Background: Both first-trimester screening and second-trimester screening for Down's syndrome are effective means of selecting women for chorionic-villus sampling or amniocentesis, but there is uncertainty about which screening method should be used in practice. We propose a new screening method in which measurements obtained during both trimesters are integrated to provide a single estimate of a woman's risk of having a pregnancy affected by Down's syndrome.
Methods: We used data from published studies of various screening methods employed during the first and second trimesters. The first-trimester screening consisted of measurement of serum pregnancy-associated plasma protein A in 77 pregnancies affected by Down's syndrome and 383 unaffected pregnancies and measurements of nuchal translucency obtained by ultrasonography in 326 affected and 95,476 unaffected pregnancies. The second-trimester tests were various combinations of measurements of serum alpha-fetoprotein, unconjugated estriol, human chorionic gonadotropin, and inhibin A in 77 affected and 385 unaffected pregnancies.
Results: When we used a risk of 1 in 120 or greater as the cutoff to define a positive result on the integrated screening test, the rate of detection of Down's syndrome was 85 percent, with a false positive rate of 0.9 percent. To achieve the same rate of detection, current screening tests would have higher false positive rates (5 to 22 percent). If the integrated test were to replace the triple test (measurements of serum alpha-fetoprotein, unconjugated estriol, and human chorionic gonadotropin), currently used with a 5 percent false positive rate, for screening during the second trimester, the detection rate would be higher 85 percent vs. 69 percent), with a reduction of four fifths in the number of invasive diagnostic procedures and consequent losses of normal fetuses.
Conclusions: The integrated test detects more cases of Down's syndrome with a much lower false positive rate than the best currently available test.