Diagnostic effectiveness of ultrasound in detection of neural tube defect. The South Wales experience of 2509 scans (1977-1982) in high-risk mothers.

2509 pregnant women were investigated between April 1, 1977, and March 31, 1980 (study 1), and April 1, 1980, and March 31, 1983 (study 2), to assess the effectiveness of diagnostic ultrasound in the diagnosis of neural tube defect (NTD) in women judged to be at high risk of NTD on the basis of raised serum alpha-fetoprotein or family history, or of other congenital malformation. The detection rate (sensitivity) for anencephaly was 100% in both studies. The detection rate for open spina bifida (OSB) improved from 33% in study 1 to 80% in study 2, specificity rose from 96% to 99%, the false-positive rate dropped from 57% to 9%, and the false-negative rate fell from 1% to 0.3%. Diagnostic ultrasound is likely to be conducted with greater vigilance and subjected to greater scrutiny than routine ultrasound because the patients have already been selected as being at high risk, the operators are more experienced, and their apparatus is likely to be more refined than that used for routine ultrasound. It is unlikely that the effectiveness levels for detection of OSB of 80% observed in this study could be sustained if diagnostic ultrasound were to be offered to all pregnant women. Consideration will need to be given to the question of whether an improved OSB detection rate will be worth the additional cost incurred in providing diagnostic ultrasound for every pregnant woman.