PB0122 - Automated algorithm for interpretation of functional Lupus anticoagulant testing

Background: Laboratory detection of antiphospholipid antibodies (aPL) consists of the detection of lupus anticoagulant (LA), anti‐cardiolipin and anti‐beta‐2-glycoprotein-I antibodies.
LA remains the most complex analysis mainly due to the lack of a gold standard, heterogeneity of aPL, availability of many different assay principles and reagents, multistep procedure testing and possible interference of anticoagulant medication. Furthermore, we can have different approaches to investigating LA that can lead to different results. Due to the abovementioned reasons, the interpretation of LA assays needs an expert evaluation.

Aims: To simplify the process, we have implemented in our middleware the latest algorithm suggested for ISTH guidelines that recommend performing two screening tests. Implementation of automatic addition of mixing and confirming tests if the screening test on the undiluted sample is prolonged.

Methods: We went through several phases from the choice of the algorithm, the creation of the different rules and the validation process where the results were compared with the manual LA interpretation.
Negativity is defined if both screening tests are normal or if confirm test is under normal range.
Positivity is defined with one of two positive tests (DRVVT/SCT) having a confirm test above the local cutoff.
We have encoded different interpretations (based also on previous LA results) with the intent to standardize the LA conclusion and reduce the intra and interlaboratory variations.
320 LA assays were evaluated in double the three possibilities implemented encoded.

Results: We found a high concordance rate. The highest rate with negative and positive LA (100%) and the lowest rate with LA present but without previous results (82,4%).

Conclusion(s): Automated processes in LA testing make it more reliable, faster, and easier to perform. Furthermore, standardizing the interpretation reduced the intralaboratory and interlaboratory variation.
This approach helps the interpretation of the results, which could be not easy in LA, especially for the less expert pathologist.