Abstract 132: Prediction of Long‐term Treatment Failure of Intracranial Aneurysms Treated With Woven EndoBridge Device: A Multi‐center Study
Bibliographic record
Abstract
Introduction Endovascular treatment of intracranial aneurysms with flow disruption devices like the Woven EndoBridge (WEB) is associated with variable occlusion success. Identifying determinants of long‐term treatment failure is vital for optimizing patient selection. This study aimed to determine morphological and clinical factors associated with treatment failure following WEB deployment for cerebral aneurysms. Methods This retrospective multicenter study analyzed intracranial aneurysm cases treated with WEB devices between 2011‐2022. Treatment failure, the primary outcome, was defined as persistent incomplete occlusion (Raymond‐Roy Occlusion Classification [RROC] grades 2/3) on final angiographic follow‐up, excluding cases with stable RROC grade 1 or worsening RROC grades. Follow‐up was categorized as short/midterm (<24 months) and long‐term (≥24 months). Univariate analyses assessed differences in baseline factors between failure and success groups. Machine learning models, including a CatBoost classifier, were developed to predict long‐term treatment failure and optimized using SHAP values. Multivariable logistic regression was performed to calculate odds ratios (OR) for factors associated with treatment failure in both follow‐up cohorts. Results Of the 813 cases, 206 (25.3%) had long‐term follow‐up (≥24 months). Treatment failure occurred in 210 (26%) cases based on persistent incomplete occlusion at final follow‐up. The CatBoost model achieved an AUC of 0.67 for predicting treatment failure, with the most influential predictors being aneurysm height, anterior communicating artery aneurysms, aneurysm‐neck‐diameter, age, worse pretreatment‐modified Rankin Scale scores, hemorrhagic complications, and WEB type DL. Multivariable logistic regression analysis revealed that antiplatelet therapy (OR 1.65, 95% CI [0.39, 6.96]; p=0.498) and compaction (minor/major) (OR 1.81, 95% CI [0.58, 5.57]; p=0.304) were not significantly associated with treatment failure at ≥ 24 months. Immediate flow stagnation was also not significant (OR 0.44, 95% CI [0.14, 1.38]; p=0.159). Aneurysm width (OR 1.31, 95% CI [0.91, 1.88]; p=0.144) and neck diameter (OR 1.24, 95% CI [0.83, 1.86]; p=0.286) did not show significant associations with treatment failure. Posterior circulation aneurysms were protective against failure at <24 months (OR 0.45, 95% CI [0.24, 0.81]; p=0.008). Older age was significantly associated with a reduced risk of treatment failure at ≥24 months (OR 0.93, 95% CI [0.88, 0.98]; p=0.005). Ruptured aneurysms and smoking were not significant predictors of treatment failure. Conclusion This study demonstrates that aneurysm morphology and patient characteristics influence long‐term outcomes with the WEB device. The machine learning model offers moderate predictive accuracy for treatment failure, while logistic regression identified key factors influencing persistent incomplete occlusion, particularly aneurysm location and patient age. These findings could inform patient selection and treatment strategies, although further external validation is required to confirm their clinical implications.
Fetched live from OpenAlex and de-inverted. Abstracts are not stored in this database: the inverted indexes are 8.6 GB of the frame’s 9.3 GB of text, and the host has 13 GB free.
How this classification was reachedexpand
Full frame machine prediction
Teacher imitationNot calibrated prevalence, not ground truth. Human validation pending. The Gemma side is a direct model label for every work in the frame, read from the title-only record. The Codex side is a classifier learned from the 10,348 direct Codex labels and calibrated to design-weighted sample rates; fields without enough sample support carry no Codex call. Candidate is the union of the two sides; consensus is their intersection. These outputs are machine_predicted_unvalidated and are not human labels.
Distilled classifier scores by category (both heads)
| Category | Codex | Gemma |
|---|---|---|
| Metaresearch | 0.002 | 0.003 |
| Meta-epidemiology (narrow) | 0.000 | 0.000 |
| Meta-epidemiology (broad) | 0.000 | 0.001 |
| Bibliometrics | 0.001 | 0.001 |
| Science and technology studies | 0.000 | 0.000 |
| Scholarly communication | 0.001 | 0.001 |
| Open science | 0.001 | 0.001 |
| Research integrity | 0.001 | 0.000 |
| Insufficient payload (model declined to judge) | 0.002 | 0.000 |
Machine scores (provisional)
The two teacher heads of the student model, read on this work. A score orders the frame for review; it never asserts a category, and the validation status ships verbatim with every row.
Baseline scores from an immature model (maturity gate not passed, 7 training rounds). Scores rank; they never assert a category.
score_only:v0-immature-baseline · verbatim from the scoring run: score_only means the number may rank works, and no category label ships from itClassification
machine, unvalidatedMachine predicted; a candidate call from one source (direct Gemma or distilled Codex), not a consensus.
How this classification was reached, model by model and score by score, is at the end of the page under "How this classification was reached".