De Novo Hepatitis B After Liver Transplantation FromHepatitis B Core Antibody–Positive Donors in an Area

With High Prevalence of Anti-HBc Positivityin the Donor Population

Martın Prieto,* Marıa D. Gomez,† Marina Berenguer,* Juan Cordoba,†

Jose M. Rayon,‡ Miguel Pastor,* Antonio Garcıa-Herola,* David Nicolas,*Domingo Carrasco,* Juan F. Orbis,§ Jose Mir,§ and Joaquın Berenguer*

Transmission of hepatitis B virus (HBV) infection fromdonors who are negative for hepatitis B surface antigen(HBsAg–) but positive for antibody to hepatitis B coreantigen (anti-HBc1) has been reported. However, previ-ous studies were generally performed in geographic re-gions with a low prevalence of anti-HBc positivity in theliver donor population. The aims of this study are (1) toassess the risk for de novo hepatitis B in recipients of liversfrom anti-HBc1 donors in an area of high prevalence ofanti-HBc positivity in the donor population, and (2) toanalyze the risk factors for acquisition of HBV infectionfrom anti-HBc1 donors. The transplantation experienceof a single center between 1995 and 1998 was reviewed.Thirty-three of 268 liver donors (12%) were HBsAg– andanti-HBc1 during the study period. The proportion ofanti-HBc1 donors increased with age; it was lowest(3.6%) in donors aged 1 to 20 years and highest (27.1%)in donors aged older than 60 years. Of the 211 HBsAg–recipients with 3 months or more of HBV serologicalfollow-up, 30 received a liver from an anti-HBc1 donorand 181 received a liver from an anti-HBc– donor. Hep-atitis B developed in 15 of 30 recipients (50%) of liversfrom anti-HBc1 donors but in only 3 of 181 recipients(1.7%) of livers from anti-HBc– donors (P < .0001).None of the 4 recipients who were antibody to HBsAg(anti-HBs)1 at the time of transplantation developedHBV infection after receiving a liver from an anti-HBc1donor compared with 15 of 26 recipients (58%) who wereanti-HBs– (P 5 .10). None of the 5 anti-HBc1 recipientsdeveloped hepatitis B compared with 15 of 25 anti-HBc–recipients (60%; P 5 0.04). Child-Pugh score wassignificantly higher in recipients of livers from anti-HBc1donors who developed HBV infection than in those whodid not (9 6 2 v 7 6 1; P 5 .03). In our area, testing liverdonors for anti-HBc is mandatory, particularly in olderdonors. With such information available, anti-HBc1donors can be safely directed to appropriate recipients,mainly those with anti-HBs and/or anti-HBc at the timeof transplantation. In the current era of donor shortage,this policy would allow adequate use of such donors.(Liver Transpl 2001;7:51-58.)

The acquisition of hepatitis B virus (HBV) infectionafter liver transplantation remains a significant

problem in the liver transplant setting. Although denovo HBV infection may occur as a result of the reac-tivation of occult HBV infection in the recipient,1,2

recent reports suggest that the vast majority of acquiredHBV infections after liver transplantation are related tothe donor liver. Liver grafts from donors who are hep-atitis B surface antigen negative (HBsAg–) but positivefor antibody to hepatitis B core antigen (anti-HBc1)have been shown to transmit HBV infection toHBsAg– liver transplant recipients at a rate that rangedfrom 33% to 78%.3-6 However, these results weremostly derived from studies performed in geographicregions with a low prevalence (3% to 4%) of anti-HBcpositivity in the liver donor population.

The prevalence of HBsAg in the adult Spanish pop-ulation ranges from 1.2% to 1.7%.7,8 Accordingly, theprevalence of anti-HBc positivity in the general popu-lation is greater than that reported in areas of low prev-alence of HBV infection. In a recent study, anti-HBcpositivity was found in 10% of the Spanish adults aged26 to 65 years.7 In Spain, the current policy of theOrganizacion Nacional de Trasplantes (ONT), theSpanish Organ Procurement Organization, is to testorgan donors only for HBsAg. To a certain extent, theliver donor population is a reflection of the generalpopulation; therefore, it is reasonable to anticipate thatthe prevalence of anti-HBc positivity in Spanish liverdonors and thus the incidence of de novo HBV infec-tion after liver transplantation would be greater in ourarea than in areas of lower prevalence. In addition, in

From the *HepatoGastroenterology, †Microbiology, and ‡PathologyServices, and the §Liver Transplantation and Surgery Unit, HospitalUniversitario La Fe, Valencia, Spain.

Presented in part at the 50th Annual Meeting of the AmericanAssociation for the Study of Liver Diseases, Dallas, TX, November 5-9,1999.

Address reprint requests to Martın Prieto, MD, Servicio de Hepato-Gastroenterologıa, Hospital Universitario La Fe, Avenida Campanar,21, 46009 Valencia, Spain. Telephone: 34-963-868-792; FAX: 34-963-987-333; E-mail: prieto_mar@gva.es

Copyright © 2001 by the American Association for the Study ofLiver Diseases

1527-6465/01/0701-0007$3.00/0doi:10.1053/jlts.2001.20786

51Liver Transplantation, Vol 7, No 1 (January), 2001: pp 51-58

previous studies, characterization of anti-HBc1 donorsand the study of risk factors for the acquisition of HBVinfection from anti-HBc1 donors have not been clearlydocumented. To clarify these issues, we undertook thisretrospective (1995 to 1998) analysis of our experienceusing anti-HBc1 liver donors.

Patients and Methods

Patients

Hepatitis B serological markers were available in 268 of 281donors of livers used in an initial liver transplant betweenMarch 31, 1995, and December 31, 1998, at our institution.For the HBV transmission study, only those 211 recipientswithout pretransplantation HBV infection and with a post-transplantation HBV serological follow-up of at least3 months were analyzed. End-stage cirrhosis was the indica-tion for liver transplantation in 203 of 211 patients. The causeof cirrhosis was as follows: chronic hepatitis C virus (HCV)infection alone (n 5 118) or associated with alcohol abuse(n 5 14), alcohol abuse (n 5 49), cryptogenic cirrhosis(n 5 7), primary biliary cirrhosis (n 5 6), secondary biliarycirrhosis (n 5 5), and autoimmune hepatitis (n 5 4). Hepa-tocellular carcinoma was present in 42 of the recipients (21%)with cirrhosis at the time of transplantation. The indicationsfor liver transplantation in the 8 patients with noncirrhoticliver disease included liver metastases (n 5 3), non–HBV-related fulminant hepatic failure (n 5 2), cholangiocarcinoma(n 5 2), and hepatocellular carcinoma arising on a noncir-rhotic liver (n 5 1).

Serological and Virological Testing

Donor evaluation. In the donor, HBsAg was determinedbefore transplantation as part of the general donor workup. Inaddition, antibody to HBsAg (anti-HBs) and anti-HBc wereretrospectively determined from serum samples collected dur-ing transplantation and stored at –20°C until analysis. Serumfrom 13 anti-HBc1 donors whose recipients acquired hepa-titis B after transplantation was also tested for HBV DNA bypolymerase chain reaction (PCR) analysis.

Recipient evaluation. All recipients were tested for HBVmarkers (HBsAg, anti-HBs, and anti-HBc) and HBV DNA byPCR before transplantation. After transplantation, HBV sero-logical markers were tested at regular intervals (1, 3, and 6months and yearly thereafter) and when clinically indicated.HBV DNA was also tested by conventional molecular hybrid-ization when HBsAg was detected in serum and at regular inter-vals thereafter and while patients were on lamivudine therapy.

Serological tests. HBsAg was measured by a commercialenzyme-linked immunoassay (Abbott Laboratories, NorthChicago, IL). Anti-HBs, anti-HBc, hepatitis B e antigen[HBeAg], and antibody to HBeAg were tested by commer-cially available enzyme-linked immunoassays (Diasorin SRL,Saluggia, Italy).

Virological tests. HBV DNA PCR analysis was performedusing primers of the precore/core region, described by Baker

et al.9 Quantification of HBV DNA in serum was performedby means of a molecular hybridization assay (Digene HBVtest Hybrid Capture I; Digene Corp, Gaithersburg, MD) thatuses chemiluminescent detection, following the manufactur-er’s instructions.

Definition of HBV infection. Acquired or de novo HBVinfection was defined as detection of HBsAg in serum on 2consecutive occasions in a recipient who was HBsAg– andHBV DNA– by PCR before liver transplantation.

Histopathological Evaluation

Initial liver biopsies were performed in patients with acquiredHBV infection when liver test results became abnormal or aspart of our protocol of annual biopsies after liver transplanta-tion. Additional liver biopsies were performed before initiat-ing lamivudine therapy and yearly while the patient was onthis therapy. Tissue HBV antigens were detected on all post-transplantation biopsy specimens of HBsAg1 patients by animmunohistochemical method. This method was applied to5-mm paraffin sections. The following antibodies were used:anti-HBs by a monoclonal assay (Biomedica, Foster City, CA)and anti-HBc by a polyclonal assay (Biomedica). The diagno-sis of hepatitis on the graft was made after excluding othercauses of graft dysfunction, such as vascular complications,acute or chronic rejection, and biliary obstruction. Acute hep-atitis was defined as hepatocyte necrosis and portal or lobularinfiltration by mononuclear cells. Liver specimens were scoredaccording to a system derived from Knodell’s histologicalactivity index, described elsewhere,10 in which histologicalgrade (activity) and stage (fibrosis) were evaluated separately.

Statistical Analysis

Categorical data were compared using Chi-squared test orFisher’s exact test when indicated. When categorical variableswere ordered, comparisons were performed using Chi-squared test for trend. Continuous variables are expressed asmean 6 SD and compared by t-test. When a normal distri-bution could not be assumed, continuous variables are sum-marized as median and ranges and compared by Mann-Whit-ney test. Kaplan-Meier curves were calculated to estimatepatient survival over time and assess the cumulated probabil-ity of acquiring HBV infection in recipients of livers fromanti-HBc1 donors using the date of detection of HBsAg asthe end point. The log-rank test or Tarone-Ware test, whenappropriate, was used to compare probability curves. P of .05or less is considered statistically significant. All statistical pro-cedures were performed using SPSS for Windows version 8(SPSS Inc, Chicago, IL).

Results

Prevalence and Demographics ofAnti-HBc1 Donors

HBV serological results were available for 268 of 281donors of livers used in an initial liver transplant duringthe study period. No donor was HBsAg1, but 33 do-

52 Prieto et al

nors (12.3%) were anti-HBc1. Of these, 18 donorswere also anti-HBs1 and 14 donors were anti-HBs–.Anti-HBs status could not be determined in 1 of theanti-HBc1 donors. Anti-HBs status was available in229 of the 235 anti-HBc– donors: 26 donors (11.4%)were anti-HBs1, and 203 donors (88.6%), anti-HBs–.

Demographics of anti-HBc1 donors did not differfrom those of anti-HBc– donors except for age, whichwas significantly older in the former group (Table 1).The proportion of anti-HBc1 donors increased withage, being lowest (3.6%) in donors aged 1 to 20 yearsand highest (27.1%) in donors aged older than 60 years(chi-squared test for trend, P 5 .0001; Fig. 1).

HBV Infection After Transplantation

Thirty of the 33 patients who received a liver from ananti-HBc1 donor were evaluated for HBV transmis-sion. The 3 recipients who were not evaluated included2 patients with a diagnosis of hepatitis B and 1 patientwho died within 3 months. One hundred eighty-one ofthe 235 recipients of a liver from an anti-HBc– donorwere evaluated regarding HBV transmission, whereas54 recipients were not. These included 25 recipientswho died (n 5 20) or underwent retransplantationwithin 3 months (n 5 5), 27 recipients with a pretrans-plantation diagnosis of hepatitis B (25 chronic,2 acute), and 2 without adequate HBV serological fol-low-up after liver transplantation.

Thus, the study population for the HBV infectiontransmission study included 211 recipients without apretransplantation diagnosis of hepatitis B who hadHBV serological follow-up of at least 3 months withoutretransplantation; 30 patients received a liver from an

anti-HBc1 donor and 181 patients received a liverfrom an anti-HBc– donor. Baseline characteristics weresimilar between both groups of patients (Table 2).

HBV infection developed in 15 of 30 recipients(50%; 95% confidence interval [CI], 32% to 68%) oflivers from anti-HBc1 donors but in only 3 of 181recipients (1.7%; 95% CI, 0.4% to 5.2%) of livers fromanti-HBc– donors (P , .00001). The indications forliver transplantation in the 15 recipients who acquiredhepatitis B from an anti-HBc1 donor were HCV–related cirrhosis alone (n 5 6) or associated with alcoholabuse (n 5 1), alcoholic cirrhosis (n 5 6), cryptogeniccirrhosis (n 5 1), and cholangiocarcinoma (n 5 1). Theoriginal liver diseases in the 3 recipients with acquiredhepatitis B from an anti-HBc– donor were alcoholiccirrhosis (n 5 2) and HCV-related cirrhosis (n 5 1).Median time from transplantation to detection of hep-atitis B was 12 months (range, 3 to 24 months) inrecipients of livers from anti-HBc1 donors. HBV in-fection was detected at 6, 6, and 24 months in the 3recipients of livers from anti-HBc– donors. The cumu-lative probability of developing hepatitis B in recipientsof livers from anti-HBc1 donors was 6.7%, 20.5%,47%, and 56% at 3, 6, 12, and 24 months after livertransplantation, respectively (Fig. 2).

Overall, 18 of 211 liver transplant recipients (8.5%)who survived at least 3 months developed posttrans-plantation hepatitis B. Fifteen of 18 cases (83%) of denovo hepatitis B occurred in patients receiving a liverfrom an anti-HBc1 donor. The relative risk for devel-oping de novo hepatitis B from an anti-HB1 donorcompared with an anti-HBc– donor was 30 (95% CI,9 to 98). Serum HBV DNA PCR analysis was retro-spectively performed in 13 of the anti-HBc1 donorswhose recipients developed hepatitis B after transplan-tation and was positive in only 1 case.

Of the 26 recipients of livers from donors positiveonly for anti-HBs, 18 recipients could be evaluated

Figure 1. Prevalence of anti-HBc positivity in the donorpopulation according to donor age. The proportion ofanti-HBc1 donors increased with donor age (chi-squaredtest for trend, P 5 .0001).

Table 1. Demographics of Liver Donors According toAnti-HBc Status

VariableAnti-HBc1

(n 5 33)Anti-HBc–(n 5 235) P

Age (yr) 51 6 17 39 6 18 .0004Sex (M/F) 21/12 165/70 NSCMV1 26/31 (84) 169/226 (75) NSAST (U/L) 60 6 74 50 6 72 NSALT (U/L) 40 6 46 43 6 54 NSBrain death cause

Stroke 20 (60.6) 111 (47.2) NSTrauma 10 (30.3) 102 (43.4) NSAnoxia 1 (3) 15 (6.4) NSOther 2 (6.1) 7 (3) NS

NOTE. Values expressed as mean 6 SD or number (%).Abbreviations: CMV, cytomegalovirus antibody immuno-globulin G; AST, aspartate aminotransferase.

53Hepatitis B Following Liver Transplantation

regarding HBV transmission. These 18 grafts weretransplanted into 10 naıve recipients, 6 anti-HBc1recipients, and 2 anti-HBs1 anti-HBc1 recipients.None of these 18 recipients has developed acquiredHBV infection after a median serological follow-up of24 months (range, 3 to 48 months).

Factors Associated With Acquisition of HBVInfection in Recipients of Livers From Anti-HBc1 Donors

To define the factors associated with HBV acquisitionin the 30 recipients of livers from anti-HBc1 donors,several variables related to the donor (Table 3), recipi-ent (Tables 4 and 5), and transplant surgery (Table 4)

were analyzed. The risk for transmission of hepatitis Bwith a liver from an anti-HBc1 donor was not affectedby donor anti-HBs status. HBV infection developed inrecipients of livers from anti-HBs1 (8 of 16 recipients;50%) and anti-HBs– (6 of 13 recipients; 46%) donors.Of the remaining variables analyzed, only Child-Pughscore and HBV status in the recipient at the time oftransplantation were significantly associated with ac-quisition of HBV infection. None of the 4 recipientswho were anti-HBs1 at the time of transplantationdeveloped HBV infection compared with 15 of 26 re-cipients (58%) who were anti-HBs– (P 5 .10). Anti-HBs titers in the 4 anti-HBs1 recipients were 18, 73,192, and greater than 1,000 mU/mL. Length of HBV

Table 2. Baseline Characteristics of the Study Population According to Donor Anti-HBc Status

Variable Anti-HBc1 Donor (n 5 30) Anti-HBc– Donor (n 5 181) P

Age (yr) 53 6 11 52 6 13 NSSex (M/F) 18/12 108/73 NSCirrhosis 28 (93.3) 175 (96.7) NSChild’s class (NA/A/B/C) 2/5/15/8 6/30/97/48 NSChild-Pugh score 8 6 2 8 6 2 NSChronic HCV infection 16 (53.3) 116 (64.1) NSAlcohol abuse 11 (37) 52 (29) NSHCC 8 (27) 35 (19) NSIMS (cyclosporine/tacrolimus) 27/3 151/30 NSPre-OLT anti-HBs1 4 (13) 33 (18) NSPre-OLT anti-HBc1 5 (17) 54 (30) NSHBV follow-up (mo) 24 6 14 21 6 14 NS

NOTE. Values expressed as mean 6 SD or number (percent).Abbreviations: NA, not applicable; HCC, hepatocellular carcinoma; IMS, induction immunosuppression; OLT, orthotopicliver transplantation.

Figure 2. Cumulative proba-bility of developing hepatitisB in 30 recipients of liversfrom anti-HBc1 donors.

54 Prieto et al

serological follow-up without HBsAg detection in theseanti-HBs1 recipients was 12, 24, 6, and 24 months,respectively. In addition, 0 of 5 anti-HBc1 recipients(2 with associated anti-HBs positivity) developed HBVinfection compared with 15 of 25 anti-HBc– recipients(60%; P 5 .04). HBV serological follow-up withoutdetection of HBsAg in the 3 recipients with isolatedanti-HBc was 24, 24, and 24 months, respectively.Child-Pugh score was significantly higher in recipientsof livers from anti-HBc1 donors who developed HBVinfection than in those who did not (9 6 2 v 7 6 1; P 5.03). HBV infection developed in 1 of 5 recipients(20%) with Child’s class A, 7 of 15 recipients (47%)

with Child’s class B, and 6 of 8 recipients (75%) withChild’s class C (chi-squared test for trend, P 5 .14).

Outcome of Acquired HBV Infection inRecipients of Livers From Anti-HBc1 Donors

None of the 15 patients with acquired HBV infectionfrom an anti-HBc1 donor spontaneously lost HBsAgduring follow-up. At the time of detection of HBVinfection or shortly thereafter, all these patients hadHBV DNA detectable by molecular hybridization tech-nique. Serum HBeAg was positive in 14 subjects.

Initial liver biopsy samples were obtained at amedian of 97 days (range, 0 to 397 days) after detectionof HBsAg. Histological findings consisted of 3 patterns:chronic hepatitis of variable degree (12 patients), mas-sive hepatic necrosis (2 patients), and minimal changes(1 patient). In the patients with chronic hepatitis,median activity of hepatitis in the initial liver specimens

Table 3. Relation Between Donor Characteristicsand Acquisition of HBV Infection in Recipients

of Livers From Anti-HBc1 Donors

VariableHepatitis B(n 5 15)

No Hepatitis B(n 5 15) P

Age (yr) 51 6 15 48 6 20 NSSex (M/F) 11/4 8/7 NSCMV1 13 (87) 11 (79) NSAST (U/L) 74 6 99 55 6 48 NSALT (U/L) 43 6 56 41 6 40 NSBrain death cause

Stroke 7 (47) 11 (73.3) NSTrauma 7 (47) 2 (13.3) .11Other 1 (6) 2 (13.3) NS

NOTE. Values expressed as mean 6 SD or number(percent).Abbreviations: CMV, cytomegalovirus antibody immuno-globulin G; AST, aspartate aminotransferase.

Table 4. Relation Between Recipient and Transplant Variables and Acquisition of HBV Infection in Recipientsof Livers From Anti-HBc1 Donors

Variable Hepatitis B (n 5 15) No Hepatitis B (n 5 15) P

Age (yr) 50 6 9 52 6 12 NSSex (M/F) 10/5 8/7 NSChronic HCV infection 7 (47) 9 (60) NSAlcohol abuse 7 (47) 4 (27) NSHCC 3 (20) 5 (33) NSChild-Pugh score 9 6 2 7 6 1 .03IMS (cyclosporine/tacrolimus) 14/1 13/2 NSPatients with rejection 6 (40) 2 (13) NSIntraoperative RBC (units) 2 (0-11) 3 (0-8) NSIntraoperative platelets (units) 12 (0-36) 18 (0-36) NSIntraoperative FFP (units) 0 (0-8) 0 (0-6) NS

NOTE. Values expressed as mean 6 SD, number (percent), or median (range).Abbreviations: HCC, hepatocellular carcinoma; IMS, induction immunosuppression; RBC, red blood cells; FFP, fresh frozen plasma.

Table 5. Relation Between Pretransplantation RecipientAnti-HBs/Anti-HBc Status and Acquisition of HBV

Infection From Anti-HBc1 Donors

RecipientHBV Status

No. ofPatients

No. DevelopingHepatitis B (%) P

Anti-HBsPositive 4 0 (0) .10Negative 26 15 (58)

Anti-HBcPositive 5* 0 (0) .04Negative 25 15 (60)

* Two of these recipients were also anti-HBs positive.

55Hepatitis B Following Liver Transplantation

was 7 (range, 2 to 14). Fibrosis was present in 5 liverspecimens (4 specimens with fibrous portal expansion,1 specimen with bridging fibrosis) and absent in 7 spec-imens. In the 12 subjects with chronic hepatitis B,median serum alanine aminotransferase (ALT) levelswere 99 U/L (range, 28 to 131 U/L) at the time of theinitial liver biopsy. Serum ALT levels were elevated(.40 U/L) in 10 patients. Two patients developedmassive hepatic necrosis at 12 and 17 months aftertransplantation. Their serum ALT levels were 436 and1,089 U/L, respectively.

Eleven of the 12 patients with chronic hepatitis and2 patients with massive hepatic necrosis were treatedwith lamivudine (100 mg/d orally) at a median of64 days (range, 0 to 239 days) after the first liver biopsydocumenting hepatitis and a median of 205 days(range, 7 to 452 days) after detection of HBsAg inserum. Despite developing HBV infection, 12 of the 15patients were alive at the time of last follow-up 17 to 58months after transplantation (median, 39 months).These included the 2 patients with de novo HBV infec-tion presenting as acute liver failure. One patient whoresponded to lamivudine therapy with disappearance ofserum HBV DNA and loss of HBsAg died of hepaticfailure attributed to recurrent HCV-related graftcirrhosis 35 months after transplantation. A second pa-tient who responded to lamivudine therapy with disap-pearance of HBV DNA but without loss of HBsAgeventually died of disseminated lymphoma 28 monthsafter transplantation. A third patient died of sepsis ofurinary origin 1 month after detection of HBsAg,13 months after transplantation.

Survival Analysis

Mean follow-up after transplantation was 29 months(range, 0.2 to 58 months) for the 33 recipients of liversfrom anti-HBc1 donors and 28 months (range, 0 to 60months) for the 235 recipients of livers from anti-HBc–donors (P 5 not significant [NS]). Patient survival ratesat 1, 2, and 4 years were 88%, 85%, and 68% forrecipients of livers from anti-HBc1 donors and 85%,80%, and 76% for recipients of livers from anti-HBc–donors (P 5 NS, log-rank test). When analysis wasrestricted to the 211 individuals who survived at least3 months after transplantation, 2- and 4-year patientsurvival rates were 94% and 78% for those who devel-oped HBV infection compared with 90% and 83% forthose who did not (P 5 NS, log-rank test).

Discussion

The risk for a recipient to develop HBV infection froman anti-HBc1 donor has been estimated to be 33% to

78%.3-6 However, these results were mostly derivedfrom studies performed in the United States, where theprevalence of anti-HBc positivity in the liver donorpopulation is low (3% to 4%). In the present study, weanalyzed the risk for acquiring HBV infection in recip-ients of livers from anti-HBc1 donors in Spain, a coun-try in which the proportion of anti-HBc1 donors inthe general population is 10%. In contrast to previousstudies, either multicenter or involving a limited num-ber of patients, we analyzed the experience of a singlecenter with a large series of patients.

The proportion of anti-HBc1 liver donors at ourinstitution during 1995 to 1998 was 12.3%, a figuresignificantly greater than that reported previously.3,5,6

A significant and previously unreported association wasfound between donor age and the prevalence of anti-HBc positivity; a greater proportion of anti-HBc1donors had increased donor age. This greater rate ofanti-HBc positivity in older donors is likely related tothe decline in the prevalence of HBV infection in Spainin recent years, probably caused by improved social andhealth conditions. In Spain, the current policy of theONT is that all potential organ donors are tested onlyfor HBsAg as a marker of HBV infection. There are noguidelines for additionally testing potential liver donorsfor anti-HBc. Until routine testing for anti-HBc in thedonor is recommended by the ONT, the results re-ported here clearly show that donors aged older than60 years should be considered a high-risk group regard-ing anti-HBc positivity.

HBsAg– recipients of livers from anti-HBc1 donorsare at a significant risk for developing de novo HBVinfection after liver transplantation. This occurred in50% of our recipients of livers from anti-HBc1 donors,a figure in the range of previous series. Interestingly,although HBV infection was diagnosed in the majorityof recipients within the first year posttransplantation,HBsAg appeared in 2 individuals 2 years after trans-plantation, which underlines the importance of regularmonitoring of these high-risk transplant recipients forHBV infection even beyond the first year, regardless ofthe recipient’s clinical condition. In addition, this pol-icy resulted in the detection of HBV infection at earlystages when therapeutic interventions theoretically maybe of greatest benefit. In contrast to anti-HBc1 donors,donors positive only for anti-HBs do not yield an in-creased risk for HBV transmission, which supports thatthese donors can be used without restrictions, as re-ported elsewhere.5

The prevalence of acquired hepatitis B after livertransplantation was 8.5% during the study period(1995 to 1998), greater than the 2% to 3% reported inprevious investigations. Factors that may account for

56 Prieto et al

this difference include (1) our strict policy of posttrans-plantation monitoring for HBV infection regardless ofthe patient’s clinical situation, but mainly (2) the highincidence of anti-HBc1 donors in our area. Althoughanti-HBc1 donors represented only 12.3% of thetransplanted livers, 83% of acquired HBV infectionscould be ascribed to these donors. The difference inincidence of HBV infection between recipients of liversfrom anti-HBc1 and anti-HBc– donors (50% v 1.7%)suggests that other sources were not important in HBVacquisition. In recent years, efforts have been made toincrease the donor pool to cope with the increasingnumber of patients entering the liver transplant waitinglist. One such measure has been to increase the donorage limit; therefore, many centers are now acceptingdonors aged older than 60 years as suitable donors.However, this study shows that this policy may result ina greater prevalence of acquired hepatitis B, at least inareas where the prevalence of anti-HBc positivityamong liver donors is high.

An important decrease in the prevalence of acquiredHBV infection might have resulted from donor anti-HBc testing before transplantation and subsequent ex-clusion of anti-HBc1 donors. However, this policymight also have led to the inappropriate exclusion of15 liver donors who, although anti-HBc1, did nottransmit HBV infection to the recipients. In the currentera of donor organ shortage, the loss of these otherwisevalid donors does not seem justified. Investigation offactors that explain why some but not all anti-HBc1donors transmit HBV infection has not been clearlyestablished in previous studies. As noted in some re-ports,6 the risk for transmission of hepatitis B fromanti-HBc1 donors was independent of donor anti-HBsstatus. HBV infection developed in recipients fromanti-HBs1 (50%) and anti-HBs– (46%) donors. Sim-ilarly, the search for HBV DNA by PCR in serum ofanti-HBc1 donors who transmitted HBV infection totheir recipients was of no value because it was negativein nearly all cases. Testing donor livers for HBV DNAby PCR may be of value in predicting the risk for HBVtransmission11; however, PCR testing is usually un-available at the time of transplantation.

Of the many variables analyzed in this study, only 2related to the recipient were significantly associatedwith the acquisition of HBV infection from anti-HBc1donors: anti-HBs/anti-HBc status and Child-Pughscore at the time of transplantation. None of the recip-ients who were anti-HBs1 and/or anti-HBc1 devel-oped HBV infection after receiving a liver from ananti-HBc1 donor. It therefore appears that the pres-ence of anti-HBs in a liver transplant recipient confersresistance to HBV infection; thus, it would be safe to

direct hepatic allografts from anti-HBc1 donors tothese recipients without specific anti-HBV prophylaxis.The protective role of anti-HBs underlines the impor-tance of pretransplantation vaccination of all anti-HBs– recipients. However, given the low response rateto HBV vaccine even using high doses in patients withcirrhosis, it is important to verify the antibody responseto the vaccine before transplantation. Although oursample is too small to draw significant conclusions, itseems that recipients positive only for anti-HBc are alsoimmune to HBV infection when given an anti-HBc1liver allograft. However, Dodson et al5 reported that2 of 15 recipients (13%) positive only for anti-HBc whoreceived a hepatic allograft from an anti-HBc1 donoreventually developed hepatitis B. The need for specificHBV prophylaxis with hepatitis B immune globulin,lamivudine, or a combination of both agents in theserecipients remains to be determined in future studiesthat include a larger number of these patients.

The second variable associated with HBV acquisi-tion was the recipient’s Child-Pugh score at the time oftransplantation, a reflection of the integrity of the im-mune system, with cirrhotic patients with more ad-vanced liver disease at greater risk for developing denovo HBV infection when receiving a liver from ananti-HBc1 donor. Sicker patients would have a moredepressed immunity that would render them more sus-ceptible to acquire HBV infection. This previously un-reported finding needs to be confirmed in further stud-ies with a larger number of patients.

Liver transplant recipients acquiring de novo HBVinfection rarely, if ever, clear HBsAg. This is in contrastto the situation observed in the immunocompetentnaıve adult population in which the majority of patientsdevelop antibodies coincidentally with the eliminationof HBsAg from serum. Persistence of HBV infection isaccompanied in all instances by high viremia levels andgraft damage in most patients. In initial studies, ac-quired HBV infection was believed to follow a ratherbenign course compared with the aggressive evolutionof recurrent hepatitis B. However, some instances ofacquired hepatitis B presenting as fulminant hepatitisor progressing rapidly to cirrhosis have been reported.12

De novo HBV infection presented as acute liver failurein 2 of our patients and as chronic hepatitis of variabledegree in most of the remaining individuals. Despitedeveloping HBV infection, only 3 of the 15 subjectswith acquired hepatitis B from an anti-HBc1 donordied after a median follow-up of 36 months of causesunrelated to HBV infection. In addition, midterm sur-vival of recipients who acquired hepatitis B after receiv-ing a liver from an anti-HBc1 donor was not differentfrom that of individuals who did not acquire hepatitis

57Hepatitis B Following Liver Transplantation

B. Lamivudine therapy was started as soon as histolog-ical hepatitis was diagnosed, a situation that has proba-bly altered the natural history of acquired HBV infec-tion. Although few data exist on the efficacy oflamivudine in de novo hepatitis B,13 this drug has beeneffective in the treatment of recurrent hepatitis B.14

We conclude that in an area in which the proportionof anti-HBc1 liver donors is 12%, the prevalence of denovo HBV infection is greater than that reported inareas of low prevalence, and most de novo HBV infec-tions are related to the use of livers from anti-HBc1donors. However, in the current era of donor shortage,although anti-HBc1 donors transmit HBV infectionto at least 50% of their recipients and HBV infectionleads to graft damage in most cases, exclusion of thesedonors for liver grafting does not seem necessary. Ourresults suggest that in our area, testing of donors foranti-HBc is mandatory, particularly in older donors.With such information available before transplanta-tion, anti-HBc1 donors can be directed to appropriaterecipients with little or no risk for transmitting HBVinfection, provided certain strategies are followed.Grafts from anti-HBc1 donors can be offered first topatients who are anti-HBs1 or HBsAg1 at the time oftransplantation, the latter group because they are goingto be administered HBV prophylaxis routinely. Liversfrom anti-HBc1 donors can be directed next to recip-ients with isolated anti-HBc, although a low risk forHBV transmission appears to exist if no specific HBVprophylactic measures are taken. When no individualsmeeting these criteria exist on the waiting list, anti-HBc1 donor livers can be offered to naıve recipientswith critical clinical situation or with hepatocellularcarcinoma, although in such cases, prophylaxis of HBVinfection with either lamivudine, hepatitis B immuneglobulin, or a combination of both agents is necessary.

Acknowledgment

The authors thank Dr Sergio Rojter (Los Angeles, CA) for hisadvice in the early stages of this study.

References1. Chazouilleres O, Mamish D, Kim M, Carey K, Ferrell L,

Roberts JP, et al. “Occult” hepatitis B virus as source of infectionin liver transplant recipients. Lancet 1994;343:142-146.

2. Roche B, Samuel D, Gigou M, Feray C, Virot V, Schmets L, etal. De novo and apparent de novo hepatitis B virus infection afterliver transplantation. J Hepatol 1997;26:517-526.

3. Douglas DD, Rakela J, Wright TL, Krom RA, Wiesner RH. Theclinical course of transplantation-associated de novo hepatitis Binfection in the liver transplant recipient. Liver Transpl Surg1997;3:105-111.

4. Wachs ME, Amend WJ, Ascher NL, Bretan PN, Emond J, LakeJR, et al. The risk of transmission of hepatitis B from HBsAg(–),HBcAb(1), HBIgM(–) organ donors. Transplantation 1995;59:230-234.

5. Dodson SF, Issa S, Araya V, Gayowski T, Pinna A, Eghtesad B,et al. Infectivity of hepatic allografts with antibodies to hepatitisB virus. Transplantation 1997;64:1582-1584.

6. Dickson RC, Everhart JE, Lake JR, Wei Y, Seaberg EC, WiesnerRH, et al. Transmission of hepatitis B by transplantation of liversfrom donors positive for antibody to hepatitis B core antigen.The National Institute of Diabetes and Digestive and KidneyDiseases Liver Transplantation Database. Gastroenterology1997;113:1668-1674.

7. Suarez A, Viejo G, Navascues CA, Garcia R, Diaz G, Saro C, etal. The prevalence of hepatitis A, B and C viral markers in thepopulation of Gijon between 26 and 65 years old. GastroenterolHepatol 1997;20:347-352.

8. Salleras L, Bruguera M, Vidal J, Taberner JL, Plans P, Bayas JM,et al. Prevalence of hepatitis B markers in the population ofCatalonia (Spain). Rationale for universal vaccination of adoles-cents. Eur J Epidemiol 1992;8:640-644.

9. Baker B, Di Bisceglie AM, Kaneko S, Miller R, Einstone S,Waggoner JG, et al. Determination of hepatitis B virus DNA inserum using the polymerase chain reaction: Significance andcorrelation with serological and biochemical markers. Hepatol-ogy 1991;13:632-636.

10. Prieto M, Berenguer M, Rayon JM, Cordoba J, Arguello L,Carrasco D, et al. High incidence of allograft cirrhosis in hepa-titis C virus genotype 1b infection following transplantation:Relationship with rejection episodes. Hepatology 1999;29:250-256.

11. Uemoto S, Sugiyama K, Marusawa H, Inomata Y, Asonuma K,Egawa H, et al. Transmission of hepatitis B virus from hepatitisB core antibody-positive donors in living related liver trans-plants. Transplantation 1998;65:494-499.

12. Crespo J, Fabrega E, Casafont F, Rivero M, de las Heras G, de laPena J, et al. Severe clinical course of de novo hepatitis B infec-tion after liver transplantation. Liver Transpl Surg 1999;5:175-183.

13. Prieto M, Cordoba J, Rayon JM, Garcıa-Herola A, Berenguer M,Carrasco D, et al. Good response to lamivudine therapy in pa-tients with de novo hepatitis B following orthotopic liver trans-plantation: A pilot study [abstract]. Hepatology 1998;28:348A.

14. Perrillo R, Rakela J, Dienstag J, Levy G, Martin P, Wright T, etal. Multicenter study of lamivudine therapy for hepatitis B afterliver transplantation. Lamivudine Transplant Group. Hepatol-ogy 1999;29:1581-1586.

58 Prieto et al