Evaluation of the risk of rupture of intracranial aneurysms in patients with aneurysmal subarachnoid hemorrhage according to the PHASES score

MR Imaging of the Cerebral Aneurysmal Wall for Assessment of Rupture Risk

The evaluation of unruptured intracranial aneurysms requires a comprehensive and multifaceted approach. The comprehensive analysis of aneurysm wall enhancement through high-resolution MRI, in tandem with advanced processing techniques like finite element analysis, quantitative susceptibility mapping, and computational fluid dynamics, has begun to unveil insights into the intricate biology of aneurysms. This enhanced understanding of the etiology, progression, and eventual rupture of aneurysms holds the potential to be used as a tool to triage patients to intervention versus observation. Emerging tools such as radiomics and machine learning are poised to contribute significantly to this evolving landscape of diagnostic refinement.

Optimal PHASES Scoring for Risk Stratification of Surgically Treated Unruptured Aneurysms

OBJECTIVE

The PHASES (Population, Hypertension, Age, Size, Earlier subarachnoid hemorrhage, Site) score was developed to facilitate risk stratification for management of unruptured intracranial aneurysms (UIAs). This study aimed to identify the optimal PHASES score cutoff for predicting neurologic outcomes in patients with surgically treated aneurysms.

METHODS

All patients who underwent microneurosurgical treatment for UIA at a large quaternary center from January 1, 2014, to December 31, 2020, were retrospectively reviewed. Inclusion criteria included a modified Rankin Scale (mRS) score of ≤2 at admission. The primary outcome was 1-year mRS score, with a "poor" neurologic outcome defined as an mRS score >2.

RESULTS

In total, 375 patients were included in the analysis. The mean (SD) PHASES score for the entire study population was 4.47 (2.67). Of 375 patients, 116 (31%) had a PHASES score ≥6, which was found to maximize prediction of poor neurologic outcome. Patients with PHASES scores ≥6 had significantly higher rates of poor neurologic outcome than patients with PHASES scores <6 at discharge (58 [50%] vs. 90 [35%], P = 0.005) and follow-up (20 [17%] vs. 18 [6.9%], P = 0.002). After adjusting for age, Charlson Comorbidity Index score, nonsaccular aneurysm, and aneurysm size, PHASES score ≥6 remained a significant predictor of poor neurologic outcome at follow-up (odds ratio, 2.75; 95% confidence interval, 1.42-5.36, P = 0.003).

CONCLUSIONS

In this retrospective analysis, a PHASES score ≥6 was associated with significantly greater proportions of poor outcome, suggesting that awareness of this threshold in PHASES scoring could be useful in risk stratification and UIA management.

A novel clinical-radscore nomogram for predicting ruptured intracranial aneurysm

Objectives

Our study aims to find the more practical and powerful method to predict intracranial aneurysm (IA) rupture through verification of predictive power of different models.

Methods

Clinical and imaging data of 576 patients with IAs including 192 ruptured IAs and matched 384 unruptured IAs was retrospectively analyzed. Radiomics features derived from computed tomography angiography (CTA) images were selected by t-test and Elastic-Net regression. A radiomics score (radscore) was developed based on the optimal radiomics features. Inflammatory markers were selected by multivariate regression. And then 4 models including the radscore, inflammatory, clinical and clinical-radscore models (C-R model) were built. The receiver operating characteristic curve (ROC) was performed to evaluate the performance of each model, PHASES and ELAPSS. The nomogram visualizing the C-R model was constructed to predict the risk of IA rupture.

Results

Five inflammatory features, 2 radiological characteristics and 7 radiomics features were significantly associated with IA rupture. The areas under ROCs of the radscore, inflammatory, clinical and C-R models were 0.814, 0.935, 0.970 and 0.975 in the training cohort and 0.805, 0.927, 0.952 and 0.962 in the validation cohort, respectively.

Conclusion

The inflammatory model performs particularly well in predicting the risk of IA rupture, and its predictive power is further improved by combining with radiological and radiomics features and the C-R model performs the best. The C-R nomogram is a more stable and effective tool than PHASES and ELAPSS for individually predicting the risk of rupture for patients with IA.

Clinical and Morphological Factors for Ruptured Anterior Communicating Artery Aneurysms

Introduction

The anterior communicating artery (ACoA) aneurysm, the most frequent cerebral aneurysm to rupture, carries a significant clinical burden, yet the factors influencing its rupture are limited in Indonesia. This study aims to determine the clinical and morphological features associated with ruptured ACoA compared to non-AcoA aneurysms among Indonesians.

Patients and Methods

We retrospectively reviewed our center's aneurysm patient registry from January 2019 to December 2022, and compared the clinical and morphological features between ruptured ACoA aneurysms and ruptured aneurysms elsewhere with univariate and multivariate analyses.

Results

Of the 292 patients with 325 ruptured aneurysms, 89 were from ACoA. The mean age of patients was 54.99 years, with female preponderance in the non-ACoA group (non-ACoA: 73.31%, ACoA: 46.07%). On univariate analysis, ages ≥60 [ages 60-69: OR = 0.311 (0.111-0.869), p=0.026; ages ≥70: OR = 0.215 (0.056-0.819), p=0.024], female gender [OR = 0.311 (0.182-0.533), p<0.001], and smoking [OR=2.069 (1.036-4.057), p=0.022] exhibited significant association with ruptured ACoA aneurysm. On multivariate analysis, only the female gender was independently associated with ruptured ACoA aneurysm (aOR 0.355 [0.436-1.961], p=0.001).

Conclusion

In our study, ruptured ACoA aneurysm was inversely associated with advanced age, female gender, presence of daughter aneurysm, and directly associated with smoking. After multivariate adjustment, the female gender showed an independent association with ruptured ACoA aneurysm.

RNA Expression Signatures of Intracranial Aneurysm Growth Trajectory Identified in Circulating Whole Blood

After detection, identifying which intracranial aneurysms (IAs) will rupture is imperative. We hypothesized that RNA expression in circulating blood reflects IA growth rate as a surrogate of instability and rupture risk. To this end, we performed RNA sequencing on 66 blood samples from IA patients, for which we also calculated the predicted aneurysm trajectory (PAT), a metric quantifying an IA's future growth rate. We dichotomized dataset using the median PAT score into IAs that were either more stable and more likely to grow quickly. The dataset was then randomly divided into training (n = 46) and testing cohorts (n = 20). In training, differentially expressed protein-coding genes were identified as those with expression (TPM > 0.5) in at least 50% of the samples, a q-value < 0.05 (based on modified F-statistics with Benjamini-Hochberg correction), and an absolute fold-change ≥ 1.5. Ingenuity Pathway Analysis was used to construct networks of gene associations and to perform ontology term enrichment analysis. The MATLAB Classification Learner was then employed to assess modeling capability of the differentially expressed genes, using a 5-fold cross validation in training. Finally, the model was applied to the withheld, independent testing cohort (n = 20) to assess its predictive ability. In all, we examined transcriptomes of 66 IA patients, of which 33 IAs were "growing" (PAT ≥ 4.6) and 33 were more "stable". After dividing dataset into training and testing, we identified 39 genes in training as differentially expressed (11 with decreased expression in "growing" and 28 with increased expression). Model genes largely reflected organismal injury and abnormalities and cell to cell signaling and interaction. Preliminary modeling using a subspace discriminant ensemble model achieved a training AUC of 0.85 and a testing AUC of 0.86. In conclusion, transcriptomic expression in circulating blood indeed can distinguish "growing" and "stable" IA cases. The predictive model constructed from these differentially expressed genes could be used to assess IA stability and rupture potential.

Profiling of Circulating Gene Expression Reveals Molecular Signatures Associated with Intracranial Aneurysm Rupture Risk

BACKGROUND

Following detection, rupture risk assessment for intracranial aneurysms (IAs) is critical. Towards molecular prognostics, we hypothesized that circulating blood RNA expression profiles are associated with IA risk.

METHODS

We performed RNA sequencing on 68 blood samples from IA patients. Here, patients were categorized as either high or low risk by assessment of aneurysm size (≥ 5 mm = high risk) and Population, Hypertension, Age, Size, Earlier subarachnoid hemorrhage, Site (PHASES) score (≥ 1 = high risk). Modified F-statistics and Benjamini-Hochberg false discovery rate correction was performed on transcripts per million-normalized gene counts. Protein-coding genes expressed in ≥ 50% of samples with a q value < 0.05 and an absolute fold-change ≥ 2 were considered significantly differentially expressed. Bioinformatics in Ingenuity Pathway Analysis was performed to understand the biology of risk-associated expression profiles. Association was assessed between gene expression and risk via Pearson correlation analysis. Linear discriminant analysis models using significant genes were created and validated for classification of high-risk cases.

RESULTS

We analyzed transcriptomes of 68 IA patients. In these cases, 31 IAs were large (≥ 5 mm), while 26 IAs had a high PHASES score. Based on size, 36 genes associated with high-risk IAs, and two were correlated with the size measurement. Alternatively, based on PHASES score, 76 genes associated with high-risk cases, and nine of them showed significant correlation to the score. Similar ontological terms were associated with both gene profiles, which reflected inflammatory signaling and vascular remodeling. Prediction models based on size and PHASES stratification were able to correctly predict IA risk status, with > 80% testing accuracy for both.

CONCLUSIONS

Here, we identified genes associated with IA risk, as quantified by common clinical metrics. Preliminary classification models demonstrated feasibility of assessing IA risk using whole blood expression.

SMART Registry: Safety and Performance of the Penumbra SMART COIL System for Patients With Intracranial Aneurysms 4 mm and Smaller

BACKGROUND

The Penumbra SMART COIL System includes a novel generation of embolic coils composed of complex and WAVE shape properties with varying levels of softness.

OBJECTIVE

To assess safety and efficacy of the SMART COIL System through a 1-year follow-up in patients with small intracranial aneurysms.

METHODS

This subset analysis of the SMART Registry, a prospective, multicenter study, includes patients with small intracranial aneurysms (≤4 mm) treated with the SMART COIL System. Registry end points include retreatment rates through 1 year, procedural device-related serious adverse events, and adequate occlusion postprocedure.

RESULTS

Of 905 enrolled patients with aneurysms, 172 (19.0%) had small (≤4 mm) aneurysms (75.6% female; mean age 57.2 ± 13.4 years). 30.8% (53/172) of small aneurysms were ruptured, of which 50.9% (27/53) had Hunt and Hess ≥3. 79.5% (132/166) were wide-necked. Stent-assisted coiling and balloon-assisted coiling were performed in 37.2% (64/172) and 22.1% (38/172) of patients, respectively. The mean packing density for very small aneurysms was 44.9 (SD 25.23). Raymond Class I and Class II were achieved in 89.5% (154/172) postprocedure and 97.2% (137/141) at 1 year. The retreatment rate through 1 year was 5.6% (8/142), and the recanalization rate was 7.1% (10/141). The periprocedural device-related serious adverse event rate was 2.9% (5/172). Intraprocedural aneurysm rupture occurred in 0.8% of patients.

CONCLUSION

This analysis suggests that the SMART COIL System is safe and efficacious in small aneurysms with satisfactory occlusion rates and low rates of rupture or rerupture. At 1 year, patients had low retreatment rates and good clinical outcomes.

A web-based dynamic nomogram for rupture risk of posterior communicating artery aneurysms utilizing clinical, morphological, and hemodynamic characteristics

Background

Predicting rupture risk is important for aneurysm management. This research aimed to develop and validate a nomogram model to forecast the rupture risk of posterior communicating artery (PcomA) aneurysms.

Methods

Clinical, morphological, and hemodynamic parameters of 107 unruptured PcomA aneurysms and 225 ruptured PcomA aneurysms were retrospectively analyzed. The least absolute shrinkage and selection operator (LASSO) analysis was applied to identify the optimal rupture risk factors, and a web-based dynamic nomogram was developed accordingly. The nomogram model was internally validated and externally validated independently. The receiver operating characteristic (ROC) curve was used to assess the discrimination of nomogram, and simultaneously the Hosmer–Lemeshow test and calibration plots were used to assess the calibration. Decision curve analysis (DCA) and clinical impact curve (CIC) were used to evaluate the clinical utility of nomogram additionally.

Results

Four optimal rupture predictors of PcomA aneurysms were selected by LASSO and identified by multivariate logistic analysis, including hypertension, aspect ratio (AR), oscillatory shear index (OSI), and wall shear stress (WSS). A web-based dynamic nomogram was then developed. The area under the curve (AUC) in the training and external validation cohorts was 0.872 and 0.867, respectively. The Hosmer–Lemeshow p > 0.05 and calibration curves showed an appropriate fit. The results of DCA and CIC indicated that the net benefit rate of the nomogram model is higher than other models.

Conclusion

Hypertension, high AR, high OSI, and low WSS were the most relevant risk factors for rupture of PcomA aneurysms. A web-based dynamic nomogram thus established demonstrated adequate discrimination and calibration after internal and external validation. We hope that this tool will provide guidance for the management of PcomA aneurysms.

The majority of ruptured aneurysms are small with low rupture risk scores

BACKGROUND

Understanding the rupture risk of unruptured intracranial aneurysms has important clinical implications given the morbidity and mortality associated with subarachnoid hemorrhage (SAH). The ISUIA, UCAS, and PHASES studies provide rupture risk calculations.

OBJECTIVE

We apply the risk calculations to a series ruptured intracranial aneurysms to assess the rupture risk for each aneurysm (had they been discovered in the unruptured state).

METHODS

This is a retrospective study of 246 patients with SAH from a ruptured saccular aneurysm. The ISUIA, UCAS, and PHASES calculators were applied to each patient/aneurysm to demonstrate a theoretical annual risk of rupture dichotomized by aneurysm location.

RESULTS

The average diameter of the aneurysms was 5.5 ± 3.1 mm. Three quarters (75%) of the aneurysms measured <7 mm and 48.8% were <5 mm. The anterior communicating artery (Acomm) was the most common location of rupture (24.7%). Posterior communicating artery aneurysms (Pcomm) were the third most common at 16.2%. The average ISUIA 1-year rupture risk was 0.46 ± 0.008%. The average UCAS 1-year rupture risk was 0.93% ± 0.01. The annualPHASESrupture risk was0.32 ± 0.004%. The highest risk locations were the vertebral artery (up to 10.3% per year) and superior cerebellar artery (up to 2.7% per year). On average, Acomm aneurysms had 1 year risk no higher than 1.1% and Pcomm aneurysms no higher than 1.2% per year.

CONCLUSION

We observed that in a small retrospective series of ruptured aneurysms, the majority were <7 mm and that the theoretical rupture risk of these aneurysms, had they been discovered in the unruptured state, is low (<1% per year). Our study has a number of limitations and these results should be validated in a larger multicenter study.

Identification of ruptured intracranial aneurysms using the aneurysm-specific prediction score in patients with multiple aneurysms with subarachnoid hemorrhages- a Chinese population based external validation study

Background

For patients with aneurysmal subarachnoid hemorrhages (SAHs) and multiple intracranial aneurysms (MIAs), a simple and fast imaging method that can identify ruptured intracranial aneurysms (RIAs) may have great clinical value. We sought to use the aneurysm-specific prediction score to identify RIAs in patients with MIAs and evaluate the aneurysm-specific prediction score.

Methods

Between May 2018 and May 2021, 134 patients with 290 MIAs were retrospectively analyzed. All patients had an SAH due to IA rupture. CT angiography (CTA) was used to assess the maximum diameter, shape, and location of IAs to calculate the aneurysm-specific prediction score. Then, the aneurysm-specific prediction score was applied to RIAs in patients with MIAs.

Results

The IAs with the highest aneurysm-specific prediction scores had not ruptured in 17 (12.7%) of the 134 patients with 290 MIAs. The sensitivity, specificity, false omission rate, diagnostic error rate, and diagnostic accuracy of the aneurysm-specific prediction score were higher than those of the maximum diameter, shape, and location of IAs.

Conclusions

The present study suggests that the aneurysm-specific prediction score has high diagnostic accuracy in identifying RIAs in patients with MIAs and SAH, but that it needs further evaluation.

Supplementary Information

The online version contains supplementary material available at 10.1186/s12883-022-02727-w.

Rupture Risk Assessment for Anterior Communicating Artery Aneurysms Using Decision Tree Modeling

Background

Although anterior communicating artery (ACoA) aneurysms have a higher risk of rupture than aneurysms in other locations, whether to treat unruptured ACoA aneurysms incidentally found is a dilemma because of treatment-related complications. Machine learning models have been widely used in the prediction of clinical medicine. In this study, we aimed to develop an easy-to-use decision tree model to assess the rupture risk of ACoA aneurysms.

Methods

This is a retrospective analysis of rupture risk for patients with ACoA aneurysms from two medical centers. Morphologic parameters of these aneurysms were measured and evaluated. Univariate analysis and multivariate logistic regression analysis were performed to investigate the risk factors of aneurysm rupture. A decision tree model was developed to assess the rupture risk of ACoA aneurysms based on significant risk factors.

Results

In this study, 285 patients were included, among which 67 had unruptured aneurysms and 218 had ruptured aneurysms. Aneurysm irregularity and vessel angle were independent predictors of rupture of ACoA aneurysms. There were five features, including size ratio, aneurysm irregularity, flow angle, vessel angle, and aneurysm size, selected for decision tree modeling. The model provided a visual representation of a decision tree and achieved a good prediction performance with an area under the receiver operating characteristic curve of 0.864 in the training dataset and 0.787 in the test dataset.

Conclusion

The decision tree model is a simple tool to assess the rupture risk of ACoA aneurysms and may be considered for treatment decision-making of unruptured intracranial aneurysms.

Shedding the Light on the Natural History of Intracranial Aneurysms: An Updated Overview

The exact molecular pathways underlying the multifactorial natural history of intracranial aneurysms (IAs) are still largely unknown, to the point that their understanding represents an imperative challenge in neurovascular research. Wall shear stress (WSS) promotes the genesis of IAs through an endothelial dysfunction causing an inflammatory cascade, vessel remodeling, phenotypic switching of the smooth muscle cells, and myointimal hyperplasia. Aneurysm growth is supported by endothelial oxidative stress and inflammatory mediators, whereas low and high WSS determine the rupture in sidewall and endwall IAs, respectively. Angioarchitecture, age older than 60 years, female gender, hypertension, cigarette smoking, alcohol abuse, and hypercholesterolemia also contribute to growth and rupture. The improvements of aneurysm wall imaging techniques and the implementation of target therapies targeted against inflammatory cascade may contribute to significantly modify the natural history of IAs. This narrative review strives to summarize the recent advances in the comprehension of the mechanisms underlying the genesis, growth, and rupture of IAs.

Development and validation of an institutional nomogram for aiding aneurysm rupture risk stratification

Rupture risk stratification is critical for incidentally detected intracranial aneurysms. Here we developed and validated an institutional nomogram to solve this issue. We reviewed the imaging and clinical databases for aneurysms from January 2015 to September 2018. Aneurysms were reconstructed and morphological features were extracted by the Pyradiomics in python. Multiple logistic regression was performed to develop the nomogram. The consistency of the nomogram predicted rupture risks and PHASES scores was assessed. The performance of the nomogram was evaluated by the discrimination, calibration, and decision curve analysis (DCA). 719 aneurysms were enrolled in this study. For each aneurysm, twelve morphological and nine clinical features were obtained. After logistic regression, seven features were enrolled in the nomogram, which were SurfaceVolumeRatio, Flatness, Age, Hyperlipemia, Smoker, Multiple aneurysms, and Location of the aneurysm. The nomogram had a positive and close correlation with PHASES score in predicting aneurysm rupture risks. AUCs of the nomogram in discriminating aneurysm rupture status was 0.837 in a separate testing set. The calibration curves fitted well and DCA demonstrated positive net benefits of the nomogram in guiding clinical decisions. In conclusion, Pyradiomics derived morphological features based institutional nomogram was useful for aneurysm rupture risk stratification.

The Subarachnoid Hemorrhage-Weather Myth: A Long-Term Big Data and Deep Learning Analysis

Objective: The frequency of aneurysmal subarachnoid hemorrhage (aSAH) presents complex fluctuations that have been attributed to weather and climate changes in the past. In the present long-term big data and deep learning analysis, we have addressed this long-held myth.

Methods: Bleeding dates and basic demographic data for all consecutive patients (n = 1,271) admitted to our vascular center for treatment of aSAH between January 2003 and May 2020 (6,334 days) were collected from our continuously maintained database. The meteorological data of the local weather station, including 13 different weather and climate parameters, were retrieved from Germany's National Meteorological Service for the same period. Six different deep learning models were programmed using the Keras framework and were trained for aSAH event prediction with meteorological data from January 2003 to June 2017, with 10% of this dataset applied for data validation and model improvement. The dataset from July 2017 to May 2020 was tested for aSAH event prediction accuracy for all six models using the area under the receiver operating characteristic curve (AUROC) as the metric.

Results: The study group comprised of 422 (33.2%) male and 849 (66.8%) female patients with an average age of 55 ± 14 years. None of the models showed an AUROC larger than 60.2. From the presented data, the influence of weather and climate on the occurrence of aSAH events is extremely unlikely.

Conclusion: The myth of special weather conditions influencing the frequency of aSAH is disenchanted by this long-term big data and deep learning analysis.

A nomogram to predict rupture risk of middle cerebral artery aneurysm

BACKGROUND

Determining the rupture risk of unruptured intracranial aneurysm is crucial for treatment strategy. The purpose of this study was to predict the rupture risk of middle cerebral artery (MCA) aneurysms using a machine learning technique.

METHODS

We retrospectively reviewed 403 MCA aneurysms and randomly partitioned them into the training and testing datasets with a ratio of 8:2. A generalized linear model with logit link was developed using training dataset to predict the aneurysm rupture risk based on the clinical variables and morphological features manually measured from computed tomography angiography. To facilitate the clinical application, we further constructed an easy-to-use nomogram based on the developed model.

RESULTS

Ruptured MCA aneurysm had larger aneurysm size, aneurysm height, perpendicular height, aspect ratio, size ratio, bottleneck factor, and height-width ratio. Presence of a daughter-sac was more common in ruptured than in unruptured MCA aneurysms. Six features, including aneurysm multiplicity, lobulations, size ratio, bottleneck factor, height-width ratio, and aneurysm angle, were adopted in the model after feature selection. The model achieved a relatively good performance with areas under the receiver operating characteristic curves of 0.77 in the training dataset and 0.76 in the testing dataset. The nomogram provided a visual interpretation of our model, and the rupture risk probability of MCA aneurysms can be directly read from it.

CONCLUSION

Our model can be used to predict the rupture risk of MCA aneurysm.

Comparison of Unruptured Intracranial Aneurysm Treatment Score and PHASES Score in Subarachnoid Hemorrhage Patients With Multiple Intracranial Aneurysms

Objective: Unruptured Intracranial Aneurysm (UIA) Treatment Score (UIATS) and PHASES score are used to inform treatment decision making for UIAs (treatment or observation). We assessed the ability of the scoring systems to discriminate between ruptured aneurysms and UIAs in a subarachnoid hemorrhage (SAH) cohort with multiple aneurysms. Methods: We retrospectively applied PHASES and UIATS scoring to the aneurysms of 40 consecutive patients with SAH and multiple intracranial aneurysms. Results: PHASES score discriminated better between ruptured aneurysms and UIAs than UIATS. PHASES scores and the difference between the UIATS subscores were higher for ruptured aneurysms compared with UIAs, which reached significance for the PHASES score. PHASES score estimated a low 5-year rupture risk in a larger proportion of the UIAs (≤0.7% in 62.3%, ≤1.7% in 98.4%) than of the ruptured aneurysms (≤0.7% in 22.5%, ≤1.7% in 82.5%). In the 40 ruptured aneurysms, UIATS provided recommendation for treatment in 11 (27.5%), conservative management in 14 (35.0%), and was inconclusive in 15 cases (37.5%). In the 61 UIAs, UIATS recommended treatment in 16 (26.2%), conservative management in 29 (47.5%), and was inconclusive in 16 (26.2%) cases. Conclusion: Similar to previous SAH cohorts, a significant proportion of the ruptured aneurysms exhibited a low-rupture risk. Nevertheless, PHASES score discriminated between ruptured aneurysms and UIAs in our cohort; the lower discriminatory power of UIATS was due to high weights of aneurysm-independent factors. We recommend careful integration of the scores for individual decision making. Large-scale prospective trials are required to establish score-based treatment strategies for UIAs.

Application of unruptured aneurysm scoring systems to a cohort of ruptured aneurysms: are we underestimating rupture risk?

The predictive values of current risk stratification scales such as the Unruptured Intracranial Aneurysm Treatment Score (UIATS) and the PHASES score are debatable. We evaluated these scores using a cohort of ruptured intracranial aneurysms to simulate their management recommendations had the exact same patients presented prior to rupture. A prospectively maintained database of ruptured saccular aneurysm patients presenting to our institution was used. The PHASES score was calculated for 992 consecutive patients presenting between January 2002 and December 2018, and the UIATS was calculated for 266 consecutive patients presenting between January 2013 and December 2018. A shorter period was selected for the UIATS cohort given the larger number of variables required for calculation. Clinical outcomes were compared between UIATS-recommended "observation" aneurysms and all other aneurysms. Out of 992 ruptured aneurysms, 54% had a low PHASES score (≤5). Out of the 266 ruptured aneurysms, UIATS recommendations were as follows: 68 (26%) "observation," 97 (36%) "treatment," and 101 (38%) "non-definitive." The UIATS conservative group of patients developed more SAH-related complications (78% vs. 65%, p=0.043), had a higher rate of non-home discharge (74% vs. 46%, p<0.001), and had a greater incidence of poor functional status (modified Rankin scale >2) after 12-18 months (68% vs. 51%, p=0.014). Current predictive scoring systems for unruptured aneurysms may underestimate future rupture risk and lead to more conservative management strategies in some patients. Patients that would have been recommended for conservative therapy were more likely to have a worse outcome after rupture.

External Validation of the PHASES Score in Patients with Multiple Intracranial Aneurysms

OBJECTIVES

This study sought to assess whether the Population, Hypertension, Age, Size, Earlier Subarachnoid Hemorrhage, Site (PHASES) score can do risk stratification of patients with multiple aneurysms (MIAs).

MATERIAL AND METHODS

Patients between January 1, 2016 and January 1, 2019 were recruited retrospectively. The PHASES score was applied to assess the theoretical risk of IA rupture. For patients-level analyses, four modes of the application of the score were used: largest IA PHASES score, highest PHASES score, sum PHASES score, and mean PHASES score.

RESULTS

A total of 701 patients with 1673 IAs were included in this study. At aneurysm-level analysis, the average PHASES score was 3.0 ± 3.0 points, with 2.8 ± 3.0 points and 4.1 ± 2.9 points in the unruptured and ruptured groups, respectively (p < 0.001). At the patient-level analysis, for the largest IA PHASES score, the areas under the curves (AUC) was 0.572. The discrimination performance of the largest IA PHASES score decreases as IA number increases, with AUCs were 0.597, 0.518, and 0.450 in the 2 IAs, 3 IAs and, 4 or more IAs subgroups, respectively. For highest PHASES score, sum PHASES score, and mean PHASES score, the AUCs were 0.577, 0.599, and 0.619, respectively.

CONCLUSIONS

In this study, PHASES score only serve as a weak tool in decision-making settings for MIAs patients; as such, more accurate models should be developed for MIAs patients and the cumulative effect of MIA may should be considered.

Management of unruptured intracranial aneurysms: correlation of UIATS, ELAPSS, and PHASES with referral center practice

Concordance between the Unruptured Intracranial Aneurysm Treatment Score (UIATS), Earlier Subarachnoid Hemorrhage, Location, Age, Population, Size, Shape (ELAPSS) score, and Population, Hypertension, Age, Size, Earlier Subarachnoid Hemorrhage, Site (PHASES) score with real-world management decisions in unruptured intracranial aneurysms (UIAs) remains unclear, especially in current practice. This study aimed to investigate this concordance, while developing an optimal model predictive of recent decision practices at a quaternary referral center. A prospective database of patients presenting with UIAs to our institution from January 1 to December 31, 2018, was used. Concordance between the scores and real-world management decisions on every UIA was assessed. Complications and length of stay (LOS) were compared between aneurysms in the UIATS-recommended treatment and observation groups. A subgroup analysis of concordance was also conducted among junior and senior surgeons. An optimal logistic regression model predictive of real-world decisions was also derived. The cohort consisted of 198 patients with 271 UIAs, of which 42% were treated. The UIATS demonstrated good concordance with an AUC of 0.765. Of the aneurysms in the UIATS-recommended "observation" group, 22% were discordantly treated. The ELAPSS score demonstrated good discrimination (AUC = 0.793), unlike the PHASES score (AUC = 0.579). Endovascular treatment rates, complications, and LOS were similar between aneurysms in the UIATS-recommended treatment and observation groups. Similar concordance was obtained among junior and senior surgeons. The optimal predictive model consisted of several significantly associated variables and had an AUC of 0.942. Cerebrovascular specialists may be treating aneurysms slightly more than these scores would recommend, independently of years in practice. Wide variation still exists in management practices of UIAs.

Stability Assessment of Intracranial Aneurysms Using Machine Learning Based on Clinical and Morphological Features

Machine learning (ML) as a novel approach could help clinicians address the challenge of accurate stability assessment of unruptured intracranial aneurysms (IAs). We developed multiple ML models for IA stability assessment and compare their performances. We enrolled 1897 consecutive patients with unstable (n = 528) and stable (n = 1539) IAs. Thirteen patient-specific clinical features and eighteen aneurysm morphological features were extracted to generate support vector machine (SVM), random forest (RF), and feed-forward artificial neural network (ANN) models. The discriminatory performances of the models were compared with statistical logistic regression (LR) model and the PHASES score in IA stability assessment. Based on the receiver operating characteristic (ROC) curve and area under the curve (AUC) values for each model in the test set, the AUC values for RF, SVM, and ANN were 0.850 (95% CI 0.806-0.893), 0.858 (95 %CI 0.816-0.900), and 0.867 (95% CI 0.828-0.906), demonstrating good discriminatory ability. All ML models exhibited superior performance compared with the statistical LR and the PHASES score (the AUC values were 0.830 and 0.589, respectively; RF versus PHASES, P < 0.001; RF versus LR, P = 0.038). Important features contributing to the stability discrimination included three clinical features (location, sidewall/bifurcation type, and presence of symptoms) and three morphological features (undulation index, height-width ratio, and irregularity). These findings demonstrate the potential of ML to augment the clinical decision-making process for IA stability assessment, which may enable more optimal management for patients with IAs in the future.

Bifurcation Configuration Is an Independent Risk Factor for Aneurysm Rupture Irrespective of Location

Background: Bifurcation and sidewall aneurysms have different rupture risks, but whether this difference comes from the location of the aneurysm is not clear. The objective of this study is to illustrate the rationality of ranking bifurcation configuration as an independent risk factor for aneurysm rupture.

Methods: Morphological features of 719 aneurysms (216 ruptured) were automatically extracted from a consecutive cohort of patients via PyRadiomics. Rupture risks and morphological features were compared between bifurcation and sidewall aneurysms, and lasso regression was applied to explore the morphological determinants for rupture in bifurcation and sidewall aneurysms. Rupture risks and morphological features of bifurcation aneurysms in different locations were analyzed. Multivariate regression was performed to explore the risk factors for aneurysm rupture.

Results: Twelve morphological features were automatically extracted from PyRadiomics implemented in Python. The rupture risks were higher in bifurcation aneurysms (P < 0.01), and morphological features Elongation and Flatness were much lower in ruptured bifurcation than sidewall aneurysms (P = 0.036, 0.011, respectively). Elongation and Flatness were the morphological determinants for rupture in bifurcation aneurysms, whereas Elongation and SphericalDisproportion were determinants for sidewall aneurysms. Different rupture risks and morphological features were found between sidewall and bifurcation aneurysms of the same location, and among bifurcation aneurysms of different locations. In multivariate regression, bifurcation configuration was an independent risk factor for aneurysm rupture (OR 3.007, 95% CI 1.752–5.248, P < 0.001).

Conclusions: Sidewall and bifurcation aneurysms and bifurcation aneurysms of different locations have different rupture risks and morphological features. Bifurcation configuration is an independent risk factor for aneurysm rupture irrespective of location.

Prediction of Aneurysm Stability Using a Machine Learning Model Based on PyRadiomics-Derived Morphological Features

Background and Purpose- Discrimination of the stability of intracranial aneurysms is critical for determining the treatment strategy, especially in small aneurysms. This study aims to evaluate the feasibility of applying machine learning for predicting aneurysm stability with radiomics-derived morphological features. Methods- Morphological features of 719 aneurysms were extracted from PyRadiomics, of which 420 aneurysms with Maximum3DDiameter ranging from 4 mm to 8 mm were enrolled for analysis. The stability of these aneurysms and other clinical characteristics were reviewed from the medical records. Based on the morphologies with/without clinical features, machine learning models were constructed and compared to define the morphological determinants and screen the optimal model for predicting aneurysm stability. The effect of clinical characteristics on the morphology of unstable aneurysms was analyzed. Results- Twelve morphological features were automatically extracted from PyRadiomics implemented in Python for each aneurysm. Lasso regression defined Flatness as the most important morphological feature to predict aneurysm stability, followed by SphericalDisproportion, Maximum2DDiameterSlice, and SurfaceArea. SurfaceArea (odds ratio [OR], 0.697; 95% CI, 0.476-0.998), SphericalDisproportion (OR, 1.730; 95% CI, 1.143-2.658), Flatness (OR, 0.584; 95% CI, 0.374-0.894), Hyperlipemia (OR, 2.410; 95% CI, 1.029-5.721), Multiplicity (OR, 0.182; 95% CI, 0.082-0.380), Location at middle cerebral artery (OR, 0.359; 95% CI, 0.134-0.902), and internal carotid artery (OR, 0.087; 95% CI, 0.030-0.211) were enrolled into the final prediction model. In terms of performance, the area under curve of the model reached 0.853 (95% CI, 0.767-0.940). For unstable aneurysms, Compactness1 (P=0.035), Compactness2 (P=0.036), Sphericity (P=0.035), and Flatness (P=0.010) were low, whereas SphericalDisproportion (P=0.034) was higher in patients with hypertension. Conclusions- Morphological features extracted from PyRadiomics can be used for aneurysm stratification. Flatness is the most important morphological determinant to predict aneurysm stability. Our model can be used to predict aneurysm stability. Unstable aneurysm is more irregular in patients with hypertension.