Pathomorphometry of ruptured intracranial vertebral arterial dissection: adventitial rupture, dilated lesion, intimal tear, and medial defect.

OBJECT: Subarachnoid hemorrhage (SAH) due to ruptured intracranial vertebral artery (VA) dissection is a life-threatening disease. Angiographic and symptomatic prognostic factors for rupture and rerupture have been investigated, but the pathological characteristics have not been fully investigated. The authors aimed to investigate these features by performing a pathomorphometic study of ruptured intracranial VA dissections.
METHODS: This study included 50 administrative autopsy cases of fatal SAH due to ruptured intracranial VA dissection among 517 fatal nontraumatic cases of SAH occurring between March 2003 and May 2011. Pathomorphometry was performed using serial 5-μm histological cross-sections with elastica van Gieson staining from each 0.2-mm segment around the ruptured intracranial VA. The longitudinal lengths of 4 types of vascular lesions-adventitial ruptures, dilated lesions where the internal elastic lamina (IEL) was ruptured with adventitial extension, intimal tears where the IEL was ruptured, and medial defects-were calculated based on the numbers of the slides in which these lesions were continuously detected (minimum 2 adjoining slides). The distance from the vertebrobasilar junction to the center of adventitial rupture was also calculated in 37 cases.
RESULTS: All cases showed one adventitial rupture with a mean length of about 1.9±1.1 mm. The center of the adventitial rupture was located 5.0-26.8 mm (mean 14.6±5.5 mm) from the intracranial VA bifurcation. Adventitial ruptures existed in the centers of dilated lesions, where the adventitia was highly extended. Other vascular lesions were serially observed surrounding the adventitial rupture. The mean lengths of dilated lesions, intimal tears, and medial defects were 9.4±4.8 mm, 13.2±6.3 mm, and 15.6±7.2 mm, respectively. The lengths between proximal lesions and distal lesions from the center of the adventitial rupture for both medial defects and intimal tears were significantly longer at proximal lesions than at distal ones (chi-square test, p<0.01).
CONCLUSIONS: Every ruptured intracranial VA dissection has a single point of adventitial rupture where the adventitia was maximally extended, so dilation appears to be a valuable predictive factor for hemorrhagic intracranial VA dissections. The adventitial ruptures were as small as 2 mm in length, and clinically detectable dilated lesions were about 9 mm in length. However, vascular vulnerability caused by IEL ruptures and medial defects existed more widely across a length of VA of 1.3-1.5 cm. Comparatively broader protection of the intracranial VA than the clinically detected area of dissection might be desirable to prevent rebleeding. Broader protection of proximal lesions than distal lesions might be effective from the viewpoint of site distribution of vascular lesions and blood flow alteration to the pseudolumen caused by the dissecting hematoma. Medial defects are the most widely seen lesions among the 4 types of vascular lesions studied. Medial degenerative disease, known as segmental arterial mediolysis, is suspected in the pathogenesis of intracranial VA dissections.