The simultaneous detection of autoantibodies and alloantibodies to red blood cells (RBC) following allogeneic transfusion is poorly understood. The incidence of autoimmune hemolysis in these patients is not well established (Young et al., Transfusion 2004; 44: 67-72).


Patients with a positive antibody screen, an identified alloantibody and a positive direct antiglobulin test (DAT) were selected from our hospital Blood Bank database from January 2002 to June 2013. Sex, age, main clinical diagnosis, transfusion history, alloantibody and autoantibody specificity and hemolysis parameters were analysed retrospectively in those patients with simultaneous first detection of auto and alloantibodies.

In our Blood Bank, all patients to be transfused underwent a three-cell antibody screen by gel technology (Ortho Clinical Diagnosis). If the screen was positive, two microcolumn panels with eleven cells (DianaGel) and an autocontrol were performed in Liss-Coombs and with papain-treated erythrocytes for antibody identification. Test-tube testing at different temperatures with Reagent RBC (Makropanel) and a polyethyleneglycol (PEG) antiglobulin technique were carried out to confirm antibody specificity when deemed appropriate. Polyspecific and monospecific (anti-IgG and C3d) DAT (Menarini reagents) and a 6% albumin negative control were performed in all patients presenting a positive autocontrol. Acid glycine elution (Gamma ELU-KIT II) was the method of choice to prepare the eluates. Adsorption techniques were used in some cases to remove autoantibodies.


A total of 2915 patients formed alloantibodies; 4.5% (n=132) presented both auto- and alloantibodies. The temporal relationship between the antibodies detected had the following distribution: 96.2% (n=127) simultaneous auto- and alloantibody, 2.3% alloantibody identification and on subsequent studies autoantibody and 1.5% had autoantibody previous to alloimmunization. The following results will refer to the group with simultaneous detection of auto and alloantibodies.

In the selected group, 55% had a complete register of previous RBC transfusion with a median of 8.5 transfusions lifelong (range: 2 to 81) and a median of 38.5 days (range: 4 to 270 days) between last negative and first positive antibody screen. The ratio male to female was 1:1 and the median age was 67 years-old (range: 20 to 91 years-old). The most common diagnosis in these patients were: solid neoplasia (21.3%), oncohematologic disease (13.4%; mielodysplastic syndromes were included in this subgroup), chronic hepatopathy (15%; of which 58% are HVC-related), cardiovascular disease (9.5%), traumatism (7%), pregnancy (3.2%) and autoimmune disease (3.1%). The most frequent alloantibodies identified were: anti-E (39.3%) > anti-C (28.3%) > anti-K (24.5%). Most autoantibodies had no apparent specificity (67%), those that did have a specific pattern were related to Rh antigens (anti-e 7%; anti-c 1.5%; anti-D 1.5%), and the remainder 23% were crioagglutinins.

A total of 16 patients (12.5%) had concomitant hemolysis at the time of antibody detection of which 43.7% (n=7) had an active cancer (57% due to oncohematologic disease), 25% (n=4) HCV hepatopathy and 12.5% autoimmune hepatitis (n=2).


Patients with de novo alloimmunization showed concomitant antierythrocytic autoantibodies in 4.5% of our population. Most cases appear to be related to previous RBC transfusion and medical conditions affecting normal immunological functioning. However, hemolysis was only found in 12.5% of these patients at the time of antibody detection and all of them had diseases with a known associated risk of developing autoimmune anemia.


No relevant conflicts of interest to declare.