Clinical Question: Can a combination of markers in the first trimester accurately screen for trisomies 18 and 21?
Setting: Outpatient (any)
Study Design: Cohort (prospective)
Synopsis: In a previous POEM (N Engl J Med 1999;341:461), a mathematical model was used to argue that it is possible to screen for trisomies 18 and 21 in the first trimester with a combination of markers and ultrasonographic findings. Earlier screening would allow for earlier diagnosis and earlier termination of pregnancy, if desired. This study attempted to validate that approach in real patients.
Investigators enrolled women at 74 to 97 days of gestation from 12 centers. Participants were offered first-trimester screening if they had a singleton gestation, no significant recent vaginal bleeding, no indications for prenatal diagnosis other than risk of trisomy, and no pregestational diabetes. The test included free β human chorionic gonadotropin and pregnancy-associated plasma protein A, and ultrasonographic measurement of fetal nuchal translucency, all interpreted in light of the maternal age and the pretest probability of trisomies 18 and 21.
Of 8,816 eligible patients, 302 did not complete screening, 102 were analyzed separately because of a previous pregnancy with trisomy, and 196 were lost to follow-up. This left 8,216 patients (93.2 percent) for the final analysis. The cutoff value to define an abnormal test result was chosen to identify a risk of one in 270 for trisomy 21 and one in 150 for trisomy 18. These values are consistent with the cutoff values used for standard second-trimester screening.
The investigators found that a combination of maternal age, the two serum markers, and nuchal translucency in the first trimester was 89 percent sensitive and 89 percent specific for identification of trisomy 18 or 21. The combination was more sensitive in women 35 years and older (91.2 versus 80 percent in younger women) but less specific. The false-positive rate was 16.8 percent in women 35 years and older and only 4.7 percent in women younger than 35 years. In women 35 years and older, this protocol identified all 11 cases of trisomy 18 and 90 percent of those involving trisomy 21; only 15 percent of women would have required amniocentesis to confirm a true- or false-positive result.
Bottom Line: A first-trimester protocol involving maternal age, two biochemical markers, and ultrasonographic measurement of fetal nuchal translucency is at least as accurate as standard second-trimester screening. (Level of Evidence: 1b)