Indirect Antiglobulin Test

The indirect antiglobulin test (indirect Coombs test) detects free antibody in the patient's serum or plasma that is directed against red cell antigens. Serum is incubated at 37 °C with group O reagent red cells of known antigen profile. The cells are washed to remove unbound immunoglobulin, and antihuman globulin is added. Agglutination, or its equivalent in gel and solid phase systems, shows that antibody has bound. The test underlies the antibody screen, antibody identification, crossmatching and antibody titration.

Clinically significant red cell antibodies are mostly IgG, react at 37 °C and cause haemolytic transfusion reactions and haemolytic disease of the foetus and newborn. Cold reacting IgM antibodies are common and rarely significant. A positive screen is found in a small minority of patients, and more often after multiple transfusions or pregnancies.

Indications. Pre-transfusion testing, antenatal screening at booking and at 28 weeks, any sensitising event in pregnancy, suspected haemolytic transfusion reaction, a new positive direct antiglobulin test, suspected haemolytic disease of the newborn, regular transfusion in sickle cell disease or thalassaemia, transplantation, and before and during anti-CD38 therapy.

Specimen. EDTA or clotted blood. The sample must be recent. Usually this means within 3 days when the patient has been transfused or pregnant in the previous 3 months, because new antibodies can appear quickly. Up to 7 days is accepted otherwise, depending on local policy.

Screening. Two or three group O cells that together carry the clinically important antigens (D, C, c, E, e, K, k, Fya, Fyb, Jka, Jkb, S, s, M, N, Lea, Leb, P1) are incubated with serum. Reactivity is enhanced with low-ionic-strength solution or polyethylene glycol, then read with antihuman globulin in tube, gel or solid phase systems.

Identification. A panel of about 10 to 16 phenotyped cells is used. Antigens present on non-reacting cells are excluded, preferably on cells with double dose expression. Specificity is confirmed when at least three antigen positive cells react and three antigen negative cells do not. The patient must lack the corresponding antigen. An autocontrol is run alongside.

Titration and quantification. Serial doubling dilutions of serum are tested against cells carrying the target antigen. The titre is the reciprocal of the highest dilution giving 1+ agglutination. Anti-D can be quantified in IU/mL on automated analysers. Titres are comparable only within the same laboratory and method, and a rise of two or more doubling dilutions is significant.

Clinically significant antibodies. Anti-D, anti-C, anti-c, anti-E, anti-e, anti-K, anti-Fya, anti-Fyb, anti-Jka, anti-Jkb, anti-S, anti-s and other antibodies reactive at 37 °C. Antigen negative red cells are needed for transfusion, and the antibody needs monitoring in pregnancy.

Usually not clinically significant. Cold reactive IgM antibodies (anti-M, anti-N, anti-P1, anti-Lea, anti-Leb, anti-I) that are inactive at 37 °C. Any reactivity at 37 °C changes this and needs identification.

Haemolytic disease of the foetus and newborn. Anti-D is the best known cause, and anti-c, anti-E, anti-K and others also cause disease. Referral thresholds are laboratory specific. Anti-D quantification of 4 IU/mL or more, and titres of about 16 to 32 for other antibodies, usually prompt foetal medicine assessment. Anti-K suppresses foetal erythropoiesis and can cause severe anaemia at lower titres (about 8 or more) with little haemolysis, so bilirubin is an unreliable marker. A rising titre matters more than a single value.

Passive anti-D. Anti-D immunoglobulin given at 28 weeks or after a sensitising event is detectable in the screen for weeks to months. The history of immunoglobulin administration and quantification distinguish it from immune anti-D.

Other causes of reactivity.

• Warm autoantibody gives panreactivity with a positive autocontrol, and adsorption is needed to exclude underlying alloantibody.
• Daratumumab and other anti-CD38 antibodies cause panreactivity because reagent cells express CD38. It is resolved with dithiothreitol treated cells. This treatment also destroys Kell antigens, so Kell negative blood is selected by phenotype.
• Antibodies to high frequency antigens cause panreactivity with a negative autocontrol.
• Rouleaux, reagent additive antibodies and intravenous immunoglobulin (passive ABO and other antibodies) can cause reactions.
• Evanescent antibodies fall to undetectable levels, while immune memory persists. The record of a previous antibody must be kept, and antigen negative blood is still given.

Critical findings needing immediate action.

• A new clinically significant antibody in pregnancy, or a rising titre, needs urgent referral to foetal medicine. Foetal anaemia is assessed by middle cerebral artery peak systolic velocity, and a value over 1.5 multiples of the median indicates moderate to severe anaemia.
• In an emergency with a positive screen, group O Rh D negative blood (Kell negative in women of childbearing potential) is issued, and antibody identification continues.
• A new antibody 3 to 14 days after transfusion with a falling haemoglobin suggests delayed haemolytic transfusion reaction, and needs a direct antiglobulin test and urgent transfusion laboratory review.

Pitfalls.

• Dosage effect: antibodies react more weakly with cells from heterozygotes (Rh, Duffy, Kidd, MNS systems).
• Multiple antibodies mask one another, and each needs to be excluded separately.
• Enhancement media can produce false positive reactions.
• Wrong-blood-in-tube and labelling errors are the most dangerous failures, and a second sample is needed to confirm group.
• An antibody that has fallen to undetectable levels leads to a falsely negative screen.

Special groups.

• Pregnancy: testing at booking, at 28 weeks and after sensitising events. Foetal RHD genotyping from maternal plasma cell free DNA can avoid unnecessary prophylaxis in some settings.
• Sickle cell disease and thalassaemia: high alloimmunisation rates, so extended phenotype matching (at least Rh and K) is advised.
• Infants under 4 months: maternal and infant samples are used, because maternal antibody is the relevant risk.
• Patients on anti-CD38 therapy: a pre-treatment phenotype or genotype is needed.

Follow up tests. Antibody identification panel, direct antiglobulin test, antibody titre or anti-D quantification, red cell phenotype and genotype, Kleihauer-Betke test, foetal RHD genotyping, middle cerebral artery Doppler, neonatal haemoglobin and bilirubin.