Comparison of CT angiography collaterals for predicting target perfusion profile and clinical outcome in patients with acute ischemic stroke

Objectives: To compare collateral status on single-phase CT angiography (sCTA) and multiphase CT angiography (mCTA) and their ability to predict a target mismatch on CT perfusion (CTP) and clinical outcome in patients with acute ischemic stroke (AIS).

Methods: Seventy-three AIS patients with stroke onset between 5 and 15 h or with unclear onset time and occlusions in the M1/M2 segment of the middle cerebral artery and/or intracranial internal carotid artery underwent head non-contrast CT and CTP. Simulated sCTA and mCTA were reconstructed from CTP data and were compared for collaterals assessment. The ability to predict target mismatch on CTP (an ischemic core < 70 ml, a mismatch ratio ≥ 1.8, and an absolute difference ≥ 15 ml) and 90-day modified Rankin Scale (mRS) score of 0-2 was compared between sCTA and mCTA by using receiver operating curve analysis.

Results: sCTA underestimated the collateral status when compared with mCTA (p < 0.01). The ability of mCTA to predict target mismatch (AUC = 0.902, 95% confidence interval [CI] 0.809, 0.959) and clinical outcome (AUC = 0.771; 95% CI, 0.655, 0.864) was better than that of sCTA (p < 0.05 overall). A mCTA collateral score of > 3 best identified the target mismatch (sensitivity, 78.4%; specificity, 90.9%) and predicted 90-day mRS score of 0-2 (sensitivity, 84.8%; specificity, 69.4%).

Conclusions: The collaterals were better estimated by mCTA compared with sCTA. A mCTA collateral score of > 3 optimized the prediction of a target mismatch on CTP and a good clinical outcome in patients with AIS.

Key points: • Collateral circulation is a key determinant of ischemic core and penumbra. Better collaterals are associated with smaller ischemic core volumes and larger mismatch ratios on CT perfusion. • The collaterals can be better estimated by multiphase CTA compared with single-phase CTA. • A collateral score of > 3 on multiphase CTA best identifies patients with target mismatch on CT perfusion and predicts 90-day mRS score of 0-2.