The comprehensive midtrimester test: High- sensitivity Down syndrome test

OBJECTIVE: The purpose of this study was to develop a highly sensitive algorithm for midtrimester Down syndrome detection. STUDY DESIGN: Urine (hyperglycosylated human chorionic gonadotropin, beta-core fragment of human chorionic gonadotropin), serum (alpha-fetoprotein, human chorionic gonadotropin and unconjugated estriol [uE(3)]), and ultrasound biometry (nuchal thickness, humerus length, the presence of gross ultrasonographic anomalies), and maternal age were measured at genetic amniocentesis. Stepwise logistic regression analysis was used to identify the most significant markers. A multivariate Gaussian algorithm plus age was used to derive patient-specific Down syndrome risk. Sensitivity and false-positive rates at different risk thresholds and the area under the receiver-operating characteristic curve were determined. A probability value of <.05 was significant. RESULTS: There were 568 study cases with 17 Down syndrome cases (3.0%). The mean (+/- SD) maternal and gestational ages for the study group were 36.9 (+/- 3.5) years and 16.2 (+/- 1.4) weeks, respectively. The significant markers were nuchal thickness (P =.0001), hyperglycosylated human chorionic gonadotropin (P <.001), and beta-core fragment (P <.002). Neither maternal age nor gross sonographic anomaly contributed significantly to Down syndrome detection. The comprehensive midtrimester test was extremely efficient for Down syndrome detection in advanced maternal age only cases with a sensitivity of 92.3% at a 0.8% false-positive rate. In women <35 years old, all the Down syndrome cases were detected at 2.2% false-positive rate. For the overall population, the sensitivity was 93.7% at 5% false-positive rate. CONCLUSION: In a preliminary study, the comprehensive midtrimester test appeared highly sensitive in different age groups. Gross anomaly detection was not required for high performance, which makes the comprehensive midtrimester test potentially suitable for low-risk screening and as an alternative to amniocentesis in women who wish to avoid the procedure. This was a small study; thus, the clinical value of this test can only be established in large trials.