Guideline for blood grouping and red cell antibody testing in pregnancy

The purpose of the guideline is to make evidence-based recommendations for the application of blood grouping and red cell antibody testing in pregnancy. The aim is to predict the potential for, and where possible, prevent, haemolytic disease of the fetus and newborn (HDFN). The blood group and antibody status of a pregnant woman should be tested at booking and at 28 weeks gestation to identify the ABO group and D status and to detect red cell antibodies that have the potential to be clinically significant. Some antibodies (including anti-D, anti-K and anti-c) are associated with significant fetal and neonatal risks, such as anaemia, jaundice or perinatal loss. There are antibodies that are unlikely to significantly affect the fetus but that can cause neonatal anaemia and hyperbilirubinaemia, and others may cause problems for the screening and timely provision of appropriate blood for the woman or baby. This guideline updates the previous guidance published in BCSH et al. (2007) and takes into account recent developments in fetal medicine, such as the widespread use of non-invasive monitoring for fetal anaemia by middle cerebral artery (MCA) Doppler ultrasound scanning, together with the facility to determine the relevant genotype of the fetus from DNA in maternal blood samples in many potential cases of HDFN as described in the recent Royal College of Obstetricians and Gynaecologists (RCOG) Green-top Guidelines (2014) for the management of women with red cell antibodies during pregnancy. Information from these investigations has changed the requirement for ongoing serological monitoring of red cell antibodies identified in pregnancy, and this has been taken into consideration in developing updated recommendations. The National Institute for Health and Care Excellence, (NICE, 2010) Guideline 62, ‘Antenatal Care’ was reviewed and re-published in 2010, but updated recommendations for blood grouping and antibody screening were however not included. The challenges flagged in the BCSH guideline for the use of anti-D immunoglobulin for the prevention of haemolytic disease of the fetus and newborn (BCSH, 2014) of distinguishing between passive and immune anti-D are now covered in detail. The annual Serious Hazards of Transfusion (SHOT) reports continue to highlight errors in the administration of anti-D immunoglobulin (anti-D Ig) prophylaxis, some of which can be attributed to shortcomings in the clinical information provided with screening samples. The testing protocols recommended here are designed to provide clarity for practice in order to protect pregnant women and their babies. This guideline was developed in accordance with the British Committee for Standards in Haematology (BCSH) methodology. The guideline group was selected to be representative of medical and scientific experts. A search of published literature was undertaken using the Cochrane Library, Pubmed, MedLine, Embase and internet searches using the following key words and relevant MeSH terms: anti-D, anti-D Ig immune globulin, pregnancy, antibodies in pregnancy, antenatal prophylaxis, rhesus, RhD, RhD haemolytic disease, erythroblastosis fetalis. This search covered the period 1999 to July 2014 and was limited to the English language and humans. In addition, appropriate non-published literature, published policy documents and knowledge from experts in the field were incorporated and utilised. The papers included were subjected to critical reading by the authors using the CASP appraisal tool (CASP website) and were ranked according to the hierarchy of evidence. This approach took account of the National Institute for Health and Care Excellence's (NICE) systematic review (Chilcott et al., 2003) and the NICE Health Technology Assessment report published in 2007. The writing group produced the draft guideline, which was subsequently revised by consensus by members of the Transfusion Task Force of the British Committee for Standards in Haematology. The guideline was reviewed by a sounding board of UK haematologists, the British Committee for Standards in Haematology (BCSH) and the British Society for Haematology (BSH) Committee as well as representatives from the Royal College of Obstetrics and Gynaecology. Reviewers' comments were incorporated where appropriate. Criteria used to assign levels of evidence, and grades of recommendations are as outlined by the Grading of Recommendations Assessment, Development and Evaluation (GRADE) working group (www.gradeworkinggroup.org), as outlined in Table 3. Providing information about any blood test and obtaining consent is a clinical responsibility, and informed consent should be obtained and documented prior to samples being taken (NICE, Guideline 62, 2008). It is essential that samples from pregnant women are correctly identified and that request forms are accurately completed. Misidentification at the time of sampling could lead to an incorrect blood group being assigned to the transfusion record. This could result in errors in anti-D Ig prophylaxis (missed or inappropriate administration of prophylactic anti-D Ig) and errors in the selection of blood components (SHOT, 2011; BCSH, 2014). It is essential that the request form and sample conform to the requirements described in the guidelines on the administration of blood components (BCSH, 2009a,b). In addition, it is essential that any previous administration of prophylactic anti-D Ig in the current pregnancy, including date and dose, is recorded on the laboratory request form. A clinical history, particularly of previous children who were affected by HDFN and of previous transfusions, is essential information and should be stated on the request form. Pre-printed labels should not be used to label pre-transfusion blood sample tubes for compatibility testing or antenatal screening. Samples should be hand-labelled, or labels that are printed ‘on demand’ (at the patient's bedside) are acceptable as an alternative to handwritten labels. All laboratory testing procedures must be validated in compliance with published guidelines (BCSH, 2012b). Wherever possible, testing should be performed on automated equipment, which ensures positive sample identification, and with electronic transfer of results to the Laboratory Information Management System (LIMS). A record of the pregnant woman's ABO and D type performed at booking is useful as confirmation of any subsequent testing on another sample taken at the point of need should the woman or her baby require blood transfusion at a later date. Maternal D typing is also undertaken to identify D-negative women who require anti-D Ig prophylaxis. Maternal antibody screening is undertaken to detect clinically significant antibodies, which might affect the fetus and/or newborn, and to detect antibodies that may cause problems with the provision of compatible blood components for the woman and for the fetus/newborn. Approximately 1% of pregnant women are found to have clinically significant red cell antibodies (Howard et al., 1998; Koelwijn et al., 2008; Smith et al., 2013). Of these, the most common specificity is still anti-D, although the universal introduction of routine ante-natal anti-D Ig prophylaxis (RAADP) has reduced the sensitisation rate. There has been an increase in the number of positive antibody screens as a result of passive anti-D Ig. A recent meta-analysis has suggested that there is a correlation between IgG anti-A/B titres and outcomes (Li et al., 2015). Due to poor reproducibility in individual cases and a relatively low incidence of severe disease in a UK population, however, testing for a high concentration of immune anti-A and/or anti-B in pregnant women is not recommended (Mollison et al., 1997c). There is no additional value in using an enzyme technique in routine antibody screening because additional clinically insignificant antibodies might be detected, resulting in unnecessary follow-up testing (Clark et al., 1999). When a positive antibody screen is obtained and red cell antibodies are detected, further testing of maternal blood should be undertaken to determine the specificity(ies) and to determine the concentration/strength of antibodies (using titration or a method of quantification) and the likelihood of HDFN. Once red cell antibodies have been identified in pregnancy, the identification process should be repeated with each additional sample taken to identify or exclude any additional clinically significant maternal alloantibodies, as women who have developed an alloantibody are at greater risk of developing additional antibodies. This will ensure that all antibodies that have potential to cause HDFN are monitored and will facilitate the timely provision of compatible blood if required for the woman and/or for the baby. The frequency of repeat tests for antibody screening and identification will be determined by the specificity and strength of antibody and whether an intrauterine transfusion (IUT) has been administered. The concentration of each clinically significant red cell antibody is measured throughout pregnancy, initially to guide the need for referral to a fetal medicine specialist and subsequently to guide management of the pregnancy, including investigations and intervention (RCOG, 2014). Quantification requires specific equipment and measures antibody concentration against a national standard [National Institute for Biological Standards and Control (NIBSC)]. Anti-D and anti-c are the only antibodies that are currently quantified, and they are reported as IU per millilitre. Where possible, each sample should be tested in parallel with the previous sample and the results compared to identify significant changes in antibody concentration. Titration is used to assess the concentration of clinically significant red cell antibodies other than anti-D and anti-c. Doubling dilutions (1 in 2, 1 in 4, etc) of plasma prepared in phosphate-buffered saline are tested by IAT using reagent red cells, where possible, showing heterozygous expression of the corresponding antigen(s). Care must be taken in selecting cells for titration where more than one antibody specificity is present (including prophylactic anti-D Ig and clinically insignificant antibodies) to ensure that the concentration of each specificity is assessed independently, e.g. where anti-K+Fya are present titrate against K−, Fy(a+b+) and K+k+, Fy(a−) cells. Careful attention to technique is necessary to minimise the variables in the methodology employed, and it is recommended that the NIBSC anti-D standard (NIBSC, 2010a) be titrated in parallel, as an internal control, to ensure reproducibility of results in-house. The reported titre is the reciprocal of the highest dilution that gives a positive reaction, and the grade of reaction taken as the end point for this should be defined in the standard operating procedure, e.g. the last dilution giving a 1+ reaction. An increase in titre of more than one dilution (e.g. a previous titre of 2 rising to a subsequent titre of 8) is considered to be a significant rise, and the titration of the previous sample in parallel is recommended wherever possible to verify that the change in titre is not due to variability in the method. Evidence from external quality assessment (EQA) exercises shows variability in titration results both between and within technologies in common use [e.g. tube, column agglutination (gel and bead), and solid phase], and this should be taken into consideration when assessing results from different institutions (UK External Quality Assessment Scheme for Blood Transfusion Laboratory Practice – reports of antibody titration exercises and questionnaires distributed with exercises 07E7 July 2007 and 11E8 November 2011). When a clinically significant antibody capable of causing HDFN is present in a maternal sample, determining the father's phenotype can provide useful information to predict the likelihood of the fetus expressing the relevant red cell antigen and for counselling the couple regarding future pregnancies. It should be recognised that in any pregnancy, the partner may not be the biological father. Furthermore, in cases where the pregnancy has been facilitated by assisted conception with sperm donation from a donor panel, the pregnant woman's partner will not be the biological father. It is reasonable to omit paternal testing and proceed directly to fetal genotyping using cffDNA, where available, in order to avoid issues of non-paternity where indicated (RCOG, 2014). Fetal DNA for genotyping by polymerase chain reaction can be obtained by amniocentesis or chorionic villus sampling. However, such invasive techniques carry a risk of miscarriage and may boost maternal antibodies if present. It is now possible to determine fetal RHD, RHCE and KEL*01 genotypes at or after 16 weeks gestation using cell-free fetal DNA (cffDNA) from maternal blood samples, thereby avoiding the need for invasive fetal blood sampling (Daniels et al., 2009). This is useful both in predicting HDFN in individual cases where clinically significant red cell antibodies are present and as mass screening to guide anti-D Ig prophylaxis for D-negative women with no immune anti-D. The false negative rate for such tests may be approximately 0·1–0·3% based on large volume testing (Finning et al., 2008; Chitty et al., 2014). For low throughput testing, the figure is less well defined. The KEL*01 assay detecting a single nucleotide polymorphism in the fetus has a particularly close cut-off between positivity and negativity. It is important that reference laboratories are provided with feedback on the blood group of the baby at delivery to provide additional quality assurance. Given the low rate of response to requests for feedback on the neonatal blood group, there may be additional assay failures that are not known about or investigated by the reference laboratory (E. Massey, Personal communication). It is important that samples are not sent too early in pregnancy as the levels of cffDNA rise with gestation. At the time of writing, there are two laboratories in the United Kingdom providing cffDNA blood grouping. This situation may change, and it is important to ensure that samples are taken and referred in accordance with the defined requirements of the laboratory performing the testing. False negative results of cffDNA typing have been reported for KEL*01 (K) at 17 weeks gestation (Finning et al., 2007). The false negative rate for RHCE (for c) and RHD (for D) genotyping performed at or after 16 weeks gestation is less than 1% (Finning et al., 2008). Testing even later than 16 weeks may be advised for certain assays by individual reference laboratories, and it is important to ensure that laboratory-specific guidance on the earliest reliable gestation for testing is followed. Similarly, repeat sampling may need to be undertaken, where this is advised by the reference laboratory (SNBTS, 2013, IBGRL website). Non-invasive fetal genotyping using cell free fetal DNA (CffDNA) should be performed in pregnant women who have a history of HDFN or where quantification values or titres suggest that the pregnancy is at risk of HDFN. These investigations should be requested by obstetricians or fetal medicine specialists who have the expertise to evaluate and explain the implications of the test results in the context of clinical history and the outcome of any other investigations. These test results may then be used to guide the frequency and nature of further monitoring. Currently, approximately 40% of D-negative women (40 000 in the UK/ per annum) will be given anti-D Ig prophylaxis unnecessarily as they are carrying a D-negative fetus. Routine fetal RHD typing for all D-negative pregnant women has been introduced in Denmark, Finland and the Netherlands to allow selective use of anti-D Ig prophylaxis. This helps reduce unnecessary exposure of young women and their fetuses to a blood product, with reduction in costs related to the provision of anti-D Ig prophylaxis and tests related to fetomaternal haemorrhage (Clausen et al., 2012; Chitty et al., 2014; Soothill et al., 2015). This service is now provided by IBGRL in England but not routinely implemented. The NICE diagnostics assessment programme will assess the clinical and cost-effectiveness of high-throughput, non-invasive prenatal diagnosis of fetal RHD status in order to make recommendations on its routine use within England (www.nice.org.uk). For mass throughput screening of all D-negative pregnant women, cffDNA testing for RHD is sufficiently accurate from 11 weeks gestation with false negative rates of 0·1–0·3% (Finning et al., 2008; Daniels et al., 2009; Chitty et al., 2014). It is, however, advisable to test the fetal RHD status using a more sensitive and specific methodology at a later gestation for alloimmunised pregnant women as described above. All pregnant women should have samples taken early in pregnancy, ideally at booking, typically at 8–12 weeks gestation, for ABO and D grouping and for screening for the presence of red cell alloantibodies (Fig. 1 and 2). When an antibody screen is positive, further tests should be carried out to determine the antibody specificity and significance (see Section 2). All pregnant women, whether D positive or D negative, should have a further blood sample taken at 28 weeks gestation for re-checking the ABO and D group and further screening for red cell alloantibodies (NICE clinical guidance 62, 2008). D-positive women are just as likely as D-negative women to form antibodies (other than anti-D) late in pregnancy (Koelewijn et al., 2008; Thompson et al., 2003). Local policies must ensure that D-negative women who are eligible for routine antenatal anti-D Ig prophylaxis (RAADP) have the 28-week antibody screening sample taken before the first dose of RAADP anti-D Ig is administered. Samples taken after the injection could result in passive anti-D being detected, which may be mistaken for immune anti-D and, conversely, potentially dangerous immune anti-D being mistaken for passive anti-D Ig (New et al., 2001) (see Section 4.1.1). There is evidence that antibodies first detected after 28 weeks gestation are less likely to cause clinically significant HDFN (Rothenberg et al., 1999; Heddle et al., 1993). Further, and significantly, the introduction of RAADP has resulted in the detection of anti-D Ig in samples taken after 28 weeks gestation from D-negative women (Cambic et al., it is not possible to between prophylactic anti-D Ig and immune anti-D the has a high concentration in reference quantification to exclude the there is the potential for between the two (New et al., 2001) (see Section anti-c and anti-K are the antibodies most in causing haemolytic disease severe to antenatal intervention (Koelewijn et al., 2008). in women with a previous history of significant HDFN however, be considered at risk and referral to a fetal medicine specialist of the antibody specificity(ies) In all other follow-up testing protocols are by the specificity and concentration of the antibodies of antenatal follow-up antibody identification should be performed to for the of additional prior to maternal or fetal Anti-D Ig prophylaxis has been in the number of to but it has in determining whether anti-D detected in pregnancy is passive to anti-D Ig prophylaxis given as RAADP or for a potentially or The associated with the of the nature of anti-D are if passive anti-D Ig is as immune anti-D, then further anti-D Ig prophylaxis may be the women from immune anti-D is as passive anti-D appropriate follow-up of the antibody concentration during pregnancy may be and that might be required to HDFN not The annual report included a the in between immune anti-D and passive anti-D Ig as there were cases reported where women with immune anti-D were not as as they should have been because the anti-D detected was to be passive anti-D and in of these the were with some of HDFN (SHOT, anti-D Ig and immune anti-D be The concentration of passive anti-D Ig will with the concentration of immune anti-D will or rise if there is (Fig. The concentration of passive anti-D Ig in maternal samples IU anti-D Ig more than IU has been administered. The concentration of anti-D Ig detected after IU of anti-D Ig in a in pregnant women was to IU and the following anti-D Ig IU et al., 2003). These values are based on the stated in the administration of an injection of anti-D a concentration of anti-D Ig is potentially present within and the blood concentration is within The of passive anti-D Ig is approximately weeks et al., anti-D Ig can be detected by serological tests for by an test at weeks or more following injection of IU and for more than weeks where more sensitive techniques are used or following of anti-D Ig. anti-D approximately weeks after exposure to D-positive cells and a concentration after weeks if there is no further exposure (Mollison et al., 1997c). of the nature of anti-D anti-D or passive anti-D Ig) based on the strength of reaction with D-positive cells is as this will with the technique and phenotype of the reagent red cells have been affected by HDFN as a result of on the of the strength of using that have not been validated (SHOT, Quantification by gives an of antibody concentration in IU per anti-D. All anti-D detected in pregnancy should be by with reference to the NIBSC anti-D standard (NIBSC, or tested by a method that has been validated against and that gives a result that is in or can be to IU per anti-D et al., et al., 2013). The only is where anti-D is detected for the first time prior to or at the time of e.g. in a group and screen sample, in which the sample need not be sent for but the baby should be monitored for of HDFN as will need to be on the of the of any anaemia and/or The results of quantification are unlikely to have a on clinical at this and are not as as measures of and Maternal anti-D quantification may be performed at a later if The quantification results should be in the context of the and dose of any anti-D Ig given the for its the of administration and the antibody status at the time of The clinical history and knowledge of the results of previous laboratory testing are in clinical where anti-D is detected in pregnancy, and should be to this if anti-D Ig has been for a potentially prior to 28 it must be that anti-D at or before 28 weeks may be immune as levels can rise between 28 weeks and monitoring should be undertaken as if the antibody may be immune anti-D is no to anti-D Ig as indicated in the BCSH guideline for the use of anti-D Ig to HDFN (BCSH, 2014). with anti-D should not be with an antibody the of anti-D it is that the anti-D is Anti-D is the most cause of HDFN. Blood samples from pregnant women with immune anti-D should be tested at 28 weeks gestation and 2 weeks delivery to the concentration of anti-D reference to the NIBSC anti-D standard (NIBSC, and to identify any additional antibodies that may Where anti-D has been to be immune (see Section an increase in concentration of or compared with the previous a significant of the period of gestation, as this be a greater than in a validated test (BCSH, When account has been taken of previous history of the of anti-D in Table 1 have been used to guide the management of and Non-invasive Doppler can then be used to detect and fetal anaemia et al., 2014). women with an anti-D concentration of IU or greater and/or a rising anti-D concentration and/or a history of affected by HDFN should be referred to a fetal medicine specialist for further assessment including It should also be that HDFN has also been reported with anti-D at less than IU et al., Where the anti-D concentration is greater than that to the presence of passive anti-D but IU anti-D quantification should be repeated weeks 28 weeks gestation and 2 weeks Once referral to a fetal medicine specialist has been and are being the value of anti-D quantification on subsequent samples is if the concentration is in the high risk (RCOG, 2014). a however a sample should still be tested for the presence of further red cell antibodies at 28 weeks gestation and blood is required for e.g. for The including the should be informed of the testing to and unnecessary e.g. the laboratory follow-up samples in cases where the clinical have that they are not at risk of HDFN should be at weeks gestation to minimise the of exposure to maternal blood group antibodies (RCOG, 2014). The D status of the fetus of pregnant women with a significant concentration of anti-D should be determined using cffDNA from a maternal blood This technique currently has a false negative rate of (Daniels et al., The result of this test can be used as a guide to the frequency and nature of ongoing monitoring Section the sample in which anti-D is detected is for routine antenatal antibody screening or is D-negative screening cells, selected to provide all red cell as for screening in the BCSH compatibility guidelines (BCSH, can be useful to detect or exclude the presence of alloantibodies of other However, if this approach is the risk of using such a screening for other in and potentially an immune anti-D should be A of antibodies with but not with high can be by serological to be or than et al., who have produced or but not anti-D, at risk of immune anti-D and should be anti-D Ig prophylaxis in accordance with the BCSH guideline for the use of anti-D Ig to HDFN 2014). It is important to identify these cases to ensure appropriate follow-up and to anti-D Ig prophylaxis, and a reference should of Where is the phenotype of the partner may be useful as of testing at a reference women with anti-c should be with the frequency as women with immune anti-D, at to 28 weeks gestation and 2 weeks Quantification of anti-c is useful in monitoring any increase in the antibody concentration. Samples from pregnant women with anti-c should be with reference to the NIBSC anti-c standard (NIBSC, and, where possible, the previous sample should be tested in parallel, as for immune anti-D. identification should be undertaken to exclude or the presence of other clinically significant antibodies. In with any previous history of the of anti-c in Table 2 are of the requirement to to a fetal medicine specialist et al., Once referral to a fetal medicine specialist has been and are being the value of anti-c quantification on