Jv. Jones et al., ANTIPHOSPHOLIPID ANTIBODIES IN A HEALTHY POPULATION - METHODS FOR ESTIMATING THE DISTRIBUTION, Journal of rheumatology, 22(1), 1995, pp. 55-61
Objective. To determine the distribution of antibodies to cardiolipin
(IgG, IgM and IgA) in a healthy population. To classify subjects accor
ding to a cutpoint in order to determine what level of antibody measur
ement would yield 95% specificity. Methods. Antiphospholipid (aPL) ant
ibodies were measured using a conventional ELISA technique, with isoty
pe specific antibody, to determine levels of IgG, IgM and IgA binding
to cardiolipin. The subjects were 282 Red Cross blood donors, 82 under
graduate medical and nursing students, and 13 laboratory volunteers. R
esults. The distribution of aPL antibodies measured in units of optica
l density is significantly skewed, while the distribution of aPL antib
odies measured on the log scale is nearly symmetric. Neither distribut
ion is normally distributed. There was no evidence that aPL antibodies
differed significantly by age (up to age 50) or sex. We estimate the
90th, 95th, and 99th percentiles for IgG, IgM and IgA antibodies for t
he optical densities and binding indices. Conclusion. For 95% specific
ity we conclude that an adjusted IgG optical density above 0.386, an a
djusted IgM optical density above 0.301, or an adjusted IgA optical de
nsity above 0.085 be considered positive.