Correlation between intracranial vertebral artery stenosis diameter measured by digital subtraction angiography and cross-sectional area measured by optical coherence tomography

Abstract

BackgroundIntracranial vertebral artery (V4 segment) stenosis quantification traditionally uses the narrowest stenosis diameter. However, the stenotic V4 lumen is commonly irregularly shaped. Optical coherence tomography (OCT) allows a more precise calculation of V4 geometry. We compared the narrowest diameter stenosis (DS), measured by digital subtraction angiography (DSA), with the area stenosis (AS), measured by OCT. We hypothesized that DS is the gold standard for measuring the degree of stenosis.MethodsFive neuroradiologists evaluated 49 stenosed V4 segments in a blinded protocol. V4 stenosis was measured in millimeters on DSA at its narrowest diameter. OCT was used to estimate the cross-sectional luminal area. We also used automated software to measure DS. Three different angles (anterior, lateral, and oblique views) were used for calculations, and the North American Symptomatic Carotid Endarterectomy Trial (NASCET) and Warfarin–Aspirin Symptomatic Intracranial Disease (WASID) methods were used in all measurements. Spearman’s R values were calculated. Non-linear regression analysis was performed between the DS and AS, with statistically different correlations.ResultsA high correlation was observed between the WASID and NASCET methods to measure DS with observer measurement and automated software. A good correlation was found between DS measured by observers and AS measured by OCT. Non-linear regression analysis showed that only observer measurement using the oblique view and the WASID method could attain statistically significant differences, but it was weak (r=0.389).ConclusionMeasurement of the narrowest diameter was not a reliable predictor of the cross-sectional area of V4 stenosis. Larger studies are therefore needed to develop a new evaluation system based on V4 stenosis.