Background and Purpose Different imaging paradigms have been used to select patients for endovascular therapy (ET) in large vessel occlusion stroke (LVOS). We sought to determine whether CT perfusion (CTP) selection improves ET outcomes as compared to non-contrast CT (NCCT) alone.

Methods Review of a prospective single-center interventional database of consecutive patients between September 2010 and March 2016. Patients with anterior circulation strokes undergoing stent-retriever thrombectomy were categorized according to imaging selection: (1) CTP and (2) NCCT alone. Two separate analyses were performed to assess the impact of CTP selection on outcomes: (1) Uni- and Multivariate analysis of the overall cohort and (2) Matched case-control analysis based on age, baseline NIHSS, and glucose levels.

Results The overall cohort included 602 patients. CTP-selected patients (n=365; 61%) were younger (mean age 63.9 v±15.2 vs. 67.8±14.5, p=0.02) and had less comorbidities. On univariate analysis, CTP-selection was associated with higher rates of full reperfusion (mTICI-3: 54.8% vs. 40.1%, p<0.001), increased rates of good outcomes (90 day mRS 0–2: 52.9% vs. 40.4%, p=0.005), smaller final infarct volumes (24.7 cc [9.8–63.1] vs. 34.6 cc [13.1–88], p=0.017), lower mortality rate (16.6% vs. 26.8%, p=0.005), and a favorable shift in the overall distribution of 90 day mRS (p<0.001) as compared with NCCT alone. The rates of any parenchymal hematoma were comparable between groups (9% vs. 10.1%, p=0.671). Multivariate logistic regression showed that CTP was independently associated with full reperfusion (OR=1.79 95% CI [1.27–2.53], p=0.001) and good outcomes (aOR=1.72 95% CI [1.10–2.67], p=0.017). In the matched case-control analysis (n=424 patients), CTP-selection was associated with a favorable shift in the distribution of 90 day mRS (p=0.016), lower 90 day mortality (15.7% vs. 23.6%, p=0.02), higher rates of TICI 3 reperfusion (54.8% vs. 40.1%, p<0.001), and a trend towards higher rates of 90 day independence (53 .% vs. 40%, p=0.06). There was an advantage in the ability of CTP to determine functional outcomes in patients presenting later than 6 hour (Akaike information criterion (AIC) 199.35 vs. 287.49 and Bayesian information criterion (BIC) 196.71 vs 283.27) and with an ASPECTS≤7 (AIC 216.69 vs 334.96 and BIC 213.6 vs 329.94).

Conclusion CTP-based selection is associated with a favorable shift in functional outcomes in patients undergoing stent-retriever thrombectomy. Future prospective studies are warranted.

Disclosures J. Grossberg: None. M. Bouslama: None. D. Haussen: None. L. Rebello: None. M. Frankel: None. R. Nogueira: None.