|
|
||||||||
Articles |
1
Irwin Memorial Blood Centers, 270 Masonic Ave., San Francisco, CA 94118.
2
Department of Laboratory
Medicine, University of California, San Francisco, CA 94143.
a Author for correspondence. Fax 415-775-3859; e-mail ltobler{at}ccnet.com
| Abstract |
|---|
|
|
|---|
| early history of posttransfusion hepatitis (pth1 ) |
|---|
|
|
|---|
|
PTH was first reported in the US by Beeson in 1943 (2). Seven cases of PTH occurring 14 months after transfusion of blood or plasma were reported. The author commented that the expanding number of blood and plasma transfusions could lead to the occurrence of a considerable number of PTH cases. In 1964 Grady and Chalmers (3) reported the results of a retrospective study of PTH in nine Boston teaching hospitals, 19521962. In one of the hospitals 29% of the blood transfused was from commercial sources, while in the other eight hospitals blood only from volunteer blood donors was transfused. The incidence of clinically overt (i.e., symptomatic or icteric) PTH in recipients of blood products from volunteer blood donors was 0.6 cases/1000 units compared with 2.8 cases/1000 units in recipients of blood products from a mixture of volunteer and commercial blood donors.
In 1970 scientists at the NIH reported the results of a prospective
study to determine the incidence of icteric and anicteric hepatitis in
patients undergoing open-heart surgery (4). During surgery
the patients were given blood from either commercial or volunteer blood
donors. Icteric and anicteric hepatitis developed in 51% of the
recipients of commercial blood, whereas no hepatitis occurred in
patients who received blood from volunteer donors. These authors
estimated the hepatitis carrier rate for commercial blood donors to be
6.3% and for volunteer donors to be <0.6%. By 1971, the association
between a paid blood donor and an increased risk of PTH was accepted by
most, but not all, experts in the field. In early 1972, two states
(California and Illinois) considered legislation to specifically
eliminate payment for blood. The first law effective in eliminating
paid blood donations was passed on October 1, 1972 (i.e., The Blood
Labeling Act of Illinois). By the end of 1975, the Food and Drug
Administration (FDA) mandated an all-voluntary blood donor system in
which blood donors could not receive monetary payment for donations
(Fig. 1
) (1).
|
role of hepatitis b virus (hbv) in pth
A viral etiology for PTH was long suspected. In 1965, Blumberg et
al. (5) first described the Australia antigen and stated
that this antigen could be identified in the sera of many hemophiliacs
who had received multiple transfusions. Subsequently retrospective
studies indicated that the presence of the Australia antigen in donor
blood seemed to be clearly associated with the occurrence of PTH
(6). During an NIH Conference in 1970, additional evidence
was presented that linked the presence of the Australia antigen to the
occurrence of PTH. Furthermore, during this conference it was stated
that the Australia antigen was part of an infectious agent, presumably
a hepatitis virus (7). Nevertheless, the screening of
blood donors was not recommended. In a position paper written by Alter
et al. (8), the authors recommended testing and deferral
of blood donors based on the presence of the Australia antigen. (The
Australia antigen is now referred to as hepatitis B surface antigen,
HBsAg; ultimately the HBV was classified as a DNA virus in the
Hepadnaviridae.) These recommendations were not implemented
because of lack of consensus and concern over the general availability
and variable detection limits of early-generation HbsAg assays.
In 1970, Gocke et al. (9), using retrospective studies, estimated that the exclusion of HBsAg-positive blood donors through either first- or second-generation assays (e.g., agar gel diffusion or counterelectrophoresis) would decrease the rate of PTH by ~25%. The results of a prospective study on the effectiveness of exclusion of commercial and HBsAg-positive blood donors on the incidence of PTH was reported in 1972 (10). The exclusion of HBsAg-positive blood resulted in a 25% reduction in the PTH rate, as predicted, while the elimination of commercial donors resulted in a 70% reduction in the PTH rate. With the simultaneous exclusion of commercial and HBsAg-positive donors, the PTH rate was reduced to 7.1% of the prior rate.
On the basis of these studies, screening of blood donations for HBsAg began in 1971 and became a US federal regulation in July 1972. The initial methods used to detect HBsAg were relatively insensitive (i.e., immunodiffusion and counterelectrophoresis), and transfusion-associated HBV cases continued to occur. In 1975, federal regulations were implemented requiring the screening of all donor blood for HBsAg by one of the third-generation tests [i.e., radioimmunoassay or enzyme-linked immunoassay (EIA)], as well as the labeling of blood components with respect to the volunteer or paid status of the donor. Since that time, virtually 100% of all blood used in single-component transfusions has been collected from volunteer donors. The transition to volunteer whole-blood donors and routine third-generation HbsAg testing of all blood donations resulted in a marked decrease in HBV-PTH, although occasional cases continue to occur (11). In contrast to whole-blood donors, donors of plasma for further manufacture into plasma derivatives continue to be paid. The Cohen fractionation procedure used to prepare plasma derivatives dramatically reduces viral infectivity, and over the last decade, additional viral inactivation procedures (e.g., heating and detergent treatment) performed on these products render contemporary plasma derivatives virtually risk-free.
non-a, non-b pth and surrogate markers
In 1973, the causative agent of hepatitis A was detected by immune
electron microscopy on stools from patients with acute food-borne
hepatitis (12). The hepatitis A virus (HAV) was later
classified as a picornavirus closely related to the genus
Enterovirus (13). HAV is associated with acute
resolving hepatitis without a chronic carrier state. Retrospective
studies involving multiply transfused thalassemia patients indicated
that these patients were at no greater risk of HAV infection than
nontransfused children (14). Similarly, analysis of donor
and recipient samples from cases of PTH showed no evidence of
involvement by HAV (15).
After implementation of specific screening tests for HBV and exclusion of HAV as a cause of PTH, it became clear that a substantial proportion of PTH cases continued to occur that were not caused by infections with HBV or other known viral agents. Termed non-A, non-B hepatitis (NANBH), this entity represented 90% of residual PTH cases in the US. In the late 1970s and early 1980s, rates of NANBH in multiply transfused patients were reported to be as high as 10%.
The results of the multicenter prospective Transfusion-Transmitted Viruses Study were published in 1981 (16). Two major goals of this study were to define the incidence of PTH and to evaluate the factors influencing its occurrence. The data reported indicated a substantial association between recipients with NANBH and donor alanine aminotransferase (ALT) concentrations. Another independent study conducted at NIH confirmed an important association with an increased ALT concentration in donor blood and the development of NANBH in recipients of that blood (17). Furthermore, in 1984 Stevens et al. (18), representing the Transfusion-Transmitted Viruses Study, reported an important association between the occurrence of NANBH in recipients of blood that tested positive for antibody to the core protein of HBV (anti-HBc). The research group at NIH confirmed this observation (19). This study showed that anti-HBc testing of donors, in concert with ALT testing, might eliminate 3050% of recipient NANBH. Primarily on the basis of these studies, in 198687 blood collection agencies began screening donated blood for surrogate markers of NANBH (anti-HBc and ALT) (20). The rate of PTH among recipients dropped subsequently to as low as 23% (21).
hcv as major etiological agent of nanbh
Extensive research was conducted in the 1970s and 1980s to
identify the etiological agent(s) of NANBH. More than 25 preliminary
reports of associated agents were determined to be false. Then, in
1988, the hepatitis C virus (HCV) was identified with molecular biology
techniques by M. Houghton and associates at Chiron, in collaboration
with D. W. Bradley of the Hepatitis Branch of the Centers for Disease
Control (22). The process of virus discovery involved
construction of a cDNA expression library in the bacteriophage
gt11
by using high-titer plasma from a chimpanzee inoculated with PTH
plasma. The library was then screened for rare clones expressing viral
antigen with serum from a chronic NANBH patient as a presumed source of
viral antibodies. Screening of this library led to the identification
of the positive cDNA clone 5-1-1 (Fig. 2
) (23).
|
HCV is a single-stranded positive-sense RNA virus with a genome of ~9500 bases coding for ~3000 amino acids. This small RNA lipid-envelope (24) virus has been classified in the family Flaviviridae. The entire viral genome was sequenced within 1 year, and antigens were expressed for development of antibody detection assays. Early studies established that HCV was the etiological agent of at least 8090% of residual NANBH (21).
hcv screening assays and residual risk of hcv-pth
Blood collection agencies in the US implemented donor screening
immediately after licensure of the first-generation anti-HCV enzyme
immunoassay (HCV 1.0 EIA) in 1990. This EIA detected antibodies to an
antigenic protein (c100-3) of HCV (Fig. 2
). Even though this assay
facilitated the screening of blood donors for anti-HCV antibodies, it
did not detect all infectious blood donations (25) and had
a protracted window of infectivity ranging from 12 weeks to >26 weeks
postinfection (26). Nevertheless, a prospective study of
patients receiving transfusions before and after mid-1990 found that
the risk of transfusion-associated HCV infection per unit dropped from
0.36% (1 in 274) before anti-HCV screening to 0.07% (1 in 3300) for
donations screened with both surrogate markers and first- generation
anti-HCV tests (27)(28). It has since been
estimated that this test prevented transmission of HCV to 40 000
patients per year in the US (21).
A second-generation anti-HCV EIA (HCV 2.0 EIA) was licensed and rapidly
implemented in 1992. This test incorporated two additional proteins,
one structural (c22-3) and one nonstructural (c33c) (Fig. 2
). This test
was substantially more sensitive than HCV 1.0 EIA in detecting acute
and chronic HCV infections. Antibodies to these proteins generally
appear much earlier than those to c100-3, so the seroconversion window
period could be shortened by 1020 days. In addition, parallel studies
comparing first- and second-generation tests showed that HCV 2.0 EIA
detected additional HCV chronically infected donors at a rate of 1 in
1000 (25). This increased yield translated into prevention
of an additional 13 000 HCV transmissions per year by transfusion
(21). Conflicting estimates existed regarding the residual
risk of HCV after implementation of the second-generation assay.
Projections based on incidence-window period models suggested that the
risk of HCV PTH had dropped to as low as 1 in 100 000 per unit
(29), whereas extrapolation from historical PTH studies
suggested a risk of up to 1 in 5000 (30).
In 1996, a third-generation screening test (HCV 3.0 EIA) was licensed
that detected antibodies to an even greater number of HCV-encoded
epitopes. This test differs in antigen content from the
earlier-generation screening assays with deletion of the c-100
recombinant protein and addition of the NS-5 recombinant protein (Fig. 2
). This allows detection of antibodies to a greater number of
HCV-encoded epitopes. HCV 3.0 EIA has consequently narrowed the
seroconversion window by about 10 days relative to the HCV 2.0 EIA
(31)(32). This window period reduction is
projected to detect 12 additional seroconverting donors per million
units screened. Although the impact of the HCV 3.0 EIA on residual HCV
risk has not been prospectively quantified, the ongoing NIH prospective
study of PTH has failed to detect any HCV transmissions among >650
patients transfused with >2500 units of blood screened by
second-generation HCV EIA since late 1992 (H. Alter, personal
communication).
hcv confirmatory assays
Given the low prevalence of viral infection in preselected
volunteer donors, it is important that confirmatory assays are
developed and used to discriminate between true and falsely positive
EIA-reactive donors. The Chiron RIBATM HCV 2.0 strip
immunoassay (RIBA 2.0) was licensed on June 27, 1993. The RIBA 2.0 is
an immunoblot assay in which four recombinant HCV-encoded antigens
fused to human superoxide dismutase are immobilized on nitrocellulose
strips. A negative, indeterminate, or positive interpretation is based
on the reaction pattern present on the strip. About 89% of RIBA
2.0-positive specimens are HCV RNA-positive by PCR tests, while ~19%
of RIBA 2.0-indeterminate specimens are HCV RNA-positive by PCR tests
(25).
The HCV 3.0 EIA companion supplemental test, Chiron RIBA HCV 3.0 SIA (RIBA 3.0), is now pending licensure from the FDA. This assay uses a mixture of peptides, rather than recombinant antigens, for the 5-1-1/c-100 and c22-3 regions of the HCV genome. The use of peptides, when appropriate, eliminates those amino acid sequences that are a source of nonspecific cross-reactivity with other antigens (33)(34). On the basis of data from Europe (35), these changes have substantially improved this assay's lowest detection limits and specificity.
The prevalence of anti-HCV antibody in US blood donors, as determined with HCV 2.0 and 3.0 EIA confirmed by RIBA 2.0 and 3.0, respectively, ranges from 0.2% to 0.4% (21). At least 70% of seropositive donors are viremic (25). Similar rates have been observed in Europe, and higher rates in Japan and other Asian countries. Importantly, these rates are in highly selected blood donors (26). The prevalence of HCV infection in the general population of the US is currently estimated to be between 1% and 2% (36).
retention of surrogate tests
With virtual elimination of NANBH by HCV screening, the issue of
whether surrogate tests should be retained has surfaced. Although the
direct cost of ALT testing is low, the indirect cost is very high
because >1% of units test positive and are discarded. Anti-HBc
screening has a higher direct cost and results in ~0.5% unit loss.
Furthermore, the donors of these units are either temporarily or
permanently deferred. The preponderance of available data indicates
that, in the presence of anti-HCV testing, retention of ALT testing
prevents few if any residual PTH cases
(30)(37). Therefore, a 1995 NIH Consensus
Statement recommended that ALT testing of volunteer blood donors be
discontinued. On the other hand, the consensus conference recommended
that anti-HBc testing be retained, at least temporarily, for the
following reasons: possible prevention of some PTH HBV cases and
possible prevention of some cases of transfusion-transmitted HIV from
donors who test negative for anti-HIV during the window phase of
infection (20). Recent studies suggest that these
applications of anti-HBc have a very low predictive value and very poor
cost effectiveness (37). Therefore, discontinuation of
anti-HBc is being actively reconsidered.
hepatitis g virus (hgv) and other putative pth agents
Evidence indicates that additional viruses that explain rare
residual PTH cases may exist. Indeed, before HCV was identified there
was strong suspicion that more than one blood-borne viral agent was
causing NANBH (26). However, it has now become clear that
reinfection by different HCV strains is not uncommon, which explains
many of the cases of recurrent PTH that led to speculation regarding
multiple agents. In addition, the rate of non-ABC hepatitis in
recipients is now comparable with that in nontransfusion controls
(i.e., <0.8% in both US and Canadian studies, similar to background
values) (20). Nevertheless, the search for additional
agents has continued.
In 1995, researchers at Abbott Laboratories reported the isolation of three viral agents [GB viruses (GBV) A, B, and C] from tamarins inoculated with serum from a surgeon "GB" who had contracted acute icteric hepatitis 20 years earlier (38)(39). Only one of the isolates (GBV-C) proved to be a human viral hepatitis candidate; the other two were tamarin agents incidentally infecting the animals at the time of the experiments. A second independent group of researchers from GeneLabs performed molecular cloning with plasma from a patient designated PNF2161, who was originally identified by the CDC as having NANBH. The virus cloned by these researchers was designated HGV (40). Given the amino acid sequence matching of 95% between GBV-C and HGV, these viruses are therefore considered to be different strains of the same virus, tentatively termed GBV-C/HGV (41). GBV-C/HGV is a positive-stranded RNA virus classified in the Flaviviridae, the same family as HCV. Only 32% amino acid homology exists between GBV-C/HGV and HCV. Therefore, although GBV-C/HGV is distantly related to HCV, this agent is not a different serotype of HCV.
GBV-C/HGV RNA has been detected at a high prevalence (225%) among various high-risk groups, e.g., intravenous drug users, hemophiliacs, and hemodialysis patients. About 12% of blood donors have also been found to be RNA-positive (42). A recent study, using PCR technology, clarified the relationship of GBV-C/HGV and the occurrence of hepatitis (43). PCR was performed on selected patients from a surveillance study of acute viral hepatitis in four US counties. The patients included in this study were: 45 patients with a diagnosis of non-A-E hepatitis, 116 patients with hepatitis C, 100 patients with hepatitis A, and 100 patients with hepatitis B. The results of this study indicated that HGV was implicated as a potential etiological agent in only 0.3% of cases. These HGV cases had only mild liver enzyme increases. Furthermore, coinfection by HGV did not affect the clinical course in patients with hepatitis A, B, or C.
Initial studies involving the prevalence of GBV-C/HGV infection were hampered by the absence of an antibody detection system; therefore, the diagnosis of GBV-C/HGV infection depended on the use of PCR to detect GBV-C/HGV RNA. Recently an antibody test has been developed that has demonstrated seroconversion associated with viral RNA clearance in about two-thirds of infected persons (44)(45).
GBV-C/HGV is unequivocally a highly prevalent transfusion-transmissible agent (46). However, a causal relationship between HGV infection and hepatitis or other diseases has not been established (41)(43). GBV-C/HGV appears to explain only a small fraction of either transfusion- or community-acquired hepatitis cases, and these cases are very mild (43)(46). Thus at present, it is unclear whether screening of blood donors for GBV-C/HGV will be recommended.
The overall risk of PTH has declined markedly over the past 2 decades with virtual elimination of transmission of HBV and HCV infections. The advances in prevention of PTH have been not only extraordinary with regard to recipient safety, but also highly cost-effective. Indeed, the cost of screening the blood donor population for HCV is more than offset by the reduced healthcare costs achieved by prevention of PTH cases (30). Unfortunately, many infections did occur before the availability of screening assays. The tragedy of the transfusion AIDS epidemic also echos loudly. Thus, the current priority of infectious disease blood banks is further enhancement of blood safety, through elimination of rare residual cases of virally transmitted hepatitis by implementation of nucleic acid screening assays (47) and avoidance of complacency through proactive surveillance of the blood supply for new or emerging infectious agents.
| Footnotes |
|---|
| References |
|---|
|
|
|---|
The following articles in journals at HighWire Press have cited this article:
![]() |
S. L. Stramer, S. A. Glynn, S. H. Kleinman, D. M. Strong, S. C. M.T. (A.S.C.P.), D. J. Wright, R. Y. Dodd, M. P. Busch, and the National Heart, Lung, and Blood Institute Nucl Detection of HIV-1 and HCV Infections among Antibody-Negative Blood Donors by Nucleic Acid-Amplification Testing N. Engl. J. Med., August 19, 2004; 351(8): 760 - 768. [Abstract] [Full Text] [PDF] |
||||
![]() |
J. A. Salomon, M. C. Weinstein, J. K. Hammitt, and S. J. Goldie Empirically Calibrated Model of Hepatitis C Virus Infection in the United States Am. J. Epidemiol., October 15, 2002; 156(8): 761 - 773. [Abstract] [Full Text] [PDF] |
||||
![]() |
L. R. Fischer, D. H. Tope, K. S. Conboy, B. D. Hedblom, E. Ronberg, D. K. Shewmake, and J. C. Butler Screening for Hepatitis C Virus in a Health Maintenance Organization Arch Intern Med, June 12, 2000; 160(11): 1665 - 1673. [Abstract] [Full Text] [PDF] |
||||
![]() |
M. Vogt, T. Lang, G. Frosner, C. Klingler, A. F. Sendl, A. Zeller, B. Wiebecke, B. Langer, H. Meisner, and J. Hess Prevalence and Clinical Outcome of Hepatitis C Infection in Children Who Underwent Cardiac Surgery before the Implementation of Blood-Donor Screening N. Engl. J. Med., September 16, 1999; 341(12): 866 - 870. [Abstract] [Full Text] [PDF] |
||||
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |