Combined sonographic and biochemical markers for Down syndrome screening

To evaluate the efficacy of fetal nuchal fold thickness and proximal long bone biometric measurements in modifying Down syndrome serum screening risk in a population of women referred for second-trimester sonography.

Sonographic biometric measurements and biochemical markers were combined retrospectively for 2533 women with known pregnancy outcomes. Four different screening methods were compared: (1) advanced maternal age; (2) biochemical serum screening markers; (3) modification of serum screening risks on the basis of categorical cutoffs for nuchal fold and femur and humerus length; and (4) a combined approach in which the sonographic measurements were treated as multiples of the medians and entered, together with the serum screening results, into a multivariable algorithm. The efficacy was compared at second-trimester risk cutoffs of 1:270 and 1:100.

Down syndrome was present in 30 of the 2533 pregnancies (1 in 84). With the use of the 1:270 cutoff, biochemical screening had 93% sensitivity and a 40% false-positive rate. With application of the categorical method of fixed cutoffs to incorporate fetal biometry, the false-positive rate was reduced to 33% with no loss of sensitivity. The combined model had 83% sensitivity and a 19% false-positive rate. The combined method had the highest positive predictive value (1 in 20). Similar gains in efficacy could be shown with the 1:100 cutoff.

For this high-risk group, the multivariate model that combines serum screening and sonography can result in a substantial reduction in the number of amniocenteses. Although the addition of the sonographic biometric measurements resulted in some Down syndrome cases being missed, the net effect was a large improvement in the overall positive predictive value of the screening.