A novel swine model of selective middle meningeal artery catheterization and embolization.

BACKGROUND: Middle meningeal artery (MMA) embolization is a promising intervention as a stand-alone or adjunct treatment to surgery in patients with chronic subdural hematomas. There are currently no large animal models for selective access and embolization of the MMA for preclinical evaluation of this endovascular modality. Our objective was to introduce a novel in vivo model of selective MMA embolization in swine. METHODS: Diagnostic cerebral angiography with selective microcatheter catheterization into the MMA was performed under general anesthesia in five swine. Anatomical variants in arterial meningeal supply were examined. In two animals, subsequent embolization of the MMA with a liquid embolic agent (Onyx-18) was performed, followed by brain tissue harvest and histological analysis. RESULTS: The MMA was consistently localized as a branch of the internal maxillary artery just distal to the origin of the ascending pharyngeal artery. Additional meningeal supply was observed from the external ophthalmic artery, although not present consistently. MMA embolization with Onyx was technically successful and feasible. Histological analysis showed Onyx material within the MMA lumen. CONCLUSIONS: Microcatheter access into the MMA in swine with liquid embolic agent delivery represents a reproducible model of MMA embolization. Anatomical variations in the distribution of arterial supply to the meninges exist. This model has a potential application for comparing therapeutic effects of various embolic agents in a preclinical setting that closely resembles the MMA embolization procedure in humans.

The first endovascular rat glioma model for pre-clinical evaluation of intra-arterial therapeutics.

BACKGROUND: Several translational animal models have been described assessing intra-arterial (IA) treatments for malignant gliomas. We describe the first endovascular animal model that allows testing of IA drug delivery as a first-line treatment, which is difficult to do in actual patients. We report a unique protocol for vascular access and IA delivery in the rat model that, unlike prior reports, does not require direct puncture and opening of proximal cerebrovasculature which carries risk of ischemia in the animal brain post-delivery. METHODS: Wistar rats underwent left femoral artery catherization with a Balt Magic 1.2F catheter or Marathon Flow directed 1.5F Microcatheter with an Asahi Chikai 0.008 micro-guidewire which was navigated to the left internal carotid artery under x-ray. 25% mannitol was administered to test blood brain barrier breakdown (BBBB). Additional rats were implanted with C6 glioma cells in the left frontal lobe. C6 Glioma-Implanted Rats (C6GRs) were monitored for overall survival and tumor growth. Tumor volumes from MRI images were calculated utilizing 3D slicer. Additional rats underwent femoral artery catheterization with Bevacizumab, carboplatin, or irinotecan injected into the left internal carotid artery to test feasibility and safety. RESULTS: A successful endovascular access and BBBB protocol was established. BBBB was confirmed with positive Evans blue staining. 10 rats were successfully implanted with C6 gliomas with confirmed growths on MRI. Overall survival was 19.75 ± 2.21 days. 5 rats were utilized for the development of our femoral catheterization protocol and BBBB testing. With regards to IA chemotherapy dosage testing, control rats tolerated targeted 10 mg/kg of bevascizumab, 2.4 mg/kg of carboplatin, and 15 mg/kg of irinotecan IA ICA injections without any complications. CONCLUSIONS: We present the first endovascular IA rat glioma model that allows selective catheterization of the intracranial vasculature and assessment of IA therapies for gliomas without need for access and sacrifice of proximal cerebrovasculature.

Tectonic infarct analysis: A computational tool for automated whole-brain infarct analysis from TTC-stained tissue.

Background: Infarct volume measured from 2,3,5-triphenyltetrazolium chloride (TTC)-stained brain slices is critical to in vivo stroke models. In this study, we developed an interactive, tunable, software that automatically computes whole-brain infarct metrics from serial TTC-stained brain sections. Methods: Three rat ischemic stroke cohorts were used in this study (Total n = 91 rats; Cohort 1 n = 21, Cohort 2 n = 40, Cohort 3 n = 30). For each, brains were serially-sliced, stained with TTC and scanned on both anterior and posterior sides. Ground truth annotation and infarct morphometric analysis (e.g., brain-Vbrain, infarct-Vinfarct, and non-infarct-Vnon-infarct volumes) were completed by domain experts. We used Cohort 1 for brain and infarct segmentation model development (n = 3 training cases with 36 slices [18 anterior and posterior faces], n = 18 testing cases with 218 slices [109 anterior and posterior faces]), as well as infarct morphometrics automation. The infarct quantification pipeline and pre-trained model were packaged as a standalone software and applied to Cohort 2, an internal validation dataset. Finally, software and model trainability were tested as a use-case with Cohort 3, a dataset from a separate institute. Results: Both high segmentation and statistically significant quantification performance (correlation between manual and software) were observed across all datasets. Segmentation performance: Cohort 1 brain accuracy = 0.95/f1-score = 0.90, infarct accuracy = 0.96/f1-score = 0.89; Cohort 2 brain accuracy = 0.97/f1-score = 0.90, infarct accuracy = 0.97/f1-score = 0.80; Cohort 3 brain accuracy = 0.96/f1-score = 0.92, infarct accuracy = 0.95/f1-score = 0.82. Infarct quantification (cohort average): Vbrain (ρ = 0.87, p < 0.001), Vinfarct (0.92, p < 0.001), Vnon-infarct (0.80, p < 0.001), %infarct (0.87, p = 0.001), and infarct:non-infact ratio (ρ = 0.92, p < 0.001). Conclusion: Tectonic Infarct Analysis software offers a robust and adaptable approach for rapid TTC-based stroke assessment.