Screening for Down's syndrome based on individual risk

To evaluate the effectiveness of biochemical screening of individual pregnancies for Down's syndrome risk.

Retrospective determination of risk.

Obstetric and cytogenetic services in Tayside, Scotland.

3436 pregnant women who had screening for neural tube defects in the second trimester during November 1988 to March 1990 and whose pregnancies were dated by ultrasonography. Three women with pregnancies associated with Down's syndrome reported later in 1990.

Individual risk calculated from age at estimated date of delivery; chorionic gonadotrophin and alpha fetoprotein concentrations in serum samples obtained at precisely determined gestational ages in second trimester. Results of karyotype determination and outcome of pregnancy.

During November 1988 to March 1990 karyotypes were determined for 5% of pregnancies for reasons of maternal age and genetic history and one of the eight affected fetuses was detected. Individual risk could not be calculated for 347 pregnancies, but screening on this basis would have detected five of the cases and required screening in 194 out of 3089 (6.3%) pregnancies; all three affected pregnancies reported later in 1990 would also have been detected, giving a success rate of 73% (95% confidence interval 39% to 94%). The age distribution of women according to individual risk suggests that women over 35 would be screened effectively.

Screening based on individual risk would use resources more effectively than screening based on maternal age and genetic history without affecting detection rates in older women.