Processes and Software Building 13: Validation of Electronic Crossmatch

The electronic/computer crossmatch is vital to the hospital blood bank’s operation.  It allows us to reduce unnecessary work and redeploy a very limited, precious workforce.  As demonstrated in the previous post, it has many rules for current and previous testing and historical transfusion records.

This is NOT a one-time validation:  it must be repeated to check for regression errors when updating the software version, regardless if there are any changes to its logic.  It required many Super-Users to perform:

  1. A test environment distinct from the live was used.
  2. Patients had to be created and specific test results entered to match the algorithm below.
  3. Donors had to be created and components made (including testing, production, release, inter-depot transfer)
  4. Hospital blood bank had to receive the created units and enter them into active stock (and repeat the ABO/D testing to do this)

This is a sample validation plan created by me and used at HMC Doha for the Medinfo Hematos IIG software:

  1. Special Electronic Crossmatch Testing
    1. Make 100 patients:
    2. 60 patients matching electronic crossmatching criteria:
      1. 2 ABO/D determinations, one current within 72 hours
      2. Second determination from history or second current sample
      3. No ABO/D typing discrepancies
      4. Negative antibody screen
      5. No history of antibodies
    3. 40 patients with contraindications to electronic crossmatch:
      1. 10 with history of antibody
      2. 10 with current antibody
      3. 10 without 2 ABO/D determinations
      4. 10 with emergency release
    4. Try to release RBC unit without further testing in each case.

Super Users had to capture screen shots of the transactions and send them to me as Division Head, Transfusion Medicine/LIS for review and acceptance.

Criteria for Acceptability:

  1. All tested processes must complete without error, including recognition of ABO/D discrepancies as such.
  2. System only allows computer release (electronic crossmatch) in those cases meeting computer-crossmatch criteria as shown in the following interim policy.


Revised Date:  1/2/18


In selected patient categories, no classical crossmatch may be required for release of RBC components.  The criteria are specified here as applicable in our Medinfo Hematos IIG computer system HIIG.


  1. An electronic crossmatch without antiglobulin or immediate-spin phase testing may be used for the following patient categories:
    1. The current ABO/D type matches the historical ABO/D type.
    2. The ABO/D type (forward and reverse) is clearly defined without any discrepancies.
    3. Two determination of the ABO/D group must be made:
      1. One from a current specimen (within the past 72 hours)
      2. The second by one of the following methods:
        1. Testing a second current specimen
        2. Comparison with previous records of ABO/D typing
        3. Retesting the same specimen
    4. The current antibody screen is negative
    5. There is no history of RBC antibodies or a non-negative antibody screen.
  2. General computer system safeguards:
    1. The system contains the donor unit number, component name, confirmed ABO/D typing, two unique recipient identifiers, recipient ABO/D, antibody screen typing, and interpretation of compatibility
    2. A method exists to verify correct entry of data before release of blood or blood components.
    3. The system contains logic to alert the user to discrepancies between donor ABO/D group on the unit label and those determined by blood confirmatory tests and to ABO incompatibility between the recipient and donor unit.
  3. HIIG will enforce the above rules.


  1. HIIG Workflow 1004, Patient Testing, 2013
  2. Section 5.16, Standards for Blood Banks and Transfusion Services, Current Edition, AABB, Bethesda, MD, USA
  3. FDA Guidance for Industry:  Computer Crossmatch (Computer Analysis of the Compatibility between the Donor’s Cell Type and the Recipient’s Serum or Plasma Type), April, 2011

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s