Screening for AG neg units versus ordering

[kirkaw]

Greetings! We are a medium sized community hospital but get quite a few patients with antibodies. We keep all the usual anitsera, except for anti-Lea , anti-Leb and anti-s.

We try to find anitgen negative units for patients with antibodies when reasonable. Do other medium to small hospitals have a 'cut-off' for screening either in volume or percentages. In other words, if you have a patient with anti-e and anti-C, would you screen for that or automatically order antigen negative units from your blood supplier. How about for anti-S, anti-E and anti-K?

I was wondering if you have a standard that when you have screened 'X' number of units, do you then order? If so, what does 'X' equal? Or do you say that if % compatible is less than 'X', you order units? Again, what would 'X' equal?

Thanks, Amelia

[David Saikin]

I am a small hospital with a decent inventory (overstock for my area). I will screen within reason. I usually have 20 O+/A+: I will screen for anything if it is a single ab (like Jk/Fy/S/s). If thiese 70+% ags are in combination with anything else I have my blood supplier screen. The Rh= units are fewer (10 O=/4A=). I might screen these but only for myself. If my local hospitals need Rh= screened for the higher incident ags I have them contact the Red Cross. Multiple ags of lesser frequency I will screen for both myself and my neighbors. If possible I will use the pt specimen (my own or referred) for a primary screen and type the negative units with reagent antisera. If my referral hospitals want me to screen units for them without a pt specimen to do the initial screen I tell them I will bill them for all the units I have to screen . . . reagent antisera prices are exhorbitant so I leave it up to them if they want to pay me or the blood supplier.

[kirkaw]

Thanks David. Your process sounds similar to mine. I actually keep a pretty large inventory of O+ and A+. 50-60 of the former and 35-40 of the latter. (I think we could stand to keep a lower minimum inventory, but my coworkers and boss disagree). I like the idea of using a patient sample to screen, especially for expensive antisera. We usually get full 7 ml tubes so typically have plenty of plasma.

I don't know if my coworkers' reticence to screen is because they don't feel like they have the time or they feel that the antisera is too expensive to 'waste' on screening lots of units.

[David Saikin]

If you use gel you can screen your units with 25uL of antisera (after you validate of course). I have found that the Lewis antisera did not work in gel and different manufacturer's sometimes did and sometimes did not work well. Most work great in gel.

[Eagle Eye]

Same like David. We screen using gel (duffy, kid, S,s). Use Gel Rht yping card & automation to screen.

Any request involving e- unit, do not screen.

C-, E-, K- & S- : screen in house.

C, E, K, one duffy & one kid antigen: we screen all our Rh neg. then is we can not find then order. The charges per antigen are so high from our supplier that we try our best to screen in house.

[Hooper]

We are still doing tube antigen typing. I am curious as what you did for validation for making the change to antigen testing in the gel card. Do you use the IgG gel card for all antigen typings? Do you use the buffered gel card for the RT? How many did you test (due to the expense of the anti-sera)? Thank you.

[David Saikin]

If I have a coombs' reacting antisera I use the IgG card; the buffered gel card is used for all the others; incubation temp is determined by the pkg insert (some ahg incubate at RT); 50uL of 0.8% cells/25uL of antisera; incubate 15 minutes(deends on the pkg insert)/spin for 10 minutes . I ran negative, homozygous and heterozygous cells. You will have to determine how many you need to run for your own validation. I only ran a few. If I purchase a different manufacturer I have to validate that it works in gel - some do and some don't.

[mollyredone]

If you use gel you can screen your units with 25uL of antisera (after you validate of course). I have found that the Lewis antisera did not work in gel and different manufacturer's sometimes did and sometimes did not work well. Most work great in gel.

David, how many samples did you use to validate using antisera in gel? Do you use heterozygous cells? I think it's a great idea! We have used patient samples to screen units preliminarily as well.

We screen for antigens if the probability is >50%, otherwise we will order ag negative units from ARC.

[David Saikin]

[David, how many samples did you use to validate using antisera in gel? Do you use heterozygous cells? I think it's a great idea! We have used patient samples to screen units preliminarily as well.

That is something you have to determine yourself - as I said above, I did not run a lot of cells.

It will be better if you use a different manufacturer's cells than Ortho. I have found the Ortho cells are very sensitive (to many things). I purchase 3 panels only one of which is Ortho's.

Edited August 17, 2012 by David Saikin

[mollyredone]

David, guess I posted at the same time you did! thanks for the info.

Mari

[OxyApos]

I have a local ARC with a reference lab so if there are multiple antibodies I let them do it. I found out that our center pulls known e and c negative units and keeps them for the reference lab, so the chances of me finding one from my inventory , even if >25 units, is slim. CEK and single others I try to find, if we have the tech time.

[Eagle Eye]

we used 20 donors (pos & negative) when we validated ours. (with each anti-sera). We tested 40 for kell as we had discrepancy between tube & Gel. (it was due to positive DAT on donor by gel). When we switched to Ortho anti-Jka & anti-Jkb we validated again as immucor is AHG (IgG gel card & 37C) & Ortho is RT (Buffered gel card & RT). In that case we had several patient with anti-Jka so we saved segments at the time of screening units for patient. That way we saved on regent (there was no need for repeating tube).

[Mabel Adams]

You can calculate how many units in your inventory are likely to be neg. If you keep only 40 O+ units and have an anti-e, you would need 100 units to find 2 e neg--assuming your supplier did not hold out the e neg units before shipping inventory. The donor centers can refer to historic records of donors and select those known to be negative and only have to confirm on this donation--unless they can't find any historic types on current donors--then they have to type a bunch just like we do. You could also calculate how much it costs you to type your own vs. how much the donor center charges you. It's always hard to know how to account for staff time since you are paying them to be there anyway (unless they have to stay overtime to do the testing).

[pbaker]

We have a great deal with our blood center regarding antigen negative units. We request historically typed units and then we only have to retype what they send. If we send them a list of our inventory, they can look to see if we have any on our shelf and, if we do, we don't have to wait for them to be shipped. This reduces greatly the amount of reagent and time used versus screening.

If we are looking for something simple like K, then we will screen our own inventory. I did have a really bad day once where 6 out of 7 of the O= I typed were K+. And my QC worked just fine, so that wasn't the problem. Just really bad luck!!!!!!!

[Dr. Pepper]

We do not have a hard and fast rule, but rely on what you can reasonably expect to find within our existing inventory (40 or so A+ and O+, etc) when calculating the chances of finding negative units as Mabel suggests. Our blood center charges $70/antigen, even when just using historical typing results, so it behooves us to try and find the units if we can ourselves. We sometimes use patient sera to screen first, then verify negatives with commercial antisera.

[tbostock]

We screen units for C,c,E,e, and K on the Tango. We do Jka, Jkb, and Fya in tube testing. All other antibodies, we order screened units from the Ref Lab.

[L106]

We do not have a hard and fast rule, but rely on what you can reasonably expect to find within our existing inventory (40 or so A+ and O+, etc) when calculating the chances of finding negative units as Mabel suggests. Our blood center charges $70/antigen, even when just using historical typing results, so it behooves us to try and find the units if we can ourselves. We sometimes use patient sera to screen first, then verify negatives with commercial antisera.

We pretty much follow what Dr. Pepper has outlined.

We also take into consideration two other factors:

1) How urgent is the patient's need for the blood? (Our blood supplier is 2 hours away.)

2) How busy is the Blood Bank tech? (Do they have time to do a lot of antigen typings?)