Optimal blood donation screening annotation.

Abstract

Blood products from human blood donations can cause a number of side-effects that pose a considerable risk to recipients. One of the most serious transfusion-associated risks is the transmission of viruses. Comprehensive prevention and screening procedures have been developed in order to minimize the transmission of these transfusion-associated diseases. The establishment of a long-term, volunteer and unpaid donor population has proved very effective in combating blood-borne diseases. Other effective safety measures include the elimination of specific risk groups through a questionnaire that screens for risk-producing behaviour, as well as the thorough medical examination of prospective blood donors. However, blood donation screening for the most important transfusion-associated infectious agents, such as human immunodeficiency virus (HIV)-1 and -2, hepatitis B virus (HBV) and hepatitis C virus (HCV), still reflects the main pillar of said security measures. In many developed countries, these tests are accompanied by an additional screening for Treponema pallidum-specific antibodies and the determination of increased liver enyzmes, i.e. ALT (alanine aminotransferase); antibodies to the hepatitis B virus core antigen (anti-HBc) are only analysed for in some countries. In contrast, testing for cytomegalovirus (CMV) is only recommended for defined patients at risk, such as bone marrow transplant patients with compromised immune systems. Today, screening for the presence of HIV antibodies is based upon immunoassays which are highly sensitive in the detection of both HIV-1 and HIV-2 antibodies. The introduction of HIV p24 antigen tests in the USA could not significantly improve the safety provided by the conventional test systems. Among 18 million blood donors tested, three were p24 antigen-positive anti-HIV antibody negative (Lackritz, 1998). This was primarily due to the lower incidence rate than that expected, but also to the incapacity of this test to reduce the diagnostic window period significantly (Schreiber et al, 1996). Although the viral load in the blood was shown to be highest during the preseroconversion window phase, the p24 antigen test is not sensitive enough to detect the virus in the early phase of entry into the blood circulation. By testing seroconversion panels, we have confirmed that the polymerase chain reaction (PCR), as calculated by Schreiber, would provide a higher sensitivity and would therefore detect significantly more infectious donations from window period donors than the p24 antigen test. Only nucleic acid amplification techniques (NAT), such as the PCR, are sensitive enough to trace the virus in the blood stream as early as 1–2 weeks after infection. It is therefore obvious that the implementation of NAT is a suitable method to close the diagnostic window in blood donation screening (Schreiber et al, 1996; Roth et al, 1999). Within 2 (and in some blood banks 3) years of NAT blood donor screening in Germany, over 6 million blood donations have been screened by PCR testing of minipools for HIV, leading to the detection of one PCR-positive antibody-negative donation. This donation was positive by p24 antigen testing, which is not implemented in routine donor screening in Germany. The transmission of HIV by two components prepared from this donation could be prevented, albeit only through the expense of much effort. However, at the same time PCR failed to detect an additional donation in the preseroconversion window phase because of primers with reduced sensitivity for the individual HIV subtype and a low viral titre of the donation, resulting in the infection of the recipient. This donation was negative in p24 antigen testing. Without PCR testing, the true remaining risk for an HIV transmission in the German donor population (reflecting 80% unpaid volunteer donors) is estimated as 2 in 6 million and thus corresponds exactly to the residual risk calculated before the introduction of PCR testing (Schreiber et al, 1996; Glück et al, 1997). Using PCR testing, the residual risk could be reduced by at least 50% to 1 in 6 million. With calculated costs for PCR testing of US$ 2 per donation and an incidence of one infected donor that could be additionally detected within 3–6 million donations, the expense necessary to detect one infectious donor can be estimated at 6–12 million dollars. However, PCR screening is not only performed for HIV but also for HCV and HBV, lowering the proportionate costs for HIV screening to less than $ 1 per donation in Germany. With this comes the reduction of costs per detected infectious donor by 3–6 million dollars. The costs per prevented infection are lowered if random platelet concentrates are produced from whole blood donations. In the near future it will be possible, without much difficulty, to increase the sensitivity of the HIV PCR test so that fewer donations will be overlooked in the diagnostic window phase and simultaneously the cost per PCR will be curtailed significantly. In this way, the cost–benefit ratio will be significantly improved. These efforts could be diminished by licence fees for PCR testing. It has been suggested that HIV PCR screening of blood donations could replace antibody screening in the near future. Results obtained by Harvey Alter suggest that the introduction of HIV PCR testing could close the diagnostic window completely (Murthy et al, 1999). The authors were successful in transmitting the HIV infection to a chimpanzee through transfusion of PCR-positive blood, but were not successful, however, with PCR-negative blood from an HIV-infected chimp before seroconversion. Even if it were possible to completely close the diagnostic window, it is still not likely that we can do without the antibody testing. The highly active anti-retroviral therapy (HAART) for AIDS patients attempts to lower the concentration of the virus below the threshold of the most sensitive nucleic acid amplification tests (Carpenter et al, 1998; Pilcher et al, 1999). Therefore, a blood donation service using only the NAT test cannot identify these patients if they are donating blood by circumventing the exclusion measures. The optimal screening strategy for HIV would therefore include the highly sensitive antibody testing for all HIV subtypes including HIV-0 and HIV-2 as well as the HIV NAT testing. With this concept, the diagnostic window could be almost completely closed. In the presence of third generation antibody assays and the HIV NAT, the use of Treponema pallidum haemagglutination assay (TPHA) as a surrogate test for HIV is thus rendered superfluous because it may generate false positive results, require significant effort and, with respect to HIV, not contribute to the safety of blood products (Herrera et al, 1997). Although the individual test costs little, the benefit–cost ratio is only minimal. In order to prevent even one HIV transmission not being detected through antibody testing with the help of this test, at least US$ 16 million must first be spent (Herrera et al, 1997). When using PCR tests for detection of HIV, it is pertinent that the sensitivity and detection of the subtypes are significantly improved in the near future. The discovery of the hepatitis C virus in 1989 and the successive introduction of antibody screening tests have reduced the rate of post-transfusion hepatitis infection of unknown origin to under 10% world-wide (Choo et al, 1989; Aach et al, 1991; Donahue et al, 1992). In contrast to HIV, the preseroconversion window phase lasts longer – anywhere from 2 to 6 months. During this phase, antibodies have not yet had a chance to develop – but infectious virus is already prevalent in the blood. To shorten the diagnostic window, NAT procedures were introduced. In the period before induction of an antibody response, HCV usually appears at very high titres and quickly increases in concentration because of the high replication rate of the virus. Therefore, the German Paul-Ehrlich Institute imposed HCV PCR testing with a sensitivity of 5000 WHO units (equivalent to 10 000–30 000 virus genomes), sufficient to reduce the diagnostic window significantly. However, blood from a PCR-negative and antibody-negative donation was recently shown to transmit hepatitis C virus to the recipient of the blood component, indicating a viral concentration below the detection limit of the PCR and proving that PCR negativity does not in all cases imply non-infectivity (Schuttler et al, 2000). By comparison, hepatitis C has a higher prevalence in blood donors than HIV. This fact coupled with the longer diagnostic window causes the benefit–cost ratio of the HCV PCR to be superior to that of the HIV PCR. PCR screening of 6 million German Red Cross blood donors yielded the isolation of eight HCV PCR-positive antibody-negative blood donations. From an estimated $ 2 per test and 16 potential infections and resulting from two components per donation, the cost per prevented infection is $ 750 000. In Germany, the residual risk for HCV differs regionally to a great extent. Some university blood donor services within an urban setting and a high number of paid first-time donors reported one HCV PCR-positive antibody-negative donation out of 10 000–20 000 donors, indicating costs of $ 10 000–20 000 per prevented infection. By contrast, there are some rural Red Cross donation services with over 95% multiple time, unpaid donors who, in 1 million tested donations, have not had one single HCV PCR-positive antibody-negative donation at this time (Cardoso et al, 1998). This example shows the different benefit–cost ratios in relation to the donor population. Our Blood Donor Service detected one additional infectious donor out of 350 000 using the PCR method. In the case of HCV, an optimal screening strategy would imply both antibody testing and a sensitive PCR test. Alternatively, the newly introduced hepatitis core antigen test combined with the antibody test could be used, provided the high sensitivity claimed by the manufacturers is independently confirmed as it has recently been in France. Unlike PCR testing, this approach has not yet been used for mass testing. In contrast to other viruses, the HBV produces envelope protein (HBsAg) in vast excess during the period before seroconversion (Hollinger, 1996). For this reason, antigen tests with sufficient sensitivity were available for blood donation screening at a very early stage. With exposure to the neutralizing anti-HBs antibodies, the HBs antigen disappears. Antibodies against the core antigen (anti-HBc) persist for life and reveal, when isolated, that an infection has been overcome. These antibodies appear very early after the HBsAg and before the anti-HBs antibody. Although the value of anti-HBc in blood donor screening is controversial, several countries, including the USA, routinely run anti-HBc screening in addition to the HBsAg test. The primary goal was not to identify HBV-infected persons who were HBsAg negative, but rather to identify persons at risk for non-A/non-B hepatitis and for AIDS, who were likely also to be infected with HBV (Desforges et al, 1995). Because of the transient HBs antigenaemia, the HBsAg test is not applicable to the latter group. In contrast, anti-HBc shows the highest prevalence in all groups because of its longevity. Similar to other surrogate markers, anti-HBc has lost its significance as a surrogate marker since the introduction of highly sensitive antibody testing for HCV and HIV and especially since the implementation of NAT. On the other hand, the relatively high transmission risk for HBV of 1:20 000–1:50 000 by HBsAg-negative blood in the absence of anti-HBc screening shows clearly that the HBsAg test does not provide optimal donor screening (Schreiber et al, 1996). This residual risk is only acceptable because of the fact that HBV is less pathogenic than HIV and HCV. Whereas 75% of those infected with HCV suffer from a chronic infection, the percentage of those with a chronic infection of HBV is below 10% (Alter et al, 1992; Hollinger, 1996). Furthermore, rather than screening for HBV infection we can also opt for vaccination of donors. A vaccination for the general public would also be a realistic alternative providing protection not only for donors but also for recipients. The efficiency of this strategy was shown in Italy and Japan, where mass vaccination of children significantly reduced the prevalence and incidence rates among the younger generation. There are many reports that prove that a small proportion of HBsAg-negative and anti-HBc-positive donors are indeed viraemic and thus infectious. This means that countries with a high prevalence of HBV should be advised to use the anti-HBc test in addition to the HBsAg test when performing blood donor screening. In contrast to HIV and especially to HCV, the concentration of HBV in the blood after infection increases relatively slowly and over an extended period of time. This means that the concentration of HBsAg also increases relatively slowly. Although the most sensitive HBsAg tests can achieve a sensitivity of up to 3000 genome equivalents/ml, they are not sufficient either to reduce significantly or to close the diagnostic gap altogether. Because of the relatively high sensitivity of the HBsAg test, the benefit of the PCR method is less than that observed with the other viruses where reduction of the diagnostic window is more efficient. It may come to several days or even weeks (Schreiber et al, 1996; Drosten et al, 2000). A prerequisite for the efficacy of PCR is that it achieves the highest possible sensitivity, which must be significantly below 1000 genome equivalents/ml donor plasma. With a PCR of the above-mentioned sensitivity, we detected two donations in the preseroconversion phase and two from chronic carriers that were anti-HBc positive after the screening of more than 900 000 blood donations. The ratio of 1:250 000 is clearly below the calculated risk and seems, at least for the donor population of the German Red Cross, to be similar or even higher than that for HCV. Our data show that with PCR testing the diagnostic window in which HBsAg is negative could be significantly reduced and that a tiny proportion of anti-HBc-only positive donations may be PCR positive and probably infectious. This observation, together with reports on low-titre viraemia found in anti-HBc-only-positive donors and even in anti-HBs-positive HBsAg-negative donors, must inevitably force countries to reconsider current test strategies that only test donors for HBsAg and not for anti-HBc (Lai et al, 1989; Whang et al, 1991; Saito et al, 1999). Anti-HBc does not have significance as a surrogate marker but as a marker for persistent low-titre hepatitis B viruses. Therefore, an optimized donor screening for HBV would use PCR in conjunction with anti-HBc screening, instead of the sole HBsAg screening, in order to detect contaminated donations as quickly as possible after infection (PCR) and to identify chronically infectious donors, whether HBs antigen positive or not (anti-HBc). As HBs antigenaemia and HB viraemia overlap with the appearance of anti-HBc antibodies, a new diagnostic gap would not emerge. The disadvantage to such optimized donor screening is a drastic cost increase. The anti-HBc test is currently more expensive than the HBsAg test, which, together with the cost of PCR testing, burdens the balance of the cost–benefit ratio. Further costs develop because more donors are anti-HBc than are HBsAg positive, including a significant proportion who are not infectious (Sun et al, 1988). To realize this optimized donor screening, it is mandatory to develop validated and practical HBV PCR procedures. Alternative socioeconomic strategies include mass vaccination programmes that would render donor screening for HBV over time superfluous, provided all persons of a given population are vaccinated. The TPHA test as evidence of infection with the syphilis bacterium Treponema pallidum is of the utmost importance in those countries with a high prevalence of sexually transmitted diseases. This test identifies not only the donors who are infected with Treponema pallidum but also those who are at risk of HIV and HBV infection. In this way, a group of donors at high risk of a variety of transfusion-related diseases can be prevented from blood donation (Desforges et al, 1995). At this time, the test is practically of no importance in countries with proper health care. As already mentioned, the use of surrogate tests to identify persons at risk of AIDS or hepatitis in industrial countries is of little help (Desforges et al, 1995; Herrera et al, 1997). The function of this test in preventing the transmission of syphilis comes into play here (Desforges et al, 1995). Treponema pallidum is, however, a highly sensitive bacterium that loses its infectivity shortly after exposure to oxygen. There are only a few indications and little proof in the literature of the transmission of Treponema pallidum through blood or blood products. Even in platelet concentrates that are stored at 22°C, the bacteria lose their infectivity through the effect of oxygen and plasticizer (van der Sluis et al, 1985). For these reasons, the test would be dispensible in industrial countries. Alanine aminotransferase (ALT), an intracellular enzyme, is released upon the destruction (either toxic or infectious in nature) of hepatocytes and other cells. As specific antibody testing for HBV and HCV has been introduced, the ALT test is only indicated for unknown infectious agents or viruses that are not tested for [cytomegalovirus (CMV), Epstein–Barr virus (EBV)] and that induce hepatitis. With the introduction of PCR for the detection of HCV and HBV, the test loses all significance as a marker for these two virus types. Moreover, because of unknown causes, the ALT often increases non-specifically in many donors. Strictly applying the normal value would lead to an enormous deferral rate of blood donors, which is medically and economically unjustified. For that reason high limits were set that, however, only occur during the acute phase of massive infection very shortly before the antibody or antigen screening tests positive. ALT remains in the normal range for a long period of time during the preseroconversion window phase of HBV and HCV, although viruses replicate and donors are infectious. Moreover, in cases of chronic low-titre HBsAg-negative HBV infections, the ALT usually reflects normal values. Whereas in donors screened with second generation tests ALT testing only detects three or four additional HCV-infected donations – the PCR would find 5–7 (Busch et al, 1996). The only remaining potential usage lies in the identification of unknown infectious agents, which cause up to 10% of all cases of post-transfusion hepatitis (Alter, 1994). The risk of transmission of unknown infectious agents from donors with elevated ALT seems extremely low because ALT usually returns to normal levels after two or three donations without signs of viral infection. The same holds true for recipients of blood products manufactured from previous ‘window’ donations. (Desforges et al, 1995; Mosley et al, 1996). ALT testing is made superfluous by PCR screening for HCV and HBV – a viewpoint that is currently being favoured by blood donation services in the USA. In recent years, several new viruses have been discovered with the help of molecular biological methods, especially the representional difference analysis (RDA), that were initially regarded as hepatitis viruses (Simons et al, 1995; Nishizawa et al, 1997). These were held accountable for the small percentage of post-transfusion hepatitis cases of unknown origin. The GBvirus-C/hepatitis G virus (GBV-C) was discovered in 1995 and 1996, and its prevalence in blood donors was rather low compared with recipients of blood products and hepatits B and C patients (Simons et al, 1995, Linnen et al, 1996). This indicated a similar transmission route for the newly discovered virus as for HCV and HBV. Initially, higher ALT values were found in GBV-C-infected patients that could not be confirmed in later studies (Alter, 1997). An association of this virus with hepatitis appears to be rather insignificant as viraemic blood donors were healthy with normal ALT values and the infection generally takes an asymptomatic course in the recipients of the contaminated blood products. Screening of blood donations for GBV-C is not indicated as: (i) there is no evidence for any disease that could be caused by the virus; (ii) it is highly prevalent in the general population when seropositive and viraemic individuals are considered; and (iii) the primary transmission most probably occurs intrafamilially by vertical and sexual routes. The only reasonable screening method would be PCR as antibodies are associated with viral clearance. The TT virus, first described in 1997, was isolated from a Japanese patient (T.T.) with hepatitis of unclear origin (non-A to non-E) (Nishizawa et al, 1997). Initially, an association with hepatitis was also reported with figures similar to those of GBV-C. Studies in Western countries could not, however, confirm the Japanese findings. An association with hepatitis and elevated ALT could not be found and nor could a corresponding illness for this virus. Here too, the prevalence in the healthy population was initially reported to be low, but with more sensitive PCR techniques and better primers it was shown to reach over 90% in certain countries. The obvious ubiquity of this virus indicates that blood transfusion does not play the major (if any) role in its spread. Because it is so widespread and no specific pathology has been attributed to the virus, it is most probably an innocent bystander. Screening for this virus is not necessary. The most recent virus that could be isolated from patients with hepatitis of unknown origin is the SEN-V. This is the only new virus that might play a role in transfusion-associated hepatitis of unknown origin. The first studies published in newspapers – not peer review journals – by the pharmaceutical company that discovered the virus show a prevalence of under 1% in blood donors and a very high occurence of this virus among patients with chronic non-A/E hepatitis of unclear origin. Furthermore, a strict association with the transfusion of blood and blood products (as is the case for HCV) can be proved. Should further studies indicate the significance of this virus in the occurence of post-transfusion hepatitis, it would be a major candidate for additional screening in blood donation. The prevalence of HTLV-1 in most Western industrialized countries may be as low as 1 in 100 000. Most of the donors testing HTLV positive originate from high-risk areas or have had sexual or parental contact with persons who live in such areas. Only a small percentage of those infected developed specific symptoms. Screening in these countries is therefore unnecessary. It is, however, indicated in areas of high prevalance. CMV is widespread and is transmitted through normal human contact. Because of its restricted pathogenicity (in immune-suppressed patients), only those donations destined for transfusion to immunocompromised patients need be tested. Prestorage leucocyte reduction will further reduce the necessity for HTLV-1 and CMV testing.