Screening and detection of congenital malformation.

Congenital malformations often occur sporadically and unpredictably, and a method of population screening to identify at-risk pregnancies would be of great value. Routine ultrasonography has some appeal in this regard. Unfortunately, the literature indicates a relatively poor sensitivity, with usually 30% or less of malformations identified or suspected by this means. Substantially better results have been described with ultrasonography targeted on the basis of screening ultrasonography or clinical findings, but these studies involve highly selected subjects examined carefully by experienced ultrasonographers, and their results are not relevant to population screening. A questionnaire survey of experts in the field indicated disagreement with the separation of screening and targeted ultrasonography, widely varying estimates of the accuracy of either, and average changes of $209 and $263, respectively. Theoretical considerations of the various types of malformations, their frequencies, and the likelihood of ultrasonographic detection reveal a minority of defects in which ultrasonography is both accurate and useful clinically. Ultrasonography seems to be of little potential as a means of screening for congenital malformation because of relatively low sensitivity and uncertain cost effectiveness.