Hypoattenuation on CTA images with large vessel occlusion: timing affects conspicuity

Abstract

Purpose

Parenchymal hypoattenuation distal to occlusions on CTA source images (CTASI) is perceived because of the differences in tissue contrast compared to normally perfused tissue. This difference in conspicuity can be measured objectively. We evaluated the effect of contrast timing on the conspicuity of ischemic areas.

Methods

We collected consecutive patients, retrospectively, between 2012 and 2014 with large vessel occlusions that had dynamic multiphase CT angiography (CTA) and CT perfusion (CTP). We identified areas of low cerebral blood volume on CTP maps and drew the region of interest (ROI) on the corresponding CTASI. A second ROI was placed in an area of normally perfused tissue. We evaluated conspicuity by comparing the absolute and relative change in attenuation between ischemic and normally perfused tissue over seven time points.

Results

The median absolute and relative conspicuity was greatest at the peak arterial (8.6 HU (IQR 5.1–13.9); 1.15 (1.09–1.26)), notch (9.4 HU (5.8–14.9); 1.17 (1.10–1.27)), and peak venous phases (7.0 HU (3.1–12.7); 1.13 (1.05–1.23)) compared to other portions of the time-attenuation curve (TAC). There was a significant effect of phase on the TAC for the conspicuity of ischemic vs normally perfused areas (P < 0.00001).

Conclusion

The conspicuity of ischemic areas distal to a large artery occlusion in acute stroke is dependent on the phase of contrast arrival with dynamic CTASI and is objectively greatest in the mid-phase of the TAC.

http://ift.tt/2oG2g0c