Hugh Chaplin

Prevalence of HLA sensitization in female apheresis donors

BACKGROUND: Transfusion‐related acute lung injury (TRALI) is a serious complication of plasma‐containing blood components. Studies have implicated HLA antibodies along with biologically active lipids in stored blood in the pathogenesis of TRALI. It has been proposed that the exclusion of HLA‐untested, multiparous donors of plasma‐rich components, including plasma and single‐donor apheresis platelets, would substantially reduce the risk of TRALI.

Alloimmunization in patients with warm autoantibodies: A retrospective study employing three donor alloabsorptions to aid in antibody detection

We examined the value of performing alloabsorptions to detect clinically significant alloantibodies in patients with warm autoantibodies who must receive crossmatch‐incompatible blood. One hundred and twenty‐five (125) patients were evaluated using alloabsorption with red cells (RBCs) from three donors: R1R1, R2R2, and rr, whose phenotypes other than Rh were selected to exclude 98 percent of clinically significant alloantibodies. This technic was selected rather than autoabsorption due to insufficient quantities of patient cells available and to the possible presence of transfused cells in some instances. Patients were divided into three risk categories: I‐no prior pregnancy or transfusion; II‐history of pregnancy and/or one to five transfusions; and III‐greater than five transfusions. No significant alloantibodies were found in 32 category I patients. Of 74 category II patients, 13 (17.5%) had significant alloantibodies detectable after absorption. Six of 19 (31.5%) category III patients had alloantibodies. The majority showed Rh specificity: anti‐E (13), ‐C (6), ‐c (2), ‐D (1). Anti‐K was found in five samples. Forty‐two (42%) percent of the alloantibodies were undetectable prior to the alloabsorptions. We conclude that category II and particularly category III patients are at significant risk of allosensitization and should be evaluated by an absorption procedure prior to the transfusion of crossmatch‐incompatible red cells.

Pregnancy and Idiopathic Autoimmune Haemolytic Anaemia: A Prospective Study during 6 Months Gestation and 3 Months Post‐Partum

A 31‐yr‐old woman with a 12 yr history of relapsing idiopathic autoimmune haemolytic anaemia was studied prospectively during her first pregnancy. Her serum contained a warm incomplete autoantibody as well as an elevated cold agglutinin; her red blood cells were strongly coated with IgG and complement (chiefly α2D). Haemolysis was active throughout pregnancy, accelerating from the 34th to 40th week, with developing thrombocytopenia. Amniocentesis in the 8th and 9th months suggested minimal foetal haemolysis. The maternal haemolytic process went into complete clinical remission following delivery of a healthy appearing infant whose red cells were coated with IgG. The infant developed mild hyperbilirubinaemia within 48 hr and experienced a fall in haemoglobin to 50% of the cord level by the 8th week. Abnormalities of maternal and infant C4 levels were observed. Review of 19 reported instances of presumed autoimmune haemolysis during pregnancy revealed life‐threatening anaemia in nearly 50% of mothers, with four still‐births, one neonatal death, and three seriously affected infants. A programme for prospective management of this serious clinical problem is discussed.

Quantitation of Red Blood Cell‐bound C3d in Normal Subjects and Random Hospitalized Patients

Summary. A sensitive radiolabelled anti‐antiglobulin method was devised and applied to quantitating red blood cell‐bound C3d (RBC‐C3d) in samples from 174 normal blood donors. C3d was demonstrable on all RBC examined; 98% of values fell over a broad range, with the highest values being approximately 3·5 × the lowest values (equivalent to 50–160 molecules of C3d per cell). RBC‐C3d did not correlate with sex or age (over 18–65 years); indirect evidence suggests that values for the paediatric age group will fall in the same normal range. Studies on samples obtained weekly for 10–12 weeks from six adult males and six adult females indicated stable levels of RBC‐C3d for individual subjects; i.e. high normals, mid normals and low normals remained in their characteristic range levels over the period of observation. For comparison, RBC‐C3d was measured in samples from 313 randomly selected hospitalized adult patients. 33% of the values were above the normal range; 8% were elevated to a level likely to have been detectable by a direct anti‐C3d antiglobulin test. The great majority of elevated values occurred in patients not ordinarily considered to have autoimmune conditions. The results provide background for studies of the aetiology and significance of RBC‐C3d in health and disease.

Transfusion of Buffy Coat‐Poor Red Cell Suspensions Prepared by Dextran Sedimentation: Description of Newly Designed Equipment and Evaluation of Its Use

Modified equipment is described for use in preparation of buffy coat‐poor red cell suspensions by the dextran sedimentation procedure. The chief advantages over previously available equipment are: (a) Minimized contamination hazard, (b) Administration of the final red cell suspension from the original blood container, thereby assuring maintenance of correct donor identification. Experience with the modified equipment in preparation of 70 units of buffy coat‐poor red cells for transfusion is described. Experience with transfusion of 297 units of buffy coat‐poor red cell suspensions at Barnes Hospital during the past five years is briefly reviewed.

Persistent Polyagglutinability in Vivo Unrelated to T‐antigen Activation

Red blood cells of a type B patient became polyagglutinable eight years ago and have remained so to the present. No infection or other cause for the phenomenon has been discovered and laboratory studies have established conclusively that the polyagglutinability is distinguishable from T‐antigen activation and from the effect of periodate treatment in vitro. Combined 51Cr and Ashby differential agglutination studies of the survival in vivo of normal type B donor blood demonstrated that the transfused cells did not become polyagglutinable during more than four months in the patients circulation. It is concluded that the underlying abnormality of the patients red blood cells originates during their formative stage in the bone marrow. The clinical implications of these findings and their possible relationship to the patients accompanying leukopenia and thrombocytopenia are discussed briefly.

The Occasional Fallibility of in Vitro Compatibility Tests

A patient is described in whom rapid destruction of transfused red cells occurred repeatedly despite entirely compatible cross‐match results by a wide variety of dependable laboratory procedures. The report includes detailed serologic investigations of these phenomena, as well as the results of 17 measurements of in vivo red cell survival. A striking clinical observation was the regular onset of typical sickle cell “crisis” coincident with the rapid destruction of large volumes of donor erythrocytes.

Low retention of white cell fragments by polyester fiber white cell-reduction platelet filters

BACKGROUND: There is intense interest in the potential of current white cell (WBC)‐reduction filters to prevent the alloimmunization of patients by the residual donor WBCs in filtered blood components transfused to them. Little attention has been paid to the capacity of current synthetic fiber filters to remove WBC membrane fragments bearing detectable leukocyte antigens.

Studies on anti-eluate sera. I. The production of antiglobulin (Coombs) sera in rabbits by the use of antibodies eluted from sensitized red blood cells.

1 Rabbits were successfully immunized with eluates prepared from human red blood cells. 2 The presence of demonstrable antibody activity in the eluate was not a prerequisite to provoking an immune response in rabbits. 3 Immune responses were obtained with eluates from red cells whose sensitization appeared to be non‐gamma globulin in type as well as from gamma globulin sensitized cells. 4 Weak but definite immune responses were obtained with eluates from normal red cells. 5 Broad cross‐reactivity was observed among the various anti‐eluate antisera when tested against a panel of cells whose sensitization covered the range of gamma globulin, mixed and non‐gamma globulin types.

Red cell-bound immunoglobulin as a predictor of severity of hemolysis in patients with autoimmune hemolytic anemia

When a patient is suspected of having autoimmune hemolytic anemia (AIHA), a simple rapid test to predict the severity of the hemolytic process would be helpful. Because thc process is immunologic, i t is reasonable to quantitate the autoantibody bound to the patient’s red cells (RBCs), as well as the autoantibody free in the patient’s serum. There are various methods for attempting these quantitations. On the basis of studies of these methods, highly competent investigators have disagreed about whether severity of hemolysis is predictable by the amounts of bound and free autoantibody. Various aspects of this dilemma are evident in the report by Garratty and Nance’ in this issue of TRANSFUSION. To understand why quantitation of autoantibody may not reliably predict the severity of AIHA, one must examine the conventional criteria for severity of hemolysis and the complexity of the immune processes contributing to a shortened life span for RBCs. Conventional criteria such as hemoglobin concentration, percentage of reticulocytes, unconjugated bilirubin concentration, lactate dehydrogenase activity, and haptoglobin concentration may collectively be of value in assessing the severity of the process, but, individually, these variables may be abnormal for reasons other than hemolysis, which introduces a potentially compounding level of uncertainty and imprecision. Some investigators simply lump antiglobulin test-positive subjects into “hemolyzing” and “nonhemolyzing” groups on the basis of arbitrary values for in vitro tests such as the above. This has the advantage of blurring the effects of atypical cases but the disadvantage of making no distinction between mild and severe cases (both of which will appear in the hemolyzing group), thereby obscuring any general correlation that may exist between the number of autoantibody molecules bound and the severity of hemolysis. Actual measurement of autologous RBC survival (directly by a radioisotope-labeling method or indirectly by endogenous carbon monoxide production) is the most meaningful index of severity of hemolysis, but the methodology is often cumbersome; isotope studies may require a week or more to obtain sufficient data points for accurate calculation of half-life values; rapid fluctuations in blood volume can increase data scatter; and intcrpretation may be uncertain if the patient has bcen transfused recently (coincident, undetected alloimmune RBC destruction is possible). Also, for a test carried out over several days, actual changes in the rate of hemolysis may compound the difficulty of interpreting RBC survival curves. Some of these limitations are of much less importance to survival studies of potentially incompatible normal donor RBCs in alloimmunized recipients, in which definitive results are frequently obtained in lcss than 24 hours, such as the elegant and enlightening studies recently summarized by Mollison.2.3 I n vitro tests aimed at quantitating the amount of autoantibody bound to autologous or normal RBCs have taken one of two general forms. The less sophisticated but more readily accessible approach involves titration of the antiglobulin serum (vs. the patient’s RBCs) or of the patient’s serum (vs. ABO-compatible normal RBCs, carried through the antiglobulin phase employing a single dilution of antiglobulin serum). These tests require multiple RBC washes, which for low-affinity antibodies, provide opportunities for loss of RBC-bound autoantibody prior to the antiglobulin step. Since titer endpoints may be difficult to determine and give no indication of the strength of reactions at lower serum dilutions, a variety of scoring systems have been devised to provide more meaningful quantitation than titer endpoints alone. Scoring improves quantitation in some instances, but i t gives greater weight to the vagaries of subjective readings of individual reactions. In circumstances of multiple maximum-strength reactions (sometimes accompanied by weaker prozone reactions at low dilutions), the scoring may substantially underestimate the number of autoantibody molecules bound.