Blood donation testing and the safety of the blood supply - Practice in Blood Centres and Hospitals

Practical Transfusion Medicine 4th Ed.

20. Blood donation testing and the safety of the blood supply

Richard Tedder¹, Simon J. Stanworth² & Mindy Goldman³

¹NHSBT/HPA Epidemiology Unit, NHS Blood and Transplant, Colindale, London, UK

²University of Oxford and NHS Blood and Transplant and Department of Haematology, John Radcliffe Hospital, Oxford, UK

³Canadian Blood Services, Ottawa, Ontario, Canada

Introduction

This chapter describes the aims and methods of laboratory testing of blood donations. It focuses not only on the range of tests currently employed but also on operational aspects crucial for the safe and efficient application of this process to the thousands of samples received in a blood centre laboratory each day [1,2]. Testing is dealt with under three headings:

· red cell serological testing;

· microbiological testing and donor follow-up;

· operational and quality control issues.

Red cell serological testing

It is mandatory to test every blood donation for:

· ABO blood group;

· RhD blood group; and

· presence of irregular red cell antibodies.

The results from these tests are necessary for safe transfusion practice in order to reduce the risk of premature destruction of the transfused donor red cells in a recipient's circulation due to immunological incompatibility towards the major red cell antigens. Correct ABO blood group typing is critical, since naturally occurring antibodies can cause intravascular hemolysis and severe transfusion reactions if incompatible components, particularly red cells, are transfused. The RhD antigen is highly immunogenic and RhD negative recipients should only be transfused with RhD antigen negative red cells to avoid alloimmunization.

A more extensive red cell phenotype, including full Rh and Kell typing, may be performed on the entire inventory or a subset of donations, with variable practice between blood services. More extensively phenotyped red cells are needed for transfusion support of particular patient groups (e.g. thalassaemia, sickle disorders) where risks of alloimmunization are high because of the patient requirement for multiple red cell transfusions. Some blood services, such as the National Health Service Blood and Transplant (NHSBT) supplying hospitals in England and North Wales, have current policies to perform full Rh and Kell phenotyping on all donations, largely for operational reasons; others operate phenotyping on selected units for specific patient groups to prevent alloimmunization by selecting Rh and Kell compatible blood.

Samples

Tests are carried out on anticoagulated venous blood samples collected at the time of donation. The samples are identified by a unique bar-coded identification system, which in most countries is an International Society for Blood Transfusion (ISBT) 128 number consistent with the aims of international conformity in blood group labelling and which ensures that each donation has a unique number. Detailed specifications and guidance on the testing reagents required for blood grouping can be found in appropriate documents, such as the UK Blood Transfusion Service Guidelines (‘Red Book’) or the AABB Standards for Blood Banks and Transfusion Services. The following paragraphs highlight some key operational principles.

ABO grouping

Donor red cells are tested with monoclonal anti-A and anti-B antibodies, which are capable of detecting all subgroups of these red cell glycoproteins. A reverse grouping is also performed by testing the donor plasma with A and B reagent cells. The exact cells specified varies by blood transfusion service (e.g. NHSBT uses Group A1 and B reagent red cells for new donations).

Most blood services make use of automated systems for serology testing where batched samples are divided into separate microtitre plate wells. The test results are read photometrically and the pattern of results obtained from testing donor red cells and donor plasma analysed by microprocessors to establish the ABO blood group result for a particular donation. The forward and reverse ABO group must be concordant in order to assign a donor blood group. In the case of repeat donors, such a system also allows the results for ABO groupings to be compared with those generated previously.

RhD grouping

RhD grouping is performed by testing donor red cells with two different highly sensitive monoclonal anti-D reagents. In many countries, RhD-negative first-time donors undergo further testing using an alternative method to confirm that they are D-negative. Use of sensitive reagents and repeat testing are done to optimize the detection of weak or partial D-bearing red cells. This would include all the weaker Rh variants, including category D^VI. It is felt to be essential that blood services identify and consider all such donors as RhD positive in view of the highly immunogenic capability of the D antigen, even if some of these donors may be considered as D-negative if they were transfusion recipients or prenatal patients.

Detection of irregular blood group antibodies

Donor samples are tested to exclude the presence of red cell antibodies that could cause reduced red cell survival or haemolysis when transfused into recipients whose red cells are positive for the relevant antigen(s). This test is termed an antibody screen and involves the mixing and testing of donor plasma with group O R₁R₂ K-positive red cells, which are also positive for the majority of other red cell antigens thought to be clinically significant.

Blood services are largely concerned with the detection of high levels of clinically significant antibodies in donations. Methods and reagent cells used to detect antibodies are less sensitive than those used for pretransfusion testing in hospitals. Red cell units for transfusion are suspended in anticoagulant nutritive solutions and, depending on the manufacturing and processing process in use, will contain only a small volume of plasma. Weak antibodies in donor plasma will therefore be considerably diluted during processing or transfusion. In contrast, hospital blood bank practice initially requires stringent detection of any antibodies in a potential recipient, irrespective of the level, in part due to the possibility of developing a secondary immune response on re-exposure to the same antigen.

In some blood services, such as in the UK, blood for neonatal transfusion is tested for irregular antibodies to a higher level of sensitivity than standard testing for all other blood in order to further minimize the very small risk of transfusion reactions due to passive transfer of antibodies in this specific group of patients.

High titre anti-A and anti-B

Some group O donors may have (unexpectedly) high titres of anti-A and anti-B in their plasma that could cause lysis of A and/or B cells, particularly where large volumes of plasma are transfused. Recipients should receive group-specific or AB plasma to avoid haemolytic reactions.

Standard practice in hospitals is to transfuse group-specific red cells to all recipients. However, group O red cells may be transfused to certain groups of patients, such as neonates and patients requiring urgent transfusion before their blood group is known. In practice, because most red cell units are stored in additive solutions for preservation, the amount of plasma ultimately transfused is very small, so any risks of haemolysis for a transfusion of group O red cells with high-titre anti-A and anti-B are very low. It might be considered important to screen for hemolysins for large-volume red cell transfusions when planned for neonates.

Since platelets for transfusion have a short shelf life, it is not always possible to provide group-specific platelet transfusions. Several cases of haemolysis have been reported in group A recipients receiving group O platelets and, more rarely, in group B recipients receiving group O platelets; paediatric/neonatal patients appear to be at highest risk.

Plasma containing high-titre haemolysins can be screened in the blood service laboratory by observing the reactions between donor plasma and a diluted sample of reagent A₁B red cells; products that are negative for testing for high titre haemolysins are then labelled for issue. This can be done using automated systems.

Very occasionally, high titre anti-A may be found in group B donations (and vice versa). Recent refinements to testing for high titre haemolysins include methods to assess only the more clinically relevant IgG (rather than a combination of IgG and IgM) fraction. There is no standard method of testing for high titre haemolysins and the acceptable cut-off titre varies greatly with the technique used, thereby requiring local assessment of the procedures used.

Supplementary testing

Not infrequently, anomalies appear in some of the above test results and will preclude accurate conclusions based on the test results. For example, it has been estimated that 1 in 10 000 blood donors have a positive direct antiglobulin test (DAT) at the time of donation, which could interfere with some of the above assays. Weakly positive DAT may also not be detected in the routine grouping tests, which include a control for donor red cells mixed with inert serum. These donations may cause problems in hospital blood banks, since they would appear incompatible after crossmatching by indirect antiglobulin test (IAT). Subsequent donations from these donors will be ‘flagged’ and monitored as the positive DAT may be transient. Donors with positive DAT results on several donations may be deferred and advised to see their physician in order to have an evaluation done for possible clinical significance and underlying disorders.

In some cases where samples from donations give anomalous automated blood grouping results, the blood service laboratory may have to resort to manual techniques to identify the blood group or antibodies correctly. In general, only antibodies reacting in the IAT are considered to be clinically significant. In the case of some donations with, for example, identified anti-D or anti-c at low levels, the red cells may still be released for transfusion, since during component preparation the amount of antibody-containing plasma will be very small and diluted with an additive storage solution.

Phenotyped red blood cells

Many blood services undertake a more comprehensive red cell antigen phenotyping service in order to identify donors whose red cells could be used for transfusion to alloimmunized recipients or to patients at high risk of forming multiple alloantibodies, e.g. sickle cell disease. This involves phenotyping for Rh C/E and Kell, if not routinely done on all units, as well as for S, s, Fy^a/b, Jk^a/b, Kp^a and Lu^a. Where donations are tested and found negative for the antigens listed above, this information may be printed on the blood group label to aid hospitals with selection of blood for patients with antibodies.

In other selected groups of donations, different specific red cell phenotyping may be arranged. Testing may be done on individuals of Afro-Caribbean origin, to meet the needs of sickle cell anaemia patients for antigen-matched units. For example, the U antigen is far more likely to be absent in Afro-Caribbean individuals than in Caucasians. This facilitates the provision of U-negative blood required for transfusion to those individuals who have developed anti-U, which is a clinically relevant antibody.

Increasingly, molecular biology microarray technology is being used to perform mass genotyping of donors for multiple blood group systems. Results for donors missing a high frequency antigen, or with a combination of antigen negative alleles, are then reconfirmed using serological methods, since genotyping methods are not currently licensed.

Testing for HbS may be performed on a subset of units with particular phenotypes likely to be used for transfusion support for patients with sickle cell disease or to neonates during exchange transfusions. The need for a sickle cell screening test depends on the prevalence of HbS within the donor population. An additional consideration is the need to provide counselling support to inform donors found to be carriers of HbS. Of recent interest, it has been found that sickle trait (HbAS) blood significantly interferes with the function of some filters currently used for leucocyte reduction (Chapter 21). Such ‘failed’ donations would be discarded, but the pattern of red cell antigens in these individuals could be unique and very useful as a transfusion resource. In addition, HbAS units do not freeze well using current methods.

Microbiological testing of blood donations and donor follow-up

A wide range of infectious agents have been documented as transmissible by blood transfusion and these are described in Chapter 13. Donor selection criteria and the use of established guidelines to defer individuals at risk of infection are the important first steps aimed at reducing the risk of collecting blood donations with the potential to transmit infection. For some agents, no laboratory testing is currently available and donor criteria are the only means of deferring at-risk donors. Where testing is available, donor criteria are still important, particularly with respect to the collection of blood from donors who may be in the ‘window period’ of an infection where they are asymptomatic and have negative testing results, but are still infectious. Laboratory screening tests form the core of the process to identify infected blood components prior to transfusion [3].

Samples

Tests are usually carried out on serum or plasma venous samples collected at the time of donation and sent to highly automated centralized donor testing laboratories. In certain European countries, such as the Netherlands, donors undergo an initial screening process where eligibility is assessed using a questionnaire and blood samples are taken for testing without the collection of a unit of blood. In other countries this happens at the time of the first and all subsequent donations. As with samples for serological blood grouping, correct labelling of microbiological samples to ensure traceability of results is extremely important. Most tests are performed on individual donor samples, but nucleic acid testing (NAT) is often performed on small pools of samples from 6 to 24 donors, termed minipool testing; serological testing for HTLV infection may also be conducted on minipools.

Testing process and donor management

Sensitivity and specificity are important test attributes. Sensitivity refers to the ability of the test to identify truly infected individuals correctly. From the perspective of the transfusion recipient, sensitivity is the most important criterion for a laboratory screening test, i.e. the test will accurately identify infected donors. Specificity refers to the ability of the test to identify correctly donors who are not infected. Specificity is important both to avoid discarding donations from donors who are actually not infected and to reduce the resulting confirmatory workload necessary to provide appropriate donor counselling of those whose samples are reactive in a screening assay. Although most currently used screening assays show remarkably high levels of both specificity and sensitivity it is essential that additional assays are available to confirm infection in the donor.

Principles of investigating a repeat-reactive sample

If an initial screening test is reactive, it will be repeated on the same sample. If the repeat test, in duplicate, is negative, the overall result is considered negative, the blood donation will be used and the donor may continue to donate. If the repeat test is again reactive, a confirmatory or supplementary test is performed to establish whether the screening test result represents a true positive donor sample. Since donors usually constitute a low prevalence population, despite the high specificity of screening tests, there will be significant numbers of samples from donors being identified as repeatedly reactive on screening but who are not confirmed to be positive on supplementary testing (termed false-positive or nonspecific reactions). These donors are deferred from further donation, although some blood services permit donors with false-positive results to be re-tested after a defined deferral period and to be re-integrated as donors if all test results are negative or where supplemental testing indicates that a reaction is falsely positive and an alternative assay of similar sensitivity is negative. These algorithms are termed donor re-entry protocols.

Blood services must have policies for notifying donors with repeat reactive test results and particularly for those where the reactions indicate a donor to be infected. A post-test discussion may be carried out by blood service personnel or information may be forwarded to the donor's general practitioner or physician for further discussion of the results and advice regarding any personal, family and public health measures. In addition, in some countries, the law requires forwarding of details on a first identification of an infected person for some infections, such as HBV, to public health authorities. For donors with false-positive test results, it can be difficult to explain that although the test almost certainly represents a false-positive reaction, the individual may be deferred as a blood donor. This may be explained by an understandable wish to avoid all the very necessary activity undertaken by a transfusion service every time a ‘reactive’ donor is encountered.

When a sample from an established donor is found to be repeat reactive for one of the mandatory microbiology tests, e.g. HIV or HBV, components from previous donations that may still be in the inventory will be retrieved and discarded on the basis of wishing to exclude a potential window infection at the time of the earlier donation. If the donor is confirmed to be infected, any components remaining in the inventory will be recalled. On a case-by-case basis recipients of earlier components from the donor will be identified, notified and offered relevant testing and management, a process termed ‘lookback’. Archived samples of plasma from the last negative donation may also be retrieved and tested with assays more proficient in identifying window infections. Where it is found that an infective plasma donation may have entered a plasma pool prior to fractionation, the fractionators should be informed. Again, on a case-by-case basis, all involved products should be identified and consideration given to their removal from inventory and notification of recipients. With improvements in testing and shortening of the window period, the likelihood of identifying an infected recipient and an infectious component/product on lookback investigation has decreased substantially.

Principles of infectious disease testing methodology

Screening tests may detect the host immune response to the microbial agent (such as antibody to HCV), a microbial antigen (such as the hepatitis B surface antigen, HBsAg) or the nucleic acid of the microbe (NAT). For bacteria, although NAT could be applied, screening tests may detect a component of the organism or a by-product of bacterial growth. Testing for bacterial growth is covered in Chapter 14.

Immunoassays

Immunoassay principles, using enzyme or chemiluminescent techniques of detection, have formed the basis for infectious disease testing [4,5]. Traditionally, donor plasma at a fixed dilution is incubated over a solid phase where an antigen–antibody interaction occurs. After incubation and then washing, the products of the antigen–antibody interaction are detected by a revealing agent. Detection may involve a conjugate linked to an enzyme, usually peroxidase, which can be detected photometrically after addition of substrate, which produces colour, or by chemiluminescence, in which the optical measuring device detects photons emitted by the chemiluminescent reaction. Until relatively recently, serological assays were constructed to detect either antibody, the most commonly used modality, or antigen, a modality almost exclusively used for detection of HBsAg. The recognition of viral antigenaemia as a feature of early window infections spawned the development initially of antigen-only assays for HIV (p24Ag) and then for HCV (p22Ag). Though not analytically as sensitive as NAT they have proved useful diagnostically.

The more recent development of combined antigen and antibody assays, often referred to as ‘Combo assays’, provides operationally convenient single well tests for the detection of infection in the acute window phase as well as the antibody response to infection. Many blood services adopted the HIV Combo assay but few, if any, have taken up the HCV Combo assay in view of already having established HCV NAT. The sensitivity reduction of NAT through pooling renders sensitive antigen-only assays on individual donation testing an analytic advantage, such that HBsAg assays are similar in proficiency to HBV DNA NAT for detecting acute phase infections. However, the assay conditions for combined antigen/antibody detection disfavour the antigen modality and the resulting analytic sensitivity for antigens in the Combo assays appear somewhat reduced. Nevertheless, in situations where the frequency of incident (i.e. acute) infections is common Combo assays may still have merit if NAT assays are not available.

Antibody assays for donor screening must be of the highest sensitivity for detecting early serologic responses in the acutely infected donor. Since the in vivo development of detectable antibody is a host response to microbial antigens this marker is invariably delayed in the infection time course, which has led to the use of a term ‘antibody window’ – hence the phrase ‘window infection’ used previously. Proficiency at this stage requires characterization of the early antibody response and enhancement of the detection of the specific early antibodies, e.g. anti-p24 in HIV and in some cases enhancing detecting IgM class antibody. Once the antibody response has matured the remaining concerns centre around whether the antibody response to extreme microbial variants can still be detected on the ‘routine’ assay. One example was the realization that HIV 2 infection could not be reliably detected by early HIV tests based on HIV 1 components; similarly, the detection of malarial antibodies is influenced by the infecting species.

All immunoassays depend on the interaction between microbial antigens and antibodies. Where an antibody is a component reagent of the detection system there has been a shift to the use of monoclonal antibodies. These are chosen to have high avidity and are often directed at ‘conserved’ antigenic epitopes. The resulting narrow specificity is an advantage in terms of driving down background rates of nonspecific reaction, but this comes at a cost of susceptibility to mutations in the target epitope, rendering the antigen undetectable. Assays for HBsAg detection used in donor screening must be secure in detecting HBsAg mutants, especially the classical vaccine escape variant G145R.

Nucleic acid testing (NAT)

In NAT, nucleic acid is extracted from the donor plasma. A nucleic acid amplification test such as polymerase chain reaction (PCR) or transcription-mediated amplification (TMA) is then used to amplify and detect microbial genetic sequences. Testing is usually done on small pools of from 6 to 24 donor samples, termed minipools, depending on the methodology used. Single-donor testing may be considered in particular circumstances to enhance sensitivity but such a policy will increase unit costs. Single-sample testing may also be required for resolution of a reactive pool to determine which donor sample contains the microbial target. Testing may be done for each agent individually or to identify several agents (HIV, HBV and HCV) simultaneously in a single reaction, when they are termed multiplex assays. Newer, completely automated platforms have reduced the operational complexity of testing. NAT, like antigen testing, reduces the window period when donors may be infectious but have negative serologic testing results in the period before the immune response becomes detectable. Window periods using serological assays are estimated in the order of 59 days for anti-HCV, 15 days for anti-HIV and 67 days for HBV. Window periods using minipool NAT are estimated in the order of 8 days for HCV, 9.5 days for HIV and 38 days for HBV. The utility of NAT depends on the incidence of these acute infections in the donor population, which in turn determines the incidence of window period donations that would be missed on antibody-only serologic testing. In countries such as the UK, Canada and the USA, where incidence rates are extremely low, the NAT yield, i.e. the number of infectious donations detected by NAT alone, has been extremely low, in the order of 1 in 1 million donations or lower. In contrast, the NAT yield has been considerably higher in countries such as South Africa, with a higher incidence of HIV infection in donors. The changing epizoology of arthropod-borne infections, e.g. West Nile and Chikungunya viruses, and other zoonotic infections, e.g. hepatitis E virus, are targets for discretionary NAT, the introduction being mandated by disease activity in any location. In each of these examples the infectious donors are those in the preclinical preantibody phase of acute infection, accessible only to either antigen or NAT assays.

Screening tests and donor–recipient matching

Table 20.1 lists the screening tests used in transfusion microbiology in different countries [6–8]. Some tests are mandatory and used to screen all donations. Other tests may be discretionary and used on selected groups of donors who are identified as being at particularly high risk for infection. In some situations additional testing may be required to mitigate transfusion transmission risks. CMV antibody testing may be done on a subset of donations in order to provide CMV seronegative components for immunosuppressed or other susceptible patients at risk of severe CMV infection on the assumption that this strategy removes the risk CMV viraemia. Such a policy does not, however, remove the risk posed by incident CMV and some services prefer to use blood from CMV seropositive donors who are known to have been seropositive from previous donations. Other herpes viruses including the gammaviruses EBV and HHV8 might justify screening on occasion. The conditions of storage of blood or blood components, in particular plasma, and the manufacture of components into blood products will materially affect the transmissibility of an infection; so too will leucocyte reduction, which removes the majority of leucocytes that may harbour cell bound viruses such as the HTLV agents. Pooling of plasma may also require additional NAT testing of the start pool for a range of agents including HAV, HEV and parvovirus B19. The decision to implement a particular screening test in a country depends on consideration of a number of factors, including the incidence and prevalence of the infectious disease in the donor population, the available testing technologies and the known or anticipated morbidity associated with transfusion-transmitted infection. Regulatory requirements for testing and the availability of test kits specifically licensed for donor screening also play an important role.

Table 20.1 Screening tests on blood donations in five countries as of 2011 (adapted from O'Brien et al. [2]).

Table020-1

Quality framework and operational issues

Ultimately, the microbiological and blood group safety of the blood supply depends on the input and interaction of a number of quality and operational factors.

A formal quality management system is an important part of ensuring that blood donation testing is adequately performed. The quality system needs to meet the requirements of a ‘Competent Authority’ under EU blood safety directives. Inspections are carried out by the Medicines and Healthcare Products Regulatory Authority (MHRA) in the UK. In the USA, there are both Food and Drug Administration (FDA) regulations and extensive requirements from professional accrediting organizations such as the AABB regarding quality requirements. Testing must be performed only by staff trained in approved standard operating procedures (SOPs). Document control systems must be in place to ensure that only current procedures are used and any changes documented and approved. Any errors that occur in laboratory procedures must be logged using a quality incident report (QIR) system, which requires corrective and preventative action to be taken.

The levels of process control now employed by the blood centre donation testing laboratories give a very high level of confidence that the test result for a donation is both valid and correct and that any potentially hazardous material will be discarded.

Most transfusion services have surveillance programmes for monitoring the rate of transmissible infections in blood donors, while haemovigilance schemes in place in several countries monitor transmission of transfusion-transmissible agents. The reporting of serious adverse events and reactions resulting from transfusion is an essential component of blood safety and is regarded as such by international agencies including the WHO and in Europe by its Commissioners.

Key points

1. It is mandatory to test every blood donation for ABO blood group, RhD blood group and the presence of irregular red cell antibodies.

2. A wide range of infectious agents have been documented as transmissible by blood transfusion, and laboratory screening tests form the core of the process to identify infected blood components prior to transfusion.

3. A knowledge of the donor panel demography and protocols for donor deferral and investigation of potentially infectious donors remain important components for maintaining blood safety.

4. Processes must be in place to communicate infectious disease marker results to donors.

5. Many of the newer techniques and kits currently used in blood centres to identify infected blood components show high levels of both specificity and sensitivity.

6. Both immunoassays and nucleic acid testing are used to identify possibly infectious units.

7. A quality framework is important for the accuracy of all laboratory testing.

Acknowledgements

Pat Hewitt, Ian Reeves and Richard Moule (NHS Blood and Transplant).

References

1. Guidelines for the UK Blood Transfusion Services, 7th edn. London: HMSO; 2005.

2. Safe Supplies: Focusing on Epidemiology. Annual Review from the NHS Blood and Transplant/Health Protection Agency Colindale Epidemiology Unit, 2010, London, September 2011. Available at: http://www.hpa.org.uk/Topics/InfectiousDiseases/ReferenceLibrary/BIBDReferences/.

3. Galel SA. Infectious disease screening. In: D Roback (ed.), AABB Technical Manual, 17th edn. Bethesda, MD: AABB Press; 2011, pp. 239–270.

4. Barbara J, Ramskill S, Perry K, Parry J & Nightingale M. The National Blood Service (England) approach to evaluation of kits for detecting infectious agents. Transfus Med Rev 2007; 21(2): 147–158.

5. UK Infection Surveillance Annual Report. Available at: http://www.hpa.org.uk/infections/topics_az/BIBD/publications.htm.-.

6. Transfusion-transmissible infections in Australia, Surveillance Report; 2011. Available at: www.med.unsw.edu.au.

7. Zou S et al. Donor testing and risk current prevalence, incidence, and residual risk for transfusion-transmissible agents in US allogeneic donations. Transfus Med Rev 2012; 26: 119–128.

8. O'Brien SF, Zou S, Laperche S, Brant LJ, Seed CR & Kleinman SH. Surveillance of transfusion-transmissible infections – comparison of systems in five developed countries. Transfus Med Rev 2012; 26: 38–57.