Sebastian Ekkehard Beyer, Presenter: Nothing to Disclose

Louisa von Baumgarten, Abstract Co-Author: Nothing to Disclose

Kolja Thierfelder MD, MSc, Abstract Co-Author: Nothing to Disclose

Thorsten R. C. Johnson MD, Abstract Co-Author: Nothing to Disclose

Birgit Betina Ertl-Wagner MD, Abstract Co-Author: Nothing to Disclose

Maximilian F. Reiser MD, Abstract Co-Author: Nothing to Disclose

Wieland H. Sommer MD, Abstract Co-Author: Nothing to Disclose

The aim was (a) to compare the predictive value of the region of hypoattenuation vs. different collateral vessel grading scores and (b) to determine the optimal time-point for reconstruction of the dynamic CTA images in order to best predict follow-up lesion volume.

We included all patients with an M1±ICA occlusion, who had follow-up imaging, from an existing cohort of 1791 consecutive patients who underwent multimodal CT, including whole brain CT perfusion (WB-CTP), for suspected stroke. Collateralization was assessed in conventional and dynamic CTA using three different collateral vessel grading scores and segmentation of the region of hypoattenuation. To determine the optimal time-point for collateral assessment, arterial, arteriovenous, and venous phases were reconstructed for dynamic CTA. Follow-up lesion volume was assessed by MRI or NECT. Different collateral gradings systems were compared using the model fit of multivariate regression analyses (that were corrected for type of treatment) and calculation of the Bayesian information criterion (BIC). Differences in BIC of 2 to 6 were regarded as “positive”, 6 to 10 “strong”, and greater than 10 “very strong”.

Our study comprised 119 patients. In the multivariate analysis, models containing region of hypoattenuation showed a significantly better model fit than models containing collateral grading scores for conventional and dynamic CTA (ΔBIC > 10 for both). All collateral grading systems showed the best model fit for the arteriovenous phase. For region of hypoattenuation, model fit was signicifantly higher for the arteriovenous phase compared to the venous phase (ΔBIC = 6.2), the arterial phase (ΔBIC > 10), and conventional CTA (ΔBIC > 10). Also for collateral grading scores, model fit was higher for the arteriovenous phase compared to the venous phase (ΔBIC > 10 for each), the arterial phase (ΔBIC between 4 and 9), and conventional CTA (ΔBIC between 3 and 13).

Quantification of the region of hypoattenuation has a higher predictive value for follow-up lesion volume than collateral vessel grading scores. Arteriovenous phase is the optimal time-point for assessment of collateralization.

The assessment of collateral blood flow on CT angiographies requires methodological standardization in order to be used to predict tissue outcome.

Beyer, S, von Baumgarten, L, Thierfelder, K, Johnson, T, Ertl-Wagner, B, Reiser, M, Sommer, W, Region of Hypoattenuation vs. Different Leptomeningeal Collateral Grading Scores: Predictive Value for Follow-up Infarct Volume in Conventional and Dynamic CTA.  Radiological Society of North America 2014 Scientific Assembly and Annual Meeting, - ,Chicago IL. http://archive.rsna.org/2014/14015501.html