A Systematic Review and Meta-Analysis of the Pathology Underlying Aneurysm Enhancement on Vessel Wall Imaging

Abstract

Intracranial aneurysms are common, but only a minority rupture and cause subarachnoid haemorrhage, presenting a dilemma regarding which to treat. Vessel wall imaging (VWI) is a contrast-enhanced magnetic resonance imaging (MRI) technique used to identify unstable aneurysms. The pathological basis of MR enhancement of aneurysms is the subject of debate. This review synthesises the literature to determine the pathological basis of VWI enhancement. PubMed and Embase searches were performed for studies reporting VWI of intracranial aneurysms and their correlated histological analysis. The risk of bias was assessed. Calculations of interdependence, univariate and multivariate analysis were performed. Of 228 publications identified, 7 met the eligibility criteria. Individual aneurysm data were extracted for 72 out of a total of 81 aneurysms. Univariate analysis showed macrophage markers (CD68 and MPO, p = 0.001 and p = 0.002), endothelial cell markers (CD34 and CD31, p = 0.007 and p = 0.003), glycans (Alcian blue, p = 0.003) and wall thickness (p = 0.030) were positively associated with enhancement. Aneurysm enhancement therefore appears to be associated with inflammatory infiltrate and neovascularisation. However, all these markers are correlated with each other, and the literature is limited in terms of the numbers of aneurysms analysed and the parameters considered. The data are therefore insufficient to determine if these associations are independent of each other or of aneurysm size, wall thickness and rupture status. Thus, the cause of aneurysm-wall enhancement currently remains unknown.

Keywords: aneurysm-wall enhancement; histology; intracranial aneurysm; magnetic resonance imaging; pathology; vessel wall imaging.

Figure 1

Flow diagram of the search strategy and study selection.

Figure 2

Forest plots of markers of enhancement. CD68, MPO, CD34, CD31 and Alcian blue were all significantly associated with enhancement when pooled. Although Alcian blue and CD31 were statistically significant, their odds ratio is extremely large, owing to the uncertainty of the data, as both only had 9 observations for each. Thrombus, vasa vasorum, CD3 and CD20 were not statistically significant. Thrombus and vasa vasorum included 4 ruptured aneurysms in their analysis [19,20,21,22,23,24,25]. Where the data is not available NA is shown.

Figure 3

A correlation matrix of histological features. The Cramer V association is shown in the lower left in red. The associated p value from a X² test is shown in blue on the top right of the matrix. NA is shown where there are no data that cross over to assess correlation.