Purpose We planned to compare the prognostic value of CT based parameters of infarct volume, collateral status, and thrombus burden in acute stroke patients treated with endovascular reperfusion therapy.
Methods Consecutive patients treated with anterior circulation occlusions confirmed by multiphase CT angiography (mCTA) that performed endovascular reperfusion therapy were included. MCTA was performed as previously reported. Collaterals were evaluated by single phase CT angiography (sCTA) and mCTA. Alberta Stroke program early CT score (ASPECTS) was analyzed for infarct burden measurement. Clot burden score was assessed for thrombus burden. Factors associated with good outcome (3 month mRS of 0˜2) and poor outcomes (3 month mRS of 5˜6) were analysed.
Results 84 patients were included. Intravenous thrombolysis was performed in 25%. Recan success was achieved in 75/84 (89.3%). A total of 49/84 (58.3%) were functionally independent at 3 months. For receiver operating curve analysis of good outcomes, C-statistics for sCTA was 0.683, followed by mCTA 0.645, and ASPECTS 0.620. However, when collateral status was incorporated in a multivariate analysis for predicting good outcomes including ASPECTS among other parameters, its significance faltered, while the significance of ASPECTS remained. For receiver operating curve analysis of good outcomes, c-statistics for sCTA was 0.758, followed by mCTA 0.740, and ASPECTS 0.603. In the multivariate analysis, both sCTA (odd ratio: 0.225, 95% confidence interval: 0.082–0.617, p=0.004) and mCTA (0.123, 0.027–0.565. p=0.007) predicted poor outcomes including ASPECTS among other parameters. Thrombus burden did not correlate with 3 month functional status.
Conclusions Infarct volume accessed by ASPECTS strongly predict positive outcome, while collateral status accessed by CT are better markers of negative outcomes. The added benefit of multiphase CT in grading of collaterals for prognostication is not evident in this study.
Disclosures J. Choi: None. S. Lee: None. M. Choi: None. S. Lee: None. J. Lee: None. J. Hong: None.
Statistics from Altmetric.com
If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.