Modifiable Cardiovascular Risk Factors in Patients With Sporadic Cerebral Cavernous Malformations: Obesity Matters

Cerebral cavernous malformations - An overview on genetics, clinical aspects and therapeutic strategies

Cerebral cavernous malformations (CCMs) are abnormally packed blood vessels lined with endothelial cells, that do not exhibit intervening tight junctions, lack muscular and elastic layers and are usually surrounded by hemosiderin and gliosis. CCMs may be sporadic or familial autosomal dominant (FCCMs) caused by loss of function mutations in CCM1 (KRIT1), CCM2 (MGC4607), and CCM3 (PDCD10) genes. In the FCCMs, patients have multiple CCMs, different family members are affected, and developmental venous anomalies are absent. CCMs may be asymptomatic or may manifest with focal neurological deficits with or without associated hemorrhage andseizures. Recent studies identify a digenic "triple-hit" mechanism involving the aquisition of three distinct genetic mutations that culminate in phosphatidylinositol-3-kinase (PIK3CA) gain of function, as the basis for rapidly growing and clinically symptomatic CCMs. The pathophysiology of CCMs involves signaling aberrations in the neurovascular unit, including proliferative dysangiogenesis, blood-brain barrier hyperpermeability, inflammation and immune mediated processes, anticoagulant vascular domain, and gut microbiome-driven mechanisms. Clinical trials are investigating potential therapies, magnetic resonance imaging and plasma biomarkers for hemorrhage and CCMs-related epilepsy, as well as different techniques of neuronavigation and neurosonology to guide surgery in order to minimize post-operatory morbidity and mortality. This review addresses the recent data about the natural history, genetics, neuroimaging and therapeutic approaches for CCMs.

Identifying potential (re)hemorrhage among sporadic cerebral cavernous malformations using machine learning

The (re)hemorrhage in patients with sporadic cerebral cavernous malformations (CCM) was the primary aim for CCM management. However, accurately identifying the potential (re)hemorrhage among sporadic CCM patients in advance remains a challenge. This study aims to develop machine learning models to detect potential (re)hemorrhage in sporadic CCM patients. This study was based on a dataset of 731 sporadic CCM patients in open data platform Dryad. Sporadic CCM patients were followed up 5 years from January 2003 to December 2018. Support vector machine (SVM), stacked generalization, and extreme gradient boosting (XGBoost) were used to construct models. The performance of models was evaluated by area under receiver operating characteristic curves (AUROC), area under the precision-recall curve (PR-AUC) and other metrics. A total of 517 patients with sporadic CCM were included (330 female [63.8%], mean [SD] age at diagnosis, 42.1 [15.5] years). 76 (re)hemorrhage (14.7%) occurred during follow-up. Among 3 machine learning models, XGBoost model yielded the highest mean (SD) AUROC (0.87 [0.06]) in cross-validation. The top 4 features of XGBoost model were ranked with SHAP (SHapley Additive exPlanations). All-Elements XGBoost model achieved an AUROCs of 0.84 and PR-AUC of 0.49 in testing set, with a sensitivity of 0.86 and a specificity of 0.76. Importantly, 4-Elements XGBoost model developed using top 4 features got a AUROCs of 0.83 and PR-AUC of 0.40, a sensitivity of 0.79, and a specificity of 0.72 in testing set. Two machine learning-based models achieved accurate performance in identifying potential (re)hemorrhages within 5 years in sporadic CCM patients. These models may provide insights for clinical decision-making.

Natural course of cerebral and spinal cavernous malformations: a complete ten-year follow-up study

Knowledge of the bleeding risk and the long-term outcome of conservatively treated patients with cavernous malformations (CM) is poor. In this work, we studied the occurrence of CM-associated hemorrhage over a 10-year period and investigated risk factors for bleeding. Our institutional database was screened for patients with cerebral (CCM) or intramedullary spinal cord (ISCM) CM admitted between 2003 and 2021. Patients who underwent surgery and patients without completed follow-up were excluded. Analyses were performed to identify risk factors and to determine the cumulative risk for hemorrhage. A total of 91 CM patients were included. Adjusted multivariate logistic regression analysis identified bleeding at diagnosis (p = 0.039) and CM localization to the spine (p = 0.010) as predictors for (re)hemorrhage. Both risk factors remained independent predictors through Cox regression analysis (p = 0.049; p = 0.016). The cumulative 10-year risk of bleeding was 30% for the whole cohort, 39% for patients with bleeding at diagnosis and 67% for ISCM. During an untreated 10-year follow-up, the probability of hemorrhage increased over time, especially in cases with bleeding at presentation and spinal cord localization. The intensity of such increase may decline throughout time but remains considerably high. These findings may indicate a rather aggressive course in patients with ISCM and may endorse early surgical treatment.

Cerebral cavernous malformations: Typical and atypical imaging characteristics

Cavernous malformations (CMs) are benign vascular malformations that maybe seen anywhere in the central nervous system. They are dynamic lesions, growing or shrinking over time and only rarely remaining stable. Size varies from a few millimeters to a few centimeters. CMs can be sporadic or familial, and while most of them are congenital, de novo and acquired lesions may also be seen. Etiology is still unknown. A genetic molecular mechanism has been proposed since a cerebral cavernous malformation gene loss of function was found in both familial and sporadic lesions. Additionally, recent studies suggest that formation of CMs in humans may be associated with a distinctive bacterial gut composition (microbioma). Imaging is fairly typical but may vary according to age, location, and etiology. Follow-up is not well established because CMs patients have a highly unpredictable clinical course. Angiogenic and inflammatory mechanisms have been implicated in disease activity, as well as lesional hyperpermeability and iron deposition. Imaging and serum biomarkers of these mechanisms are under current investigation. Treatment options, including surgery or radiosurgery, are not well defined and are dependent upon multiple factors, including clinical presentation, lesion location, number of hemorrhagic events, and medical comorbidities. Our purpose is to review the imaging features of CMs based on their size, location, and etiology, as well as their differential diagnosis and best imaging approach. New insights in etiology will be briefly considered. Follow-up strategies, including serum and imaging biomarkers, and treatment options will also be discussed.

Functional outcome after initial and multiple intracerebral hemorrhages in children with cerebral cavernous malformations

BACKGROUND AND PURPOSE

We aimed to assess the course and predictors of functional outcome after single and multiple intracerebral hemorrhage (ICH) in pediatric patients with cerebral cavernous malformations (CCMs) and to conduct a risk assessment of a third bleed during the first follow-up year after second ICH.

METHODS

We included patients aged ≤18 years with complete baseline characteristics, a magnetic resonance imaging dataset, ≥1 CCM-related ICH and ≥1 follow-up examination, who were treated between 2003 and 2021. Neurological functional status was obtained using modified Rankin Scale scores at diagnosis, before and after each ICH, and at last follow-up. Kaplan-Meier analysis was performed to determine the cumulative 1-year risk of third ICH.

RESULTS

A total of 55 pediatric patients (median [interquartile range] age 12 [11] years) were analyzed. Univariate analysis identified brainstem cavernous malformation (BSCM; p = 0.019) as a statistically significant predictor for unfavorable outcome after second ICH. Outcome after second ICH was significantly worse in 12 patients (42.9%; p = 0.030) than after first ICH and in five patients (55.6%; p = 0.038) after a third ICH compared to a second ICH. Cumulative 12-month risk of rebleeding during the first year after a second ICH was 10.7% (95% confidence interval 2.8%-29.37%).

CONCLUSIONS

Pediatric patients with a BSCM have a higher risk of worse outcome after second ICH. Functional outcome improves over time after an ICH but worsens following each ICH compared to baseline or previous ICH. Second bleed was associated with neurological deterioration compared to initial ICH, and this deteriorated further after a third ICH.

Effects of medication intake on the risk of hemorrhage in patients with sporadic cerebral cavernous malformations

Objective

Recurrent intracerebral hemorrhage (ICH) poses a high risk for patients with cerebral cavernous malformations (CCMs). This study aimed to assess the influence of medication intake on hemorrhage risk in sporadic CCMs.

Methods

From a database of 1,409 consecutive patients with CCM (2003-2021), subjects with sporadic CCMs and complete magnetic resonance imaging data were included. We evaluated the presence of ICH as a mode of presentation, the occurrence of ICH during follow-up, and medication intake, including beta blockers, statins, antithrombotic therapy, and thyroid hormones. The impact of medication intake on ICH at presentation was calculated using univariate and multivariate logistic regression with age and sex adjustment. The longitudinal cumulative 5-year risk for (re-)hemorrhage was analyzed using the Kaplan-Meier curves and the Cox regression analysis.

Results

A total of 1116 patients with CCM were included. Logistic regression analysis showed a significant correlation (OR: 0.520, 95% CI: 0.284-0.951, p = 0.034) between antithrombotic therapy and ICH as a mode of presentation. Cox regression analysis revealed no significant correlation between medication intake and occurrence of (re-)hemorrhage (hazard ratios: betablockers 1.270 [95% CI: 0.703-2.293], statins 0.543 [95% CI: 0.194-1.526], antithrombotic therapy 0.507 [95% CI: 0.182-1.410], and thyroid hormones 0.834 [95% CI: 0.378-1.839]).

Conclusion

In this observational study, antithrombotic treatment was associated with the tendency to a lower rate of ICH as a mode of presentation in a large cohort of patients with sporadic CCM. Intake of beta blockers, statins, and thyroid hormones had no effect on hemorrhage as a mode of presentation. During the 5-year follow-up period, none of the drugs affected the further risk of (re-)hemorrhage.

Central nervous system cavernous malformations: cross-sectional study assessing rebleeding risk after a second haemorrhage

BACKGROUND AND PURPOSE

The purpose of this study was to investigate the 5-year risk of a third bleeding event in cavernous malformations (CMs) of the central nervous system.

METHODS

Patients with cerebral or spinal CMs treated between 2003 and 2021 were screened using our institutional database. Patients with a complete magnetic resonance imaging dataset, clinical baseline characteristics, and history of two bleeding events were included. Patients who underwent surgical CM removal were excluded. Neurological functional status was obtained using the modified Rankin Scale score at the second and third bleeding. Kaplan-Meier and Cox regression analyses were performed to determine the cumulative 5-year risk for a third haemorrhage.

RESULTS

Forty-two patients were included. Cox regression analysis adjusted for age and sex did not identify risk factors for a third haemorrhage. 37% of patients experienced neurological deterioration after the third haemorrhage (p = 0.019). The cumulative 5-year risk of a third bleeding was 66.7% (95% confidence interval [CI] 50.4%-80%) for the whole cohort, 65.9% (95% CI 49.3%-79.5%) for patients with bleeding at initial diagnosis, 72.7% (95% CI 39.3%-92.7%) for patients with a developmental venous anomaly, 76.9% (95% CI 55.9%-90.3%) for patients with CM localization to the brainstem and 75% (95% CI 50.6%-90.4%) for patients suffering from familial CM disease.

CONCLUSIONS

During an untreated 5-year follow-up after a second haemorrhage, a significantly increased risk of a third haemorrhage compared to the known risk of a first and second bleeding event was identified. The third bleeding was significantly associated with neurological deterioration. These findings may justify a surgical treatment after a second bleeding event.

Medication intake and hemorrhage risk in patients with familial cerebral cavernous malformations

OBJECTIVE

The objective of this study was to analyze the impact of medication intake on hemorrhage risk in patients with familial cerebral cavernous malformation (FCCM).

METHODS

The authors' institutional database was screened for patients with FCCM who had been admitted to their department between 2003 and 2020. Patients with a complete magnetic resonance imaging (MRI) data set, evidence of multiple CCMs, clinical baseline characteristics, and follow-up (FU) examination were included in the study. The authors assessed the influence of medication intake on first or recurrent intracerebral hemorrhage (ICH) using univariate and multivariate logistic regression adjusted for age and sex. The longitudinal cumulative 5-year risk of hemorrhage was calculated by applying Kaplan-Meier and Cox regression analyses adjusted for age and sex.

RESULTS

Two hundred five patients with FCCMs were included in the study. Multivariate Cox regression analysis revealed ICH as a predictor for recurrent hemorrhage during the 5-year FU. The authors also noted a tendency toward a decreased association with ICH during FU in patients on statin medication (HR 0.22, 95% CI 0.03-1.68, p = 0.143), although the relationship was not statistically significant. No bleeding events were observed in patients on antithrombotic therapy. Kaplan-Meier analysis and log-rank test showed a tendency toward a low risk of ICH during FU in patients on antithrombotic therapy (p = 0.085), as well as those on statin therapy (p = 0.193). The cumulative 5-year risk of bleeding was 22.82% (95% CI 17.33%-29.38%) for the entire cohort, 31.41% (95% CI 23.26%-40.83%) for patients with a history of ICH, 26.54% (95% CI 11.13%-49.7%) for individuals on beta-blocker medication, 6.25% (95% CI 0.33%-32.29%) for patients on statin medication, and 0% (95% CI 0%-30.13%) for patients on antithrombotic medication.

CONCLUSIONS

ICH at diagnosis was identified as a risk factor for recurrent hemorrhage. Although the relationships were not statistically significant, statin and antithrombotic medication tended to be associated with decreased bleeding events.

Five-year symptomatic hemorrhage risk of untreated brainstem cavernous malformations in a prospective cohort

Hemorrhage of brainstem cavernous malformation (CM) would cause various symptoms and severe disability. The study aimed to elaborate on the 5-year actuarial cumulative hazard of symptomatic hemorrhage. Patients diagnosed in our institute between 2009 and 2013 were prospectively registered. All clinical data were obtained, follow-up was performed, and risk factors were evaluated. Four hundred sixty-eight patients (217 female, 46.4%) were included in the study with a median follow-up duration of 79.0 months. A total of 137 prospective hemorrhages occurred in 107 patients (22.9%) during 1854.0 patient-years. Multivariate Cox analysis found age ≥ 55 years (hazard ratio (HR) 2.166, p = 0.002), DVA (HR 1.576, p = 0.026), superficial-seated location (HR 1.530, p = 0.047), and hemorrhage on admission (HR 2.419, p = 0.026) as independent risk factors for hemorrhage. The 5-year cumulative hazard of hemorrhage was 30.8% for the overall cohort, 47.8% for 60 patients with age ≥ 55 years, 43.7% for 146 patients with DVA, 37.9% for 272 patients with superficial-seated lesions, and 37.2% for 341 patients with hemorrhage on admission. As a stratified analysis, within subcohort of 341 patients with a hemorrhagic presentation, age ≥ 55 years (HR 3.005, p < 0.001), DVA (HR 1.801, p = 0.010), and superficial-seated location (HR 2.276, p = 0.001) remained independently significant. The 5-year cumulative hazard of hemorrhage was 52.0% for 119 patients with both DVA and hemorrhagic presentation. The 5-year cumulative hemorrhagic risk was 30.8% and was higher in subgroups if harboring risk factors that helped to predict potential hemorrhagic candidates and were useful for treatment decision-making.Clinical Trial Registration-URL: http://www.chictr.org.cn Unique identifier: ChiCTR-POC-17011575.

Natural Course of Cerebral Cavernous Malformations in Children: A Five-Year Follow-Up Study

BACKGROUND AND PURPOSE

The purpose of this study was to investigate the natural course of cerebral cavernous malformations (CCM) in the pediatric population, with special emphasis on the risk of first and recurrent bleeding over a 5-year period.

METHODS

Our institutional database was screened for patients with CCM treated between 2003 and 2020. Patients ≤18 years of age with complete magnetic resonance imaging data set, clinical baseline characteristics, and ≥1 follow-up examination were included. Surgically treated individuals were censored after CCM removal. We assessed the impact of various parameters on first or recurrent intracerebral hemorrhage (ICH) at diagnosis using univariate and multivariate logistic regression adjusted for age and sex. Kaplan-Meier and Cox regression analyses were performed to determine the cumulative 5-year risk for (re)hemorrhage.

RESULTS

One hundred twenty-nine pediatric patients with CCM were analyzed. Univariate logistic regression identified brain stem CCM (odds ratio, 3.15 [95% CI, 1.15-8.63], P=0.026) and familial history of CCM (odds ratio, 2.47 [95% CI, 1.04-5.86], P=0.041) as statistically significant predictors of ICH at diagnosis. Multivariate logistic regression confirmed this correlation (odds ratio, 3.62 [95% CI, 1.18-8.99], P=0.022 and odds ratio, 2.53 [95% CI, 1.07-5.98], P=0.035, respectively). Cox regression analysis identified ICH as mode of presentation (hazard ratio, 14.01 [95% CI, 1.80-110.39], P=0.012) as an independent predictor for rehemorrhage during the 5-year follow-up. The cumulative 5-year risk of (re)bleeding was 15.9% (95% CI, 10.2%-23.6%) for the entire cohort, 30.2% (20.2%-42.3%) for pediatric patients with ICH at diagnosis, and 29.5% (95% CI, 13.9%-51.1%) for children with brain stem CCM.

CONCLUSIONS

Pediatric patients with brain stem CCM and familial history of CCM have a higher risk of ICH as mode of presentation. During untreated 5-year follow-up, they revealed a similar risk of (re)hemorrhage compared to adult patients. The probability of (re)bleeding increases over time, especially in cases with ICH at presentation or brain stem localization.