Donation testing and transfusion transmissible infections

Abstract

The accurate testing of blood donations plays a vital role in the provision of safe blood for transfusion. This section covers testing that should be carried out on every unit of blood donated. All donations should be routinely tested for the ABO and Rh blood groups, screened for red cell antibodies and should be tested for transfusion transmissible infections (TTIs) including the human immunodeficiency virus (HIV), hepatitis B and C and syphilis. Anomalous test results should be investigated and resolved before blood is deposited in available stock. A range of systems, equipment and techniques are available for both the red cell serological testing and the testing for TTIs. Fully automated computerised systems are used for all aspects of the testing in large-scale testing. In some organisations manual techniques may be used, e.g. for red cell serology tests, test tubes, microplates or microcolumns, (gel technologies) are used. Anticoagulated samples are required when testing is automated. Clotted samples are suitable for blood grouping performed manually. Each donation must have a unique identification number (see Section 9: Blood collection). When the donation samples taken from the blood donors at the time of collection arrive at the testing laboratory, they should be correlated with the respective donations to check that the batch of blood donations received relates exactly to the accompanying samples. Samples should be transported to the testing laboratory in accordance with procedures that ensure a constant approved temperature and secure confinement. Samples should be stored at 4 ± 2°C and tested as soon as possible, and within the time limits specified in test or kit manufacturer’s instructions. The tests described in this section are based on manual testing. The principles, however, apply to the testing in general. An overview of automation will be provided at the end of this section. The ABO and D typing should be carried out on every donation. Results should be compared with previous results if possible, and anomalies brought to the immediate attention of the individual member of staff responsible for investigation and resolution. Saline methods are used to perform ABO grouping tests. Both forward and reverse grouping is performed. The results of forward and reverse grouping must correlate for the result to be acceptable. Reagent antisera should be standardised for both rapid spin and longer incubation techniques, as a rapid spin technique is not practical when dealing with large numbers of samples. Therefore, batches of tests may be incubated at room temperature for about an hour (depending on manufacturer’s instructions) before reading. The ABO grouping result for a repeat or lapsed donor should be compared with the result from the previous donation to check that it correlates, and action taken if it does not. In the case of an anomaly, the senior technologist should be responsible for investigating and resolving the problem. If the anomaly cannot be resolved, the donation should not be used. Group A or AB individuals, lacking the A1 antigen, sometimes develop anti-A1, usually as a cold IgM antibody. When the A antigen is very weak and is not readily detectable on initial testing, even when using avid blood grouping reagents, the presence of anti-A1 in the donor serum/plasma may complicate the interpretation of the ABO blood group. On repeat testing of anomalous groups, it is therefore preferable to use both group A1 and A2 reagent red cells, together with the reagent group B cells. Using both group A1 and A2 reagent red cells would enable clarification of subgroups of A as shown in Table 1. A very weak subgroup of A, with anti-A1 in the serum/plasma may be mistaken for a Group O if the weak agglutination is not detected in forward ABO grouping, and Group A1 and B reagent red cells are used for the reverse grouping. However, an anomalous grouping result is noted when Group A2 and B reagent red cells are used instead. The use of Group A2 reagent red cells alerts the technologist to the fact that further testing is needed to resolve the ABO group, and prevents the incorrect interpretation of the blood group as Group O. Similarly, if the very weak AB is not detectable in forward ABO grouping, and anti-A1 is present in the serum/plasma, it appears as Group B when group A1 and B reagent red cells are used and gives an atypical grouping result when Group A2 and B reagent red cells are used instead. Reverse or serum grouping is indispensable as a means of confirming the forward or cell grouping of blood donations. Isoagglutinin tests should be read blind with no preconceived idea of what results should be. This also applies to records of ABO groups – the technologist reading the test should not have knowledge of the previous ABO group at the time of reading. Every atypical result should be investigated, including weak agglutination results where strong agglutination should be apparent. Once the ABO group of a donor has been confirmed, his/her record should be appropriately documented for future reference. Unusual or anomalous ABO types on donations need to be resolved before blood may be labelled and placed in available stock. With problematical groups, a sample from the pilot tube of the donation should also be tested. On rare occasions when the blood type cannot be interpreted with confidence, the donation should not be used, and the anomaly may need to be resolved at the time of the next donation. Table 2 shows an example on an anomalous ABO grouping result. Even though it appears from the forward grouping that the result is Group A, the group cannot be interpreted as the reverse grouping does not confirm the Group A type. Blood grouping records for repeat donors should be consulted to check that the group on record is the same as the group for the current donation. If it is not, then the pilot tube of the blood unit should be grouped. If this differs from the group on the sample, then a switch is confirmed. However, if the group using the pilot tube is the same as the group using the sample, then the group on record may be incorrect, or the current donation misidentified. The implications of such errors could be far reaching, and results on other units of blood donated at the same session, put into doubt. The individual in charge is responsible for carrying out an investigation in such cases. If there is no conclusive resolution, the blood cannot be used. Donations are separated into two D types: D positive and D negative based on the presence or absence of the D antigen, respectively. D typing results that indicate a D variant should also be classified D positive. This is to ensure that D negative recipients do not receive blood containing the D antigen. Some D variants, such as DEL, are not detected by routine serologic methods. Prevalence of the DEL phenotype differs significantly among populations, with 10–30% of D negative East Asians having the DEL phenotype. In the Caucasian population, only 0.1% of D negative people express a DEL phenotype and among African populations, there are no reports of DEL phenotypes. D typing results on repeat donors should be compared with the results from previous donations to check that they are the same, and action taken if they are not. In the case of an anomaly, the individual in charge should be responsible for investigating and resolving it. If it cannot be resolved, the donation should not be used. Such donations should be flagged so that further testing may be carried out at the donor’s next donation to investigate the anomaly. D typing is usually carried out using IgM (saline reacting) commercial reagents. Donors should be typed with a minimum of one selected monoclonal anti-D reagent and negative results should be confirmed using a second appropriate reagent, which could be an anti-D blend reagent, using the indirect antiglobulin test (IAT) or a saline reactive monoclonal reagent known to detect weak D. If the initial D typing is negative and additional testing using the IAT is positive, further testing should be performed to determine whether the red cells were sensitised prior to testing, in which case the result would be invalid. This may involve performing a direct antiglobulin test (DAT) on the donor red cells or repeating the test with the appropriate typing control according to manufacturer’s instructions. Table 3 gives examples of additional testing carried out to determine the D type. Depending on the anti-D reagents in use, the original test may be taken to the IAT phase or the test may be repeated including the first saline phase. There are many variations in procedure depending on the source of the monoclonal anti-D selected for use in the laboratory. It is important to follow the manufacturer’s instructions/laboratory protocol. Positive and negative controls should be performed on all reagents used, in parallel with every batch of tests, or at least once a day at the same time of the day, for standard ABO and D typing reagents. If the reagents perform as expected, this will provide the assurance that the reagents meet sensitivity and specificity requirements and that test results produced using the reagents can be accepted. Any reagent, in which controls show anomalous results, should be removed from circulation to avoid its continued use until an investigation has been carried out to resolve the problem. When faulty reagents are used for testing, all results for that batch are invalid. Irregular red cell antibodies in donor plasma may have an adverse effect on a recipient with the corresponding antigen especially when whole blood or plasma components are transfused. Plasma containing strong irregular antibodies may not be suitable for the preparation of fresh frozen plasma or for transfer to a fractionation facility for the preparation of plasma derivatives. A set of reagent Group O screening cells is used to test for irregular red cell antibodies, as described in Section 4: Principles of laboratory techniques. When large numbers of samples are tested manually, it may not be feasible to use the IAT method, which may be the most sensitive but is also the most laborious. There is a variety of different techniques for automated systems depending on the instrument in use. These include the use of bromelin-treated cells, gel cards or solid phase technology. When the red cell antibody screen is positive, antibody identification should be performed to determine antibody specificity. Suitable records should be kept for donors with antibodies so that they are recognised at each donation and antibodies of known specificity are not re-identified at every donation. Donations showing the presence of cold autoantibodies, which are not of clinical significance, should be tested to ensure that the cold autoantibodies do not mask a clinically significant antibody. Whole blood donations with strong autoantibodies should not be transfused to patients undergoing hypothermia. To detect potentially harmful anti-A and anti-B, samples may be tested for haemolysins or for high titre anti-A and/or anti-B antibodies. High titre antibodies can be detected by using a single dilution (e.g. 1 in 128, in saline). At this dilution, strong isoagglutinins should still be able to agglutinate group A1 and B reagent red cells and should be classified as ‘High Titre’. In the haemolysin test, complement is triggered by the reaction of donor immune anti-A and/or immune anti-B with reagent red cells, causing the red cells to haemolyse. It is important that donors with haemolytic or high titre anti-A and/or anti-B are detected so that their blood/plasma is not used for heterologous group transfusions. Group A1B (or A1 and B) reagent red cells are used for the selected tests as they contain the strongest A (A1) and B antigens to react with either anti-A or anti-B donor isoagglutinins. When used for haemolysin tests, reagent red cells may have to be washed to remove preservative fluid if it is anti-complementary. Plasma samples are unsuitable for haemolysin tests, as anticoagulants are anti-complementary. Time also negatively affects complement. Provided that serum is used, and the sample is less than 24 h old, it is not necessary to add complement from an external source. When testing samples older than this, the addition of extraneous complement is required. Without active complement, haemolysins are indistinguishable from non-haemolysing isoagglutinins. Donations identified as containing potentially harmful anti-A and anti-B, either by titration or haemolysin test, are labelled ‘high titre’ for homologous group transfusion only. Other donations, that do not demonstrate high titre/haemolysing isoantibodies, are labelled ‘low titre’. As an ongoing proactive measure, selected donations (e.g. Group O donations) may be screened for additional red cell antigens, such as those within the Rh and Kell systems. This may be carried out by using reagents with specificities such as anti-C, anti-c, anti-E, anti-e and anti-K. Results of extended antigen screening should be added to the records, and subsequent donations from the same donor flagged so that the donations can be identified from available stock, and selected if they fit the type required to resolve a compatibility problem. This avoids time-consuming screening that may be needed at the time of crossmatch, to find compatible blood. Alternately the donors may later be traced to donate blood for patients with antibodies but there is a time delay between calling the donor to give blood, and the fully tested donation being available for crossmatch and issue. Ongoing screening of donations may be performed to find donors who lack high incidence antigens so that rare donations identified in this way may be stored frozen in glycerol in a low temperature freezer for future use. When a patient with antibodies to a high frequency antigen requires blood, suitable blood can be requested from the rare donor registry. The blood will need to be thawed and deglycerolised prior to crossmatch and issue. All the tests described previously may be fully automated or partially automated according to the systems and modules selected. For information on automation, see Section 4: Principles of laboratory techniques. Automated testing of all donations in batches to perform the red cell serology tests is much quicker than manual testing and suits laboratories that handle many samples. Most automated instruments require special reagents with optimum reaction temperatures and techniques. It is important to note that some reagents require dilution before loading into the instrument, and manufacturer’s instructions should be closely followed. Automation saves time as the instrument is programmed to rapidly interpret all standard reaction patterns for the tests. This process categorises the majority of donations into one of the four ABO groups (A, B, O or AB), one of the two D types (D positive or D negative) and identifies antibody screen positive and high titre donations. The computer program linked to the automation set-up should be designed to detect and flag anomalous results so that they can be investigated further. Automated instruments require the inclusion of specific quality control samples of known ABO and Rh type, in the batch being tested, to ensure that all reagents are working correctly. Controls should be included in every batch and when large numbers of samples are tested, after a certain number of tests (such as every 200 tests) and should always be repeated when reagents are replaced, or new dilutions made. When automated control results fail, it is important to establish the root cause of the failure. Each laboratory should have a troubleshooting procedure in place that is applicable to the specific instrument in use. The training and operating manuals should provide such information. Problems with automated instruments may need to be resolved by technical personnel from the supplier or manufacturer of the instrument. Common problems may, however, be identified by suitably trained technologists working with the instrument in the laboratory, but it is critical that qualified and experienced support from the supplier or manufacturer is readily available. When blood grouping is automated, technologists should regularly perform manual testing, so if the instrument fails and is not operational for an extended period, they retain competency in manual techniques and are able to continue testing blood donations. Blood safety depends on both donor health screening (deferring donors who are aware they are infected by a TTI or are in a high-risk group for a TTI), and donation testing. The strategy employed for a given infectious agent will depend on the epidemiology of the particular agent in a given donor population, blood processing steps that might reduce transmission (such as pre-storage universal leucoreduction), and the availability of testing equipment and kits adapted for donor screening. Sometimes a combination of approaches may be used. For example, in countries where there is no local transmission of malaria, donors may be asked about their travel history, and malaria testing may be performed only on at risk donors (selective testing). See Section 8: Donor selection, for details about donor health screening, and Section 4: Principles of laboratory techniques, for more details about donation testing. Testing for TTIs is subject to ongoing change and improvement, as additional and more sensitive tests and automated test systems become available or new risks of possible infections are identified. Tests may detect viral nucleic acid (NAT testing), a viral component (Hepatitis B surface antigen, or HBsAg; the p24 antigen of HIV), or the host’s immune response to the infection (antibody testing performed using Enzyme linked immunosorbent assay (ELISA) or enzyme immunoassay (EIA)/chemiluminescence assay (CLIA) for antibodies such as anti-HIV, and anti-HBV). A donation with a confirmed reactive result for a TTI may not be used for transfusion. The window period refers to the period of time during which the donor may transmit infection, but testing performed will be negative. It is a latent period of immunosilence shortly after infection when laboratory tests for markers of the infection are non-reactive. Figure 1 shows the window period of immunosilence. A blood donation given during the window period will test negative, be classified safe for use and be placed in available stock yet may be infectious. The term seroconversion refers to the time when the blood of an infected individual shows the presence of antibodies to the infection. The biological attributes of host–virus interaction, replication times, infectivity dose, natural history of infection, serum volume used for in vitro testing, and the relative sensitivity of each assay employed to detect an infection present in a blood donor are highly variable. As illustrated in Table 4, the lag time (cumulative window period) for ability of each serological test to detect an infection present in a donor’s blood is based on the window period data for HIV, HBV and HCV published in 2016 by WHO (see reference related to Table 4). In general, window periods are shortest for NAT testing, longer for antigen testing, and longest for antibody detection. Thus, the lag time for anti-HIV to detect HIV infection can be as long as 21 days, for HBsAg test to detect HBV infection as long as 42 days, and for anti-HCV to detect HCV infection as long as 60 days. This lag period gets truncated (shortened) by direct tests for viral gene amplification with NAT. Single unit NAT testing, where each donation is tested separately, is slightly more sensitive than minipool testing, when a small number of donor samples are pooled for testing. The window period is of great concern to blood services, as it is not possible to detect an infected donation of blood during this phase. Although laboratory tests today are very sensitive, and the window period may be narrowed to a few days, no matter how sensitive the test may be, there could still be a window period. The interpretation of results and the strategy for donor deferral or exclusion as well as the way in which donors are notified of anomalous or reactive results, is dynamic. Each laboratory/service should have protocols in place for both initial and confirmatory testing. In the context of TTIs, an algorithm is the term used for a sequence of steps that is documented and followed in order when a donation is initially found to be reactive. Depending on the TTI concerned, the algorithm is unique. For example, the steps taken when a sample is HIV reactive may not be the same as steps taken when the sample is syphilis or malaria reactive. In general, initially reactive tests are usually repeated in duplicate on the same sample, using the same testing platform. If one or both of the repeat tests is positive, the sample is considered ‘repeat reactive’, the donation should be discarded, and the donor should be deferred. Repeat reactive donations are discarded using the appropriate disposal protocol/methods and according to local regulations for the disposal of biohazardous material. Algorithms are usually complex and individualised for each TTI. In this publication, however, a single algorithm is shown that generalises and summarises the suggested flow of actions when a donation is reactive for a TTI. Figure 2 is a guideline algorithm summarising the sequence of actions that may be taken when a laboratory test is initially reactive for HIV, HBV or HCV. The algorithm does not describe the procedure to follow for a specific TTI. Partly because of the need for sensitivity in testing systems, TTI screening may lead to false reactive results, and biological false positives do occur. Many TTI test systems rely on cut-off values to determine whether a result is reactive or not, so some test results may fall close to a range of uncertainty (i.e. ‘indeterminate’ result). Extreme care needs to be taken with follow-up action, including confirmatory testing on the donation. The confirmed reactivity to certain TTIs such as HIV, HBV and HCV leads to permanent exclusion of the donor, whereas the risk of other TTIs such as malaria, may have specified time deferrals. For purposes of donor notification and counselling and possible actions to be taken regarding previous donations from the donor, confirmatory tests are done. Whenever possible, it is preferable to confirm initially reactive results by using a different confirmatory test system. The pilot tube from the donation is ideally used for follow-up testing, to check that blood bags and samples were not switched at the time of donation. However, this step is more difficult if testing is performed at a central laboratory and blood donations retained in the peripheral blood bank. Discrepancies between initial test result and confirmatory test result must be fully investigated and resolved before the final decision is made regarding the donation. If there is any doubt about the result on repeat testing, the donation should also be discarded. Confirmatory testing may also be performed using a second sample from the same donor, if this is available, in case the sample used originally became contaminated in the laboratory or in transit. The information on all reactive units must be referred to the appropriate department for authorised donor follow-up (for more information, see Section 9: Blood collection). Of all the transfusion transmissible infections, HIV, HBV, HCV and to a lesser extent syphilis, are the most important with regard to universal screening of blood donations. All blood services should test every donation for HIV, HBV, HCV and syphilis, all of which may result in chronic, asymptomatic infections in donors. Every donation should be tested for antibodies to HIV-1 and HIV-2 and common subtypes, using a test system such as ELISA or EIA/CLIA. Laboratory markers for HIV include anti-HIV, p24 antigen, and viral RNA. In order to reduce the window period for HIV, many blood centres use combination tests for HIV antibody and p24 antigen, and/or perform NAT testing in addition to antibody testing. Western blot is not used routinely for screening blood donations for viral markers. It is a gel electrophoresis technique and requires specialised equipment. It may be useful for confirmatory testing for HIV. Every donation should be tested for hepatitis B, using a test system such as ELISA, EIA/CLIA or NAT. Laboratory markers for HBV include HBsAg and viral DNA. Screening for anti-HBc is used in some countries and reactive donations excluded. However, in hyper-endemic countries, introduction of this marker to the screening algorithm might adversely affect the adequacy of the blood supply. Every donation should be tested for hepatitis C, using a test system such as ELISA, EIA/CLIA or NAT. Laboratory markers for HCV include anti-HCV and viral RNA. Blood services may decide to use either RPR or TPHA as a first-line screening test and retest reactive specimens using FTA, which is less likely to give false positive results. However, if a confirmatory test is not performed, donors should not be notified of the infection solely on the result of a screening test, although reactive donations are not transfused. If feasible, the donor could be referred to their own doctor for further testing and possible treatment. The donor record should be flagged (in code) to draw attention to the testing result when a subsequent donation is given. Serum/plasma from the donor is tested for antibodies to reagin and not for antibodies to T. pallidum. Reagin levels are raised in certain infectious conditions such as syphilis, and this causes an antibody response in the host. Reactive results are demonstrable as flocculation; a form of precipitation between antibody to reagin in the donor sample and the reagin reagent. The TPHA test is used to detect antibodies to T. pallidum. The test uses avian (i.e. related to birds) erythrocytes coated with antigenic components of the T. pallidum organism. This test can be automated, and results produced are read either by the technologist or by instrument. Some blood services may use an FTA test, which measures specific antibody for T. pallidum, to confirm screen reactive. However, once reactive, an individual remains reactive for life, so the test does not differentiate between infectious and non-infectious individuals. Serum/plasma from the donation sample is mixed with T. pallidum and after processing, an anti-human globulin reagent labelled with a fluorescent indicator is added. If antibodies are present in the sample, the labelled AHG will act as an indicator, and fluorescence will be detected when using a specially designed microscope or appropriate reader. The FTA is costly and requires a degree of technical skill to perform. Screening donations for markers to infections that are endemic to certain geographical regions should be performed in addition to HIV, HBV, HCV and to a lesser extent, syphilis. For example, screening for Chagas’ disease should be carried out in endemic regions of South America and is performed as a one-time donor screening in the USA. Screening for West Nile Virus (WNV) and Zika virus may also be required in some countries. Human T-lymphotropic virus (HTLV) -I and -II screening should be performed in areas, where it is regularly encountered in the population, such as in the Far East, and other countries where the prevalence warrants it. These and other TTIs are not described in detail in this publication, but some are mentioned briefly in succeeding discussions. Table 5 provides a list of microorganisms and other agents that may cause TTI. Because of the increase in international travel, other transmissible infections too, are more likely to be spread amongst communities in which they were previously not detected. Individuals who are particularly vulnerable due to a lack of immunity to an infection rarely encountered in their home environment become increasingly exposed to these infections as a result of travel. As the following extract from a World Health Organization (WHO) report shows, infection with multiple pathogens is becoming of increasing concern, particularly in developing countries. Extract from a WHO World Health report 2007: the deadly interaction: HIV/AIDS and other diseases. The interaction of HIV/AIDS with other infectious diseases is an increasing public health concern. In sub-Saharan Africa, for example, malaria, bacterial infections and tuberculosis have been identified as the leading causes of HIV-related morbidity. HIV infection increases both the incidence and severity of clinical malaria in adults. In some parts of Africa, falciparum malaria and HIV infection represent the two most important health problems. Testing for malaria Although it is possible to test donations for malaria, some blood services do not have the facilities to do so or are operating in countries where the test is not licensed for use in blood donor screening. Blood components provide a hospitable, conducive environment for bacterial growth. This is especially true for platelets, since they are stored at room temperature. Bacteria present in the blood component may multiply during storage and cause severe transfusion reactions and septic shock when transfused; see Section 14: Risks of transfusion and haemovigilance. Blood collection kits are sterile, and steps are taken in donor screening, venepuncture, and processing and storage to ensure sterility; see Section 9: Blood collection. Despite precautions, occasionally bacterial contamination and proliferation does occur. The main mechanism of contamination is inadequate skin disinfection, with primarily Gram positive organisms that are part of normal skin bacterial flora, such as Staphylococcus epidermidis and Staphylococcus aureus introduced into the blood collection bag at the time of phlebotomy. Other mechanisms include intermittent donor bacteraemia with Gram-negative organisms, sometimes related to gastroenteritis (Yersinia enterocolitica) or chronic intestinal pathology, such as colon cancer (Streptococcus bovis). Yersinia enterocolitica in particular is cold tolerant and may survive and proliferate in refrigerated whole blood and red cells. More rarely, contamination may occur with environmental contaminants (Serratia liquifaciens, Pseudomonas) during storage or processing of components, particularly if there is a defect in the storage bags, such as a micro-puncture or leaky seal. Some blood services test all platelet components for bacteria. Most commonly, after a 24–36 h hold period, an aliquot of the main platelet component is expressed into an integral sampling pouch, which is then removed from the component in a sterile fashion. The sample is injected into one or several blood culture bottles that are incubated in an automated culture system. The platelets are put into inventory but removed from inventory at the blood centre or the hospital blood bank if the blood culture becomes positive. Co-components from the same donation are also recalled. Further confirmatory testing may be done to confirm that bacteria are present and identify the precise microorganism. An emerging pathogen is either a microorganism that has expanded in geographic range or pathogenicity, or an entirely novel agent. A blood service should be constantly on the alert for the reported detection of emerging pathogens that may impact on the safety of the blood supply. When there is a risk that an infectious agent may be transmitted by blood transfusion, and this microbe is known to have entered the donor population, this may involve additional screening of donors and testing of blood donations. The best example in this regard was the introduction of HIV testing in the mid-1980s. Other examples of potential risks to the safety of the blood supply occurred with the advent of vCJD, a novel agent. Pathogens with expanded geographic range include WNV, Zika virus, and babesiosis. To mitigate the impact of these emerging pathogens on the safety of the blood supply in a timely and effective manner, a close relationship between blood services, regulatory authorities, public health, industry and other stakeholders is important. Pathogen reduction is a proactive strategy to deal with these threats (for more information, see Section 11: Blood processing and components). A blood service should develop an action plan in case of pandemics that may occur, such as the outbreak of coronavirus disease in 2019/2020 (COVID-19) which was later declared a pandemic by the WHO. Although it was not immediately apparent whether or not the causative virus (designated Severe Acute Respiratory Syndrome Coronavirus 2 or SARS-CoV-2) was transmissible via blood transfusion, blood services acted quickly to implement strategies to identify at risk donors and defer them from donating. Pandemics not only threaten the safety of the blood supply, but affect personnel within the blood service, and the availability of safe donors to provide blood. In the case of COVID-19, for example, it was necessary for blood services to introduce additional precautions to prevent possible transmission of the virus from occurring within their premises, between visiting donors and also amongst staff. Due to the need for confidentiality of medical information, it is important that there is no visible link within the testing laboratory, between a donation that is reactive for a TTI, and the identity of the donor. As few individuals as possible should be able to link the donor with the sample that is reactive for a TTI; either in the laboratory or in the blood collection centre. Only authorised personnel should have access to this information. A designated individual should gather all the relevant results and refer them to a trained counsellor (with the consent of the medical director) who should be the only one authorised to make contact with the donor or with a clinician identified by the donor. Look-back investigations may be done when a donation from a donor who had given blood before, and which was then non-reactive for TTIs and therefore considered safe for transfusion into a patient, now tests positive for a TTI. Recipients of previous transfusions from the donor may be contacted and tested for the specific TTI. When a recipient becomes positive for a TTI following a transfusion, donors who contributed components to the transfusion may be contacted and tested, if they have not already returned to donate. National regulatory authorities should establish guidelines for blood services, to define the scope and algorithm for managing lookback and traceback investigations, and how earlier donations and recipients of blood products made from these donations are handled, when a donor has seroconverted and may have been in the window period of infection when the earlier donation was given. Both lookbacks and tracebacks require good records in both the blood service and hospital to ensure vein to vein traceability of blood, from donor to recipient. The yield of lookbacks and tracebacks will be highest when new tests are introduced, or there has been a major advance in test sensitivity. Yields are low when donations have been tested using sensitive methods, such as NAT, and donors were very unlikely to have been in the window period leading to infection of a recipient.