Nuchal translucency measurement at different crown-rump lengths along the 10- to 14-week period for Down syndrome screening.

Objectives: To evaluate the screening accuracy for Down syndrome of nuchal translucency (NT) measurement at different crown-rump length (CRL) subgroups along the 10- to 14-week period.

Methods: NT was classified 'enlarged' if greater than or equal to 1.5 and 2.0 multiples of the regressed median. Accuracies for Down syndrome (formula=[(TP + TN)/(TP + TN + FP + FN)]x100, where TP: true positive, TN: true negative, FP: false positive, FN: false negative) were evaluated in four classes of CRL: 38-44 mm, 45-54 mm, 55-70 mm, and 71-84 mm, and compared.

Results: Of 20 743 fetuses, 20 611 were with no chromosomal abnormalities and 132 were with Down syndrome. Down syndrome fetuses with enlarged NT were 99 (greater than or equal to 1.5 MoM) and 86 (greater than or equal to 2.0 MoM). Sensitivity decreased with gestation, while specificity increased, resulting in increasing likelihood ratios with gestation for each of the CRL groups (8.1, 14.1, 16.3, 17.1 with the use of the 1.5 MoM cut-off, and 13.2, 27.1, 50.1, 84.1 for the 2.0 MoM cut-off). The accuracy increased with gestation (89%, 95%, 95%, 96% with the use of the 1.5 MoM cut-off, and 94%, 97%, 98%, 99% for the 2.0 MoM cut-off, for each of the CRL groups), differences being statistically significant between periods in half of the comparisons.

Conclusions: Although sensitivity of NT assessment for Down syndrome screening decreased as gestation advanced from the 10th to the 14th week, accuracy showed a remarkable increase. These changes should be taken into account in defining and improving the Down syndrome screening policies.