SMILLER

This is pretty much exactly what we do and neither the FDA or JCAHO inspectors have a problem with it. Scott

If I am reading these responses correctly, I would suggest that it is at least possible that there could be a weak developing allo-anitbody that would easily be obscured by a 'mere" 2 or 3 cell screening (that has positivity that matches the previous sscreen). In which case, just because a new positive screen re4activity matches the previous one, one cannot rule out other newer significant antibodies, as you may be able to do with a full antibody ID workup involving panels. Getting back to the original posts, we would attempt to have the patient phenotyped to start with. Then for transfusions, we would only give those phenotypically-matched units. As long as we keep this up, we can be reasonably certain that no new antibodies are going to sneak in (as long as we are the only facility transfusing the patient!). This is the approach we try to take for particular patients with multiple antibodies, warm autos, Daralex patients etc. Scott

I do not think that it's really that hard to get consistent results from tech to tech with tube testing. But it requires adequate training and oversight, just as you need for automated testing, in order to provide appropriate patient care. Interesting note above about platelet-poor-plasma and gel though. Scott

Reverse ABO typing, and IS crossmatch. Scott

We have had this done from time to time. Like David's facility, above, the surgery department has a procedure that leaves the blood bank out of it. Scott

Not a bad idea! There are a few other things that could be gone over as well, like why one has to do a smear review when you have a high neutrophil or RDW. I believe our Hema manager actually included questions like this on our last annual competency. They try to ask about stuff that associates are slipping on. Scott

The problem we have here from time to time, is that we will have something like an obvious iron-deficiency anemia where the MCV, MCH, MCHC are all low, as you would expect. But when certain techs (who should know better) see that H&H check fail flag, they immediately incubate the specimen at 37, thinking this will correct "something". It is a waste of time, of course. So that with many pathological conditions, the H&H "rule of three" is supposed to fail, and further manipulation to "correct" the H&H fail flag is not indicated. Scott

OK. But I would say that results for MCH and MCHC are different because they are measuring two different things. And I am not so sure that using the "rule of three" or "H&H check fail" to check results is very useful when we can just look to the indices for troubleshooting. For example: I see the MCH as the Mean Cell Hemoglobin, that is: the total amount of hemoglobin in the cell (NOT the concentration of Hgb). So for high MCVs, the MCH will tend to be high, and for low MCVs, the MCH will tend to be low. The MCHC, on the other hand, is the Mean Cell Hemoglobin Concentration. Cell size, by itself, does not matter here. However, consider an iron-deficiency anemia where not only are the cells small (low MCV and MCH) but the concentration of Hgb is also low, which will result in a low MCHC. However, the concentration of Hgb can only be so high (no more than 36 or 37), as it is physically impossible to have a higher density of Hgb beyond that. (Again, the MCV or MHC is not the issue here.) This is why when one has a very high MCHC you have to try to resolve it. Too much lipemia skews hemoglobinometer results upward, resulting in a higher Hgb and no change to the RBC or MCV. Cold agglutinins skew the RBC count downward, with no change in Hgb. In either case, if you do the calculations for MCHC, you get a impossibly high result. Scott

Generally just CSFs for RBC counts. But a physician here can order a "miscellaneous body fluid cell count" that would include RBC, WBC and diff (no reference ranges, and since the counts are done on a hemacytometer, we can go down to zero.) Scott

What? You don't already have one? Just kidding... CONGRATS! IT MAKES US ALL HERE VERY HAPPY! WE KNOW THAT THEY APPRECIATE YOU OVER THERE! (here too!) Scott

We use them. No plans to drop them. There are just too many problems relying on the chart-label armbands for patient ID. Scott

Just one more note. While the MCH represents the total Hgb in RBCs, the MCHC reflects the concentration of Hgb in the RBCs. So for hypochromic RBC patients, the MCHC will be low. But it is physically impossible to have a concentration much above 37 for the MCHC. That is why anything higher than that must be resolved. Scott

Ah yes! Al Chemy! An old school chum of mine. He still owes me $20... Scott

Oh! I had that backwards! (should have re-read the posts from the beginning!) What I meant was suppose that some of the Ca++ decayed into titanium-40, thereby releasing the EDTA to further chelate excess Ca++ ? Scott

Does old EDTA in solution break down after a while, releasing Ca++ back into the solution? Somebody must have written a paper on this! Scott

For interference with the hemoglobinometer due to lipemia, we would do a saline replacement. If due to strong cold agglutinins, we would warm for 10 minutes or so. For really really string cold agglutinins that cannot be resolved, we would blank out those parameters that are affected. Scott

Sure why not. The difference between 6.0, 7.0 and 7.1 is only a variance of about 4% so as far as precision, the three results are pretty much the same clinically. Practically, that is a really low Hgb either way. However, there is the question of the critical (or near-critical) value to report. My personal preference would be to report (and call) the lower critical value just to make sure someone knows about it right away. Scott

Again, the acceptable error rate, regardless of where you work or what published literature you can or can not find, has to be zero. Your quality management project just needs to show how you improve on whatever your baseline rate is. The article I cited above does indeed involve entry error rates. ( But I would be surprised if you can get someone at your facility to agree that your error rate is OK because you found a paper that gives an "acceptable' rate that is worse than yours! ) Scott

But...but...but... polyspecific reagent DOES contain anti-C3... (And I do not think that Ms. Fabian would suggest ignoring any manufacturer's recommendations contrary to her quote above.) Scott

We use a point-of-care ACT analyzer to monitor these patients. Similar to what we use in the cardiac cath lab. Scott

I am thinking now, that if you want to know where the US healthcare system is going, you would want to look at whatever Russia is doing... Scott

...or false negatives, which would be even worse. You would need to have something from Diamed to allow the use of non-standard reagents, diluents, etc. Scott

Regardless of a particular regulatory requirement concerning DAT QC, I believe that one must at the least follow manufacturer's recommendations for a particular system. For our Ortho BioClone polyspecific, the procedure requires both types of sensitized cells for positive controls. This makes sense to me. And besides showing that the cell-washer is working properly, one has to prove that the reagent can give a positive with both compliment- and IgG-sensitized cells. Scott

With a positive poly on a DAT, we only run the anti-IgG here. But we send the specimen out to a nearby lab for the Compliment. We do not do the anti-compliment as we do not want to pay for the QC material that we would have to buy (and mostly waste when it outdates). Scott

You run antibody ID panels, just as you would for any positive screen, and ID what is there. Scott