An evaluation of statistical process control techniques applied to blood component quality monitoring with particular reference to CUSUM.

BACKGROUND: Statistical process control (SPC) is used to monitor the performance of blood component collection and production processes in the UK and elsewhere. The sensitivity of the applied technique(s) needs to be matched to the clinical importance of the parameter being monitored such that significant deviations in the process mean and/or variability of critical parameters (e.g. the leucocyte content of leucodepleted components) are detected and investigated immediately. AIMS: This study assessed the sensitivity and specificity of a range of techniques for variable and attribute (proportion non-conforming) data. MATERIALS AND METHODS: Comparison was based on a range of simulated and 'live' blood component quality monitoring data including X/R, cumulative sum (CUSUM) procedures, the scan statistic and np charts. RESULTS: X/R and CUSUM could detect shifts of two standard deviations in the process mean within 5 days. Current leucocyte count data (substantially skewed even after log transformation) was found to be better suited to attribute analysis. CUSUM alone was able to detect shifts on the same day when based on 20 or more samples and achieved acceptable specificity. CONCLUSIONS: CUSUM procedures for proportion non-conforming can usefully augment existing X/R techniques for leucodepletion monitoring, provide valid control limits and the required sensitivity. The scan statistic and 'np' charts offered no obvious advantages.

Eurobloodpack: a common European design for blood bag systems with integral leucodepletion filters.

Three EBA specified blood bag configurations ('Eurobloodpack') are described which are capable of meeting >80% of its member's requirements. These include a 'top-and-top' and two 'bottom-and-top' packs enabling aseptic, pre-donation collection of up to 40 ml of samples, 427.5-522.5 ml of whole blood and the preparation of an extensive range of blood components. Features currently beyond the scope of ISO standardisation have been controlled including: anticoagulant and additive volumes; collection needle and sampling system; transfer tubing; cross-match line; base label; leucodepletion filter performance; compatibility of access ports and transfusion sets. Eurobloodpack has significant advantages for blood services and blood bag manufacturers.

Screening of platelets for bacterial contamination at the Welsh Blood Service.

BACKGROUND AND OBJECTIVE: This report details the results of the implementation of a bacterial screening system at the Welsh Blood Service and provides an estimate of the levels of bacterial contamination at the time of sampling. MATERIALS AND METHODS: Apheresis (Caridian BCT) and buffy coat-derived pooled platelet components were sampled on day 1 for bacterial contamination and the sample was monitored throughout the lifespan of the platelet component. Unused platelet components were re-tested to determine the effectiveness of the screening. Results from the BacT/ALERT are uploaded to the in-house Blood Establishment Computer System (BECS) every 12 min. Positive alerts are automatically sent to staff, facilitating a timely intervention. RESULTS: Between February 2003 and March 2010 the screening system tested 54 828 platelets and detected 257 (1 in 213) initial positives of which 35 (1 in 1567, 0·06%) were confirmed [95% confidence interval (CI), 0·04-0·08%]. Additionally, screening of 6438 unused platelet components detected another 6 (1 in 1073, 0·09%) confirmed positives not detected during initial testing (95% CI, 0·02-0·16%). Analysis of the data suggests that on day 1 the number of bacteria in such platelet component packs was between 5 and 62 cfus total. Day 1 culture has a sensitivity of 40%. CONCLUSIONS: The bacterial screening system has removed a significant number, but not all bacterially contaminated platelet components from the supply. The sample volume is an important factor in sensitivity due to the low number of bacteria in a platelet component pack on day 1. An effective notification and recall system is a critical part of the bacterial screening system.

Human Rh monoclonal antibodies: assessment of functional activity by chemiluminescence and RhD antibody quantitation.

In vitro cellular assays have been described which are capable of evaluating the interactions between sensitised red cells and monocyte or K cells. The chemiluminescence assay (CL) has several advantages over other cellular assays used to assess functional activity. The CL assay unlike the ADCC assays does not require the use of radioisotopes and therefore can be easily integrated into the work of a Reference Serology laboratory. The CL assay is an objective test and not labour intensive which is the main criticism of the monocyte monolayer assay. Seventy-four monoclonal anti-Ds and 29 other Rh specificities have been evaluated by a CL assay. The use of the chemiluminescent response produced by erythrophagocytosis of sensitised red cells has been shown to correlate well with the in vivo response to red cells sensitized with polyclonal IgG antibodies. This study aimed at investigating whether the CL assay could identify and differentiate monoclonal antibodies that are capable of eliciting a response from human monocytes. Poor correlation was obtained between the CL assay results and anti-D quantitation (r = 0.236). The chemiluminescence assay discriminated between anti-D's with high quantitation levels but low predicted functional activity and anti-Ds of low quantitation levels which produced elevated CL responses. Only 3 of the 29 non-Rh D specificities tested produced a response in the CL assay emphasising the importance of specificity in the functional activity of monoclonal antibodies. The demonstration of significant differences in the functional capabilities of monoclonal antibodies has important implications for reviewing the possible use of monoclonal anti-D preparations for Rh immune prophylaxis and highlights the requirement for factors other than the antibody concentration to be examined.

A study on Lu-null families in South Wales.

Red cells of 75614 blood donors in South Wales were screened with anti-Lu(b) revealing 54 Lu(b-) donors of which 15 were also Lu(a-) giving a frequency of 0.0002 for the Lu(a-b-) phenotype in South Wales. The families of 11 Lu-null propositi were investigated to determine which of the three known genetic backgrounds, dominant, recessive or X-linked recessive, was responsible for their Lu-null phenotype. In 10 of the 11 families the Lu-null phenotype was caused by the dominant suppressor gene In(Lu). The first reported family demonstrating independence of In(Lu) and LU, through the Au groups, is described together with the third family demonstrating suppression of P1 by In(Lu). The families showed that In(Lu) is not closely linked to HLA. The genetic background for the 11th propositus was not determined; homozygosity of the silent allele LU is a possible but unproved explanation.

A private red cell antigen, Jones, causing haemolytic disease of the newborn.

Two previously unpublished low-incidence antigens, Jones and Hol., are identical. The antigen is a dominant, autosomally inherited character that segregates independently from the loci for the ABO, MNS, Duffy and Yt blood group systems and is different from previously published infrequent antigens. The antigen is apparently unaffected by enzyme treatment and is well developed on red cells of neonates. The antibody reacts best by indirect antiglobulin testing, is IgG and has caused haemolytic disease of the newborn. This private blood group antigen, named Jones, has been assigned the ISBT number 700047.

HLA class I (Bg) antigens on red cells of SLE patients: a serological study with polyclonal and monoclonal antibodies.

The enhanced HLA class I (Bg) on red blood cells (RBC) of many patients with systemic lupus erythematosus has allowed a significant correlation to be made between their HLA-B types and haemagglutination reactivity with lymphocytotoxic anti-HLA-B sera stimulated by pregnancy alone. Therefore the class I expression on these RBC relates to classical, rather than non-classical, class I gene products. Studies of class I expression on RBC by means of monoclonal antibodies (MAb) to epitopes on the heavy polypeptide chain and beta 2-microglobulin (beta 2m) have suggested that the complete extracellular structure is present. The specific effect of chloroquine in 'stripping HLA' from RBC had been assumed to support the concept that HLA class I was adsorbed from plasma. However, from our data, we conclude that HLA class I is an intrinsic membrane component. We suggest that the action of chloroquine is to remove beta 2m alone, which prevents normal class I expression and also results in conformational changes to the class I heavy chain, but that it is not capable of removing the membrane-bound heavy chain.

The elucidation of a Kell-related autoantibody using ZZAP-treated red cells.

A 61-year-old, nontransfused, Caucasian male was found to have a positive direct antiglobulin test (DAT) and an autoantibody in his serum prior to total hip replacement. Autoabsorption with the patient's ZZAP-treated red cells failed to absorb the autoantibody, giving a clue to its possible specificity, which was subsequently found to be Kell-system related. In addition, his red cells were found to have a slightly weakened expression of Kell system antigens. The patient was fit and healthy with normal hematological indices at the time of the operation. One year the hemlogical findings remained the same despite the persistence of the autoantibody and a positive DAT. This suggested that the autoantibody was benign, which was supported by a negative in vitro mactophage assay.

Elucidation of alloantibodies in autoimmune haemolytic anaemia.

Forty-one patients with autoimmune haemolytic anaemia (AIHA), who had free antibody in their sera, were investigated for the presence of alloimmune erythrocyte antibodies using the ZZAP autoabsorption technique. Patients were subdivided into three risk categories: (I) no prior pregnancy or transfusion; (II) history of pregnancy and/or one to five transfusions; and (III) greater than 5 transfusions. A total of 13 (32%) of the 41 patients exhibited significant alloantibodies. Of 11 category-I patients 2 (18%) had significant alloantibodies. Eight (31%) of the 26 category-II patients had significant alloantibodies and 3 (75%) of the 4 category-III patients had significant alloantibodies after absorption. The majority showed Rh specificity: anti-E(8), -C(3), -Cw(1). Anti-K was found in 6 samples and 1 had anti-Fya. Alloantibodies had not been suspected prior to autoabsorption in 10 (77%) of the 13 patients with alloantibodies. These findings underline the importance of performing autoabsorption in AIHA when free autoantibody is present in the serum. Additionally, Rh phenotyping performed on ZZAP-treated cells showed complete agreement with that ascertained using pure IgM Rh typing sera and untreated cells.

Application of a microtitre plate antiglobulin technique to determine the incidence of donors lacking high frequency antigens.

Approximately 65,000 random blood donors have been screened by a microtitre low ionic strength saline antiglobulin technique over a 12-month period. Antisera to high frequency antigens including: Vel, Lub, k, Kpb, Yta and Era were used. Additional sera including: anti-Lan, -Coa, -Sc1 and an antibody to an unidentified high frequency antigen, Mrs C.A., have been used as they became available. This has enabled the approximate frequencies of donors lacking the corresponding antigen, to be calculated.

Transfusion significance of LWa allo-antibodies.

An example of anti-LWa, arising as a complication during a RhD immunization programme, has been studied for evidence of its likely in vivo haemolytic properties. In vitro testing of the anti-LWa showed it to be largely IgG1 acting by the antiglobulin technique. Results of antibody-dependent cellular cytotoxicity and macrophage phagocytic assays were both negative. However, 99mTc-labelled Lw(a+) donor cells showed a slight reduction in t1/2 (18 h) compared with the normal survival of autologous cells. Despite this observation, and bearing in mind the difficulties of interpreting apparently accelerated destruction of small serologically incompatible red cells, it was concluded that the presence of this example of anti-LWa should not be a bar to urgent transfusion.

The Webb (Wb) antigen in South Wales donors.

Eight Webb antigen Wb+ propositi, 2 in the same family, were found in testing 10,117 random blood donors in South Wales with anti-Wb serum. Family studies on the new propositi confirmed the Wb antigen to be inherited as an autosomal dominant character. Wb segregated independently from ABO, Rh, MNSs, K, Fy, Jk, and was not X- or Y-borne; for the first time Wb was shown to be independent of Lu. Wb antigen was destroyed by some proteolytic enzymes and by neuraminidase. Sera from 5,144 random donors were screened with Wb+ cells resulting in detection of 2 additional anti-Wb.

Two further examples of the low-frequency antigen Rea (Reid).

A positive direct antiglobulin reaction of a baby's cells, found during antenatal testing at Cardiff, was caused by a low-frequency antigen which has been identified as Rea (ISBT No. 700019). Investigations with this, only the second example, anti-Rea revealed another Re(a+) propositus in testing 6,635 random donors at Oxford; no further positive was found in 4,770 random donors at Cardiff. Study of the families of these two propositi, the first to be identified since the original Canadian family, confirmed that Rea was inherited as an autosomal Mendelian dominant. The families showed that Rea is not controlled by the ABO, Kell or Au loci and confirmed that it was not controlled by MNSs and was not X- or Y-borne.

A new low-frequency antigen, Hga (Hughes).

This paper describes a 'new' low-frequency antigen, Hga, which came to light during routine antibody identification tests. 3 families were investigated, 2 found during the screening of 5,434 random group O donors, giving a frequency of 1:2717. The families showed the antigen to be inherited as a Mendelian autosomal dominant character segregating independently of: Rh, MNS and is not linked to X or Y. No pure examples of the corresponding antibody have been discovered in testing 6,580 sera from normal donors in South Wales.