Ostium Ratio and Neck Ratio Could Predict the Outcome of Sidewall Intracranial Aneurysms Treated with Flow Diverters

Correlation of Flow Diverter Malapposition at the Aneurysm Neck with Incomplete Aneurysm Occlusion in Patients with Small Intracranial Aneurysms: A Single-Center Experience

BACKGROUND AND PURPOSE

Flow diversion treatment repairs aneurysms by altering the hemodynamics of the aneurysmal sac and providing a scaffold for endothelial cell adhesion. The purpose of this study was to investigate the correlation of flow diverter (FD) malapposition at the aneurysm neck with incomplete occlusion of small intracranial aneurysms (IAs) and investigate other factors that are possibly related to incomplete occlusion.

MATERIALS AND METHODS

From January 2019 to June 2022, the clinical and imaging data for 153 patients (175 aneurysms) with unruptured small IAs treated with flow diversion were retrospectively analyzed. FD apposition at the aneurysm neck was evaluated by high-resolution conebeam CT (HR-CBCT), and the complete occlusion rate for aneurysms was judged according to the latest follow-up conventional angiography findings (≥6 months). Multivariate logistic regression analysis was used to determine factors associated with incomplete aneurysm occlusion.

RESULTS

In total, 159 FDs were implanted in 153 patients. HR-CBCT performed after the deployment revealed FD malapposition at the aneurysm neck in 18 cases. According to the latest follow-up angiograms (average: 9.47 ± 3.35 months), the complete aneurysm occlusion rate was 66.9%. The complete occlusion rates for incomplete and complete stent apposition at the neck were 38.9% (7/18) and 70.1% (110/157), respectively. The results of regression analysis showed that an aneurysm sac with branch vessels (OR, 2.937; P = .018), incomplete stent apposition at the aneurysm neck (OR, 3.561; P = .023), and a large aneurysm diameter (OR, 1.533; P = .028) were positive predictors of incomplete aneurysm occlusion.

CONCLUSIONS

An aneurysm sac with branch vessels, a large aneurysm diameter, and malapposition at the aneurysm neck significantly affect aneurysm repair after FD stent-only treatment for small IAs.

Aneurysm Isolation is Associated with Complete Occlusion of Aneurysms After Flow Diverter Treatment

PURPOSE

Some aneurysms remain patent after treatment with flow diverters (FD) due to residual blood flow in the aneurysm. Several studies have proposed that branches and residual flow are associated with delayed aneurysm occlusion. We propose that aneurysm isolation (i.e., the complete disconnection of the aneurysm from surrounding vessels) might be a possible factor facilitating aneurysm occlusion. This study aimed to determine if aneurysm isolation was a factor associated with aneurysm occlusion after FD treatment.

METHODS

We reviewed 80 internal carotid artery (ICA) aneurysms treated with FDs between October 2014 and April 2021. Aneurysm isolation was assessed in high-resolution cone-beam computed tomograms at the end of each treatment. Aneurysms with incorporated branches and those with connections to other branches due to stent malapposition were deemed to be nonisolated. Other factors, such as patient age, sex, anticoagulant use, aneurysm size, adjunct coil use, and the presence of incorporated branches, were considered. The degree of aneurysm occlusion (complete or incomplete) was assessed by follow-up angiograms 12 months after treatment.

RESULTS

Complete occlusion was achieved in 57 of 80 aneurysms (71%). Completely occluded aneurysms had a significantly higher ratio of isolation compared to incompletely occluded aneurysms (91.2% vs. 69.6%, P = 0.032). Multivariate logistic regression analysis showed that aneurysm isolation was the sole significant predictor of complete aneurysm occlusion (odds ratio, OR 19.38; 95% confidence interval, CI 2.280-164.657; P = 0.007).

CONCLUSION

Aneurysm isolation is a significant factor contributing to complete occlusion after FD treatment.

Persistent Opacification of the Woven EndoBridge Device: A Conebeam CT Analysis of the Bicêtre Occlusion Scale Score 1 Phenomenon

BACKGROUND AND PURPOSE

Some Woven EndoBridge devices present a persistent intradevice opacification at imaging follow-up, described as the Bicêtre Occlusion Scale Score 1 (BOSS 1) phenomenon. The clinical implications remain unknown. We aimed here to analyze the factors influencing this occurrence and to precisely describe the evolution of BOSS 1 with time using conebeam CT.

MATERIALS AND METHODS

We retrospectively analyzed a prospectively maintained Woven EndoBridge database at our tertiary center and included all patients with isolated BOSS 1 and BOSS 1 associated with small neck remnant (BOSS 1 + 2).

RESULTS

Two hundred sixty-seven aneurysms were treated with a Woven EndoBridge device between July 2012 and December 2021. Follow-up with DSA was available for 220 aneurysms (median, 5 months), among which BOSS 1 and 1 + 2 were found in 9.1% (20/220) (95% CI, 5.5%-12.7%). A second DSA follow-up (median, 17 months) was performed in 15 of these 20 aneurysms, which revealed that 40% had evolved to complete Woven EndoBridge occlusion, 33% showed a decreased persistent opacification, and 27% remained stable. BOSS 1 was significantly associated with postoperative antiplatelet medication, a lower aneurysm aspect ratio, and the use of the Woven EndoBridge 17 (P  < .05). The average Woven EndoBridge shape modification was less pronounced in the BOSS 1 population (P  < .02). None of the BOSS 1 or 1 + 2 aneurysms required retreatment or were associated with hemorrhage occurrence.

CONCLUSIONS

Isolated persistent flow inside the Woven EndoBridge device at follow-up is rare and notably associated with antiplatelet prescription. It seems to present a benign course in most cases.

Rupture discrimination of multiple small (< 7 mm) intracranial aneurysms based on machine learning-based cluster analysis

BACKGROUND

Small multiple intracranial aneurysms (SMIAs) are known to be more prone to rupture than are single aneurysms. However, specific recommendations for patients with small MIAs are not included in the guidelines of the American Heart Association and American Stroke Association. In this study, we aimed to evaluate the feasibility of machine learning-based cluster analysis for discriminating the risk of rupture of SMIAs.

METHODS

This multi-institutional cross-sectional study included 1,427 SMIAs from 660 patients. Hierarchical cluster analysis guided patient classification based on patient-level characteristics. Based on the clusters and morphological features, machine learning models were constructed and compared to screen the optimal model for discriminating aneurysm rupture.

RESULTS

Three clusters with markedly different features were identified. Cluster 1 (n = 45) had the highest risk of subarachnoid hemorrhage (SAH) (75.6%) and was characterized by a higher prevalence of familiar IAs. Cluster 2 (n = 110) had a moderate risk of SAH (38.2%) and was characterized by the highest rate of SAH history and highest number of vascular risk factors. Cluster 3 (n = 505) had a relatively mild risk of SAH (17.6%) and was characterized by a lower prevalence of SAH history and lower number of vascular risk factors. Lasso regression analysis showed that compared with cluster 3, clusters 1 (odds ratio [OR], 7.391; 95% confidence interval [CI], 4.074-13.150) and 2 (OR, 3.014; 95% CI, 1.827-4.970) were at a higher risk of aneurysm rupture. In terms of performance, the area under the curve of the model was 0.828 (95% CI, 0.770-0.833).

CONCLUSIONS

An unsupervised machine learning-based algorithm successfully identified three distinct clusters with different SAH risk in patients with SMIAs. Based on the morphological factors and identified clusters, our proposed model has good discrimination ability for SMIA ruptures.

EFFECTS OF THE ANEURYSM TREATMENT METHOD ON COIL PACKING DENSITY AND ITS RELATIONSHIP WITH THE OSTIUM AREA

PURPOSE

To test the following hypotheses: (1) balloon or stent assistance increases coil packing density (CPD) in the endovascular treatment of intracranial aneurysms and (2) CPD is related to the ostium area and volume of the aneurysm.

MATERIALS AND METHODS

This retrospective study included 60 ruptured and unruptured (54 and 6, respectively) aneurysms treated between August 2017 and December 2019 in the authors' institution with simple coiling (n=18), balloon-assisted coiling (n=7), or stent-assisted coiling (n=35). Aneurysm volume and ostium area measurements were obtained using commercial software from three-dimensional digital subtraction angiography images. Coil sizes were retrieved from patient files, and coil volume measurements were obtained using https://www.angiocalc.com/. Analysis of covariance, multivariable covariance analysis, and Pearson's correlation analyses were performed.

RESULTS

The median values for the aneurysm volume, coil volume, CPD, and ostium area were 63.4 (range 5.5-1771.4) mm³, 23.13 (2.03-296.95) mm³, 33.29% (13.41%-81.02%), and 10.7 (2.7-49.9) mm², respectively. Multivariate analysis showed that the CPD values were not significantly different among the treatment groups, although the ostium area differed significantly between the simple and stent-assisted coiling groups (P<.05). Pearson's correlations showed that similar to aneurysm volume, the ostium area was negatively correlated with CPD (r=-0.321, P<.05).

CONCLUSION

In cerebral aneurysms treated with balloon- or stent-assisted coiling, the CPD value does not differ from that in aneurysms treated with simple coiling.

Automated Machine Learning Model Development for Intracranial Aneurysm Treatment Outcome Prediction: A Feasibility Study

Background: The prediction of aneurysm treatment outcomes can help to optimize the treatment strategies. Machine learning (ML) has shown positive results in many clinical areas. However, the development of such models requires expertise in ML, which is not an easy task for surgeons.

Objectives: The recently emerged automated machine learning (AutoML) has shown promise in making ML more accessible to non-computer experts. We aimed to evaluate the feasibility of applying AutoML to develop the ML models for treatment outcome prediction.

Methods: The patients with aneurysms treated by endovascular treatment were prospectively recruited from 2016 to 2020. Treatment was considered successful if angiographic complete occlusion was achieved at follow-up. A statistical prediction model was developed using multivariate logistic regression. In addition, two ML models were developed. One was developed manually and the other was developed by AutoML. Three models were compared based on their area under the precision-recall curve (AUPRC) and area under the receiver operating characteristic curve (AUROC).

Results: The aneurysm size, stent-assisted coiling (SAC), and posterior circulation were the three significant and independent variables associated with treatment outcome. The statistical model showed an AUPRC of 0.432 and AUROC of 0.745. The conventional manually trained ML model showed an improved AUPRC of 0.545 and AUROC of 0.781. The AutoML derived ML model showed the best performance with AUPRC of 0.632 and AUROC of 0.832, significantly better than the other two models.

Conclusions: This study demonstrated the feasibility of using AutoML to develop a high-quality ML model, which may outperform the statistical model and manually derived ML models. AutoML could be a useful tool that makes ML more accessible to the clinical researchers.

Prediction of atherosclerotic changes in cavernous carotid aneurysms based on computational fluid dynamics analysis: a proof-of-concept study

PURPOSE

Recent computational fluid dynamics (CFD) studies have demonstrated the concurrence of atherosclerotic changes in regions exposed to prolonged blood residence. In this proof-of-concept study, we investigated a small but homogeneous cohort of large, cavernous carotid aneurysms (CCAs) to establish the clinical feasibility of CFD analysis in treatment planning, based on the association between pathophysiology and hemodynamics.

METHODS

This study included 15 patients with individual large CCAs. We identified calcifications, which indicated atherosclerotic changes, using the masking data of digital subtraction angiography. We conducted a CFD simulation under patient-specific inlet flow rates measured using magnetic resonance (MR) velocimetry. In the post-CFD analysis, we calculated the blood residence time ([Formula: see text]) and segmented the surface exposed to blood residence time over 1 s ([Formula: see text]). We measured the decrease in volume after flow diversion using the original time-of-flight MR angiography data.

RESULTS

Calcifications were observed in the region with [Formula: see text]. In addition, the ratio of [Formula: see text] to the surface of the aneurysmal domain exhibited a negative relationship with the rate of volume reduction at the 6- and 12-month follow-ups. Post-CFD visualization demonstrated that intra-aneurysmal swirling flow prolonged blood residence time under the condition of a small inlet flow rate, when compared to the aneurysmal volume.

CONCLUSION

The results of this study suggest the usefulness of CFD analysis for the diagnosis of atherosclerotic changes in large CCAs that may affect the therapeutic response after flow diversion.

Definition and extraction of 2D shape indices of intracranial aneurysm necks for rupture risk assessment

Purpose

Intracranial aneurysms are local dilations of brain vessels. Their rupture, as well as their treatment, is associated with high risk of morbidity and mortality. In this work, we propose shape indices for aneurysm ostia for the rupture risk assessment of intracranial aneurysms.

Methods

We analyzed 84 middle cerebral artery bifurcation aneurysms (27 ruptured and 57 unruptured) and their ostia, with respect to their size and shape. We extracted 3D models of the aneurysms and vascular trees. A semi-automatic approach was used to separate the aneurysm from its parent vessel and to reconstruct the ostium. We used known indices to quantitatively describe the aneurysms. For the ostium, we present new shape indices: the 2D Undulation Index (UI\documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$_\mathrm{2D}$$\end{document}2D), the 2D Ellipticity Index (EI\documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$_\mathrm{2D}$$\end{document}2D) and the 2D Noncircularity Index (NCI\documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$_\mathrm{2D}$$\end{document}2D). Results were analyzed using the Student t test, the Mann–Whitney U test and a correlation analysis between indices of the aneurysms and their ostia.

Results

Of the indices, none was significantly associated with rupture status. Most aneurysms have an NCI\documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$_\mathrm{2D}$$\end{document}2D below 0.2. Of the aneurysms that have an NCI\documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$_\mathrm{2D}$$\end{document}2D above 0.5, only one is ruptured, which indicates that ruptured aneurysms often have a circular-shaped ostium. Furthermore, the ostia of ruptured aneurysms tend to have a smaller area, which is also correlated with the aneurysm’s size. While also other variables were significantly correlated, strong linear correlations can only be seen between the area of the ostium with the aneurysm’s volume and surface.

Conclusion

The proposed shape indices open up new possibilities to quantitatively describe and compare ostia, which can be beneficial for rupture risk assessment and subsequent treatment decision. Additionally, this work shows that the ostium area and the size of the aneurysm are correlated. Further longitudinal studies are necessary to analyze whether stable and unstable aneurysms can be distinguished by their ostia.

Evaluation of Two Fast Virtual Stenting Algorithms for Intracranial Aneurysm Flow Diversion

BACKGROUND

Endovascular treatment of intracranial aneurysms (IAs) by flow diverter (FD) stents depends on flow modification. Patient-specific modeling of FD deployment and computational fluid dynamics (CFD) could enable a priori endovascular strategy optimization. We developed a fast, simplistic, expansion-free balls-weeping algorithm to model FDs in patientspecific aneurysm geometry. However, since such strong simplification could result in less accurate simulations, we also developed a fast virtual stenting workflow (VSW) that explicitly models stent expansion using pseudo-physical forces.

METHODS

To test which of these two fast algorithms more accurately simulates real FDs, we applied them to virtually treat three representative patient-specific IAs. We deployed Pipeline Embolization Device into 3 patient-specific silicone aneurysm phantoms and simulated the treatments using both balls-weeping and VSW algorithms in computational aneurysm models. We then compared the virtually deployed FD stents against experimental results in terms of geometry and post-treatment flow fields. For stent geometry, we evaluated gross configurations and porosity. For post-treatment aneurysmal flow, we compared CFD results against experimental measurements by particle image velocimetry.

RESULTS

We found that VSW created more realistic FD deployments than balls-weeping in terms of stent geometry, porosity and pore density. In particular, balls-weeping produced unrealistic FD bulging at the aneurysm neck, and this artifact drastically increased with neck size. Both FD deployment methods resulted in similar flow patterns, but the VSW had less error in flow velocity and inflow rate.

CONCLUSION

In conclusion, modeling stent expansion is critical for preventing unrealistic bulging effects and thus should be considered in virtual FD deployment algorithms. Also endowed with its high computational efficiency and superior accuracy, the VSW algorithm is a better candidate for implementation into a bedside clinical tool for FD deployment simulation.

VICTORIA: VIrtual neck Curve and True Ostium Reconstruction of Intracranial Aneurysms

Purpose

For the status evaluation of intracranial aneurysms (IAs), morphological and hemodynamic parameters can provide valuable information. For their extraction, a separation of the aneurysm sac from its parent vessel is required that yields the neck curve and the ostium. However, manual and subjective neck curve and ostium definitions might lead to inaccurate IA assessments.

Methods

The research project VICTORIA was initiated, allowing users to interactively define the neck curve of five segmented IA models using a web application. The submitted results were qualitatively and quantitatively compared to identify the minimum, median and maximum aneurysm surface area. Finally, image-based blood flow simulations were carried out to assess the effect of variable neck curve definitions on relevant flow- and shear-related parameters.

Results

In total, 55 participants (20 physicians) from 18 countries participated in VICTORIA. For relatively simple aneurysms, a good agreement with respect to the neck curve definition was found. However, differences among the participants increased with increasing complexity of the aneurysm. Furthermore, it was observed that the majority of participants excluded any small arteries occurring in the vicinity of an aneurysm. This can lead to non-negligible deviations among the flow- and shear-related parameters, which need to be carefully evaluated, if quantitative analysis is desired. Finally, no differences between participants with medical and non-medical background could be observed.

Conclusions

VICTORIAs findings reveal the complexity of aneurysm neck curve definition, especially for bifurcation aneurysms. Standardization appears to be mandatory for future sac-vessel-separations. For hemodynamic simulations a careful neck curve definition is crucial to avoid inaccuracies during the quantitative flow analysis.

Supplementary Information

The online version contains supplementary material available at 10.1007/s13239-021-00535-w.

Woven EndoBridge (WEB) Width at the Aneurysm Neck Level Affects Early Angiographic Aneurysm Occlusion

Purpose

Endovascular therapy with the Woven EndoBridge (WEB) device is a safe treatment approach, whereby neoendothelialization at the neck area is a crucial element for aneurysm occlusion. We hypothesized that WEB sizing at the aneurysmal neck level has an impact on early aneurysm occlusion.

Methods

Patients with short-term follow-up digital subtraction angiography following WEB treatment of unruptured aneurysms were included. Aneurysms were categorized according to the Bicêtre Occlusion Scale Score (BOSS) as adequately (BOSS 0, 0′, 1) or partially occluded (BOSS 2, 3, 1 + 3). The WEB device dimensions, including the average aneurysm diameter (AADi) and the average neck diameter (ANDi) as well as baseline patient characteristics were documented.

Results

In this study 75 patients with 76 aneurysms were included and 65 aneurysms showed adequate occlusion at short-term follow-up (86%). In univariable logistic regression analysis, smaller differences in WEB size to ANDi (D-ANDi) were significantly associated with adequate aneurysm occlusion (odds ratio, OR = 0.41, 95% confidence interval, CI 0.23–0.71, p = 0.002). Receiver operating characteristic (ROC) curve analyses displayed higher discriminative power for the D‑ANDi (AUC = 0.77, 95% CI 0.66–0.86, cut-off ≤2.9 mm) compared to the difference in WEB size to the average aneurysm diameter (D-AADi, AUC = 0.65, 95% CI 0.53–0.75, cut-off ≤1.0 mm).

Conclusion

Smaller differences between the WEB width and ANDi were associated with adequate early aneurysm occlusion and might thus have a higher impact on the results than the traditional device sizing considering the mean aneurysm diameter. D‑ANDi ≤2.9 mm served as an optimal cut-off to classify occlusion after WEB treatment at the short-term follow-up. Further external validation is warranted.

Supplementary Information

The online version of this article (10.1007/s00062-021-01034-0) contains supplementary material, which is available to authorized users.

Predictive score for complete occlusion of intracranial aneurysms treated by flow-diverter stents using machine learning

BACKGROUND

Complete occlusion of an intracranial aneurysm (IA) after the deployment of a flow-diverter stent is currently unpredictable. The aim of this study was to develop a predictive occlusion score based on pretreatment clinical and angiographic criteria.

METHODS

Consecutive patients with ≥6 months follow-up were included from 2008 to 2019 and retrospectively analyzed. Each IA was evaluated using the Raymond-Roy occlusion classification (RROC) and dichotomized as occluded (A) or residual (B/C); 80% of patients were randomly assigned to the training sample. Feature selection and binary outcome prediction relied on logistic regression and threshold maximizing class separation selected by a CART tree algorithm. The feature selection was addressed by a genetic algorithm selected from the 30 pretreatment available variables.

RESULTS

The study included 146 patients with 154 IAs. Feature selection yielded a combination of six variables with a good cross-validated accuracy on the test sample, a combination we labeled DIANES score (IA diameter, indication, parent artery diameter ratio, neck ratio, side-branch artery, and sex). A score of more than -6 maximized the ability to predict RROC=A with sensitivity of 87% (95% CI 79% to 95%) and specificity of 82% (95% CI 64% to 96%) in the training sample. Accuracy was 86% (95% CI 79% to 94%). In the test sample, sensitivity and specificity were 89% (95% CI 77% to 98%) and 60% (95% CI 33% to 86%), respectively. Accuracy was 81% (95% CI 69% to 91%).

CONCLUSION

A score was developed as a grading scale for prediction of the final occlusion status of IAs treated with a flow-diverter stent.

Large Neck and Strong Ostium Inflow as the Potential Causes for Delayed Occlusion of Unruptured Sidewall Intracranial Aneurysms Treated by Flow Diverter

BACKGROUND AND PURPOSE

Flow diverter-induced hemodynamic change plays an important role in the mechanism of intracranial aneurysm occlusion. Our aim was to explore the relationship between aneurysm features and flow-diverter treatment of unruptured sidewall intracranial aneurysms.

MATERIALS AND METHODS

MR imaging, 4D phase-contrast, was prospectively performed before flow diverter implantation in each patient with unruptured intracranial aneurysm. Two postprocedure follow-ups were scheduled at 6 and 12 months. Responses were grouped according to whether the aneurysms were occluded or remnant. Preprocedural aneurysm geometries and ostium hemodynamics in 38 patients were compared between the 2 groups at 6 and 12 months. Receiver operating characteristic curve analyses were performed for significant geometric and hemodynamic continuous parameters.

RESULTS

After the 6-month assessment, 21 of 41 intracranial aneurysms were occluded, and 9 additional aneurysms were occluded at 12 months. Geometrically, the ostium maximum diameter was significantly larger in the remnant group at 6 and 12 months (both P < .001). Hemodynamically, the proximal inflow zone was more frequently observed in the remnant group at 6 months. Several preprocedural ostium hemodynamic parameters were significantly higher in the remnant group. As a prediction for occlusion, the areas under the curve of the ostium maximum diameter (for 6 and 12 months), systolic inflow rate ratio (for 6 months), and systolic inflow area (for 12 months) reached 0.843, 0.883, 0.855, and 0.860, respectively.

CONCLUSIONS

Intracranial aneurysms with a large ostium and strong ostium inflow may need a longer time for occlusion. Preprocedural 4D flow MR imaging can well illustrate ostium hemodynamics and characterize aneurysm treatment responses.

2D parametric contrast time-density analysis for the prediction of complete aneurysm occlusion at six months' post-flow diversion stent

OBJECTIVE

Indications for the treatment of cerebral aneurysms with flow diversion stents are expanding. The current aneurysm occlusion rate at six months ranges between 60 and 80%. Predictability of complete vs. partial aneurysm occlusion is poorly defined. Here, we evaluate the angiographic contrast time-density as a predictor of aneurysm occlusion rate at six months' post-flow diversion stents.

METHODS

Patients with unruptured cerebral aneurysms proximal to the internal carotid artery terminus treated with single flow diversion stents were included. 2D parametric parenchymal blood flow software (Siemens-Healthineers, Forchheim, Germany) was used to calculate contrast time-density within the aneurysm and in the proximal adjacent internal carotid artery. The area under the curve ratio between the two regions of interests was assessed at baseline and after flow diversion stents deployment. The area under the curve ratio between completely vs. partially occluded aneurysms at six months' follow-up was compared.

RESULTS

Thirty patients with 31 aneurysms were included. Mean aneurysm diameter was 8 mm (range 2-28 mm). Complete occlusion was obtained in 19 aneurysms. Younger patients (P = 0.006) and smaller aneurysms (P = 0.046) presented higher chance of complete obliteration. Incomplete occlusion of the aneurysm was more likely if the area under the curve contrast time-density ratio showed absolute (P = 0.001) and relative percentage (P = 0.001) decrease after flow diversion stents deployment. Area under ROC curve was 0.85.

CONCLUSION

Negative change in the area under the curve ratio indicates less contrast stagnation in the aneurysm and lower chance of occlusion. These data provide a real-time analysis after aneurysm treatment. If validated in larger datasets, this can prompt input to the surgeon to place a second flow diversion stents.

Aneurysm characteristics, coil packing, and post-coiling hemodynamics affect long-term treatment outcome

BACKGROUND

Recurrence of intracranial aneurysms after endovascular coiling is a serious clinical concern.

OBJECTIVE

We hypothesized that recurrence is associated with aneurysm morphology and flow, as well as the coil intervention and the induced flow modifications.

METHODS

We collected 52 primary-coiling aneurysm cases that were either occluded (n=34) or recurrent (n=18) at >1 year follow-up. We created aneurysm models from pre-coiling digital subtraction angiographic images, calculated aneurysm morphology, simulated pre-coiling hemodynamics, modeled coil deployment, and obtained post-coiling hemodynamics for each case. We performed univariable analysis on 26 morphologic, treatment-specific, and hemodynamic parameters to distinguish between recurrent and occluded groups, and multivariable analysis to identify independently significant parameters associated with recurrence. Univariable analysis was also performed on ruptured and unruptured aneurysm subcohorts separately to investigate if they shared specific significant parameters.

RESULTS

Recurrence was associated with pre-coiling aneurysm morphologic and flow parameters including larger size (maximum dimension and volume), larger neck (diameter, area, and neck-to-parent-artery ratio), and higher flow momentum and kinetic energy. Recurrence was also associated with lower coil packing (packing density and uncoiled volume), higher post-treatment flow (velocity, momentum, and kinetic energy), lower post-treatment washout time, and higher post-treatment impingement force at the neck. Multivariable analysis identified two aneurysmal characteristics (neck diameter and pre-coiling flow kinetic energy), one coil packing parameter (uncoiled volume), and one post-treatment hemodynamic parameter (flow momentum) that were independently associated with recurrence. In ruptured aneurysms, recurrence was associated with larger neck (diameter and area), whereas in unruptured aneurysms, recurrence was associated with larger size (maximum dimension and volume). In both subcohorts, recurrence was associated with higher post-coiling flow momentum and kinetic energy.

CONCLUSION

Recurrence at >1 year after coil treatment is associated with intrinsic aneurysm characteristics, coiling itself, and flow changes induced by coiling. Larger aneurysm size and neck, less coil packing, and higher intra-aneurysmal flow before and after coiling predict recurrence.

Outcome prediction of intracranial aneurysm treatment by flow diverters using machine learning

OBJECTIVEFlow diverters (FDs) are designed to occlude intracranial aneurysms (IAs) while preserving flow to essential arteries. Incomplete occlusion exposes patients to risks of thromboembolic complications and rupture. A priori assessment of FD treatment outcome could enable treatment optimization leading to better outcomes. To that end, the authors applied image-based computational analysis to clinically FD-treated aneurysms to extract information regarding morphology, pre- and post-treatment hemodynamics, and FD-device characteristics and then used these parameters to train machine learning algorithms to predict 6-month clinical outcomes after FD treatment.METHODSData were retrospectively collected for 84 FD-treated sidewall aneurysms in 80 patients. Based on 6-month angiographic outcomes, IAs were classified as occluded (n = 63) or residual (incomplete occlusion, n = 21). For each case, the authors modeled FD deployment using a fast virtual stenting algorithm and hemodynamics using image-based computational fluid dynamics. Sixteen morphological, hemodynamic, and FD-based parameters were calculated for each aneurysm. Aneurysms were randomly assigned to a training or testing cohort in approximately a 3:1 ratio. The Student t-test and Mann-Whitney U-test were performed on data from the training cohort to identify significant parameters distinguishing the occluded from residual groups. Predictive models were trained using 4 types of supervised machine learning algorithms: logistic regression (LR), support vector machine (SVM; linear and Gaussian kernels), K-nearest neighbor, and neural network (NN). In the testing cohort, the authors compared outcome prediction by each model trained using all parameters versus only the significant parameters.RESULTSThe training cohort (n = 64) consisted of 48 occluded and 16 residual aneurysms and the testing cohort (n = 20) consisted of 15 occluded and 5 residual aneurysms. Significance tests yielded 2 morphological (ostium ratio and neck ratio) and 3 hemodynamic (pre-treatment inflow rate, post-treatment inflow rate, and post-treatment aneurysm averaged velocity) discriminants between the occluded (good-outcome) and the residual (bad-outcome) group. In both training and testing, all the models trained using all 16 parameters performed better than all the models trained using only the 5 significant parameters. Among the all-parameter models, NN (AUC = 0.967) performed the best during training, followed by LR and linear SVM (AUC = 0.941 and 0.914, respectively). During testing, NN and Gaussian-SVM models had the highest accuracy (90%) in predicting occlusion outcome.CONCLUSIONSNN and Gaussian-SVM models incorporating all 16 morphological, hemodynamic, and FD-related parameters predicted 6-month occlusion outcome of FD treatment with 90% accuracy. More robust models using the computational workflow and machine learning could be trained on larger patient databases toward clinical use in patient-specific treatment planning and optimization.