Tag: Immunohematology

Immunohematology testing and processes

Opinion: Handling Incorrect Physician Orders

drzeyd

Most of the non-transfusion-medicine physicians with whom I have worked have had only limited training in placing component or hospital blood bank orders.  In previous posts I have discussed this and suggested that all physicians who may possibly order blood bank tests or blood components should have a training and documented competence on a periodic basis.  Only a very few physicians, mainly hematologists and some organ transplant physicians, have placed reliable orders.

Before we had a blood bank computer system, we received orders on a manual paper requisition.  If there were errors, my technical staff and I corrected the order.  Any changes were made by me in my capacity as the blood bank medical director.  The ordering physicians and I had good relations and they had no problem with this—in fact, many were afraid of the blood bank and felt happy to be relieved of the responsibility.

In the current software era, what happens depends on how the order is placed.  Can the technical staff and I correct the order directly or must we each and every time contact the physician to revise his/her order?  Do my staff have to cancel each erroneous order and wait for the corrected order?  This could slow down the work process and prevent release of blood components in a timely manner.

Should the non-blood bank physician be allowed to order our complicated esoteric transfusion tests directly—e.g. ordering extended antigen typings or antibody identifications or elutions?   What if they order something unnecessary or inappropriate?

At a recent hospital position as Division Head of Transfusion Medicine, I discovered that the Hospital Information System HIS was very slow for all blood bank orders.  The physicians complained, I would allow them to bypass the HIS and use the manual backup system in critical situations.  If we had been forced to cancel the order and have the physician reorder, this would have greatly disrupted the work process.

We did not use the HIS at all in Transfusion Medicine.  We had patient and donor modules in the dedicated blood bank system Medinfo.  We had a limited ordering interface from the HIS to Medinfo for blood components and some basic testing (e.g. ABO/D typing, DAT, type and screen/group and save, cord blood testing, and transfusion reaction workups).  All work was done in the dedicated blood bank system.

All components were leukodepleted to < 1E6, all platelet and plasma were Mirasol pathogen-inactivated, all platelets were in platelet additive solution PAS.

The doctors did not order the specific blood component, rather they indicated a preference, e.g.

  • Packed RBCs
  • Platelets (adult dose of 2.4 E11)
  • Plasma
  • Cryoprecipitate

The following appeared as an order comment in the blood bank system:

  • The specific number or amount of the each component type
  • Preferences for pooled vs. apheresis platelets, washed RBCs, irradiated RBCs.

The blood bank staff could review the doctor’s request and order in Medinfo as per our internal protocols under my responsibility.  In effect, I was modifying the orders when needed just as I had done under our manual system.  This included type, modification, and quantity.  Yet now, I did not have to change any orders since the doctors had only indicated preferences.

For blood bank testing orders, we had our own internal algorithms for the workups.  The doctors could not order antibody identifications, elutions, or any antigen typings other than ABO/D.  They latter were triggered by the screening test results.

This system allowed us to avoid cancellations due to physician errors.  I was very comfortable taking the responsibility to alter the orders for best patient care.  If a physician did not agree with our algorithms, they could discuss the issue with one of the transfusion physicians, or ultimately appeal to me.  I was fortunate that there were no legal issues with this approach where I practiced.

In summary, my recommendations are:

  1. Have the doctors order a preference for the type of blood component.  They still order the quantity and can request modifications such as washing or irradiating or apheresis-derived or buffy coat pool platelets.  The actual allocation is made by blood bank staff under my direction.
  2. Offer algorithmic based testing.  Doctors only order basic testing which triggers reflex algorithms.  Most of the test menu is not orderable directly by the physicians (example:  an antibody identification is triggered by a non-negative antibody screen.

9/12/20

Anti-CD47 Interference

drzeyd

Background:

Anti-CD47 (Hu5F9-G4)interferes with all phases of pretransfusion testing, including ABO reverse typing.

Anti-CD47 (Hu5F9-G4) is a human monoclonal IgG4 antibody used to treat hematologic or solid malignancies.  CD47 is a glycoprotein expressed on all cells. 

All plasma samples reacted 3-4+ in all phases with RBCs, including immediate spin.  Stronger reactivity may occur in D-negative RBCs.  Plasma reacted with DTT, trypsin, papain, alpha-chymotrypsin or WARM (warm autoantibody removal medium—Immucor). 

DAT and autocontrol are negative or weak whereas eluates are 3+ reactive.  Reactivity is removed by multiple alloadsorptions with papain-treated cells or pooled platelets.  PEG absorption is invalid due to precipitation of the antibody.  Gamma heavy-chain specific AHG reagent (Gamma-clone IgG—Immucor) does not detect IgG4 antibodies.

Process:

  1. Always check medication history for anti-CD47 treatment.
  2. If positive, use gamma heavy-chain specific AHG reagent (Gamma-clone IgG) for IAT.

The following table summarizes anti-CD47 reactivity and compares it to anti-CD38—see reference below.

Reference:

Randall Velliquette et al, Anti-CD47 Interference in Red Cell and Platelet Testing, Transfusion 2019; 59, 730-737

Liver Transplant Processes for the Transfusion Service

drzeyd

Principle:

Liver transplantation requires coordination of the Transfusion Service TS, Liver Transplant LTS , and the Blood Donor Center BDC to prepare blood components for this massive transfusion setting.

Policy:

  1. Whoever receives notification of a possible liver transplant should ensure that ALL of the following senior staff are also informed:
    1. Head, Transfusion Medicine
    2. Coordinator, Donor Recruitment
    3. Supervisor, Transfusion Service
  2. Transfusion Service Supervisor will arrange transfusion service staffing as needed to cover the surgery period and immediate post-operative period.
  3. The current, up-to-date inventory must be calculated.  The following minimum stock must be reserved:
    1. Twenty (20) packed RBCs
    2. Twenty (20) FFPs—-leukodepleted and pathogen-inactivated
    3. Three (3) platelet pools or apheresis doses (each ≥ 2E11 absolute number of platelets)—leukodepleted and pathogen-inactivated.
    4. The Medinfo Liver Transplant Protocol automatically orders all the above.
  4. Complex needs:
    1. If the patient has clinically significant antibodies, confirm the availability of the requested number of antigen-negative/compatible units.
    2. The Head, Transfusion Medicine, or the covering transfusion medicine consultant on-call will contact the liver transplant team to discuss the feasibility of proceeding after assessing the inventory of antigen-negative units.
    3. Only the transfusion medicine consultant can approve the use of antigen-incompatible RBCs in conjunction with the liver transplant team.
  5. Donor Center should increase recruitment/production to meet the anticipated usage as needed.
  6. When the procedure is confirmed:
    1. Crossmatch the RBCs by the appropriate technique according to our algorithms (i.e. computer/electronic, immediate-spin, or full AHG).
    2. Thaw the FFP:  the FFP is valid for FIVE days after thawing for use for the transplant patient or other patients with coagulopathy.
    3. If more than the above number of units is needed, HGH Transfusion Service must inform the Donor Center senior staff (Head, Donor Center; Administrative/Technical Director, and Coordinator, Donor Recruitment) to arrange for additional donations.

Enzyme Panel Details

drzeyd

Principle:

Performing antibody panels using both enzyme (ficin, bromelin, and/or papain)-treated  and routine panel cells may be necessary to detect most clinically significant antibodies

Policy:

  1. Regular (LISS) panels are to be performed using AHG reagents whereas enzyme panels done by the gel technology must use the saline (NaCl)-enzyme card.
  2. Perform BOTH enzyme and routine panels in the following situations
    1. All patients with sickle cell anemia and thalassemia
    2. Antibody pattern of panreactivity with a negative autocontrol (see attached examples)
    3. Antibody workup that does not show a specific pattern with the regular panel alone
    4. Any case with a previous history of antibodies with a current negative antibody screen
    5. Any other case that you are directed to do so by the transfusion medicine consultant, supervisor, or senior technologist.
  3. Remember the reactivities of the following antibody specificities with enzyme-treated cells:
    1. Reactions may not be the same with papain vs ficin-treated cells!
    2. Enzyme-labile with both papain- and ficin-treated cells:  Fya, Fyb, M, N, Ge2, Yta, Rg, Ch, Pr, Tn, Mg, Mia, Cla, Jea, Nya, JMH, Inb
    3. Variable, enzyme-labile or weakened, some unchanged with both papain and ficin-treated cells:  S, s, U
    4. Variable reactions with papain (labile, weakened, unchanged, or increased) but usually increased or unchanged with ficin-treated cells:  Kell system (K, k, Kpa, Kpb)
    5. Reactions are increased or unchanged with both papain and ficin-treated cells:  Rh (D, C, c, E, e), Jka, Jkb, Lea, Leb, Lua, Lub, P1, H, most cold antibodies, autoantibodies, Tja (PP1Pk)

Example 1:

Antibody to High-Incidence/Prevalence or Public Antigen:

Reactions destroyed by enzyme (typical of anti-Ge2):

Example 2:

Antibody to high-incidence antigen, reactions unchanged or enhanced by enzyme—VERY DANGEROUS PATTERN:  Examples:  Anti-H, Anti-Tja (PP1Pk), anti-k (cellano), anti-U

Enzyme Reagent Effects

drzeyd

The following table was distributed at an AABB technical seminar at its 2015 Annual Convention and comes from the Blood Center of Wisconsin.  It is especially noteworthy for discussing enzymes that are mainly used in a reference laboratory setting.

Opinion: Transfusion Education for Physicians

drzeyd

Although there are resources allocated for nursing transfusion proficiencies, the training for physicians and medical students may not be as developed.

In medical school in the United States, I had one hour dedicated to transfusion medicine and it did not discuss the actual proficiencies necessary to select and physically transfuse patients.  In my clinical rotations, especially surgery, I was given “cookbook formulae” on what to order for different procedures—without any explanation.

I had considerable training in transfusion medicine during my pathology residency, and I took it seriously.  Many/most pathology residents are not interested in the topic and use the rotation for vacations and other business.  They learn what is necessary to pass the pathology board examinations and do not intend to practice this at all.  Actually, I sense that many practicing pathologists are afraid of blood bank and minimize their activities there.

What about the non-pathology residency training?  In the Middle East, some centers rotate the hematology, surgical, and obstetrics-gynecology residents through the blood bank.  In other posts, I have discussed some training plans I have used for each group.

Sadly, however, most practicing physicians do not well understand how to select and order blood components or administer components.  I try to tell them that they should at least know if the blood components being used for their family members are correct.

In a previous position, there were mandatory training programs for physicians including infection control, fire,  and disaster training.  There was even a program on handwashing!  Yet, there was no required training for transfusion practices.  I had offered to make such a training program but there was little interest.

Minimally, such a proficiency program should be offered to all physicians who might order a transfusion.  It should include:

  • Indications for transfusion of each component type including the institution’s transfusion triggers
  • Indications for modified components, e.g. irradiated, aliquoted, etc.
  • How to order blood components—manually and in the institution’s computer system
  • How to administer blood components, including the maximum transfusion times (e.g. four hours for RBC components), use of filters, etc.
  • Recognition and treatment of transfusion reactions and other adverse effects

In my career, I received blood component orders with many errors.  In the pre-computer, paper era, the blood bank staff automatically corrected these under the direction of the blood bank medical director.  Nowadays, with computer systems, it may not be possible for the blood bank staff to “autocorrect” the orders.

In my Middle Eastern practice, we could still correct the orders—even after adopting a blood bank computer system.  That will be the subject of a future post.

Release of COVID-19 Convalescent Plasma CCP

drzeyd

Principle:

Testing, allocation, release, and transfusion of this special plasma will be similar to normal blood components except that the storage, modification, and release will handled at special quarantine location(s).  Release and modification of components is to be handled separately from the regular hospital blood bank/transfusion service.

Abbreviations:

CCP:  COVID-19 convalescent plasma

ID:  Infectious Disease Department (clinical COVID19 treatment unit)

QHBBB:  Quarantine hospital blood bank (which stores, thaws, and releases CCP)

Policy:

  1. Any potential recipient for CCP must have a CURRENT type and screen test (< 72 hours old.)
    1. A repeat type and screen should be ordered every 3 days during the time that CCP may be used.
  2. All CCP orders must be submitted to the ID senior consultant for triage.
  3. If the order is accepted by that authority, then an order and a doctor’s prescription for CCP must be both prepared and submitted to the QHBB.
  4. QHBB will allocate ABO-compatible CCP units and attached blood bank computer-generated release labels and forms to it.
  5. Designated ward staff will pick-up the CCP at the QHBB.
  6. Allow 1 hour for the plasma to be thawed.
  7. Only order when the transfusion is scheduled—1 hour before the intended transfusion date/time.
    1. Sign-out will follow normal procedures, same as other blood components.
    2. Ward staff must provide a doctor’s prescription for CCP and a fully completed request.
    3. Ward staff will directly transport the CCP to the intended transfusion site.  They should immediately transport it to the patient transfusion site.
  8. Transfusion will be performed according to standard procedures:
    1. Positively identify the patient.
    2. Use a standard 180 micron blood filter.
    3. Transfusion should be done as quickly as possible to avoid potency loss
    4. Fill out the transfusion record and return a copy to the QHBB.

Example of Dealing with HIS Interface Issues to Dedicated Blood Bank Software

drzeyd

The following is an actual working document for interactions between the Medinfo HIIG system and a monolithic hospital information system covering nursing, laboratory, ADT, etc.  For the purpose of this post, I shall name it B*.

In a previous position, I was in charge of the Laboratory Information Systems and worked both with Medinfo for donor and patient hospital blood bank processes AND B* for the general laboratory.  It was my decision NOT to use B* for the hospital blood bank because there would be no integration between the donor module in Medinfo and patient blood bank module in B*.  Also, B* had limited handling of complex immunology algorithms and fewer safeguards than Medinfo.

B* could not directly read ISBT product labels from the ISBT dictionary and required hard-coding the links for each and every type of blood component and modification.  It was also slow to order blood components so in emergency situations, I allowed physicians to bypass B* and order directly if they felt in their clinical judgment that the delay in using B* might harm the patient.

The non-transfusion physicians had to directly order blood components in B* except for emergencies as stated above.  They were only allowed to order a limited number of basic tests such as ABO/D type, direct antiglobulin test DAT, or antibody screen.  Based on those tests, we would use algorithms in Medinfo’s patient module.  Example:  an outside doctor could not order a Jka typing—that could only be done according to our internal Medinfo algorithms.  Doctors could only order antibody screens, not antibody identifications, elutions, or titrations.

Sadly, most physicians had no specific training in blood component or blood component therapy so they made many mistakes in ordering manually before the computer system.  I had requested offering training sessions for the doctors but this had never been approved by the medical administration.

Thus, my decision was to place the responsibility for the correct ordering of blood components and testing directly by Transfusion Medicine.  The non-transfusion doctors were only ordering preferences for components—the actual selection was made in Medinfo by my blood bank staff under my order.  We decided whether to irradiate.  All platelet and plasma components were pathogen-inactivated.  All components were leukodepleted to < 1E6 according to CE Standards.

In Medinfo we could thus modify or even cancel requests without having to deal with the B* system.  B* would accept Medinfo cancellations directly.

During the frequent B* downtimes, all processes would be limited to using Medinfo.  There was no way to initiate orders in Medinfo and send them back to B*.  Thus during downtimes, the only way to retrieve results was to use Medinfo.

The following is the process for the interface as was used during my tenure at that institution:

Medinfo-B* Interface Process

Principle:

Limited ordering of components and basic transfusion testing may be initiated on the B* side.  All component orders and all test results will pass into B*, including those which can only be ordered by blood bank staff.  All blood bank processes will continue to be performed within Medinfo.  Transfusion Medicine is not responsible for training physicians and nurses on how to use the B* interface.

Abbreviations:

NTMP:  Non-Transfusion Medicine physicians

TM:  Transfusion Medicine

B*:  A hospital information system including laboratory module (not used by Transfusion Medicine for either patient blood bank or donor issues)—NOT MEDINFO!!

THE FOLLOWING POLICY SUPERCEDES ANY AND ALL PROCESSES CURRENTLY DOCUMENTED IN TRANSFUSION MEDICINE.  The processes are currently being updated to reflect the Medinfo-B* interface issues.

Policy:

  1. For our hospitals, B* ordering must be used unless B* is non-functional or there is a life-threatening clinical emergency that cannot be met expeditiously by ordering in B* e.g. MTP)
  2. NTMP must place all component orders (RBC, plasma, and platelet) and the limited test menu (type and screen, transfusion reaction, ABO/D type, direct antiglobulin test, cord blood) in B*.
    1. The ordering physician must DIRECTLY enter the order, not list the transfusion as a nursing task in the B* system.
  3. NTMP may indicate a preference to the type of component and number/amount requested, but the actual selection and release will be based on internal TM algorithms under the order of the Division Head, Transfusion Medicine.
  4. Tests not listed in the B* menu cannot be directly ordered by the NTMP.
  5. It is the responsibility of TM staff to periodically check the interface from Medinfo to import requests.
  6. Transfusion service/hospital blood bank staff will integrate those order requests which are currently needed for patient care.
    1. Other requests will be kept in the B*-Medinfo queue until they are needed and will be automatically cancelled after 3 days.
  7. Internal Hospital Transfusion Service/Blood Bank Processes:
    1. All Transfusion Medicine work processes will be performed within Medinfo, nothing within B*.
    2. B* Functioning (non-MTP):
    3. Transfusion Medicine staff will accept signed specimens without requisition.
      1. The specimen bar code will provide the two unique identifiers for patient identification.
        1. We will accept even if name is truncated on the B*-generated label, relying on the full name and HC number visible on the screen
      2. Internal processes and algorithms in Medinfo are to be used for selection and reservation of components, component modification, and all testing.
    4. Documentation of work:
      1. Routine specimens (type and screen, ABO/D typings, DAT, negative cord bloods) will be paperless.
      2. Non-routine specimens require paper documentation.  These include:
        1. Abnormal Results that needs supervisor /TRM physician review
        2. If the analyzer is not interfaced to Medinfo
        3. If a manual tube technique is performed
        4. Requests received from any site not ordering through the Medinfo-B* interface
    5. Massive transfusion protocols:
      1. The ordering physician will decide whether to order in B*:  if he decides that ordering in B* will adversely affect the patient outcome, he may revert to the old system (e.g. call the blood bank hotline and request blood and then send paper requisitions and samples to blood bank as conditions permit)
      2. Physicians who use the B* interface must order each wave of the MTP separately (e.g. MTP1, MTP2, MTP3) and adjust the quantities accordingly.
        1. The B*-Medinfo interface does not include ordering for non-blood-bank items (e.g. medications) in the massive transfusion protocol.
  8. B* Results and Statuses:
    1. All results and statuses of tests and components will be available in B* for those sites using the Medinfo-B* interface.
      1. This includes non-B* orderables for components and tests.
    2. Formatting of tests and other requests are limited to the B* build’s capabilities:
      1. NTMP may see test results in the Medinfo Viewer if they prefer.
    3. During B* downtime, TM will revert to Medinfo-only processes including ordering, test-requesting, resulting, and release of components.
      1. There will be NO recovery post-down-time in B*:   test results will be viewable in the Medinfo Viewer only.  There will be no component information in the Medinfo Viewer.
  9. Downtimes and recovery:
    1. B* Downtimes:
      1. Revert to 100% Medinfo processes.
      2. Manual requisitions and specimens will be sent:  complete concordance between specimen tube and requisition is required for the specimen to be used for the purpose of selecting components.
      3. Medinfo accession numbers will be used.
      4. There will be no B* ordering recovery.
      5. Test results will be viewable in Medinfo Viewer only.
      6. No statuses will be available.
    2. Medinfo Downtimes (B* available):
      1. Paper downtime procedures will be in effect.
      2. After restoration:
        1. Retrieve orders from B* interface and proceed OR:
        2. Others use paper requisitions and order/process in Medinfo only.
  10. Training of Clinical Staff:
    1. Training of physicians and nurses to use the B* interface is the responsibility of the Hospital Information System HIS staff, NOT TRANSFUSION MEDICINE!!
    2. All questions for clinical usage should be referred to the hospital’s HIS Help Desk.
    3. Transfusion Medicine will NOT provide B* support!!!

References:

  1. Medinfo interface documentation for B* Integration, January, 2019
  2. Standards for Blood Banks and Transfusion Services, Current Edition, AABB, Bethesda, MD, USA

Antigen Typings in the Presence of a Strongly Positive DAT

drzeyd

Principle:

Antigen typing of cells with large amounts of coating antibody (i.e. strongly positive DAT 3-4+) may not always be possible because the bound antibody may block available antigen sites.  This policy is to clarify how to recognize and handle such situations.

Policy:

  1. Always follow the manufacturer’s instructions for the use of the typing reagent.
    1. In particular, note whether a control must be run with the test (e.g. D-control, D-diluent, etc.) or if it is included in the gel or glass bead card.
      1. If a control is required, use exactly what the manufacturer recommends.
      2. DO NOT SUBSTITUTE ANYTHING ELSE AS THE CONTROL!!
  2. Interpret the reactions exactly as the manufacturer indicates.
  3. If the test is invalid because of the control or any other reason, report the antigen typing as indeterminate and send for Transfusion Medicine Physician review.
  4. If the DAT is 3-4+ and the antigen typing shows no reaction (apparent negative), send the case to the Transfusion Medicine Physician for review and final interpretation.  DO NOT ENTER THE RESULT AS NEGATIVE UNLESS THE TMP INSTRUCTS YOU TO DO THIS!!
  5. To rule out a blocking antibody, a special elution to gently remove the coating antibody may be needed so that the RBCs can then be typed (not acid glycine technique—rather, gentle heat elution.)  The Transfusion Medicine Physician will decide whether to do this additional testing.

References:

  1. Standards for Blood Banks and Transfusion Services, Current Edition, AABB, Bethesda, MD, USA
  2. Technical Manual, Current Edition, AABB, Bethesda, MD, USA
  3. Guidelines to the Preparation, Use, and Quality Assurance of Blood Components, European Committee (Partial Agreement) on Blood Transfusion (CD-P-TS), Current Edition

Opinion: Software for Massive Transfusion Protocols–Pools

drzeyd

Massive transfusion protocols may contain large numbers of different blood components all to be released simultaneously as quickly as possible:

  • Packed RBCs
  • Low-titer group O whole blood
  • Platelets
  • Plasma
  • Cryoprecipitate

For an adult, it might include 6 RBCs, 1 adult platelet dose, 6 plasma, and 10 cryoprecipitate—23 components at one time.

For liver transplant, I had a protocol with 20 RBCs, 3 adult platelet doses, and 20 plasma units—43 components in all.

Even if the blood bank software prints all of these release forms, it requires considerable time to check the identity and serial number of each unit.  This is the rate-limiting step for their release to the critically ill patient.

Our clinicians wanted release within 5 minutes of ordering;  however, it was physically impossible to sign out so many components within this time limit.

Medinfo has facilitated the release by offering pools of one component type (platelets, cryoprecipitate, or plasma).  A pool number is generated to cover the components and this ONE number is used for release.

I would like to see the pooling concept expanded to allow multiple component types to be included in the pool.  Then, at release from the blood bank, only ONE number can be used to sign out the components.

Additionally, the individual units in this mixed pool should be treated the same way as if the units had been released individually including:

  • Apply all protocols for the components.
  • Return individual units back into stock or discard.
  • Use the mixed pool number to view the contents of the pool.
  • Query using the mixed-pool number for its contents.
  • Query the individual unit(s) (e.g. for look-back).
  • Quarantine the unit(s) as needed.

29/11/20