Study Design Differences and Temporal Morphology Bias in Intracranial Aneurysm Rupture Research: Toward Prospectively...

Risk stratification for unruptured intracranial aneurysms largely relies on meta-analyses that synthesize heterogeneous primary data. However, fundamental differences in study design, particularly between cross-sectional rupture-status analyses and prospective natural history cohorts, may influence the reported associations between aneurysm morphology and rupture risk. This umbrella review aimed to determine the extent to which study design affects these associations and whether the current evidence base supports clinically interpretable estimates of rupture risk.

An umbrella review of systematic reviews and meta-analyses (1999-2025) was conducted across PubMed, Embase, and Scopus. Methodological quality was assessed using AMSTAR-2. Effect estimates for aneurysm size, location, and selected morphometric parameters (irregular shape, aspect ratio, size ratio) were descriptively compared between meta-analyses of prospective cohorts (incident rupture risk) and cross-sectional rupture-status studies.

Seventeen meta-analyses encompassing more than 200 000 aneurysms were included. Moderate overlap among studies was observed (corrected covered area 8.9%), with high redundancy among natural history cohorts and minimal overlap in computational modelling studies. Effect estimates for morphological variables were generally larger in cross-sectional rupture-status syntheses than in prospective analyses. However, prospective data incorporating standardized quantitative morphometric assessments were limited, event rates were low, and potential selection bias constrained causal interpretation. Machine learning and radiomics studies predominantly focused on rupture status rather than time-to-event risk and rarely underwent external validation.

Reported associations between aneurysm morphology and rupture vary substantially according to study design. Cross-sectional rupture-status analyses and prospective natural history studies address different clinical questions and should not be interpreted interchangeably. The current evidence base is further constrained by the scarcity of contemporary prospective cohorts integrating standardized morphometric assessment. Advancing clinically meaningful rupture risk stratification will require large-scale prospective natural history studies capable of supporting externally validated, time-to-event prediction models that incorporate morphology in a standardized and biologically interpretable framework.