Predictive Accuracy and Procedural Benefits of Software-Enhanced Flow-Diverter Planning: A Systematic Review and Meta-Analysis.

Abstract

Background: Flow-diverting (FD) stents play a critical role in the management of complex intracranial aneurysms. Successful FD stent deployment requires precise planning to minimize peri- and postprocedural complications. This traditionally relied on physician expertise for precise planning; however, software simulations now offer estimative modeling.

Purpose: This systematic review and meta-analysis compares physician and software simulations for predicting stent deployment and assesses the clinical impact of software integration on procedural metrics.

Data sources: We searched PubMed, Embase, Web of Science, and Scopus (articles through July 18, 2025) using Nested Knowledge Autolit. The search combined Medical Subject Headings (MeSH) terms and keywords and followed updated Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) guidelines. A manual review of the reference lists was performed to identify gray literature.

Study selection: Fourteen studies were included. All studies that involved comparative data, a physician, and software/simulation tool regarding intracranial aneurysms with outcomes relating to accuracy metrics, interobserver agreement or apposition, or procedural metrics were included.

Data analysis: All data were analyzed using R, Version 4.5.2, and the "meta" package. Risk of bias (RoB) assessment was completed using the ROBINS-I tool for procedural metric-related studies and the COSMIN-based checklist for accuracy/agreement-related studies.

Data synthesis: The pooled analysis found software-predicted stent length to be shorter than the physicians' predicted length (mean difference [MD], 2.88 mm; 95% CI, 1.25-4.51), without a statistically significant difference in diameter (MD, 0.00 mm; 95% CI, -0.09 to 0.09); software-chosen dimensions were accurate to the postimplantation length (MD, -1.40 mm; 95% CI, -3.56 to 0.76) and diameter (MD, 0.17 mm; 95% CI, -0.63 to 0.97); software-assisted cases were associated with shorter procedure duration (MD, 9.91 min; 95% CI, 1.23-18.58), lower radiation dose (MD 305.52 mGy [172.14, 438.90]), and fewer corrective interventions (OR 3.44 [1.78, 6.64]); and there was no significant difference in hemorrhagic complications (OR 1.67 [0.65, 4.28]) and postoperative aneurysm occlusion (OR 0.52 [0.07, 3.78]).

Limitations: Given the retrospective nature and bias limitations of current literature, further prospective, randomized studies are warranted to better assess the clinical impact of simulation software in the endovascular treatment of intracranial aneurysms.

Conclusions: The findings in this paper provide preliminary evidence to support integration of virtual simulation software in the preprocedural FD treatment planning.