LAUSR

Dear Sir, Blood centres have, for many years, undertaken extended red cell phenotyping of blood donors, and in recent years, some have started a programme of genotyping donors. There are two clear benefits in adopting a genotyping programme: first, phenotyping reagents of some blood group specificities are not readily available, and second, genotyping may allow a better match of donors with patients with unusual genetic variations, particularly at the RH locus. Chou et al. (2013) reported that 87% of sickle cell disease (SCD) patients had at least one D variant RH allele and that such patients often had unexpected antibodies of Rh specificity. In a later paper, Chou et al. (2018) reported that, although RH gene matching of SCD patients was feasible, cost remained a limiting factor. Indeed, for SCD and other patients, the large majority of hospital blood banks rely on phenotyping potential recipients for the selection of suitable blood. In the UK, current guidelines for SCD patients recommend that an extended phenotype (or genotype) including C, c, E, e, K, k, Jka, Jkb, Fya, Fyb, S, s should be performed on all SCD patients at baseline, and as a minimum and in the absence of alloantibodies, red cells should be matched for D, C, c, E, e and K red cell antigens (Davis et al., 2017). Although alloantibody formation is more common in SCD patients than in other groups, with alloimmunisation rate reported to be between 2·6 and 76·2% (Zheng & Maitta, 2016), other patient groups such as thalassaemia are also associated with a significantly raised alloimmunisation rate, and some clinicians believe that these groups also merit the transfusion of more extensively phenotyped blood. The provision of phenotyped blood for women of childbearing age is somewhat controversial: in the UK, K-negative blood is usually selected, with some laboratories additionally selecting c-negative blood. Doyle et al. (2014) found that 15% of mothers formed alloantibodies following intra-uterine transfusion (IUT), where the transfused red cells expressed the cognate antigen, and recommended the use of better matched blood for IUT. It is difficult to estimate the cost-effectiveness of prospective antigen matching between donor and patient. In a comprehensive study, Kacker et al. (2014) examined the cost-effectiveness in the United States of various matching strategies for preventing alloimmunisation in sickle cell patients. They reported that implementing prospective matching would be very costly compared to history-based extensive matching but would result in 2424 fewer alloimmunisation events. A later paper reported that