Consensus and persistent uncertainty in fetal ventriculomegaly: Delphi study mapping unresolved clinical domains and research priorities

OBJECTIVES: Fetal ventriculomegaly (VM) is a common prenatal finding with heterogeneous etiology and unpredictable outcome, yet its management remains among the least standardized in fetal medicine. This study aimed to establish expert consensus on key decision points in the prenatal management of fetal VM, and to systematically map the domains of persistent uncertainty as a structured framework for future research priorities. METHODS: A two-round modified Delphi study was conducted with an international panel of experts in fetal medicine, selected through a systematic review of publications that included more than 50 cases of VM over the preceding decade. Round 1 comprised 39 multiple-choice questions, with consensus achieved when ≥ 75% of panelists selected a single answer option; items that did not reach consensus were reformulated into Likert-scale statements for Round 2, which comprised 60 statements rated on a 9-point Likert scale. In Round 2, consensus was defined by two simultaneous criteria: ≥ 75% of valid respondents scoring in the agreement (score, 7-9) or disagreement (score, 1-3) zone, with an interquartile range of ≤ 2. Non-consensus domains were classified using a post-hoc taxonomic framework into three categories: absent evidence (Type 1), divergent clinical frameworks (Type 2) and undefined construct (Type 3), which could occur independently or in combination. RESULTS: Of 56 experts who completed Round 1, 45 (80%) also completed Round 2. In Round 1, consensus was reached on 20/39 (51%) questions. In Round 2, 6/60 (10%) statements reached positive consensus and four (7%) reached negative consensus; 50 (83%) did not reach consensus. Over the two rounds, consensus was achieved on standardized measurement technique, three-tier grading, comprehensive neurosonographic and investigative work-up, individualization of the timing of fetal magnetic resonance imaging, stepwise neurosurgical referral and delivery management of fetal VM. Classification of non-consensus domains identified 15/50 (30%) items as Type 1, 16/50 (32%) as Type 2, 4/50 (8%) as Type 3 and 15/50 (30%) as Type 1+3. Ventricular asymmetry, the largest non-consensus cluster with 12 items, was classified as Type 1+3 in 11/12 (92%) statements, reflecting the simultaneous absence of prospective outcome data and of a universally accepted operational definition. CONCLUSIONS: This Delphi study provides structured expert-derived guidance, primarily applicable to specialist and referral settings, on the prenatal management of fetal VM across key clinical domains. Beyond consensus recommendations, we propose that the original contribution of this study is a hypothesis-generating taxonomic map of persistent uncertainty, classifying unresolved domains by the nature of the underlying evidence gap and identifying the specific prerequisites (definitional, methodological or empirical) required before consensus can be achieved. © 2026 International Society of Ultrasound in Obstetrics and Gynecology.