Single uterine axial fast acquisition magnetic resonance fetal survey: is it feasible?

Abstract

Objectives: To determine which non-biometric components of the ultrasound fetal survey can routinely be seen on a single fast acquisition magnetic resonance (MR) sequence aligned axial to the maternal uterus. Study design: The non-biometric components of the routine fetal ultrasound examination were applied retrospectively to the initial MR single-shot fast spin-echo acquisition aligned axial to the maternal uterus in the normal fetus to determine whether these parameters could be routinely evaluated. Nineteen women with anatomically normal fetuses had a total of 31 MR studies performed for fetal or maternal indications, either as part of an indicated examination or as part of a study protocol approved by the institutional review board. The images in these 31 MR studies were reviewed by two independent examiners who were blinded to the other's assessment; concordance was necessary for a component to be adequately assessed. The Wilcoxon rank sum test was used to determine the effect of gestational age and fetal lie on the ability to assess non-biometric parameters. Results: Assessment was possible in 85% of the non-biometric parameters. Cord insertion, external genitalia and the four-chamber view of the heart were most problematic. In only two cases was the four-chamber view of the heart identified. Longitudinal lie allowed significantly more parameters (82%) to be evaluated than transverse lie (45%) (p < 0.003). No difference based on gestational age was found. Conclusion: A single fast acquisition axial MR sequence can evaluate 85% of the non-biometric components of the fetal ultrasound survey. Fetal lie is an important confounder in the ability to resolve fetal anatomy with a single MR axial uterine acquisition.