Second trimester prenatal screening for Down's syndrome using alpha‐fetoprotein and free beta hCG: a seven year review

OBJECTIVE: To determine the value and impact over a seven year period of a second trimester screening programme for trisomy 21 and trisomy 18, using the two maternal serum markers alpha-fetoprotein and free beta human chorionic gonadotrophin. DESIGN: Retrospective review. SETTING: A biochemical screening laboratory serving three health districts with three antenatal clinics in both teaching and nonteaching hospitals. POPULATION: 67,904 pregnancies in women of all ages screened between 14 and 22 weeks of gestation between 1 April 1991 and 31 March 1998. METHODS: All women booked into three major antenatal clinics were offered biochemical screening. Women at increased risk of trisomy 21 or trisomy 18 (> or =1 in 250 at term) were offered an invasive diagnostic procedure. Follow up of the outcome of all pregnancies was performed. MAIN OUTCOME MEASURES: Detection rate for trisomy 21 and trisomy 18, false positive rates, uptake of screening, uptake of amniocentesis in women identified at increased risk, prevalence of trisomy 21 at birth, detection and false positive rates by maternal age, fetal loss rate after amniocentesis, report turn around time, and identification of other anomalies. RESULTS: Overall, 87% (67,904/78,501) of women underwent screening. The rate of detection of trisomy 21 was 75% (80/107; 95% CI 66 to 83) with a 5.1% false positive rate (3466/67,904; CI 4.9 to 5.3%). In women under 30 years of age the detection rate was 60% (18/30; CI 41 to 77) with a 2.6% false positive rate (956/36,371; CI 2.5 to 2.8). The rate of detection of trisomy 18 was 57% (8/14; CI 29 to 82) with a 0.7% false positive rate (475/67,904; CI 0.64 to 0.76). Uptake of amniocentesis was 83% (2912/3508). Women were 3.3 times more likely to refuse amniocentesis if the risk was close to the cutoff (1 in 250) than if the risk was > or =1 in 50. Fetal loss within 28 days of amniocentesis was 0.9% (25/2912). Prenatal screening identified 84 other anomalies in addition to 41 cases of impending fetal death. CONCLUSION: Second trimester prenatal screening for trisomy 21 and trisomy 18 using a simple two marker approach incorporating free beta hCG can achieve high detection rates over a long period of time. Health authorities who still have not introduced trisomy 21 screening should be encouraged by what can be achieved and should consider making such screening available to all women. Established second trimester detection rates of 75% for a 5% false positive rate will be the benchmark by which first trimester screening using nuchal translucency, PAPP-A and free beta hCG will be judged.