A hospital based transfusion medicine colleague in Georgia wonders how others perform quality control (QC) of reagent red cells that are part of an antibody identification panel, and what accreditation and/or regulatory requirements causes them to do the QC.

The following comments have been received.

Editor's NOTE: It would be helpful if accrediting organizations and regulatory agencies provided responses to this e-Network Forum question and/or commented on the input (like that below) which has been submitted by colleagues in the field:

ADDENDA May 31, 2005

1. A medical technologist and Transfusion Services Laboratory supervisor in San Antonio, Texas reports that their panel testing QC is limited to the performance of a selected cell "rule-out panel". When doing a selected cell panel they include a reagent cell that is positive for an antigen against which the patient has a known antibody, in order to demonstrate that the 'rule out' panel can detect known antibodies.

2. A medical technologist in Columbia, Missouri reports that their panel QC consists of documenting that a panel has both reactive and non-reactive results. Almost all panels they perform have a mix of reacting and non-reacting red cells, which they interpret to demonstrate that the panel is in control. For example, if they know or have a suspicion that a patient has anti-K, they make certain to test a sufficient number of K-positive and K-negative cells to demonstrate the panel is working properly. They acknowledge that their approach is problematic when a panel is completely negative in spite of a positive antibody screening test. Finally, in a case where they need to do 'rule out' cells (such as a mini-panel), they include a reagent red cell that expresses an antigen against which the patient has a known antibody, in order to demonstrate that the 'rule out' panel can detect known antibodies.

ADDENDA June 1, 2005

3. A biomedical scientist working for the national blood service in the UK reports that they use a weak antibody control to QC check their panels and screening cells. They usually use anti-Fya, because in their experience, the Duffy antigens are the "first to go when a panel deteriorates". They employ this weak antibody control at least once a day, with a batch of antibody panels. With screening cells, they use the antibody control with every batch of tests (usually on an automated machine). The NBS supplies this product through its Reagents department.

4. An immunohematologist in Michigan has the opinion that daily QC for panel cells is redundant, because the process of identifying a red cell antibody is, in and of itself, a controlled process which entails several interlocking steps. These steps include determining a pattern of reactivity, verifying a patient's autologous RBCs lack the antigen which corresponds to the antibody specificity, testing enough red cells so that there is a sufficient number of positive and negative reactions to obtain a p value of < 0.05, observing that the antibody reacts at test phases which are consistent with its specificity, and confirming that the results obtained with reagent RBCs in the antibody screen are consistent with the final antibody identification interpretation. Moreover, and perhaps most important, any RBC units selected for antiglobulin crossmatch (based on the antibody identification) are compatible. He cautions that testing each reagent RBC for loss of a particular antigen is impractical, and will not address errors in misidentifcation by failure to follow the basic principle (outline above) for the process of antibody identification.

5. A transfusion medicine technologist in Los Angeles reports that CLIA makes reference to quality control of antibody identification panel cells at 42CFR 493.1271(a)(1) in the Interpretive Guideline which can be accessed at www.cms.hhs.gov/CLIA/downloads/apcsubk2.pdf. (Please scroll down to page 27 of the pdf file to find part 1271.) According to the information provided, there are no daily quality control requirements for reagent red cell panels used in antibody identification. Panel quality control (as suggested by the colleague from Michigan in Addendum #4) is a combination of serological test results, such as: strength of reactions and patient phenotype; statistical probability, patient’s medical history; and laboratory standard of practice (i.e., how the laboratory handles compatibility testing for patients with unexpected antibodies). However, one must NOT FORGET the QC requirements for each new batch, lot, or shipment of identification systems according to 42CFR Section 493.1256(e)(1).

6. A Transfusion Service Supervisor at a hospital in Bradenton, Florida reports that based on discussions that her laboratory has had with an FDA representative, QC on antibody identification panel red cells is only done upon receipt of a new panel set or if an expired 'selected' panel cell (that expresses an unusual or special phenotype) must be used to complete a workup. When doing their routine panel QC, they test one Jk(a+b+) cell and one Jk(a-b+) cell with a 1:10 dilution of anti-Jka antisera, looking for agglutination after a 5 minute room temperature incubation. She adds that their QC protocol does not take much time nor is it burdensome.

ADDENDA Aug. 21, 2005

7. Sheryl A. Kochman, Chief, Devices Review Branch (CBER/OBRR/DBA; office phone 301-827-6123) acknowledges that in 42 CFR there are some regulations applicable to CMS/CLIA which go beyond what FDA typically expects in terms of the QC done by hospital transfusion services under 21CFR. She suggests that if any institution is faced with a citation during a CLIA inspection, they can try the following:

1. Show that you are following the QC instructions in the reagent red cells package insert.
2. Ask that the inspector call FDA (Ms. Kochman or her staff) for clarification as to what FDA expects users to do.

Please submit comments to the e-Network Forum.

Ira A. Shulman, MD
CBBS e-Network Forum Editor & Moderator