Fetal nuchal translucency can be measured in most pregnant women in the first and early second trimester. The size of translucency varies slightly with gestational age and crown rump length and is independent of maternal age. Most authors have used a nuchal thickness of > or = 2.5 mm or > or = 3 mm to define abnormal, although some have suggested that the normal variation with gestation requires that different thresholds be used at different gestational ages. The accuracy of nuchal translucency measurement varies between examiners and between patients, likely in relation to examiner skill and image resolution. The small size of a nuchal translucency, less than 3 mm in most cases, probably approximates the threshold of normal interexaminer and intraexaminer variability. The presence of a thickened nuchal translucency is associated with chromosomal abnormality and perhaps with structural abnormality even when the karyotype is normal. Because of the reported variations in the populations studied, the methods used, and the results of screening, it is inappropriate at this time to assign a numeric risk to any individual patient with this finding. However, in both high-risk and low-risk groups, the positive predictive value appears to be high enough that patients with increased nuchal translucency should be counseled by their obstetrician and prenatal diagnostic testing should be offered. Because early genetic diagnosis by CVS has a substantially higher procedure-associated loss rate than amniocentesis in the second trimester, many patients may elect to wait for chromosomal testing. If so, disappearance of nuchal thickening should not be taken as reassurance. As a screening test to be widely applied to a general or low-risk population, the utility of fetal nuchal translucency measurement is uncertain. The reported sensitivity for identification of trisomy 21 has ranged from about 40% to 80%, and the sensitivity for identification of other aneuploidies may be lower than for Down's syndrome. From a cost-risk-benefit standpoint, universal first-trimester ultrasound screening has not been appropriately compared with standard risk assessment using maternal age and multiple-marker serum screening, with amniocentesis as the predominant diagnostic method. Also, the issues of availability and reimbursement have not been addressed. Currently, measurement of nuchal translucency is not a substitute for the standard of obstetrical care, which is to offer multiple-marker serum screening to every pregnant woman at 15 to 20 weeks. Similarly, it is inappropriate to substitute nuchal translucency measurement for genetic counseling and CVS or amniocentesis in women above 35 years of age or those with a significant positive history. Finally, the data are not clear as to whether a normal nuchal translucency decreases the likelihood of chromosomal abnormality in a high-risk population, and such women should not be discouraged from invasive testing because of a normal first-trimester ultrasound study. The data supporting the association between thickened nuchal transluency and chromosomal abnormality are compelling, but further study is needed before adopting routine nuchal translucency screening. Combining first-trimester ultrasonography with early serum screening is currently being investigated and may ultimately prove to be the most efficient means of screening for chromosomal anomaly.