Blood compatibility testing

Jump to navigation Jump to search

Author: Mikael Häggström [note 1]

Blood typing

ABO antigens and antibodies

Upon direct testing by adding antibodies against A, B and/or Rh to patient blood, agglutination means that the patient has the antigen tested. Upon indirect testing by adding A or B antigen to patient plasma, agglutination means absence of the antigen in the patient (and thus the patient produces antibodies against it).

Identification of non-ABO antibodies

In the antibody screening procedure, an individual's plasma is added to a panel of two or three sets of red blood cells which have been chosen to express most clinically significant blood group antigens. Agglutination of the screening cells by the plasma, with or without the addition of anti-human globulin, indicates that an unexpected blood group antibody is present. If this occurs, further testing using more cells (usually 10–11) is necessary to identify the antibody. By examining the antigen profiles of the red blood cells the person's plasma reacts with, it is possible to determine the antibody's identity as follows:

The image above shows the interpretation of an antibody panel used to detect antibodies towards the most relevant blood group antigens. Each row represents "reference" or "control" red blood cells of donors which have known antigen compositions and are ABO group O. A + means that the antigen is present on the reference red blood cells, and 0 means it is absent; nt means "not tested". The "result" column to the right displays reactivity when mixing reference red blood cells with plasma from the patient in 3 different phases: room temperature, 37°C and AHG (with anti-human globulin, by the indirect antiglobulin test).[1]

  • Step 1; Annotated in blue: starting to exclude antigens without reaction in all 3 phases; looking at the first reference cell row with no reaction (0 in column at right, in this case cell donor 2), and excluding (here marked by X) each present antigen where the other pair is either practically non-existent (such as for D) or 0 (presence is homozygous, in this case homozygous c).
    When both pairs are + (heterozygous cases), they are both excluded (here marked by X), except for C/c, E/e, Duffy, Kidd and MNS antigens (where antibodies of the patient may still react towards blood cells with homozygous antigen expression, because homozygous expression results in a higher dosage of the antigen).[2] Thus, in this case, E/e is not excluded in this row, while K/k is, as well as Jsb (regardless of what Jsa would have shown).[note 2]
  • Step 2: Annotated in brown: Going to the next reference cell row with a negative reaction (in this case cell donor 4), and repeating for each antigen type that is not already excluded.
  • Step 3: Annotated in purple. Repeating the same for each reference cell row with negative reaction.
  • Step 4: Discounting antigens that were absent in all or almost all reactive cases (here marked with \). These are often antigens with low prevalence, and while there is a possibility of such antibodies being produced, they are generally not the type that is responsible for the reactivity at hand.
  • Step 5: Comparing the remaining possible antigens for a most likely culprit (in this case Fya), and selectively ruling out significant differential antigens, such as with the shown additional donor cell type that is known to not contain Fya but contains C and Jka.

In this case, the antibody panel shows that anti-Fya antibodies are present. This indicates that donor blood typed to be negative for the Fya antigen must be used. Still, if a subsequent cross-matching shows reactivity, additional testing should be done against previously discounted antigens (in this case potentially E, K, Kpa and/or Lua).[1]

Hemagglutination inhibition[2]
Neutralizing substance Antigen cancelled
  • Hydatid cyst fluid
  • Pigeon eggs
Saliva H, Lea
Breast milk I
Guinea pig urine Sda
Hemagglutination inhibition substances sound like they came from a witch brew!

When multiple antibodies are present, or when an antibody is directed against a high-frequency antigen, the normal antibody panel procedure may not provide a conclusive identification. In these cases, hemagglutination inhibition can be used, wherein a neutralizing substance cancels out a specific antigen.[2] Alternatively, the plasma may be incubated with cells of known antigen profiles in order to remove a specific antibody (a process termed adsorption); or the cells can be treated with enzymes such as ficain or papain which inhibit the reactivity of some blood group antibodies and enhance others (see table below).

Clinical implication

The following is a simplified classification for the main anti-erythrocyte antibodies, using mneumonics for the main involved antigen groups:

  • Anti-A/B antibodies: These will cause immediate hemolytic transfusion reaction if the red blood cells do not have compatible antigens, even if there is no previous exposure to the antigens. Therefore, blood transfusions must always be ABO compatible.
  • "Kickers"-class antibodies: Antibodies against Kidd, Kell, Rh, S and Duffy group antigens. These have a significant risk of causing hemolytic transfusion reactions when present, and therefore, patients with kickers-class antibodies should receive blood that is negative for the antigen, except for very critical situations where there is no time to find compatible blood.[3] Kickers-class antibodies generally need a previous exposure to the antigen to form, with transfusion reactions being possible upon subsequent transfusions.[2] Some patients first test positive and later test negative for a kickers-class antibody, but such patients must still be transfused with antigen-negative blood regardless.[4] They are generally of the IgG subtype, and are generally most active at 37°C. They can potentially cross the placenta and cause hemolytic disease of the newborn. They generally show increased reactivity against homozygously expressed antigens compared to heterozygously expressed ones (as mentioned in Step 1 above).
  • "Limply"-class antibodies: Antibodies against Lutheran, Ii, M/N, P1, Lewis group antigens. These almost never cause clinically significant transfusion reactions (but anti-Ii antibodies are usually the type that causes cold agglutinin disease,[5] a form of autoimmune hemolytic anemia).[2] Hence, there is generally no need to find blood that is negative for the antigen for a limply-class positive patient. These antibodies are generally naturally occurring, that is, they don't require a previous exposure to the antigen to form. They are generally of the IgM class, and are generally not reactive at body temperature, but rather most active at room temperature and below.[6] They generally pose no significant risk of hemolytic disease of the newborn (as IgM class antibodies do not cross the placenta).

Following is a comparison of clinically relevant characteristics of antibodies against the main human blood group systems:[2]

ABO Rh Kell Duffy Kidd Lutheran MNS Lewis P Ii
Most common in immediate hemolytic transfusion reactions A Yes Fya Jka
Most common in delayed hemolytic transfusion reactions E,D,C Jka
Most common in hemolytic disease of the newborn Yes D,C Yes
Commonly produce intravascular hemolysis Yes Yes Yes
Reactive at room temperature Yes M,N Lea, Leb P1
Nearly always clinically insignificant Yes M,N Yes P1
Naturally occurring Yes Yes M,N Yes Yes Yes
Enhanced by ficain[7] and papain[8] Yes Yes Yes Yes P1 Yes
Destroyed by ficain[7] and papain[8] Fya, Fyb Yes Yes
Displaying dosage Further information: Blood compatibility testing Cc, Ee Yes Yes Yes


  1. For a full list of contributors, see article history. Creators of images are attributed at the image description pages, seen by clicking on the images. See Patholines:Authorship for details.
  2. Besides from C/c, E/e, Duffy, Kidd and MNS, clinically significant dosage effects is rare but not impossible for other antigens, which thus may still be considered if subsequent cross-matching is reactive.

Main page


  1. 1.0 1.1 Justin R. Rhees, M.S., MLS(ASCP)CM, SBBCM. Introduction to Antibody Identification. University of Utah, Medical Laboratory Sciences.
  2. 2.0 2.1 2.2 2.3 2.4 2.5 Mais, Daniel (2014). Quick compendium of clinical pathology . United States: American Society for Clinical Pathology Press. ISBN 978-0-89189-615-9. OCLC 895712380. 
  3. Pamela P. Goodell, Lynne Uhl, Monique Mohammed, Amy A. Powers (2010). "Risk of Hemolytic Transfusion Reactions Following Emergency-Release RBC Transfusion ". American Journal of Clinical Pathology 134 (2). doi:10.1309/AJCP9OFJN7FLTXDB. 
  4. Uhl, L (11 Jan 2021). Pretransfusion testing for red blood cell transfusion. UpToDate.
  5. "Autoimmune hemolytic anemia: current knowledge and perspectives ". Immunity & Ageing 17 (1): 38. November 2020. doi:10.1186/s12979-020-00208-7. PMID 33292368. 
  6. Ferdowsi S, Mohammadi S, Ahmadnezhad M, Herfat F, Rezvani A, Eshghi P (2022). "Anti-M antibody and ABO blood grouping discrepancy: a report of three cases with review of literature. ". Hematol Transfus Cell Ther 44 (2): 288-290. doi:10.1016/j.htct.2020.09.150. PMID 33358685. PMC: 9123591. Archived from the original. . 
  7. 7.0 7.1 Hill, Ben C.; Hanna, Courtney A.; Adamski, Jill; Pham, Huy P.; Marques, Marisa B.; Williams, Lance A. (2017). "Ficin-Treated Red Cells Help Identify Clinically Significant Alloantibodies Masked as Reactions of Undetermined Specificity in Gel Microtubes ". Laboratory Medicine 48 (1): 24–28. doi:10.1093/labmed/lmw062. ISSN 0007-5027. PMID 28007780. 
  8. 8.0 8.1 Eric Ching. Questions and Answers on Proteolytic Enzymes Used in Blood Group Serology. Canadian Society for Transfusion Medicine.

Image sources