Falcine Sinus Diameter on Fetal MRI Predicts Neonatal Outcomes in Vein of Galen Malformations

Introduction/Purpose Recent studies have reported that mediolateral (ML) diameter of the falcine sinus (FS) on fetal MRI is highly predictive of outcomes in patients with vein of Galen malformations (VOGM). We aimed to assess the discriminative ability of ML diameter for clinical severity and mortality for neonates born at our institution. Materials and Methods All VOGM cases from October 2014 to August 2025 were reviewed, and cases in which the neonate was delivered at our institution were assessed for inclusion. Cases were divided into severity groups, with the neonatal‐at‐risk (NAR) group inclusive of cases in which urgent neonatal intervention was performed or treatment not offered due to poor prognosis, and infantile treatment (IT) groups, comprised of patients who were successfully discharged with planned first embolization as an infant. Two independent observers measured maximum ML diameters of the FS at the narrowest point in the axial and coronal plane on standard fetal MRI sequences. Agreement was assessed using Bland‐Altman analysis, reporting mean bias and 95% limits of agreement (LoA). Receiver operating characteristic (ROC) analyses was used to assess the discriminative ability of ML measurements for severity classification and mortality, with area under the curve (AUC) values and 95% confidence intervals reported. Results Of 86 VOGM cases, 57 (66.3%) were diagnosed antenatally and 21 (15 NAR, 6 IT) with fetal MRI and delivery at our institution were analyzed. There were 5 mortalities in the NAR group, 2 following neonatal embolization. Axial ML diameters were lower than coronal for both readers (Reader 1 bias ‐0.8 mm, 95% LoA ‐3.1 to 1.7; Reader 2 bias ‐0.5 mm, ‐2.5 to 1.5), and between‐reader bias was 0.4 mm (‐2.3 to 3.2) for axial and 0.6 mm (‐4.0 to 2.6) for coronal. ROC analysis demonstrated excellent discrimination of ML diameter for NAT versus IT classification (AUC = 0.9, 95% CI 0.8 ‐ 1.0) but poor discrimination for mortality (AUC = 0.6, 95% CI 0.3‐0.9). Conclusion In our cohort, ML diameter demonstrated discriminative ability for NAT versus IT classification, consistent with prior reports, but showed poor discrimination for mortality. Prior reports of 100% mortality with FS diameters >9 mm were not observed, underscoring the need for caution when applying rigid cutoffs. Coronal measurements were on average larger than axial, with small but measurable bias between planes and readers. Larger datasets will be required to establish reliable prognostic markers to assist with counseling and treatment planning.