Risk factors for the progression of unruptured intracranial aneurysms in patients followed by CT/MR angiography

The effect of clinical, bifurcation, and aneurysm morphological characteristics on the risk of rupture in internal carotid artery bifurcation aneurysms

BACKGROUND

This study aimed to examine the clinical and morphological characteristics associated with the risk of rupture of internal carotid artery (ICA) bifurcation aneurysms (ICAbifAn) by comparing ruptured and unruptured aneurysms.

METHODS

The two-center observational study included 66 patients with ICAbifAn (4.3%) identified from a database of 1,512 patients with intracranial aneurysms. The following data were collected and evaluated for their association with rupture risk: demo-graphic data, medical history, aneurysm neck and dome size, bottleneck factor, aspect ratio (AR), size ratio, dome projection and localization, ICA (D1), M1, and A1 diameters, and ICA-M1 (β), ICA-A1 (γ), and M1-A1 (α) angles.

RESULTS

Sixty ICAbifAn cases were included in the study. Of these, 26 (43.3%) were ruptured aneurysms, and 34 (56.7%) were un-ruptured aneurysms. Patients in the ruptured group were younger than those in the unruptured group (p=0.017). The ruptured group had a smaller α angle (p=0.018) and significantly narrower A1 (p=0.004) and M1 (p=0.005) vessel diameters compared to the unruptured group. Irregular shape (p=0.001), AR>1.7, and a narrow neck (p=0.007) were significant predictors of rupture. Logistic regression analysis revealed that AR, α angle, and M1 and A1 diameters were significant predictors of aneurysm rupture. In receiver operating characteristic (ROC) analysis, an α angle cutoff of 126.2° exhibited a sensitivity of 61.5% and a specificity of 67.7% (area under the curve [AUC]=0.67). A cutoff M1 diameter of 2 mm exhibited a sensitivity and specificity of 61.5% and 76.4%, respectively (AUC=0.71). Additionally, a cutoff A1 diameter of 1.5 mm exhibited a sensitivity and specificity of 73.1% and 71.1%, respectively (AUC=0.75).

CONCLUSION

This study provided insights into the impact of aneurysm and bifurcation geometry on the risk of ICAbifAn rupture, which may also be applicable to more common bifurcation site aneurysms. Simple morphological measurements at the bifurcation region, where instability prevails, may serve as useful indicators for clinicians evaluating the likelihood of ICAbifAn rupture.

Assessment of the stability of intracranial aneurysms using a deep learning model based on computed tomography angiography

PURPOSE

Assessment of the stability of intracranial aneurysms is important in the clinic but remains challenging. The aim of this study was to construct a deep learning model (DLM) to identify unstable aneurysms on computed tomography angiography (CTA) images.

METHODS

The clinical data of 1041 patients with 1227 aneurysms were retrospectively analyzed from August 2011 to May 2021. Patients with aneurysms were divided into unstable (ruptured, evolving and symptomatic aneurysms) and stable (fortuitous, nonevolving and asymptomatic aneurysms) groups and randomly divided into training (833 patients with 991 aneurysms) and internal validation (208 patients with 236 aneurysms) sets. One hundred and ninety-seven patients with 229 aneurysms from another hospital were included in the external validation set. Six models based on a convolutional neural network (CNN) or logistic regression were constructed on the basis of clinical, morphological and deep learning (DL) features. The area under the curve (AUC), accuracy, sensitivity and specificity were calculated to evaluate the discriminating ability of the models.

RESULTS

The AUCs of Models A (clinical), B (morphological) and C (DL features from the CTA image) in the external validation set were 0.5706, 0.9665 and 0.8453, respectively. The AUCs of Model D (clinical and DL features), Model E (clinical and morphological features) and Model F (clinical, morphological and DL features) in the external validation set were 0.8395, 0.9597 and 0.9696, respectively.

CONCLUSIONS

The CNN-based DLM, which integrates clinical, morphological and DL features, outperforms other models in predicting IA stability. The DLM has the potential to assess IA stability and support clinical decision-making.

Rupture risk of intracranial aneurysms: Comparison between small ruptured intracranial aneurysms and large unruptured intracranial aneurysms

To compare the differences in clinical and morphological features between small ruptured intracranial aneurysms and large unruptured intracranial aneurysms to evaluate the risk factors for the rupture of IAs. The clinical data of 189 consecutive patients with 193 IAs were reviewed. The patients and IAs were divided into ruptured (<5 mm) and unruptured groups (>10 mm). The characteristics of the patients and the intracranial aneurysms (IAs) were compared between the 2 groups, and the risk factors for rupture of IAs were assessed using multiple logistic regression. Patient age (odds ratio [OR], 0.955), IA located at the internal carotid artery (ICA, OR, 0.202), irregular shape (OR, 0.083) and parent vessel diameter (OR, 0.426) were negatively correlated with the risk of IA rupture. IAs located at bifurcations (OR, 6.766) were positively correlated with the risk of IA rupture. In addition to the size of the IAs, regardless of IAs shape, other factors, such as younger age (<63.5 years), location at a bifurcation, IAs located at the ICA and a small parent vessel diameter (<3.25 mm), can influence the risk of IA rupture.

Compare deep learning model and conventional logistic regression model for the identification of unstable saccular intracranial aneurysms in computed tomography angiography

Background

It is crucial to distinguish unstable from stable intracranial aneurysms (IAs) as early as possible to derive optimal clinical decision-making for further treatment or follow-up. The aim of this study was to investigate the value of a deep learning model (DLM) in identifying unstable IAs from computed tomography angiography (CTA) images and to compare its discriminatory ability with that of a conventional logistic regression model (LRM).

Methods

From August 2011 to May 2021, a total of 1,049 patients with 681 unstable IAs and 556 stable IAs were retrospectively analyzed. IAs were randomly divided into training (64%), internal validation (16%), and test sets (20%). Convolutional neural network (CNN) analysis and conventional logistic regression (LR) were used to predict which IAs were unstable. The area under the curve (AUC), sensitivity, specificity and accuracy were calculated to evaluate the discriminating ability of the models. One hundred and ninety-seven patients with 229 IAs from Banan Hospital were used for external validation sets.

Results

The conventional LRM showed 11 unstable risk factors, including clinical and IA characteristics. The LRM had an AUC of 0.963 [95% confidence interval (CI): 0.941-0.986], a sensitivity, specificity and accuracy on the external validation set of 0.922, 0.906, and 0.913, respectively, in predicting unstable IAs. In predicting unstable IAs, the DLM had an AUC of 0.771 (95% CI: 0.582-0.960), a sensitivity, specificity and accuracy on the external validation set of 0.694, 0.929, and 0.782, respectively.

Conclusions

The CNN-based DLM applied to CTA images did not outperform the conventional LRM in predicting unstable IAs. The patient clinical and IA morphological parameters remain critical factors for ensuring IA stability. Further studies are needed to enhance the diagnostic accuracy.

Intracranial aneurysm rupture risk in northern China: a retrospective case-control study

Background

Rupture of intracranial aneurysms (IAs) can cause subarachnoid hemorrhage (SAH), which leads to severe neurological dysfunction and even death. Exploring the risk factors for IA rupture and taking preventive measures accordingly can reduce or prevent the occurrence of SAH. Currently, there is still no consensus on the detrimental factors for IA rupture. Thus, our study aimed to investigate the risk factors of IA rupture in a population of northern China.

Methods

We systematically collected the demographic features, medical history, and imaging data of aneurysms from patients with ruptured and unruptured IAs (UIAs) who attended the Department of Neurosurgery at the Second Hospital of Hebei Medical University from 2014 to 2019. All cases had been diagnosed by digital subtraction angiography. We excluded patients with SAH resulting from injuries, as well as those with vascular dissection and incomplete data. Finally, 1,214 patients including 616 with ruptured IAs and 598 with UIAs were collected for further analysis. A case-control study was conducted, in which multivariable logistic regression was used to analyze the risk factors for IA rupture.

Results

Our multivariable logistic regression showed that anterior cerebral artery [odds ratio (OR) =2.413; 95% confidence interval (CI): 1.235-4.718], anterior communicating artery (OR =3.952; 95% CI: 2.601-6.006), posterior communicating artery (OR =2.385; 95% CI: 1.790-3.177), and anterior circulation branches (OR =3.493; 95% CI: 1.422-8.581) were risk factors for IA rupture, whereas patients with a history of cerebral infarction (OR =0.395; 95% CI: 0.247-0.631) and those with IAs located in the internal carotid artery (OR =0.403; 95% CI: 0.292-0.557) were less likely to have IA rupture.

Conclusions

IAs at specific locations are prone to rupture. These IAs should be paid particular attention and preventive measures should be taken to reduce or prevent their rupture.

Hounsfield unit as a predictor of symptomatic vasospasm and hydrocephalus in good-grade subarachnoid hemorrhage treated with endovascular coiling

Background

Patients with good-grade subarachnoid hemorrhage (SAH) often expect favorable outcomes; however, several patients may experience secondary neurological deterioration. Hydrocephalus and vasospasm are significant complications affecting SAH prognosis. We aimed to evaluate the relationship between the incidence of symptomatic vasospasm or hydrocephalus and the Hounsfield unit (HU) value of the subarachnoid space on brain computed tomography (CT) in patients with good-grade SAH treated with endovascular coiling.

Methods

We conducted a retrospective analysis of consecutive initially good-grade pure SAH patients (Hunt-Hess grade I or II, modified Fisher scale I or III) with ruptured anterior circulation aneurysms treated with endovascular coiling in a single tertiary neurosurgical center between January 2010 and December 2019. The HU value within each cisterns of enrolled patients was measured, and after setting an appropriate cutoff value, it was investigated whether it could be a predictor of the occurrence of vasospasm and hydrocephalus.

Results

The study included 108 eligible patients (34 males, mean age 60.88±12.26 years): 26 (24.1%) showed symptomatic vasospasm and 31 (28.7%) developed hydrocephalus. Patients with symptomatic vasospasm had a greater proportion of those with Hunt-Hess grade II (77% vs. 51%, P=0.021) and modified Fisher scale III scores (58% vs. 22%, P=0.001). The hydrocephalus group presented an older mean age (65.90 vs. 58.86 years, P=0.006) and a greater proportion of Hunt-Hess grade II (74% vs. 51%, P=0.025) and modified Fisher scale III cases (45% vs. 25%, P=0.037). The mean HU values of the Sylvian cistern (53.23 vs. 43.99, P<0.001) and basal cisterns (47.04 vs. 40.18, P=0.003) were higher in the vasospasm group. In the hydrocephalus group, only the basal cistern HU value was significantly higher (45.60 vs. 40.32, P=0.016). The area under the receiver operating characteristic (ROC) curve to determine the best cut-off HU value for the prediction of patients with symptomatic vasospasm revealed a Sylvian cistern HU value of 50.375 (sensitivity: 0.692, specificity: 0.683) and basal cistern HU value of 44.875 (sensitivity: 0.615, specificity: 0.659). Multivariable logistic analysis showed that age >70 years and Sylvian cistern HU value were independent predictors of any neurological complication at 1 year.

Conclusions

The HU value of the subarachnoid space on brain CT can predict vasospasm, hydrocephalus, and long-term prognosis in good-grade SAH patients.

Increased Levels of Serum IL-15 and TNF-β Indicate the Progression of Human Intracranial Aneurysm

Objective

Existing evidence suggests that chronic inflammation promotes the progression of human intracranial aneurysm (IA) and many cytokines have been detected to participate in the process of inflammation. However, rare cytokines in plasma have been used as proxies for progression of IA. This study aimed to identify novel cytokines as biomarkers to predict the development of IA.

Methods

Patients with unruptured intracranial aneurysms (UIAs) undergoing microsurgical clipping were prospectively recruited from January 2017 to June 2020 and were separated into two groups based on their ELAPSS score (low risk group < 10, intermediate-high risk group ≥ 10). Propensity score matching (PSM) was used to reduce imbalances in the baseline characteristics between groups. All blood samples were collected before surgery. A human serum 48-cytokines examination was performed to analyze the concentrations of serological cytokines. Clinical data and cytokines were compared between groups.

Results

A total of 184 patients were enrolled in this study. The low risk group contained 77 patients and 107 patients were included in the intermediate-high risk group. Finally, there were 69 patients in each group after PSM with a matching rate of 1:1. The concentrations of 3 serum cytokines were significantly increased in intermediate-high risk patients, namely, interleukin-15 (IL-15), monocyte chemoattractant protein-1 (MCP-1), and tumor necrosis factor-β (TNF-β) (P < 0.05, |log₂ fold change| > 2). The result of receiver operator characteristic (ROC)curve revealed that TNF-β had the highest predictive accuracy, with an area under the curve (AUC) value of 0.725 [95% confidence interval (CI) 0.639–0.811, P < 0.001] followed by IL-15 (AUC = 0.691, 95% CI 0.602–0.781, P < 0.001) and MCP-1 (AUC = 0.661, 95% CI 0.569–0.753, P = 0.001). Multivariate logistic analysis demonstrated high IL-15 [odds ratio (OR), 3.23; 95% CI, 1.47–7.12; P = 0.004] and high TNF-β (OR, 8.30; 95% CI, 3.25–21.25; P < 0.001) as the risk factors that correlated with intermediate-high risk of IA progression.

Conclusion

UIA patients with intermediate-high growth risk exhibited increased serum levels of IL-15, MCP-1, and TNF-β. Serum IL-15, and TNF-β could serve as biomarkers to predict the progression of UIAs.