d. Serum. As the autoantibody reacts optimally at 37C and thus is being
absorbed "in vivo," the patient's serum may contain very little free antibody. Generally
speaking, free autoantibody may only appear in the serum when all antigen sites on the
patient's RBCs are filled and no more antibody can be absorbed (for example, when the
direct antiglobulin test is strongly positive). Thus, only approximately 35 percent of
serums from warm antibody AIHA will react by the indirect antiglobulin test. If more
sensitive techniques, such as, the use of enzyme-treated RBCs, are employed, over 90
percent of the serums can be shown to contain autoantibody. The antibody may be IgG
and/or lgM and react optimally at 37C. Agglutinins and hemolysins against untreated
RBCs at 37C are extremely rare, but warm hemolysins, active only against enzyme-
treated RBCs, occur in about 20 percent of the serums. Although cold agglutinins titers
are within the normal range (for example, up to 64) at 4C, it is not unusual to find them
reacting up to room temperature in serums from patients with warm antibody AIHA. It is
important to remember that the serum may contain alloantibodies in addition to
autoantibodies or quite often only alloantibodies, the autoantibody having been
absorbed by the patient's RBCs "in vivo."
e. Specificity. The specificity of the autoantibodies associated with warm
antibody AIHA is very complex. As discussed previously, the main specificity is directed
against the Rh complex, but may only be obvious if very rare cells such as, -D-, or
Rhnull, are available. Apart from Rh specificity, there have been reports of anti-U,
anti-LW, anti-IT, anti-Kell, and anti-Wrb, being associated with warm antibody AIHA.
f. Compatibility Testing.
(1) Blood transfusions should be avoided in AIHA. The transfusion of
incompatible blood, especially in the presence of alloantibody that may be masked by
"nonspecific" autoantibody, involves substantial risk. In order to assume these risks,
there must be clinical needs sufficiently great to overcome the potential danger. It is
equally wrong to avoid all transfusions unequivocally because of incompatible blood
while a patient dies with progressive anemia. Clinical judgment must resolve these
(2) There are two main problems to be considered before carrying out
compatibility tests on AIHA patients. The first, and most important, is to determine if
alloantibodies are present in the patient's serum. The second is the relevance of the
specificity of the autoantibodies.
(3) As the presence of alloantibodies capable of causing severe transfusion
reactions can be masked by the presence of autoantibodies that react with all RBCs
tested, it is essential to use techniques such as the autoabsorption technique discussed
previously to determine their presence or absence. If alloantibodies are detected, blood
lacking antigens to those antibodies is selected for cross matching.