OBJECTIVE: Our purpose was to develop a method of calculating the individual odds of Down syndrome on the basis of a combination of maternal age and multiple ultrasonographic parameters that can be used to counsel women at high risk who initially decline amniocentesis.

STUDY DESIGN: Maternal age and ultrasonographic biometry data were collected prospectively on 3254 normal and 30 Down syndrome singleton fetuses between 15 and 24 weeks’ gestation. Humerus length data were expressed as multiples of the normal median. Log transformation of the humerus length data permitted their expression in gaussian frequency distributions and the calculation of likelihood ratios for Down syndrome on the basis of humerus length. We also developed likelihood ratios on the basis of the degree of nuchal skinfold thickening and the presence or absence of hyperechoic fetal bowel and hypoplastic fifth digit.

RESULTS: The ultrasonographic parameters and maternal age did not significantly correlate with each other and were significant independent predictors of Down syndrome. We therefore calculated the individual odds of Down syndrome by using the product of the age-related risk and the likelihood ratios associated with nuchal thickening, humerus length shortening, and the presence or absence of hyperechoic fetal bowel or fifth digit hypoplasia, respectively. At a Down syndrome risk level of >1:50, a 60.0% detection rate with 4.5% false-positive rate was observed with a screen-positive rate of 5.5%, positive predictive value of 1:10, and odds ratio (95% confidence interval) of 28.4 (12.8 to 64.0).

CONCLUSION: This is the first report of individual odds calculation based on multiple midtrimester biometry parameters and maternal age. The screening efficiency is similar to that reported with triple-analyte serum screening. These data are useful for counseling women who are at increased Down syndrome risk and initially decline amniocentesis. (Am J Obstet Gynecol 1998;179:514-9.)