Abstract
OBJECTIVE: To apply a network medicine-based approach to analyze the phenome of the prenatal fetal MRI and biometric findings in the Chiari II malformation (CM II) to detect specific patterns and co-occurrences. METHOD: A single-center retrospective review of fetal MRI scans obtained in fetuses with CM II was performed. Co-occurrence analysis was utilized to generate a phenotypic comorbidity matrix and visualized by Gephi software. Traditional univariate regression and geometric thin-plate spline methodology were used to elucidate the mechanisms underlying the relationships between morphometric measurements and geometric landmarks of the spine, skull, and brain deformations. RESULTS: The CM II phenome consists of 35 nodes interconnected by 979 edges with a density of 0.828. Key "hubs" identified within this network include spinal bony defects, reduced posterior fossa dimensions, and vermis ectopia. The brain edema phenotype appearing only in the fetal stage but disappearing after postnatal surgery, links to increased postnatal morbidity and demonstrates distinct shape patterns by geometric analysis. Traditional univariate regression reveals correlations among spinal defects, posterior fossa dimensions, and caudal extent of vermis ectopia. The degree of brain rearrangement versus spinal bony rearrangement shows a correlation (r = 0.721, p = 0.0023) by partial least-squares analysis. CONCLUSION: The CM II prenatal phenome is a multifaceted network centered around three key elements-spinal bony defects, small posterior fossa, and vermis ectopia-with strong interconnections. Fetal brain edema emerged as an exclusively prenatally detectable and transient phenotype of prognostic relevance.