Visualization of the fetal lip and palate: is brain-targeted MRI reliable?

Objective : To evaluate the ability of brain-targeted magnetic resonance imaging (MRI) to assess the anatomy of the fetal upper lip and palate. Design : Two independent readers made a blind retrospective review of 60 brain-targeted MRIs of fetuses from 20 to 38 gestational weeks (GW). Fifty-five MRIs were normal and five had orofacial anomalies, including one isolated cleft lip and four cleft lip and palate. Both normal and cleft MRIs had postnatal confirmation. The upper lip, primary palate, secondary palate, and nasal septum were scored into four levels, from evidently normal to evidently abnormal. In case of a suspected pathology, the readers attempted a diagnosis. Setting : Collaboration between a university hospital and a large private practice MRI center. Results : Interobserver agreement (weighted kappa) was 0.79 for the upper lip, 0.70 for the primary palate, 0.86 for the secondary palate, and 0.90 for the nasal septum. The scoring levels of the readers did not change significantly across gestational age. Normality was correctly scored in 96% to 100% of the normal lips and primary palates and in 93% to 97% of the normal secondary palates depending on the reader. A deviated septum was only scored in two fetuses with unilateral cleft palates. The readers identified all pathological cases. Conclusion : Brain-targeted fetal MRI in experienced hands seems to be highly accurate for the evaluation of the lip and palate in fetuses above 20 GW, regardless of gestational age. The assessment of the secondary palate may be slightly more limited than the lip or primary palate.