The predictors of clinical outcomes in brainstem arteriovenous malformations after stereotactic radiosurgery

Prediction of Obliteration After the Gamma Knife Radiosurgery of Arteriovenous Malformations Using Hand-Crafted Radiomics and Deep-Learning Methods

INTRODUCTION

Brain arteriovenous malformations (bAVMs) are vascular abnormalities that can be treated with embolization or radiotherapy to prevent the risk of future rupture. In this study, we use hand-crafted radiomics and deep learning techniques to predict favorable vs. unfavorable outcomes following Gamma Knife radiosurgery (GKRS) of bAVMs and compare their prediction performances.

METHODS

One hundred twenty-six patients seen at one academic medical center for GKRS obliteration of bAVMs over 15 years were retrospectively reviewed. Forty-two patients met the inclusion criteria. Favorable outcomes were defined as complete nidus obliteration demonstrated on cerebral angiogram and asymptomatic recovery. Unfavorable outcomes were defined as incomplete obliteration or complications relating to the AVM that developed after GKRS. Outcome predictions were made using a random forest model with hand-crafted radiomic features and a fine-tuned ResNet-34 convolutional neural network (CNN) model. The performance was evaluated by using a ten-fold cross-validation technique.

RESULTS

The average accuracy and area-under-curve (AUC) values of the Random Forest Classifier (RFC) with radiomics features were 68.5 ±9.80% and 0.705 ±0.086, whereas those of the ResNet-34 model were 60.0 ±11.9% and 0.694 ±0.124. Four radiomics features used with RFC discriminated unfavorable response cases from favorable response cases with statistical significance. When cropped images were used with ResNet-34, the accuracy and AUC decreased to 59.3 ± 14.2% and 55.4 ±10.4%, respectively.

CONCLUSIONS

A hand-crafted radiomics model and a pre-trained CNN model can be fine-tuned on pre-treatment MRI scans to predict clinical outcomes of AVM patients undergoing GKRS with equivalent prediction performance. The outcome predictions are promising but require further external validation on more patients.

Effect of Cobalt-60 Treatment Dose Rate on Arteriovenous Malformation Obliteration After Stereotactic Radiosurgery With Gamma Knife

BACKGROUND AND OBJECTIVES

Stereotactic radiosurgery (SRS) marginal dose is associated with successful obliteration of cerebral arteriovenous malformations (AVM). SRS dose rate-how old the cobalt-60 sources are-is known to influence outcomes for some neurological conditions and benign tumors. The objective of this study was to determine the association between cobalt-60 treatment dose rate and cerebral AVM obliteration in patients treated with SRS.

METHODS

We performed a retrospective cohort study of 361 patients undergoing 411 AVM-directed SRS treatments between 2005 and 2019 at a single institution. Lesion characteristics, SRS details, obliteration dates, and post-treatment toxicities were recorded. Univariate and multivariate regression analyses of AVM outcomes regarding SRS dose rate (range 1.3-3.7 Gy, mean = 2.4 Gy, median = 2.5 Gy) were performed.

RESULTS

At 10 years post-SRS, 68% of AVMs were obliterated on follow-up imaging. Dose rates >2.9 Gy/min were found to be significantly associated with AVM obliteration compared with those <2.1 Gy/min ( P = .034). AVM size, biologically effective dose, and SRS marginal dose were also associated with obliteration, with obliteration more likely for smaller lesions, higher biologically effective dose, and higher marginal dose. Higher dose rates were not associated with the development of post-SRS radiological or symptomatic edema, although larger AVM volume was associated with both types of edema.

CONCLUSION

Patients with cerebral AVMs treated with higher SRS dose rates (from newer cobalt-60 sources) experience higher incidences of obliteration without a significant change in the risk of post-treatment edema.

Embolization of midbrain arteriovenous malformation fed by the artery of Percheron in a child, the first case report and literature review

BACKGROUND

Artery of Percheron (AOP) as main feeder artery of arteriovenous malformation (AVM) is extremely rare. Two cases of thalamic AVM fed by AOP have been reported to date and only one AVM been removed by microsurgery when attempt of intervention embolization failed. Midbrain AVM fed by AOP has not been reported yet.

CASE PRESENTATION

Here, we presented the first successful embolized case of midbrain AVM supplied by the AOP in a 10-year-old boy, who suffered dual oculomotor nerve palsy and secondary hemorrhage before embolization. During endovascular embolization, selective angiography by 1.2 Fr. Magic microcatheter showed an intranidal aneurysm located on the distal AOP. Two injections of a 1:4 ratio mixture of NBCA-MS completely occlude the nidus and intranidal aneurysm with no complications occurred. The child recovered well and the oculomotor deficits improved.

CONCLUSION

This case highlighted that AOP is a clinically significant branch associated with AVM in midbrain and thalamus. Moreover, intervention embolization of midbrain AVM fed by AOP is a considerable therapeutic strategy.

Stereotactic radiosurgery outcome for deep-seated cerebral arteriovenous malformations in the brainstem and thalamus/basal ganglia: systematic review and meta-analysis

Deep-seated unruptured AVMs located in the thalamus, basal ganglia, or brainstem have a higher risk of hemorrhage compared to superficial AVMs and surgical resection is more challenging. Our systematic review and meta-analysis provide a comprehensive summary of the stereotactic radiosurgery (SRS) outcomes for deep-seated AVMs. This study follows the guidelines set forth by the Preferred Reporting Items for Systematic Reviews and Meta-Analysis (PRISMA) Statement. We conducted a systematic search in December 2022 for all reports of deep-seated arteriovenous malformations treated with SRS. Thirty-four studies (2508 patients) were included. The mean obliteration rate in brainstem AVM was 67% (95% CI: 0.60-0.73), with significant inter-study heterogeneity (tau2 = 0.0113, I2 = 67%, chi2 = 55.33, df = 16, p-value < 0.01). The mean obliteration rate in basal ganglia/thalamus AVM was 65% (95% CI: 0.58-0.72) with significant inter-study heterogeneity (tau2 = 0.0150, I2 = 78%, chi2 = 81.79, df = 15, p-value < 0.01). The presence of deep draining veins (p-value: 0.02) and marginal radiation dose (p-value: 0.04) were positively correlated with obliteration rate in brainstem AVMs. The mean incidence of hemorrhage after treatment was 7% for the brainstem and 9% for basal ganglia/thalamus AVMs (95% CI: 0.05-0.09 and 95% CI: 0.05-0.12, respectively). The meta-regression analysis demonstrated a significant positive correlation (p-value < 0.001) between post-operative hemorrhagic events and several factors, including ruptured lesion, previous surgery, and Ponce C classification in basal ganglia/thalamus AVMs. The present study found that radiosurgery appears to be a safe and effective modality in treating brainstem, thalamus, and basal ganglia AVMs, as evidenced by satisfactory rates of lesion obliteration and post-surgical hemorrhage.