INCLUSION OF SERUM MARKER MEASUREMENTS FROM A PREVIOUS PREGNANCY IMPROVES DOWN-SYNDROME SCREENING PERFORMANCE

A predisposition for high or low levels of serum marker concentrations in second trimester Down syndrome screening reflecting itself in cons ecutive pregnancies in the same woman has been demonstrated, but hithe rto the possible effect of including previous marker results in a curr ent risk evaluation has been considered negligible. Using published da ta on correlations between the markers AFP, hCG and uE3 in different n ormal pregnancies in the same women and age-related a priori probabili ties we found, that in triple marker screening the inclusion of result s from a previous pregnancy in a likelihood ratio based risk calculati on could increase the detection rate for women having had an earlier p regnancy from 68.0 per cent to 70.2 per cent at a risk cut-off=1:250. The screen positive rate for normals for the same population of women, being on average older than the total population, fell from 7.1 per c ent to 68 per cent. These figures, that are based on an assumption of the same correlations between one normal and one Down syndrome pregnan cy as between two normal pregnancies, corresponds to an expected reduc tion, in the population considered, of the number of children born wit h Down syndrome of 6.7 per cent and of the number of screen positive n ormals of 4.7 per cent. Considering that this can be achieved at no ex tra cost, it is concluded that implementation of a procedure for takin g information from previous pregnancies into account in second trimest er screening should be considered at centres that can handle the softw are problems involved in doing so. However, better data on the correla tions between a normal and a subsequent Down syndrome pregnancy in the same woman should probably be awaited before this is done. (C) 1998 J ohn Wiley & Sons, Ltd.