Objective: To evaluate the ability to confidently identify intracranial translucency (IT) in a clinical practice and following specific training of 10 operators.
Methods: Two experienced observers reviewed 11-13-week nuchal translucency (NT) images for IT visibility in (1) a series of 50 randomly selected images obtained by 10 skilled operators certified by the Collège Français d'Echographie Foetale (CFEF) (retrospective analysis) and (2) a series of 315 images obtained by 10 different operators following specific training for IT visualization (prospective analysis). We calculated proportions of images for which IT was deemed visible and the agreement between the two observers. Data were also stratified by Herman and CFEF quality-score intervals.
Results: In the retrospective analysis, IT was visualized by both reviewers in 52% of images, with a moderate level of agreement (κ = 0.63). The rate of IT visualization by both reviewers increased very slightly to 56-58% when only considering images with the best NT quality-control scores. Following specific training of the operators the proportion of images for which both reviewers could identify the fourth ventricle increased to 85%, but the level of agreement remained moderate (κ = 0.66). When considering images with the best NT quality-control scores, IT visualization by both reviewers increased to 91-92%.
Conclusions: In a clinical practice that focuses on NT measurement IT cannot be visualized in a substantial proportion of the images obtained, which limits the utility of this approach for the early prenatal diagnosis of open spina bifida. However, the ability to identify the fourth ventricle significantly increases following specific training.
Copyright © 2011 ISUOG. Published by John Wiley & Sons, Ltd.