Clinical utility of anticardiolipin antibody assays: high inter-laboratory variation and limited consensus by participants of external quality assurance programs signals a cautious approach

Abstract

Antibodies that bind phospholipid (anti-phospholipid antibodies [APA]) are a focus of major interest to both clinical and laboratory personnel across a variety of disciplines because of the assortment of disorders with which they are reportedly associated. A solid phase assay using cardiolipin as the test phospholipid (anti-cardiolipin antibody assay [ACA]) has now become a laboratory standard for the detection of APA. In the current report, data from two separate external quality assurance programs (QAP) and collected over the past four years, have been evaluated to assess the utility of the ACA. Despite attempted standardizations, exceedingly high inter-laboratory variation and a general lack of test result consensus would signal the adoption of a cautious clinical approach towards laboratory findings. For example, for a total of 21 cross-laboratory tested serum samples (tested for both IgG and IgM for 41 quantitative estimations), inter-laboratory variation for both ACA-IgG and ACA-IgM was higher than 50% in 20 of 41 testing cases (48.8%). The situation with regard to testing consensus was equally concerning. Total consensus (i.e., 100% of participating laboratories agreed that a given serum sample gave an ACA result of either negative or positive) occurred in less than 20% of cases for ACA-IgG. More importantly, in about 50% of serum testing occasions there was no general consensus in returned laboratory data (i.e., more than 80% of labs could not agree on whether a serum sample tested was either ACA-positive or ACA-negative). Thus, despite attempted international standardizations, exceedingly high inter-laboratory variation and a general lack of test result consensus would argue that the assay has limited utility. We conclude that single point laboratory results must be used with considerable caution before accepting that ACA activity is present in patient serum or not. We agree with recommendations indicating that laboratory tests should be repeated at least once prior to making a clinical diagnosis of any APS-like disorder.