nziegler

we just renewed our contract with Beckman and replaced three overworked DxH's with brand new ones. our setup is all three connected in a line with the SMS at the end (and by the end of the year connected to the Power Express). Honestly, the only other system I would even consider switching to (based on comments read on numerous other forums) is the Sysmex XN. You have a tough choice there. If your facility is anything like mine, the finance department will make the final call. You'll have a choice of creating a connected line with both Sysmex and Beckman. Scott mentioned the connection doesn't work well for his lab - and I find that true for the other labs in our health system. For my lab, it works great. The SMS has pros and cons. LOTS of maintenance. We run a Wright-Giemsa stain, and the lines that handle the stain-buffer mix have developed what I have diagnosed as "coronary artery disease". That mix gets swampy and gross and clogs things up. They finally made a modification that makes flushing with methanol easier, but it is time consuming (45 minutes). BUT the slides are awesome! In my opinion, if you're looking for a system with a slidemaker stainer, I would lean toward Sysmex.

I always have 2 stainers in operation - Beckmans SMS, and then a benchtop backup. The question has come up regarding whether or not these need to be correlated on a 6 month basis. Because I live in NYS, and while there currently is no regulation stating we need to do cross-over studies or correlation when we switch lot numbers or manufacturers of toilet paper, I'm sure it's coming. So - does anyone else out there with more than 1 stainer perform 6 month correlations between them? We are documenting stain quality on a daily basis.

Gastroccult: QC is built into the card, just like Hemoccult, so external QC should not be required. Hansel stain: we also make smears with high eosinophil counts and the techs must indicate on the log that the nucleus stained blue and the granules stained pink/orange Giemsa: we also judge quality as our QC, the techs check off that they checked stain quality and there is a statement at the bottom of the log indicating what "quality" means PFA-100: as per the manufacturer, you must identify "normal donors" from your staff to be used as a monthly control HIT: can't help you there. We will be using IL's reagent which has commercial controls available

I bought "Blood Cell Morphology Grading Guide" by Gene Gulati - an excellent reference for bench techs. One of anything seen in an entire slide is not significant. All references give a percentage of cells with the specific morphological feature as a guide for grading. I always tell people, you shouldn't have to look for morphology, it should jump out at you. (with the exception of malaria/babesia - they can be sneaky)

Go with Instrumentation Laboratories TOP550! (sorry, no experience with Siemens)

We are not CAP inspected, but in NYS this is not a requirement YET (although I heard of one lab being cited for it). Both MAS and BioRad have microscopic acceptable ranges. You have to spin the control down in order to perform microscopic (unless you run it through something like the Iris). I know of a lab that aliquots off a bottle into 1-2 mL aliquots and uses them daily. Nicole

It would be easy to see if you had poor technique resulting in a dilutional error doing the plasma replacement based on consistent or inconsistent WBC, RBC, and platelet counts. If you're really bad at it, those won't match the original specimen.

Forever ago we used a calculation. You would spin the sample down, then run the lipemic plasma for hgb measurement. Then calculate: True hgb = original hgb - ((plasma hgb x (1 - hct as a decimal)) I have no idea where this calculation came from, but it did always the plasma replacement method. In my opinion, doing any kind of calculation is just silly when you can just replace the problematic plasma with diluent/saline. Using plasma replacement, you don't have to recalculate any of the indices, either. Nicole

When I started with the DDHS500 line a few years back I had a horrible time with controls. That 30 day claim is junk - and I thought they were going to work on changing that wording in the package insert. I could only get 3 days out of mine. They did seem to work out the issue I was having, though. I believe it was a combination reagent stability AND qc stability. (We ended up switching to BioRad controls simply because in NYS I have to have a "negative" control, and IL doesn't make one that low yet.) But in all the years of using IL, that is the ONLY problem I've had. The TOP analyzers are rock stars! (and I just got my new 550's in!) The best thing to do is document EVERYTHING. I had a log at the instrument where I had the techs write down every time they changed reagent or made up new qc and why (empty vs. qc trouble). And then I would print the Levy-Jennings and send them in. Do you submit monthly to the AccuTrak? If everyone else is running high, that is in your favor.

Scott - you aren't located in NYS, are you? Definitely a requirement for us, and have gotten cited for it. I asked my apps person what other people are doing and she indicated they are doing the same thing I am. Some can be done by concentrating the normal control. Others I have to hope to come across a patient that reaches the higher levels.

to anyone that performs special coag testing - what do you use to perform the require cal ver / linearity? I looked into George king, they only have material for clot based assays, but most of mine are chromogenic or immunoturbidimetric. I have to run Protein C Activity, Free Protein S, von Willebrand Activity, von Willebrand Antigen, and ATIII Activity. The manufacturer just recommends using a different lot of control material, but that doesn't go anywhere near the upper linearity. I've been trying to catch high patients and serially dilute, but I always end up missing one or two tests (especially Free Protein S). Any help is appreciated! Nicole

We've been working up the Alcor iSED. Unfortunately, I am not the one in charge of it (for once) and have only ran one sample on it. It takes 3 minutes for the first sample and then 30 seconds for each additional sample (up to 20). So in my view, anything that takes less than an hour is a win. Plus it reads a sample barcode and provides a printout. Too bad we don't have the $$ to interface it. Apparently if you're in a nursing home here a weekly sed rate is a necessity - so we get ALOT. And don't forget the STAT ERs - because they are waiting for the result to discharge the patient. *sigh*

very well written! I'm going to save this for lab week. the public only hears when bad things happen with the lab, and most nurses seem to think our job is to compromise samples so they have to be redrawn. they don't realize that we do know patients by name, bad results have an effect on us, and we often wonder how patients are doing.

yes, SMILLER! it's true! the kit is VERY easy to use - all liquid ready-to-use reagents. I don't remember if the controls are liquid or not. It will be a great test to quickly rule out HIT, but I'm pretty sure in order to diagnose you will still need to send a confirmation to a reference lab. Our friends at the FDA have made them change it to a qualitative pos/neg instead of reporting actual values. This is why it will take until 1st quarter of 2017 before it becomes widely available in the US - they need to reprint all the package inserts to reflect that change. Start saving correlation samples!

We perform dRVVT testing for lupus-like anticoagulant/antiphospholipid testing. We automatically run both the screen and confirm, then report a normalized ratio. For any ratio >=1.20 a comment is added to the report saying, "Evidence of a lupus anticoagulant (antiphospholipid antibody) is detected ... blah blah blah." Our new hematopathologist would like for us to change the wording because it sounds too definitive - ISTH guidelines recommend performing at least two methods and to have testing repeated at 12 weeks to definitively call an antiphospholipid antibody. I agree with her, so I'm wondering what (if any) comment any of you that perform this testing attach to the result?