Concordance of Prenatal Ultrasonography and Magnetic Resonance Imaging in the Classification of Corpus Callosum Dysplasia

Background: Prenatal ultrasound has always been difficult to classify fetal corpus callosum abnormalities. This paper aims to evaluate the added value of fetal magnetic resonance imaging (MRI) to ultrasound in detecting fetal corpus callosum anomalies and the consistency of the classification of prenatal ultrasound and magnetic resonance imaging for corpus callosum anomalies. Methods: A retrospective analysis was performed of fetuses with abnormal cavum septi pellucidi who had ultrasonography and MRI in utero in Sichuan Maternal and Child Health Hospital and Wenjiang District Maternal in China and Child Health Hospital from January 2018 to December 2023. Fetal corpus callosum anomalies are classified according to the severity of the findings. The classification results of MRI were used as the diagnostic criteria. The findings detected on ultrasound (US) were compared to those detected on MRI. and the receiver operating characteristic (ROC) curve was used to analyze the sensitivity and specificity of prenatal ultrasound in diagnosing corpus callosum anomalies. The Kappa test was used to analyze the consistency of prenatal ultrasound and MRI in the classification of corpus callosum anomalies. Results: Of the 203 cases of fetuses with abnormal cavum septi pellucidi, 143 cases (70.4%) were normal, 34 cases (16.7%) had complete agenesis of the corpus callosum, and 18 cases (8.9%) had partial agenesis of the corpus callosum. Eight cases (3.9%) had dysplasia of the corpus callosum. The area under the ROC curve of prenatal ultrasound for the diagnosis of corpus callosum anomalies was 0.840 (95% confidence interval (CI) 0.782-0.888), the sensitivity was 75.00%, and the specificity was 93.01%, The positive likelihood ratio was 10.7 and the negative likelihood ratio was 0.3. Prenatal ultrasonography and MRI had moderate concordance in the classification of corpus callosum anomalies (Kappa = 0.673, p < 0.001). The concordance was good in diagnosing complete agenesis of the corpus callosum (Kappa = 0.862, p < 0.001), while the concordance was moderate in diagnosing partial agenesis of the corpus callosum (Kappa = 0.643, p < 0.001). Conclusions: Prenatal ultrasonography is accurate and effective in the diagnosis of corpus callosum anomalies. Magnetic resonance imaging can provide more information in the classification of corpus callosum anomalies and the diagnosis of neurologic complications. The MRI should be combined with ultrasonographic transverse, midsagittal, and coronal views to evaluate the classification of corpus callosum anomalies.