From Maureen Tacke, Idaho; Hi George, my colleagues and I enjoy the Fratsma Factor postings you send. Thank you for the work that you do instilling best practice across the nation. Each year at the PTT lot change, we redo the heparin response curve. We’ve done this to follow practices from years prior. We’ve noted in the CAP checklist that after the initial response curve, it is only necessary to compare the current lot with the new lot of PTT. We are considering changing practices, and wanted to hear your thoughts on the matter. CAP does not dictate the correlation coefficient range between lots. What would you recommend? 0.9 – 1.1? Smaller? Eventually, we’d like to change practice within our pharmacy so that they monitor patients via anti-Xa. But we are not quite to that point yet with our pharmaceutical colleagues. Thank you for your time.
Hello, "Mo," and thank you for your message. I support the CAP guideline and recommend you perform a simple lot-to-lot correlation, provided your suppliers' lots are relatively consistent, which is not always possible. I agree that a CC of 0.9 or above is a good limit, however be sure to also check the individual data points to ensure none is far outside of your selected comparison range, for instance, 10%.
In case you decide you need to perform a "Brill-Edwards" curve, here is my favorite reference, Marlar RA, Gausman J., The optimum number and types of plasma samples necessary for an accurate activated partial thromboplastin time-based heparin therapeutic range. Arch Pathol Lab Med. 2013;137:77–82. In this article, Marlar provides stats to support a relatively small number of samples for producing the curve.
By the way, I encourage you and your pharmacy group to move to the chromogenic anti-Xa heparin assay as soon as possible. Tell them the nurses will be happy to reduce the need for repeatedly adjusting their heparin drip rate and the assay can also be used for LMWH , fondaparinux, rivaroxaban, apixaban, and edoxaban.