Incidence of Hepatitis C in Patients Receiving Different Preparations of Hepatitis B Immunoglobulins after Liver Transplantation

Methods

Five hundred thirty-eight patients who underwent transplantation between December 1984 and March 1994 were initially considered for our study. This group comprised all patients who received transplants for HBV- or HCV-related cirrhosis and all patients who received transplants for nonviral cirrhosis before March 1990.

Blood-product exposure was assessed in every case as the total number of units of packed red blood cells, frozen plasma, and platelets transfused during transplantation and during the first year after transplantation and as the number of blood donors (that is, donors of packed red blood cells, fresh frozen plasma, and platelets). As of October 1989, 41 patients had undergone transfusion with pooled clotting factors (Centre National Transfusion Sanguine, Paris, France) to minimize perioperative bleeding. Patients positive for HBsAg were prescribed long-term HBIG therapy (Laboratoire Francais de Fractionnement et de Biotechnologie, Les Ulis, France), as described elsewhere [8]. Briefly, these patients received 10 000 IU of HBIG during the anhepatic phase and 10 000 IU every day during the first 6 days after surgery. The titer of antibody to HBsAg (anti-HBsAg) was then assessed weekly; 10 000 IU of HBIG was administered when the titer was less than 100 IU/L.

Immunosuppression

In 401 patients, basic immunosuppression was induced by a combination of cyclosporine (Sandoz, Bales, Switzerland), steroids, and azathioprine. Twenty-seven patients received a combination of steroids and tacrolimus (Fujisawa, Osaka, Japan) without azathioprine [9]. Forty-three patients had initially been enrolled in a prospective multicenter trial [10] of antilymphocyte antithymocyte globulin (Biomerieux, Lyon, France), given for renal impairment (n = 15), or Orthoclone OKT3 (Ortho, Raritan, New Jersey) (n = 28), given for primary prophylaxis of rejection. All episodes of rejection were first treated with one to three boluses of steroids and then with Orthoclone OKT3 if necessary.

Follow-up

All patients were seen at our outpatient clinic by the same team of hepatologists or surgeons at least monthly during the first year and every 2 to 6 months thereafter. Liver graft biopsies were performed when liver test results were abnormal, and all specimen slides were interpreted by the same pathologist. Protocol biopsies were done routinely 1, 2, and 5 years after transplantation in 93%, 88%, and 90% of survivors, respectively. These survivors had the same liver graft at these follow-up examinations. All patients underwent at least one liver biopsy more than 6 months after transplantation. Lobular hepatitis was defined as hepatocyte necrosis and portal and lobular infiltration by mononuclear cells. Serum samples have been stored at −30°C from 1984 to the present in the same location.

Screening of Organs, Blood, and Donors for Polyclonal Hepatitis B Immunoglobulin

In March 1990, serologic screening for HCV with a first-generation commercial kit became obligatory in France for all blood and liver donors, although it was initiated in most French blood banks as early as December 1989. All batches of HBIG prepared from the plasma of patients not tested for HCV had to be destroyed in 1991. In March 1990, however, our blood bank destroyed all batches of HBIG harvested before December 1989 (the start of effective screening in France). Therefore, all 182 patients who had transplantation after March 1990 received grafts, blood products, and HBIG from donors who were negative for anti-HCV.

Markers of Hepatitis B Virus

Until 1993, all HBsAg-positive patients were tested for serum HBV DNA before liver transplantation by using the spot-test technique [11]; after 1993, testing was done with the Hybrid capture system (Digen Sharp Signal System, Beltsville, Maryland). Patients receiving HBIG underwent weekly titration of anti-HBsAg during the first 2 years; after that point, titration was done at least bimonthly. If anti-HBsAg titers were low (<100 IU/L), routine testing for HBsAg was done.

Markers of Hepatitis C Virus

Antibodies to HCV were confirmed retrospectively in all serum samples obtained before transplantation by using a recombinant immunoblot assay (RIBA-2 and then RIBA-3, Test System, Chiron, Emeryville, California). At the end of 1991, serologic tests for HCV were done yearly in every case. In addition, serum samples harvested during the first month after transplantation from 28 patients who were negative for anti-HCV and were receiving HBIG (15 underwent transplantation before 1990 and 13 underwent transplantation after this date) were tested retrospectively with both enzyme-linked immunosorbent assay (HCV Elisa 3.0, Orthodiagnostic system) and a line immunoassay (InnoLia HCV Ab III, Innogenetics, Gent, Belgium). These assays are more sensitive than the recombinant immunoblot assay. All serum samples obtained before and after transplantation were tested for HCV RNA. Serum samples were obtained a mean of 6 weeks (range, 0 to 6 months) before liver transplantation and a mean of 12 months (range, 10 to 16 months) after liver transplantation. In patients receiving HBIG, serum samples were obtained a few hours before perfusion of HBIG. A commercial polymerase chain reaction assay (Amplicor kit, Produits Roche, Basel, Switzerland) was used to detect HCV RNA, according to the manufacturer's instructions. Samples with negative Amplicor signals were retested in a “one-tube” nested polymerase chain reaction procedure [12]. The HCV genotypes were determined by using a line probe assay (InnoLipa, Innogenetics) according to the manufacturer's instructions. In all amplification procedures, decontamination was based on classic recommendations [13] and on the routine use of uracil-N-glycosylase (AmpErase, Perkin-Elmer, Norwalk, Connecticut).

Statistical Analysis

We used BMDP statistical software (University of California Press, Berkeley, California). Analysis was done separately on the incidence of HCV viremia 1 year after transplantation and on the actuarial rate of HCV-related hepatitis. The latter was defined as a combination of histologically proven hepatitis and serum positivity for HCV RNA. Infection with HCV was defined as acquired if the patient was negative for both anti-HCV and HCV RNA before transplantation; infection was defined as recurrent if at least one of these two markers was detected. Histologic follow-up began on the date of the first liver transplantation and continued until the last available biopsy of the first liver graft. Presence of HCV viremia at 1 year and actuarial rates of hepatitis due to HCV were analyzed according to the following variables: age; sex; pretransplantation markers of HCV (serologic results and HCV RNA), HBV (HBsAg), and hepatitis delta virus (antibody to hepatitis delta virus); use of HBIG; immunoprophylaxis of rejection (antilymphocyte serum or Orthoclone OKT3); year of transplantation; amount and number of donors of each perfused blood product (red blood cells, platelets, frozen plasma, and pooled clotting factors); numbers of steroid boluses during the first year after transplantation; and steroid-resistant rejection.

Incidence of viremia 1 year after transplantation was analyzed by using the chi-square test and logistic regression (maximum likelihood). Analysis of actuarial rates of histologically defined hepatitis C was based on the log-rank test and Cox model. All data are presented with 95% CIs.

Role of Funding Source

The institutions that supported this study had no direct role in gathering, analyzing, or interpreting the data and played no role in the decision to publish the study findings.

Results

Of the 538 patients initially eligible for the study, 64 died during the first year. Death was caused by sepsis (n = 33), cancer recurrence (n = 6), cardiovascular complications (n = 10), perioperative complications (n = 4), de novo cancer (n = 5), transplant rejection (n = 5), or unexplained graft failure (n = 1). Of the 474 patients surviving more than 1 year, 428 (90%) had available serum samples obtained before and after transplantation; these patients formed the study sample. Of these 428 patients, 170 were positive for HBsAg; of the 258 HBsAg-negative patients, 172 had transplantation for HCV-related cirrhosis and 86 had transplantation for nonviral cirrhosis. Informed consent was obtained in accordance with the guidelines of our institutional review board.

Serum Hepatitis C Virus RNA before Transplantation

Serum HCV RNA was detected before transplantation in 186 of 215 (86%) anti-HCV-positive patients; these included 27 of the 43 patients (63%) who were also positive for HBsAg and 159 of the 172 HBsAg-negative patients (92%) (P < 0.001) (Table 1). Serum was positive for HCV RNA in 3 of the 213 anti-HCV-negative patients who were also positive for HBsAg. Before transplantation, 218 patients had at least one marker of HCV infection (46 were HBsAg positive) and 210 were negative for both anti-HCV and HCV RNA (124 were positive for HBsAg). The viral genotype was determined in 161 patients randomly selected from among patients who were positive for HCV RNA before transplantation. Genotype 1b was detected in 15 of 18 (80%) patients receiving HBIG and in 115 of 143 (80%) patients not receiving HBIG.

Serum Hepatitis C Virus RNA 1 Year after Transplantation

Table 2 and Table 3 report the presence of serum HCV RNA 1 year after transplantation according to HCV markers before transplantation, perfusion of HBIG after transplantation, and date of transplantation: Among the 218 patients who had at least one HCV marker before transplantation, the frequency with which HCV infection developed after transplantation was lower in patients receiving HBIG than in patients not receiving HBIG. Among patients receiving HBIG, the frequency was lower in patients who had transplantation before March 1990 than in those who had transplantation after this date. Multivariate analysis showed that the absence of HBIG and transplantation after March 1990 were independent risk factors for HCV viremia 1 year after transplantation.

Among the 210 patients who did not have markers of HCV infection before transplantation, the frequency with which HCV infection developed 1 year after transplantation was higher in patients who received the transplant before March 1990 than in those who received the transplant after that date. Before March 1990, the frequency of HCV infection was lower in patients receiving HBIG than in those not receiving HBIG.

Histologically Proven Hepatitis C

Of the 218 patients with pretransplantation HCV infection, 13 developed HBV infection after transplantation (12 patients receiving HBIG had a recurrence of HBV infection, and 1 patient not receiving HBIG developed HBV infection for the first time). These 13 patients were excluded; as a result, 205 patients (34 receiving HBIG and 171 not receiving HBIG) were analyzed. The 3-year actuarial rate of HCV-related hepatitis was lower in patients receiving HBIG (27% [95% CI, 14% to 41%]) than in those not receiving HBIG (70% [CI, 62% to 78%]) (P < 0.001) (Figure 1, top left). Among HBIG recipients, the 3-year actuarial rate of recurrent hepatitis C was lower in patients who underwent transplantation before March 1990 (10%) than in those who underwent transplantation after this date (61%) (P = 0.001) (Figure 1, bottom left). Among patients not receiving HBIG, the rate was similar in those who had transplantation before (63% [CI, 49% to 77%]) and after (71% [CI, 61% to 81%]) March 1990 (Figure 1, bottom right). Cox analysis demonstrated that the absence of HBIG therapy (relative risk, 3.5 [CI, 1.8 to 6.7]; P = 0.001) and the occurrence of transplantation after March 1990 (relative risk, 1.7 [CI, 1.1 to 2.5]; P = 0.05) were independent risk factors for post-transplantation hepatitis C.

Of the 210 patients without HCV markers before transplantation, 26 were excluded because HBV infection developed after transplantation (25 patients had recurrent HBV infection, and 1 acquired HBV infection). No cases of de novo hepatitis C occurred after March 1990. In patients undergoing transplantation before March 1990, the 3-year actuarial rate of HCV-related hepatitis was lower in patients receiving HBIG (17% [CI, 5% to 29%]) than in patients not receiving HBIG (27% [CI, 17% to 37%]) (P < 0.05) (Figure 1, top right).

Detection of Passively Transmitted Antibodies to Hepatitis C Virus in the Serum of Patients Receiving Polyclonal Hepatitis B Immunoglobulin

Twenty-eight patients who received HBIG before (n = 15) and after (n = 13) March 1990 and who were negative for both HCV RNA and anti-HCV before and 1 year after transplantation underwent testing for passively transmitted anti-HCV. Serum samples were harvested between 1 and 3 weeks after transplantation and were analyzed with enzyme-linked immunosorbent assay and line immunoassay. Results of the two assays were positive in 4 of 28 and 10 of 28 serum samples, respectively. All positive samples had been harvested before March 1990; this finding suggests the presence of passively transmitted anti-HCV in patients who received HBIG before this date.

Discussion

Our study indicates that HBIG given to HBsAg-positive transplant recipients to prevent recurrent HBV infection on the liver graft can also protect against acquired and recurrent hepatitis C. An important finding was that the protective effect of HBIG for recurrent HCV infection decreased and that passive transmission of anti-HCV through HBIG disappeared after March 1990 (the date after which persons positive for anti-HCV were prohibited from donating blood). These findings strongly suggest that the protective effect of HBIG was due to the presence of anti-HCV in HBIG.

Our findings are similar to those of previous trials of nonspecific polyclonal immunoglobulins given to prevent non-A, non-B hepatitis after transfusion [14, 15]. In particular, although it was conducted before hepatitis C was recognized, the randomized trial by Sanchez-Quijano and colleagues [15] showed that such preparations had a marked protective effect on post-transfusion hepatitis. Given the large number of donors required for such preparations and the incidence of anti-HCV-positive donors in western countries, anti-HCV was probably present in both nonspecific immunoglobulins and HBIG.

We detected passively transmitted anti-HCV in some patients receiving HBIG. These antibodies were detected during the first month after transplantation and were concomitant with high titers of infused HBIG. The antibodies were detected only in patients who had transplantation before March 1990. Unfortunately, batches of HBIG prepared from unscreened donors were destroyed and thus could not be tested retrospectively. After HCV was recognized, the first attempts to induce humoral neutralization in chimpanzees failed, suggesting an inefficient humoral response in HCV infection [4]. In more recent studies, neutralizing anti-HCV has been detected after chimpanzee vaccination [16] and has been demonstrated in cell lines that allow HCV infection [17]. Neutralization has also been reported when anti-HCV-positive plasma is incubated with infectious doses of HCV and subsequently given to chimpanzees [18]. Finally, the incubation period of experimental HCV infection in chimpanzees is prolonged when the infectious challenge is followed by injection of polyclonal immunoglobulins enriched by anti-HCV [19].

Because we administered HBIG only to HBsAg-positive recipients, HBV itself rather than HBIG administration may have been responsible for the low incidence of HCV infection after liver transplantation. Before transplantation, viral interference could theoretically explain the lower rate of recurrence after transplantation in patients infected with both HBV and HCV. In this view, the incidence of HCV viremia was significantly lower in these patients than in those infected by HCV alone. This finding confirms those of previous studies on the inhibition of HCV replication by HBV [20]. However, through the multivariate analysis of patients positive for anti-HCV, HCV viremia before transplantation did not predict recurrence. After transplantation, viral interference was also unlikely: Patients with HBV infection after transplantation were excluded, and transplant recipients using HBIG who remained negative for HBsAg have extremely low levels of hepatic or extrahepatic replication of HBV [21]. Finally, recurrent hepatitis C in patients receiving HBIG was significantly more frequent after March 1990 than before this date; thus, viral interference cannot explain our findings.

Another potential confounding factor is a lower level of immunosuppression in patients receiving HBIG, which could influence the course of HCV after liver transplantation. In a previous study, we found a lower rate of rejection in patients receiving HBIG [22]. However, we did not find a relation between the number of steroid boluses and episodes of steroid-resistant rejection (treated with Orthoclone OKT3) or an association between HCV RNA positivity 1 year after transplantation and the occurrence of hepatitis C.

The most plausible explanation for the protective effect of HBIG observed essentially before March 1990 is passive transmission of anti-HCV. This suggests that polyclonal immunoglobulins enriched in anti-HCV could have therapeutic value in patients undergoing liver transplantation for HCV-related cirrhosis. However, most potential donors of plasma containing anti-HCV are also viremic. Although these immunoglobulins could be harvested from donors with low-level viremia, additional decontamination procedures would be necessary [23], and the viral safety of such preparations would have to be tested in chimpanzees. Immunoglobulins are stable products that could theoretically withstand such procedures.

Dr. Ducot: Unite-292, Institut National de la Sante et de la Recherche Medicale, CHU Bicetre, 63 rue Gabriel Peri, le Kremlin-Bicetre 94270, France.

Dr. Maisonneuve: Centre de Transfusion Sanguine, 2 rue Jean-Louis Joffrain, Le Chesnay 78150, France.