Clinical utility and specificity of anticardiolipin antibodies.

Established solid phase assays for anticardiolipin antibodies (aCL) are often characterized by high levels of nonspecific binding. As a result, only very high levels of aCL have been reported to be associated with a variety of clinical conditions including systemic lupus erythematosus (SLE), recurrent intravascular thrombosis and unexplained recurrent fetal loss. We have developed an ELISA replacing direct evaporation of soluble cardiolipin with cardiolipin micelles in physiological saline as the antigen binding step in the assay. Levels of IgG aCL were detected in various sera at dilutions of 1/100 to 1/3200, showing improved assay sensitivity. Assay specificity was determined using double stranded DNA and ovalbumin as irrelevant binding antigens and no crossreactivity was found. The controversial use of Tween 20 in the assay was investigated and results showed it decreases nonspecific binding without interfering in antibody detection. This assay has enabled us to identify differences in the prevalence and level of aCL antibodies in sera from healthy nonpregnant controls (0/25 positive), healthy pregnant controls (5/47 positive for IgG and 8/47 positive for IgM) and from women with unexplained recurrent fetal loss (16/62 and 14/62 positive, respectively). We support the observation that aCL are not normally distributed, and therefore nonparametric methods of statistical analysis are necessary to determine population prevalence. We confirm that aCL IgM are a relatively nonspecific finding, and extreme caution must be used in basing any clinical decisions on the presence of this antibody alone.