Is measurement of nuchal translucency thickness a useful screening tool for heart defects? A study of 16 383 fetuses

Abstract

To determine the performance of nuchal translucency thickness (NT) measurement as a screening method for congenital heart defects (CHD) among fetuses with normal karyotype. An NT measurement was made in 16 383 consecutive euploid fetuses derived from an unselected pregnant population. The cut-offs for increased risk of heart defects, chosen a priori and tested prospectively, were: NT >or= 95th centile for crown-rump length, NT >or= 3 mm, and NT >or= 3.5 mm. The sensitivity and false-positive rate (FPR; 1 minus specificity) of the risk cut-offs and their positive and negative likelihood ratios (+LR and -LR) with regard to CHD were calculated. Among the 16 383 fetuses with an NT measurement there were 127 cases with a diagnosis of heart defect confirmed by cardiac investigations after birth or at autopsy. Of these, 55 defects were defined as major, of which 52 were isolated (no other defects or chromosomal aberrations), corresponding to a prevalence of major heart defects in chromosomally normal fetuses/newborns of 3.3/1000. The sensitivity, FPR, +LR and -LR for NT >or= 95th centile with regard to an isolated major heart defect were: 13.5%, 2.6%, 5.2 and 0.9, respectively. For NT >or= 3.0 mm these values were: 9.6%, 0.8%, 12.0 and 0.9, and for NT >or= 3.5 mm they were: 5.8%, 0.3%, 19.3 and 0.9. NT measurement is a poor screening method for isolated major CHD. A method with a much higher detection rate and with a reasonably low FPR is needed. However, increased NT indicates increased risk of fetal heart defect, and women carrying fetuses with increased NT should be offered fetal echocardiography in the second trimester.