From Rachel Donner:
“Hello, George. I am a POC consultant at Maine Medical Center in Portland, ME. My department has had difficulty finding guidelines for best practices for instrument-to-instrument comparisons on our ACT and TEG programs. Currently, we have 19 Medtronic ACT Plus and 4 TEG 6s (and only 2 POC consultants at this location, of course, with other responsibilities!).
In past years, our department has used historical QC for the ACTs and compared them that way, but lots were not always easily trackable between departments, so we did not find this to be very accurate. In 2023, we did run a study on ALL the ACTs to see that they were comparable. We used the manufacturer’s QC material and found the CV to be 3% for normal and 6% for abnormal. The TEG 6s is a new program started within the past year and is technically not under our department’s CLIA oversight, but we offer consultation to them when needed.
We have had difficulty finding published guidelines for either of these types of testing regarding tEA, etc., so any ideas regarding best practices for these types of programs would be greatly appreciated. Thank you so much for your amazing website and for being such a great resource for lab staff with questions.”
Hello, and thank you for the compliment and your questions. It appears you’ve done a good job validating your ACTs. Pardon a self-reference, but there is a brief section entitled Multi-site Validation in Sarkar MK, Fritsma GA. Chapter 3; Quality Assurance in Hematology and Hemostasis Testing in Keohane EM, Butina MM, Mirza KM, Walenga GM. Rodak’s Hematology; Clinical Principles and Applications, 7th Edition, Elsevier. ISBN 978-0-323-93650-7. You may access the textbook through your facility’s library or online. Watch the comments section below, as I’ve reached out to colleagues who may offer additional advice.
February 5, 2025: From colleague, author, and frequent contributor Bob Gosselin: Hey GF, et al, I think she is asking about what to do for between-instrument assessments for POC devices, including the TEG or ACT…did I get that right?
Presumably, they are trying to establish equivalence between instrumentation at least on an annual basis, but for other instruments, we do comparisons on a more frequent basis. If this were within my bailiwick, I would consider at least a biannual assessment, but if there are a lot of instruments, then rotate 25% of these instruments on a quarterly basis. There are lots more glucometers in hospitals than coag POC, so likely lean on them for some guidance about ensuring comparability between devices. I would use surrogate preparations, as I did for preparing monthly IQCP samples for TEG, TEG Platelet Mapping, and PLT aggregation testing (added by Geo: also known as viscoelastic testing [VET]). Presumably the ACT is used for high dose UFH monitoring, but surrogate preparations should reflect the monitoring conditions (i.e. ECMO/Cath Lab vs CPB). As the ACT is likely using whole blood samples, recalcifying the citrated surrogate preparations would be required.
Dear Rachel, Comparing the performance of multiple analyzers can be done by testing a set of samples across all devices at appropriate time intervals. These samples should cover a wide measurement range, including critical clinical decision points. Depending on the nature of the samples—whether they are standard or prone to error—you can use Ordinary Least Squares (OLS) regression, Deming regression, or Passing-Bablok regression (which does not require a normal distribution and is robust to outliers) in your data analysis.
The most challenging step is defining acceptance criteria, which should be established in advance based on the device’s application (clinical relevance) and analytical performance (reproducibility and imprecision). For guidance, the CLSI EP09 document provides valuable recommendations on method comparison and bias estimation. CLSI has also a series of useful guideline documents for POC devices:
https://clsi.org/standards/products/point-of-care-testing/documents/