Effect of Deviation of Nuchal Translucency Measurements on the Performance of Screening for Trisomy 21

Abstract

The single most effective marker for all major chromosomal defects including trisomy 21 (T-21) is the first trimester fetal nuchal translucency thickness (NT). Accurate measurement of the fetal NT to screen for chromosomal abnormalities requires appropriate training, adherence to a standard ultrasound technique, and regular audits of performance. The investigators theorized that an underestimate in NT measurement by individual sonographers would reduce the estimated risk of Down syndrome and thus the overall screen-positive rate; an overestimate would increase both individual risk and the screen-positive rate. This study was designed to examine the effect of deviations in both the median NT and the spread of NT measurements on Down syndrome screening performance, using either the combination of maternal age and fetal NT, or the combination of maternal age, fetal NT, maternal serum pregnancy-associated plasma protein-A (PAPP-A), and free β-human chorionic gonadotropin (β-hCG). The effects of deviations in fetal NT medians were assessed by simulating NT measurements and PAPP-A and free β -hCG multiples of the median for 500,000 euploid and 500,000 T-21 pregnancies at 12 weeks' gestation. Detection and false-positive rates were calculated without adjusting NT, and after overestimating and underestimating NT measurements by adding or subtracting values ranging from 0.1 to 1.0 mm to each observed measurement. The effects of variation in the scatter of NT measurements were examined by applying a multiplicative factor ranging from 0.5 to 2 to the SD of the simulated NT data. When the false-positive rate was fixed at 3%, the Down syndrome detection rate was 72% when determined by maternal age and fetal NT, compared to 86% when determined by maternal age, fetal NT, and maternal serum biochemistry. When the false positive rate was held constant, consistent underestimates of fetal NT reduced the detection rate, while overestimates increased the false-positive rate. For example, when NT was underestimated by 0.6 mm and the false positive rate was fixed at 3%, the detection rate using maternal age and NT fell from 71% to 68% and from 86% to 83% for maternal age, NT, PAPP-A, and β -hCG. Widening the scatter of NT measurements had small impact on the detection rate, but resulted in a major increase in the false-positive rate. By illustration, if the NT scatter was increased by 50%, at a detection rate of 71 % the false positive rate was 7.0% for maternal age and NT; at a detection rate of 85%, the false positive rate 4.3% for maternal age, NT, PAPP-A, and β -hCG. These findings illustrate the screening impact of small deviations in NT measurement. The accuracy of the NT measurement techniques of individual sonographers can be assessed by determining the deviation of those measurements from the median and SD of NT measurements obtained by experts. Failure to correct such deviations can alter the detection and false positive rates of the Down syndrome screening test. Continuous auditing and training in the correct method for measuring NT is necessary to ensure effective screening for major chromosomal abnormalities.