JACO Officially listed i-STATs as a Blood Analyzer

We have been looking at this change from all angles and were told by TJC that we also need to re-evaluate our IQCP and perform monthly QC on every device. Now that they are considered chemistry analyzers, we have to treat them as such in every way. We have been considering using our monthly data to complete our 6 month correlations. 

We only have one i-Stat in my facility, and only perform the ABG's on that analyzer, so we have never performed correlations.  We did, at one point, many moons ago, perform the 6 mth cal verify.  During one of our JC inspections, an inspector stated the cal verify was not necessary.  Does this change reimplement the need for the cal verify at 6 mths?

many thanks

I'm also trying to figure out how others are interpreting this requirement. We currently perform cal/ver on all iSTAT every 6mo, have an established IQCP, perform weekly blood gas QC, and complete method comparison studies against our GEM analyzers every 6mo on a subset of iSTAT. We have about 28 iSTATs across our health system so performing method comparisons on every analyzer every 6mo would be a pretty significant undertaking. My interpretation is that all iSTAT need to be included in the correlation process but it's not clear whether JC expects direct method comparison on every analyzer or if there is another acceptable way to demonstrate agreement across all devices. We're considering using our cal/ver data to show agreement between the iSTAT while continuing method comparison studies on a subset of analyzers against the central lab method

Our lab is CAP accredited while our hospital is JC accredited. Also curious whether anyone has heard if CAP plans to issue guidance on this topic.

We just had our JC inspection last month. The inspector told us that using QC data or Cal Ver data are not acceptable as correlation data. Basically, they are separate events. Instead, he suggested (but cannot tell us what to do) that we test 5 (at minimum) different pooled samples and average the results. Those averages are then compared to each other. I'm still working out if I want to designate a primary (the inspector didn't like the idea but my pathologist does) or use an average of all results to compare (possible with a trimmed average to remove outliers). IDK, I have all the data, just unsure how to best analyze it. 

We only need to correlate directly measured analytes (we were already doing this), and only to similar sample types (we no longer need to correlate plasma/serum methods to whole blood).

What?! Cannot use QC or Cal Ver data to crunch separately as correlation? I would highly recommend direct contact with JC again and request further investigation on this, find direct evidence in a regulation that states you cannot use QC or Cal Ver. I am not JC but the CAP does allow this. 

My facility also does the Cal Ver and QC as well as patient sample comparison on the lab analyzers. We also get the AQSQ Quality Cross Check survey from CAP for our iSTATs twice per year which is done on all in use iSTATs.  I'm wondering if this will suffice for JC regulations.

How many correlation samples are you all performing on each i-STAT meter? I know this is dependent usually on each Medical Director's decision --- just curious. Are you also correlating the ACT cartridge? 

Good morning, Everyone. Just following up if anyone has made a solid plan yet for these "meter-to-meter" correlations? 

Our plan is to continue using the CAP cross check for all Istat instrument-to-instrument comparisons, to perform Cal/Ver on all analyzers (beginning with the next CLEW update), and to perform patient correlations (5-samples) on a subset of analyzers.  

@Autilia, how many i-STAT meters do you do the instrument-to-instrument comparisons to? 

https://poct.groupsite.com/files/1152636/preview

Hopefully the link that I shared opens. What we decided to do based off that link is run all 50 of our istats on the same day using all same lots and preforming QC level 1 &2. For ACT we only used the meters that use ACT cartridges and did the ACT level 1 and 2 QC, again using all same lots of QC and cartridges. I then put them into EP evaluator and used simple precision with the TEA as the cut offs then got them signed off. 

Hello all,

I just wanted to circle back to this discussion.  In April we found out about this change (our JC inspection was in Feb), then found out CMS was coming the following week for review.  We wanted to get ahead of the new interpretation due to CMS inspection, so we gathered all 56 of our istats throughout the week and performed meter to meter testing for chemistry/BG/HH analytes (Chem8/CG8/CG4) to a "reference meter". We used 3 tricontrols and same lot of cartridges. I then compared my reference meter like I have for the past 16 years prior to this change to the BG lab and main labs using 5 patient samples.
It will be time for 6 month correlations again in October and we wanted to revamp if possible. There have been some suggestions made and I am getting caught up on lot numbers (importance). 

1. Is it acceptable to only use 3 "samples" (tricontrols) to perform meter to meter testing and then 5 patient samples for the reference meter? 
2. Is it acceptable to use historic data even if the reagent and qc lot numbers are not the same? 

Thank you for your input and suggestions on how you have addressed this interpretation with working smarter and not harder like we chose to do in April lol...

You also need to do 6-month AMR validation.  Your medical director can choose how many samples are required. Abbott recommends using 3 levels of the Cal/Ver samples for the 6-month AMR validation, but it would be fine to use the Tricontrols if the range is wide enough. We also use a Hyperbaric and Hypoxic level to increase the PO2 AMR.  You can use these results as your correlations as long as it is spelled out in your IQCP. I use the AMR results to correlate the iSTATs to each other and use the SD and CV to prove they are interchangeable/equivalent. I do not use a "reference meter" unless I am correlating to the main lab.  I do the lab correlations separately and only on analytes performed using the sample type. 

You can use historical QC results, but I would think you need to use the same lot of QC to mimic using the same patient sample for all iSTATs and run all cartridge types on every iSTAT.  Cartridge lot isn't as important in my opinion. However, you can pretty much do what you want with medical director approval.  I am unable to do it this way because we have so many iSTATs in areas that use different cartridges.  We used the CAL/VER levels and do them on all iSTATs at the time of the CLEW update. If you can't do it all at once, you can do it over the course of 6 months as long as every cartridge type is run on all iSTATs for every sample. 

Whatever you choose to do should be approved by your medical director and must be written in your IQCP.  Per TJC, the monthly QC does not need to be done on every iSTAT as long as you have the reason spelled out in your IQCP. For example, monthly QC is performed to verify the storage conditions. Therefore, it doesn't need to be done on every iSTAT that shares the cartridges stored there.