrosis) were evaluated separately. 5,8 The grade was determined by combining the HAI scores for periportal necrosis, scored 0 to 6 (0, none; 1, mild piecemeal necrosis; 3, moderate piecemeal necrosis; 4, marked piecemeal necrosis; 5, moderate piecemeal necrosis plus bridging necrosis; 6, marked piecemeal necrosis plus bridging necrosis), lobular degeneration and necrosis (0-4), and portal inflammation (0-4), and was defined as follows: 1 to 2, minimal; 3 to 6, mild; 7 to 10, moderate; 11 to 14, severe. Statistical Analyses. Categoric data were compared by using a 2 test or Fisher s exact test when indicated. Continuous variables were expressed as mean SD and compared by Student s t test. When a normal distribution could not be assumed, continuous variables were summarized as median and ranges and compared by the Mann- Whitney test. Kaplan Meier curves were calculated to estimate patient and graft survival over time, and to estimate the cumulative probability of developing liver decompensation and undergoing liver retransplantation. The log-rank test was used to compare probability curves. A P value of.05 or less was considered significant. RESULTS Clinical Features of Patients at Entry. Forty-nine patients with a diagnosis of graft cirrhosis made at a median time of 2 years (range, 4 months-7 years) were included in the study. Table 1 summarizes their clinical features. The initial laboratory results including liver function tests, median doses, and trough levels of both cyclosporine and tacrolimus are also shown in Table 1. None of the patients had been treated with either interferon or ribavirin before the diagnosis of graft cirrhosis. Thirty-nine patients (80%) were clinically compensated at the time of histologic diagnosis of fibrosis 4, whereas 10 patients had already developed a decompensation at a median of 67 days earlier (range, days). The type of decompensation was ascites in the majority of these patients (n 8). As expected, the only differences between the group of compensated patients and those with decompensated cirrhosis at the time of posttransplantation cirrhosis were the Child-Pugh score and bilirubin levels, which were significantly higher in the latter group, and AST values and prothrombin time (%), which were lower in decompensated patients (Table 2). Clinical Outcome of the Study Population. The evolution of the 49 patients is summarized in Fig. 1. The median follow-up

3 854 BERENGUER ET AL. HEPATOLOGY October 2000 TABLE 1. Clinical Features in Patients With HCV-Related Graft Cirrhosis at Posttransplantation Cirrhosis Median age at post-olt cirrhosis* (yr) 58 (28-68) Median age at transplantation (yr) 56 (25-66) Sex distribution (% men) 27/22 (55% men) Induction immunosuppression (cyclosporine vs. tacrolimus) 39 (80%)/10 (20%) HCC at transplantation 15 (31%) Alcohol intake before transplantation 2 (4%) HAI 11 (7-14) Periportal necrosis 5 (3-6) Lobular degeneration and necrosis 3 (1-4) Portal inflammation 4 (1-4) Child-Pugh score 5 (5-11) ALT (U/L) 131 (40-631) Aspartate transaminase (U/L) 141 (31-565) Bilirubin (mg/dl) 1.5 ( ) Albumin (g/dl) 3.8 ( ) Cholesterol (mg/dl) 166 (59-664) Prothrombin time (%) 89 (26-100) Cyclosporine dose (mg/d) 100 (50-400) Cyclosporine trough levels (ng/ml) 126 (60-342) Tacrolimus dose (mg/d) 1 (0.5-5) Tacrolimus trough levels (ng/ml) 9 (4.5-16) Prednisone dose (mg/d) 5 (2.5-5) Time from OLT to post-olt cirrhosis (mo) 24.9 ( ) NOTE. N 19. Normal values for alanine and aspartate aminotransferases (0-40 U/L). Normal values for total bilirubin ( mg/dl). To convert the values for total bilirubin to mol/l, multiply by Abbreviations: OLT, liver transplantation; HCC, hepatocellular carcinoma; HAI, histologic activity index; ALT, alanine transaminase. *The time when histologic diagnosis of HCV-related graft cirrhosis was made. until the last observation period, including the postretransplantation follow-up, was 1.1 years (range, 17 days-3.6 years). If we consider the time of retransplantation as the last follow-up in those who underwent retransplantation for HCVrelated graft cirrhosis, the median follow-up was slightly shorter (11.3 months; range, 7 days-3.2 years). Clinical decompensation. Twenty-eight patients (57%) developed at least 1 episode of decompensation, which happened before the histologic diagnosis of graft cirrhosis in 10 patients and after that diagnosis in 18 patients. The evidence of decompensation included ascites with/without edemas in the vast majority of patients (n 25; 93%). Three of these patients also presented with encephalopathy as the first decompensation episode. The remaining 3 patients presented with hydrothorax (n 2) and encephalopathy (n 1). The median Child-Pugh score among decompensated patients at the time of the first decompensation was 9.5, ranging from 6 to 12. Twenty-one patients with histologic diagnosis of graft cirrhosis remain clinically compensated after a median follow-up of 13.4 months (range, 46 days-3 years). Inclusion in the waiting list. Of 28 decompensated patients, 14 were considered suitable for retransplantation and included on the waiting list. So far, 7 patients have undergone retransplantation at a median of 5.6 months (range, 7 days-20.8 months) since histologic diagnosis of fibrosis 4. The median time in the waiting list in patients undergoing liver retransplantation was 2.9 months (range, 2 days-6.9 months). The remaining 7 patients have either died while awaiting the second graft (n 4) or are still on the waiting list (n 3). Evolution of patients not included in the waiting list. Retransplantation was not deemed appropriate in 14 decompensated patients mainly for age-related reasons (age older than 65, n 10; and a very aggressive course of HCV-related liver damage with development of graft cirrhosis in less than 1 year, n 4). Six patients died at a median of 2.5 months (13 days-1.7 years), and 8 are still alive and are followed up regularly at the liver clinic, after a median of 6.8 months (range, 38 days-2.5 years) since decompensation. Patient survival. Twelve patients died at a median of 1.6 years (range, 17 days-4.3 years) since entry, including 2 who underwent retransplantation. Patient survival rates from histologic diagnosis of fibrosis 4 was 100%, 87%, and 72% at 1, 6, and 12 months, respectively. Graft Survival Rates. Overall, 17 patients have lost their graft during their follow-up including 7 who underwent retransplantation. Graft survival from entry was 97% at 1 month, 82% at 6 months, and 69% at 1 year, respectively. Outcome of Patients With Compensated Cirrhosis. Cumulative probability for clinical decompensation. The cumulative probability for decompensation (n 39) was 8% at 1 month, 17% at 6 months, and 42% at 1 year, respectively (Fig. 2). The median interval between fibrosis 4 and decompensation was 7.8 months, ranging from 4 days to 2.6 years. Cumulative probability for retransplantation. The cumulative probability for retransplantation from entry was 0% at 1 month, 7% at 6 months, and 10% at 1 year. The median interval between fibrosis 4 and retransplantation was 8.3 months, ranging from 4.6 months to 1.7 years. Patient survival rates. Patient survival rates reached 100% at 1 month, 92% at 6 months, and 74% at 1 year (Fig. 3). Patient survival rates dropped significantly once decompensation developed (93%, 61%, and 41% at 1, 6, and 12 months, respectively) (Fig. 4). Patient survival rates were slightly lower among patients who were decompensated at fibrosis 4 (88%, 64%, and 64% at 1, 6, and 12 months, respectively) compared with those who were compensated (P.1, log rank test). Graft survival rates. Graft survival rates were 100%, 85%, and 71% at 1, 6, and 12 months, respectively. Graft survival rates were lower among the 10 patients who were decompensated at entry (90%, 65%, and 32% at 1, 6, and 12 months, respectively) compared with those who were compensated (P.002, log rank test). Risk Factors for Decompensation, Retransplantation, and Mortality. The following factors were analyzed as predictors of clinical decompensation, retransplantation, and mortality: HAI and its 3 components at entry, liver function tests at entry including total bilirubin, prothrombin time, albumin, AST, ALT, gamma glutamyl transpeptidase, cholesterol, age at entry, sex, and maintenance immunosuppression at entry. For clinical decompensation. Univariate analysis indicated that the risk of decompensation was significantly higher in patients with a high Child-Pugh score at the time of identification of cirrhosis in the transplanted liver, in those with low initial levels of albumin, in those with high cyclosporine trough levels, and in those with a short interval from transplantation to development of cirrhosis (Table 3). For retransplantation and mortality. Univariate analysis indicated that the only variables associated with a higher rate of retransplantation and mortality rate were lower initial levels of albumin (3.5 g/dl [ ] vs. 4 g/dl [ ], P.01; and 3.3 g/dl [2.4-4] vs. 4 g/dl [ ], P.03, respectively)

4 HEPATOLOGY Vol. 32, No. 4, 2000 BERENGUER ET AL. 855 TABLE 2. Differences at Posttransplantation Cirrhosis Between Patients With Compensated Graft Cirrhosis and Those With Decompensated Cirrhosis Variables Compensated Cirrhosis (n 39) Decompensated Cirrhosis (n 10) P Value Median age at post-olt cirrhosis* (yr) 58 (41-68) 56 (28-66) NS Median age at OLT (yr) 56 (39-66) 55 (25-65) NS Sex distribution (% men) 19 (53%) 8 (61%) NS Induction immunosuppression (cyclosporine vs. tacrolimus) 31 (80%)/8 (20%) 8 (80%)/2 (20%) NS HCC at OLT 13 (36%) 2 (15%) NS Alcohol intake pre-olt 1 (3%) 1 (8%) NS HAI 11 (7-14) 11 (7-14) NS Periportal necrosis 5 (3-6) 5 (3-6) NS Lobular degeneration 3 (1-4) 3 (1-4) NS Portal inflammation 4 (1-4) 4 (3-4) NS Child-Pugh score 5 (5-11) 9 (6-10).0001 ALT (U/L) 130 (40-630) 125 (72-477) NS Aspartate transaminase (U/L) 145 (31-565) 75 (47-200).02 Bilirubin (mg/dl) 1.5 ( ) 2.7 ( ).04 Albumin (g/dl) 3.8 ( ) 3.7 ( ) NS Cholesterol (mg/dl) 165 (84-532) 160 (59-600) NS Prothrombin time (%) 90 (50-100) 70 (25-100).02 Cyclosporine dose (mg/d) 100 (50-250) 150 (50-400) NS Cyclosporine trough levels (ng/ml) 125 (60-340) 125 (60-330) NS Tacrolimus dose (mg/d) 1.5 (1-5) 0.75 (1.5-1) NS Tacrolimus trough levels (ng/ml) 9 (4-16) 11 (8-14) NS Prednisone dose (mg/d) 5 (2.5-5) 5 (5-5) NS Time from OLT to post-olt cirrhosis (mo) 26.3 ( ) 21.4 ( ) NS Abbreviations: OLT, liver transplantation; HCC, hepatocellular carcinoma; HAI, histologic activity index; ALT, alanine transaminase. *The time when histologic diagnosis of HCV-related graft cirrhosis was made. and a shorter interval from transplantation to development of cirrhosis (256 days [112-1,348] vs. 797 days [222-2,555], P.03; and 443 days [ ] vs. 939 days [112-2,555], P.01, respectively). Outcome After Retransplantation. Seven patients underwent retransplantation at a median of 5.6 months (range, 7 days-1.7 years) since histologic diagnosis of graft cirrhosis, or 1 year (range, 9.4 months-5 years) since transplantation. The median Child-Pugh score at retransplantation was 11, ranging from 8 to 14. Only 3 of these patients were still treated with prednisone at a median dose of 5 mg daily at the time of retransplantation. None was being treated with azathioprine FIG. 1. Outcome of the study population (n 49). The outcome, including decompensation, inclusion on waiting list for retransplantation, retransplantation, and death, of 49 patients with HCV-graft cirrhosis is summarized. FIG. 2. Cumulative probability for decompensation (n 39). The cumulative probability for decompensation of 39 recipients with clinically compensated HCVgraft cirrhosis was 8%, 17%, and 42% at 1, 6, and 12 months, respectively.

5 856 BERENGUER ET AL. HEPATOLOGY October 2000 FIG. 3. Patient survival rate in patients with clinically compensated HCVgraft cirrhosis (n 39). Patient survival rates since histologic diagnosis of clinically compensated HCV-related graft cirrhosis reached 74% at 1 year posttransplantation. or mycophenolate. Maintenance immunosuppression was based on cyclosporine in 4 patients (median dose, 150 mg daily; median trough levels, 115 ng/ml) and tacrolimus in 3 patients (median dose, 0.25 mg daily; median trough levels, 8.4 ng/ml). Two of these retransplanted patients died at 3 and 10 days after retransplantation because of multiorgan failure. Of the 5 patients alive to date, only 1 has received therapy with interferon-ribavirin at the time of acute hepatitis without development of significant chronic hepatitis on the second graft. In contrast, 2 of the 4 remaining untreated patients have developed severe hepatitis with or without cirrhosis in their second graft at approximately 1 and 3 years, respectively (Table 4). DISCUSSION As in many other countries, cirrhosis secondary to chronic HCV infection is the leading indication for liver transplantation in Spain. 1 After transplantation, recurrence of viremia occurs universally 2 with reinfection of the graft and the development of graft hepatitis in a percentage that varies from 50% after 1 year to almost 100% after 5 years of follow-up. 3-5 Although several studies have reported similar patient and graft survival rates when compared with uninfected control patients, at least during the first 5 years posttransplantation, 3,4 there is a consensus emerging that the natural history of hepatitis C is shortened in immunocompromised patients when compared with immunocompetent patients. In a recent study, fibrosis progression was found to be significantly higher after transplantation than before it and the time over which patients developed a cirrhosis was found to be significantly shorter, with an estimated median of 12 years. 7 This figure suggests that the number of liver transplant recipients reaching the stage of cirrhosis will likely increase within the next years as liver transplant centers reach their second decade of activity. 12 And indeed, in a previous study, we found that the actuarial rate of developing graft cirrhosis caused by HCV reinfection rose to 28% at 5 years, 5 with a further increase to 34% when patients were followed up during a median of 3 years. 13 These data relate to the evolution of recurrent hepatitis C but, to date, no available information exists regarding the natural history of compensated HCV-related graft cirrhosis. This information is relevant to transplant centers to facilitate the indication for liver retransplantation. Data from the immunocompetent population with survival rates of at least 90% suggest that the 5-year outcome favors observation in absence of transplantation, and this therapeutic option should only be considered when decompensation develops Therefore, we undertook this study to evaluate the medium-term outcome of liver transplant recipients who had reached the stage of clinically compensated HCV-related graft cirrhosis, and the variables associated with decompensation and mortality. Our results can be summarized as follows: (1) approximately one half of the patients diagnosed with compensated HCV-related graft cirrhosis will develop at least 1 decompensation after a median of 8 months; (2) the most frequent decompensation is ascites; (3) the cumulative probability for decompensation reaches almost 50% after 1 year of follow-up; (4) the mortality rate is low in compensated cirrhotic patients followed up for a short-medium term of approximately 1 year; (5) in the absence of therapeutic interventions, patient survival rate drops significantly once decompensation develops ( 50% at 1 year); (6) variables associated FIG. 4. Patient survival rate after clinical decompensation of graft cirrhosis. Patient survival rate was extremely low once clinical decompensation developed (93%, 61%, and 41% at 1, 6, and 12 months, respectively).

6 HEPATOLOGY Vol. 32, No. 4, 2000 BERENGUER ET AL. 857 TABLE 3. Univariate Analysis of Variable Predictors of Decompensation From the Histologic Diagnosis of HCV-Related Graft Cirrhosis Variables Decompensation (n 18) No decompensation (n 21) P Value Median age at F 4* 58 (47-67) 58 (41-68) NS Sex (M/F) 8/10 13/8 NS HAI 12 (7-14) 11 (7-13).1 PN 6 (3-6) 5 (3-6).1 PI 4 (1-4) 4 (3-4).7 LD 3 (1-4) 3 (1-4).1 Child-Pugh at F Median and range 6.5 (5-11) 5 (5-7) A/B/C 9 (50%)/8 (44%)/1 (6%) 20 (95%)/1 (5%)/0 Interval from OLT to 15.3 ( ) 36.1 ( ).004 F 4 (mo) Cyclosporine Dose at F (50-250) 100 (50-250) NS Levels at F (66-342) 108 (62-281).01 Tacrolimus Dose at F 4 1 (1-2) 2 (1-5) NS Levels at F (4-15.9) 9.3 (5.7-16) NS AST (U/L) 144 (40-631) 122 (47-405) NS ALT (U/L) 129 (31-281) 148 (81-565) NS GGT (U/L) 88 ( ) 94 ( ) NS Bilirubin (mg/dl) 1.6 ( ) 1 ( ) NS Cholesterol (mg/dl) 164 (94-532) 167 ( ) NS Albumin (g/dl) 3.4 ( ) 4 ( ).002 PT (%) 90 (50-100) 86 (62-100) NS Abbreviations: PN, periportal necrosis; PI, portal inflammation; LD, lobular degeneration; PT, prothrombin time in %; AST, aspartate transaminase; GGT, gammaglutamyl transpeptidase; OLT: liver transplantation; F, fibrosis. *The time when histologic diagnosis of HCV-related graft cirrhosis was made. with clinical decompensation, retransplantation, and mortality rate include a short interval from transplantation to development of posttransplantation cirrhosis, a high Child-Pugh score at diagnosis of compensated cirrhosis, and a low albumin level at the latter time. There are several conclusions that can be drawn from these results. The first conclusion is to confirm the higher aggressiveness of hepatitis C in immunocompromised patients compared with those who are immunocompetent. 7 Indeed, the cumulative probability for decompensation was approximately 50% in 1 year, a figure significantly higher than that reported in the immune-competent population In fact, the actuarial rate of decompensation in 1 year in immune-competent cirrhotic patients is similar to that found at 1 month in our population of transplant cirrhotic recipients. Moreover, once decompensation occurs, the survival rate decreases substantially to less than 50% in 1 year. If we assume that the postretransplantation survival rate reaches 65% within 1 year among HCV-infected recipients, 12 the data presented here provide the evidence that clinical management of patients with compensated HCV-graft cirrhosis should be aimed at the prevention and early recognition of decompensation. Once decompensation develops, patients should be transplanted as soon as possible to avoid a high mortality rate after retransplantation. In this study, we also assessed the predictive value of clinical variables for morbidity and mortality rates. A short duration since transplantation to histologic diagnosis of HCVrelated graft cirrhosis, and a high Child-Pugh score or low albumin levels at the latter time, were associated with an increased risk for decompensation and mortality. These results relate to those found in the immunocompetent popula- TABLE 4. Evolution After Retransplantation Patient No. OLT Date Re-OLT Date Time Since OLT to re-olt (d) Time Since F4 to re-olt (d) Time on Waiting List (d) Child- Pugh at re-olt Status Antiviral Treatment Last Histologic Follow-up Fibrosis Stage at Last Biopsy Activity at Last Biopsy Alive No Alive Yes Alive No Dead No Alive No Alive No Dead No NOTE. Patients 4 and 7 died at 3 and 10 days, respectively, after retransplantation. Abbreviations: OLT, liver transplantation; Re-OLT, retransplantation; F, fibrosis.

7 858 BERENGUER ET AL. HEPATOLOGY October 2000 tion The higher Child-Pugh score and lower albumin levels in patients who subsequently developed a clinical decompensation versus those who did not may also reflect a subclinical decompensation at the time of histologic diagnosis of graft cirrhosis, manifested only biochemically. Based on these results, we propose that in clinically compensated HCVrelated cirrhotic patients characterized by the fast development of posttransplantation cirrhosis and a high Child-Pugh score at that time, an alternative way to improve their outcome would be their listing before the development of decompensation. Given the increased time on the waiting list, this alternative may provide an opportunity for these patients to undergo retransplantation before serious complications develop. Early and severe histologic changes of recurrent hepatitis C have been shown to predict a more aggressive evolution of hepatitis C after liver transplantation. 5,14 We also assessed the predictive value of histologic necroinflammatory activity on the outcome of HCV-cirrhotic patients, but failed to find an association between these 2 variables. These results should, however, be analyzed with a larger number of patients. Viral factors may also play a role in the evolution of posttransplantation cirrhosis, but could not be evaluated because all the patients tested were infected with HCV genotype 1b, and serum quantitation was not available. Additional studies with a more heterogeneous population should address this issue. There has been an increase in the prevalence of HCV infection among patients undergoing retransplantation in recent years. 12 This trend will likely continue in the near future, and, given the shortage of organs, places transplant teams in the dilemma of denying a valuable organ to a first-time recipient. Furthermore, because initial data showed an increased mortality rate after retransplantation in these patients, 12,15 and there is a fear of a similar evolution after the second transplant, not all transplant centers are adopting the same guidelines regarding this new indication. Although the assessment of the evolution postretransplantation was not the aim of this study, and the numbers are low, our data suggest that hepatitis C postretransplantation follows an aggressive pattern similar to that observed in the first graft unless specific measures are taken. Indeed, 2 of the 4 retransplanted patients who survived more than 1 month have developed a significant hepatitis after a median of 2 years. The 2 patients who have not reached these stages have been followed up histologically for only 3 months and 1 year, respectively. Interestingly, the only patient who has been treated with interferon-ribavirin started treatment within the first month of retransplantation, and had a nearnormal liver at the first protocol annual liver biopsy. We conclude that HCV-infected recipients who reach the stage of clinically compensated cirrhosis have a very high risk of decompensation within the first year, and that once decompensation develops, the mortality rate is high unless retransplantation is promptly performed. First-time liver recipients who are diagnosed with compensated cirrhosis early after transplantation (in the first 2 years) and who have a high Child-Pugh score ( A) or low levels of albumin at that time ( 3.5 mg/dl) are at a higher risk for prompt decompensation, and, hence, should be monitored more regularly and included early on the waiting list for retransplantation. If retransplantation is considered, prophylactic or early antiviral therapy appears the most promising measure to reduce disease progression after retransplantation. Acknowledgment: The authors would like to thank Marisa Galera for her daily assistance. The authors would also like to thank the Service of Microbiology for their participation in the serological and virological determinations. REFERENCES 1. Prieto M, Berenguer M, Rimola A, Loinaz C, Barrios C, Clemente G, Figueras J, et al. Liver transplantation in hepatitis C: a Spanish multicenter experience. Eur J Gastroenterol Hepatol 1998;10: Wright T, Donegan E, Hsu H, Ferrell L, Lake JR, Kim M, Combs C, et al. Recurrent and acquired hepatitis C viral infection in liver transplant recipients. Gastroenterology 1992;103: Gane E, Portmann B, Naoumov N, Smith HM, Underhill JA, Donaldson PT, Maertens G, et al. Long-term outcome of hepatitis C infection after liver transplantation. N Engl J Med 1996;334: Charlton M, Seaberg E, Wiesner R, Everhart J, Zetterman R, Lake J, Detre K, et al. Predictors of patient and graft survival following liver transplantation for hepatitis C. HEPATOLOGY 1998;28: Prieto M, Berenguer M, Rayón M, Córdoba J, Argüello L, Carrasco D, Garúa-Herola A, et al. High incidence of allograft cirrhosis in hepatitis C virus genotype 1b infection following transplantation: relationship with rejection episodes. HEPATOLOGY 1999;29: Feray C, Caccamo L, Alexander GJM, Ducot B, Gugenheim J, Casanovas T, Loinaz C, et al. European Collaborative Study on factors influencing the outcome after liver transplantation for hepatitis C. Gastroenterology 1999;117: Berenguer M, Ferrell L, Watson J, Prieto M, Kim M, Rayón M, Córdoba J, et al. HCV-related fibrosis progression following liver transplantation: increase in recent years. J Hepatol 2000;32: Desmet VJ, Gerber M, Hoofnagle JH, Manns M, Scheuer PJ. Classification of chronic hepatitis: diagnosis, grading and staging. HEPATOLOGY 1994; 19: Fattovitch G, Giustina G, Degos F, Tremolada F, Diodati G, Almasio P, Nevens F, et al. Morbidity and mortality in compensated cirrhosis type C: a retrospective follow-up study of 384 patients. Gastroenterology 1997; 112: Serfarty L, Aumaître H, Chazouillères O, Bonnand AM, Rosmorduc O, Poupon RE, et al. Determinants of outcome of compensated hepatitis C virus-related cirrhosis. HEPATOLOGY 1998;27: Hu K-Q, Tong MJ. The long-term outcomes of patients with compensated hepatitis C virus related cirrhosis and history of parenteral exposure in the United States. HEPATOLOGY 1999;29: Rosen HR, Martin P. Hepatitis C infection in patients undergoing liver retransplantation. Transplantation 1998;66: Prieto M, Berenguer M. [Response to a letter]. Gastroenterology 1999; 117: Rosen HR, Gretch DR, Oehlke M, Flora KD, Benner KG, Rabkin JM, Corless CL, et al. Timing and severity of initial hepatitis C recurrence as predictors of long-term liver allograft injury. Transplantation 1998;65: Sheiner PA, Schluger LK, Emre S, Thung SN, Lau JY, Guy SR, Schwartz ME, et al. Retransplantation for recurrent hepatitis C. Liver Transpl Surg 1997;3: