Is the Antiglobulin Crossmatch a Necessary Part of the Pretransfusion Testing? [ LHSC Logo ]

Slide #1: Title Slide

Is the Antiglobulin Crossmatch a Necessary Part of the Pretransfusion Testing?

Glen Dietz, ART, CLS
Director of Technical Operations
Laboratory Hematology - Blood Bank
Victoria Hospital, London, Ontario

Index

At Victoria Hospital we have omitted the antiglobulin phase of the crossmatch for patients with a negative antibody screening test over the past ten years. This article describes how decided to make this move and to comment on the need for the antiglobulin phase of the crossmatch.

Victoria Hospital is a 750 bed teaching hospital located in London, Ontario. It is a general hospital which provides most medical and surgical services for adults and pediatric patients. We transfuses about 12,500 units of red cells per year and about 27,000 units of other blood products. We perform about 25,000 crossmatches per year.

Goto: Index


Slide #2: Objectives

In this article, my objectives are to:

At the end of our discussion, I will also introduce you to a concept of pretransfusion testing which is being proposed by a few Blood Bankers which they are describing as a computer crossmatch.

The antiglobulin crossmatch is only one test of a battery that Blood Bankers use to identify donor units that will be safe to transfuse to a recipient. The battery of tests varies somewhat from hospital to hospital. The tests that are fashionable to be used have also changed considerably over the years. Today the tests usually include the following:

View the associated slide

Goto: Index


Slide #3: Pretransfusion Compatibility Testing Tests

Testing done on Donor Unit

Each component of the pretransfusion compatibility procedure has one or more purposes. They have a different level of importance in different situations. For example, if there are previous records on a recipient, comparing the results of the ABO & Rh testing to the previous records provides some assurance that the specimen has been drawn from the correct patient and there have been no labelling mixups. When there is no previous record, the ABO & Rh grouping results are used to select ABO compatible units to crossmatch.

The tests that make up this battery are universally accepted, and the tests which are fashionable to include change over a period of time. For example, in the mid 1960's it was common practice to perform a minor crossmatch as well as a major crossmatch. The minor crossmatch became less important when blood collection services started to screen the donors for blood group antibodies. In fact, the minor crossmatch is done so infrequently that many of our students and new technologists are not aware of what is involved in performing a minor crossmatch. We can come up with many more examples of how Blood Bankers have rationalized the need for certain components of the pretransfusion compatibility testing battery.

View the associated slide

Goto: Index


Slide #4: Is the Antiglobulin Phase of the Crossmatch Necessary

During today's telenar we will discuss the need for the antiglobulin phase of the crossmatch. There are situations where the antiglobulin phase of the crossmatch is probably not necessary and other situations where it should be used. At Victoria Hospital, we perform the antiglobulin phase of the crossmatch only when the patient has a positive antibody screening test or has a previous record of clinically significant antibody. If the antibody screening test is negative and there is no previous record of a positive antibody screening test, we perform only an immediate spin crossmatch.

The concept of omitting the antiglobulin phase of the crossmatch is not new. In fact it was proposed as early as 1964 by Groves- Rasmuessen. He suggested that the antiglobulin phase of the crossmatch added little extra value for improving patient safety when the antibody screen was negative. This view was not widely accepted at the time, but has been restated by others over the last three decades.

Lyle Unrau and I first started to think about using an abbreviated crossmatch procedure after Oberman published an article in the April 1978 edition of Transfusion. At this time the literature was filled with reports of the safety of the Group & Reserve Serum procedures. Oberman pointed out that there was little risk to a patient with a negative antibody screening test if blood was issued after an immediate spin crossmatch. He pointed out that the main value of performing an immediate spin crossmatch was to ensure the donor units are ABO compatible with the recipient. His arguments were so convincing that many others used his observations to question the need to perform the antiglobulin phase of the crossmatch at all when the antibody screen was negative.

Not everyone was convinced about the safety or wisdom of this approach. Taswell and others were convinced that the antiglobulin phase of the crossmatch was necessary.

Goto: Index


Slide # 5: Theoretical Reasons

They felt the antiglobulin phase of the crossmatch adds to patient safety, even when the antibody screening test is negative because:

View the associated slide

Goto: Index


Slide #6: 54 Month Study

At Victoria Hospital, we carried out a 54 month study of the antibodies detected in the crossmatch but not detected in the antibody screen to see if these were significant. The first 12 month of the study were retrospective and the next 42 months were prospective. This study involved 110,000 specimens tested between January 1978 and June 1983. During nine months of the study, we compared the antibodies detected when we used screening cells from three donors or two.

During the 54 month period we had found 3591 patients with blood group antibodies. In 195 cases these were detected only by the crossmatch. The next slide lists the specifities and numbers of examples of antibodies found only in the crossmatch.

There were only a few of the antibodies that we were concerned about missing. These are highlighted on the list on the next slide.

View the associated slide

Goto: Index


Slide #7: Antibodies detected

One example of anti-E was missed even though one of the screening cells were homozygous for the E antigen. When the screening was repeated it was still not reactive. The anti-c was probably a technical problem since it was detected when the test was repeated. There were 47 cases where we could not identify the specificity of the antibody. These may or may not be clinically important. There were 38 antibodies directed against low incidence antigens. About a third of these were due to anti-Wra. The remaining antibodies were not likely of any clinical significance.

Once we convinced ourselves that an abbreviated crossmatch was a scientifically sound approach, we developed a proposal to introduce the procedure. We discussed our findings with our Hematologists, developed a proposal for the Blood Transfusion Committee and introduced the procedure in July, 1983.

View the associated slide

Goto: Index


Slide #8: 10 year experience

We have used the abbreviated crossmatch procedure for almost ten years now. During this time we have transfused more that 125,000 units of red cells to more than 30,000 patients. Our protocol is to perform a blood group and antibody screening test on each specimen received for pretransfusion compatibility testing. If the antibody screening test is positive we identify the antibody, select units which are antigen negative if the antibody is clinically significant and crossmatch it using an indirect antiglobulin test. If the antibody screening test is negative, we perform only an immediate spin crossmatch. The main purpose of the immediate spin crossmatch is to ensure that we have selected units of the correct ABO group for the recipient. With this procedure, it takes between 5 and 10 minutes to perform the crossmatch portion of the pretransfusion compatibility testing.

When we have emergency situations, we modify our testing procedure only slightly. We set up the blood grouping and antibody screening test as soon as we can, then while the screening tests are incubating we start to crossmatch the required number of units. We read the crossmatches after immediate spin, then incubate the tests at 37 degrees C. By this time, the antibody screening tests are usually almost ready to read. We complete the antibody screening test. If the screen is negative, we discard the crossmatch tests. If the screen is positive, we carry the crossmatches to AHG. If necessary, we will issue crossmatch compatible units. We introduced LISS about a year ago and now find that the antibody screens are just about ready to read by the time the immediate spin crossmatch is complete and it is sometimes not even necessary to incubate the immediate spin crossmatches.

During the ten years we have used an abbreviated crossmatch procedure, we know of only few cases where the antiglobulin phase of the crossmatch might have prevented crossmatch incompatible blood being transfused. In each case there was no clinical evidence of red cell destruction. We had at least one case where a weak anti-Fya was missed on the initial screening test and six immediate spin compatible crossmatches were issued. Four of the units were Fya+. We found out about this case when we were asked to crossmatch additional units since the patient's hemoglobin fell to the point where additional transfusion was required. When the cause of the falling hemoglobin was recognized, transfusions were not given and the patient recovered without any problem.

We have found at least three cases where patients have experienced febrile reactions and when the reactions were investigated we unmasked an anti-Bg antibody. There were a few additional cases where anti-Sda or anti-P1 were found in specimens after immediate spin crossmatches were issued.

I believe the incidence of finding these types of occurrences is no more common than it was when we performed an indirect antiglobulin crossmatch on every patient who we transfused. We are satisfied that issuing blood which is compatible by the immediate spin crossmatch is safe for patients who have a negative antibody screening test. We also feel it is a practical and efficient method.

View the associated slide

Goto: Index


Slide #9: AABB Standards

Omitting the antiglobulin phase of the crossmatch was recognized as a sound practice in the early 1980's. In the tenth edition of the AABB standards (published in 1984), the standards were modified to support this practice. The second edition of the CSTM Standards published in 1986 also supported this practice.

View the associated slide

Goto: Index


Slide #10: Hamilton Study

For those of you who are not yet sold on the concept of omitting the antiglobulin phase of the crossmatch, I will describe a very well controlled prospective study which was carried out in Hamilton by Heddle et al which was reported in the British Journal of Hematology in 1992. They carried out a two year study involving 9128 patients. 8936 (97.9%) of these patients had a negative antibody screening test. 26.9% of the patients with a negative antibody screening test were transfused with blood which was compatible by immediate spin crossmatch. An antiglobulin crossmatch was performed after the blood was transfused.

View the associated slide

Goto: Index


Slide #11: More Detail on Hamilton Study

168 incompatible antiglobulin crossmatches using 130 specimens collected from 119 patients. In 103 cases (79.2% of specimens), the incompatible crossmatch results could not be reproduced when tested in duplicate. There were only 27 cases where the crossmatch was designated incompatible by the antiglobulin crossmatch. In spite of this, none of the patients had clinical or laboratory evidence of reduced red cell survival. Each of the recipients transfused during this study were tested approximately 24 and 96 hours post transfusion for evidence of hemolysis. Specimens collected pretransfusion were also investigated for hemolysis. The test performed included plasma hemoglobin, haptoglobin, haemopexinin-haem complex, and methemalbumin. Total bilirubin was measured if the direct or indirect antiglobulin test was negative on the pretransfusion sample but positive on one or more of the posttransfusion specimens.

Of the 27 incompatible crossmatches, 4 were caused by units with a positive DAT, 10 were caused by unidentified antibodies, 8 were due to anti-Bg or anti-H. Two were caused by antibodies to low incidence antigens that were not present on the screening cells (anti-Wra and anti-Cl). One example of anti-P1 was not detected because the screening cells had only a weak expression of the antigen. One example of anti-Jka was not detected even though the screening cells were homozygous for Jka. One example of anti-Fya was not detected due to a technical error.

Patient charts from the 27 patients receiving crossmatch incompatible blood were reviewed by two different physicians who were not aware of the laboratory test results. This chart review was a blind study since a number of additional chart reviews on patients who were transfused with crossmatch compatible blood were also carried out by the same physicians.

The direct antiglobulin test was positive in only 4 of the 27 cases where the incompatible crossmatch results had been obtained. Two had antibodies that could not be identified, one had anti-Jka and one had anti-Fya. In none of these cases was there any laboratory or clinical evidence of hemolysis.

Of the 103 cases where the antiglobulin crossmatch was initially positive but later shown to be non-reproducible, one developed a positive DAT 3 days post transfusion but had no laboratory or clinical evidence of red cell destruction. Two more patients had laboratory evidence of red cell destruction but no clinical evidence. One patient developed a positive DAT with anti-K in eluate but did not exhibit clinical or laboratory evidence of hemolysis.

View the first associated slide

View the second associated slide

Goto: Index


Slide #12: Conclusions

Nancy Heddle and her colleagues concluded that the antiglobulin phase of the crossmatch was not necessary for patients who have a negative antibody screening test, a conclusion which we support fully.

There are many more studies of the need for the antiglobulin phase of the crossmatch, however we do not have sufficient time to describe these. Please refer to the handout for a list of references that you may wish to read.

Lets look now at the reasons why you as a Blood Banker might want to omit the antiglobulin phase of the crossmatch. There are three main reasons for considering adopting this approach:

View the associated slide

Goto: Index


Slide #13: Why Consider

View the associated slide

Goto: Index


Slide #14: Cost Savings

The cost savings that can be expected from omitting the antiglobulin phase of the crossmatch are minimal. We find that it costs approximately $4.79 to perform an antiglobulin crossmatch and $4.25 to perform an immediate spin crossmatch. These costs reflect the reagents, disposable costs, technologist time and a distribution of fixed costs. At Victoria Hospital we perform about 25,000 crossmatches per year. Of these, 87.6% are immediate spin crossmatches and 12.4% are antiglobulin crossmatches. Therefore, our savings amount to about $17,000 per year on a budget of slightly more than $1 million dollars. This represents a savings of slightly less than 2% on our budget.

There will be some additional savings to the health care system if you are able to reduce the rate of outdating of blood products. In our case, we reduced our outdating from about 1.5% to less than 1%.

Cost savings should certainly not be the driving force for making the move to an abbreviated crossmatch procedure.

View the associated slide

Goto: Index


Slide #15: Reduced Requirement for Patient Specimens

When we moved to an immediate spin crossmatch, the volume of serum we used decreased from six drops to two drops for each crossmatch. This makes it possible to get more crossmatches from a blood specimen. It also means that we are less likely to require additional specimens from patients who receive multiple transfusion. There are probably a few dollars saved from this, but I have no way of estimating the amount that we were really saving (we now use a LISS method which reduces our serum requirements even further). The value of avoiding drawing extra blood samples from a patient is difficult to quantified.

View the associated slide

Goto: Index


Slide #16: Turn Around Time

Probably the best reason for considering abandoning the antiglobulin phase of the crossmatch is improved turnaround time for processing specimens. There are several reasons why you can expect an improved turnaround time when you use an immediate spin crossmatch. The incubation time and time required to wash the crossmatches, add antiglobulin serum and read the final results are not required. In many cases, this will improve the time required for processing a specimen. This is especially true if the screening test is performed and blood is crossmatched only when there is an order to transfuse.

It may also be possible to reduce the number of units that are crossmatched for surgical cases. An immediate spin crossmatch can be completed within 5 to 10 minutes so it may be possible to convince physicians not to hold blood for patients where there is only a slight chance of using blood. If the surgeon cannot anticipate the blood loss, fewer units can be crossmatched initially and more can be crossmatched if the first few units are transfused. If you are able to do this, you can expect significant savings from the reduced workload. At Victoria Hospital we were able to reduce our crossmatched to transfused ratio from 2.8 crossmatched for every unit transfused to less than 2.0 crossmatched per unit transfused.

Another reason that you can expect an improved turnaround time is because of the reduced number of false positive (non- reproducible) crossmatches. When an incompatible crossmatch is encountered, it is necessary to investigate the cause of the incompatibility. This may involve obtaining additional specimens from the patient, repeating the tests, performing an antibody investigation etc. This extra work not only increases the workload for the Blood Bank, it also prolongs the time takes to provide blood for the recipient.

In urgent cases, improving the turnaround time for providing compatible blood reduces the stress levels for physicians caring for the patients. It is likely that the patient will receive more appropriate care if blood is ready promptly when there is an urgent need. We feel that a better relationship has developed between the Blood Bank and the OR or ER because of our improved turnaround time. They truly appreciate having the blood available when they need it.

View the associated slide

Goto: Index


Slide #17: Computer Crossmatch

By now I trust you are all convince that omitting the antiglobulin phase of the crossmatch is not only safe, it is also good management! For those who still have sweaty palms when they think about this move, I am going to increase your stress level by asking "Is the Crossmatch a Necessary Part of Pretransfusion Testing?" An innovation that may become common practice in Blood Banks where a computerized record keeping system is in use may soon be using a Computer Crossmatch.

I know of at least two American hospitals with very active Blood Banks where this is being used routinely. John Judd from the University of Michigan Medical Centre is one of the most vocal advocates. John argues that the most important reason for performing any form of a crossmatch on a recipient with a negative antibody screening test is to ensure the donor and recipient are ABO compatible. If a computer is used for managing Blood Bank records and it has records on both the recipients and the donors, it can be programmed to check that the donor's blood group matches the recipient. John feels the safety of the computer crossmatch is at least as great as if a serological crossmatch were performed. He advocated a number of safeguards to further improve the safety of this procedure. First, there must be two or more ABO & Rh test results on file for the recipient, and the ABO group of the donor units must be retested by the hospital Blood Bank. A bar code scanner must be used to enter blood unit numbers and the computer must be used to print the blood bag labels.

This approach to pretransfusion compatibility testing was discussed at the 1992 AABB convention and seems to be gaining in support. Look for changes in the next AABB Standards to legitimize this practice.

View the associated slide

Goto: Index

Return to Programs and Services Questions? Comments? Contact David Schaus
Copyright ©1996-2003 London Health Sciences Centre. Please read our disclaimer

[ GOTO ] [ HOME ]
[ Goto Hematology-Blood Bank Home Page ]
[ Victoria Hospital Home Page ]
Last Updated: December 8, 1995
Copyright ©1995, Glen Dietz, Technical Director, Hematology-Blood Bank, Victoria Hospital