Background and purpose Perfusion imaging with RAPID post-processing analysis estimates volume thresholds/mismatch ratios of infarcted versus potentially salvageable tissue, providing standardization in the selection of acute ischemic stroke (AIS) patients for mechanical thrombectomy. Although CTP/MRP RAPID selection has been associated with improved clinical outcomes in several randomized controlled trials, variability of imaging selection across multiple trials and methodological concerns of perfusion metrics to define ischemia has resulted in an uncertainty of excluding patients that could benefit from endovascular thrombectomy, especially in the early <6 hour intervention window. We aimed to study the differences, if any, in outcomes of AIS patient populations selected for thrombectomy with standard CT/CTA versus selective CT perfusion imaging techniques.

Material and methods We compared separate multi-institutional prospective neurointerventional AIS databases (same operators) of consecutive populations that underwent either standard CT/CTA imaging without or with CTP selection for mechanical thrombectomy from November 2015-March 2018. AIS patients met study inclusion criteria if presenting with NIHSS >5 and within <6 hours of symptom onset. CT/CTA imaging inclusion criteria required ASPECTS >5 and internal carotid artery (ICA) and/or M1 middle cerebral artery (MCA) occlusion in both cohorts. In the CTP selection cohort, thrombectomy inclusion criteria required automated RAPID software to determine appropriate target mismatch profile as per DEFUSE-2 criteria. We assessed demographics, comorbidities, treatment times, efficacy/safety, mortality, and overall clinical outcomes of both CT/CTA and CTP cohorts (combining included and excluded populations). Statistical analysis for clinical outcomes was performed on percentage of patients achieving functional independence (90 day modified Rankin Score 0–2) using a normal approximation of the Wilcoxon-Mann-Whitney test (the generalized likelihood ratio test).

Results In total 167 patients met clinical inclusion criteria, 75 patients underwent standard CT/CTA imaging selection/intervention without exclusion versus 92 patients that underwent CTP selection (50 thrombectomy:42 excluded). Both cohorts demonstrated equivalent demographics (mean age: 70 vs 69, p=0.48), comorbidities, neurological presentations (median NIHSS: 18 vs 16, p=0.13), IV tPA utilization (82% vs 74%, p=0.42), ICA occlusions (34.2% vs 37.2%, p=0.62), median symptom onset to revascularization times (270 min vs 265 min, p=0.5), and TICI 2b/3 reperfusion rates (88% vs 86%, p=0.55). There was no significant difference in SICH complications (10.7% vs 8%, p=0.19) or 90 day mortality (11.7% vs 12%, p=0.99). Although patients who underwent mechanical thrombectomy revealed that CTP selection offered higher independence rates at 90 days than standard CT/CTA selection (42% vs 29.3%, p=0.02), overall CTP selection cohorts (included and excluded) independence rates at 90 days were similar (36.9% vs 29.3%, p=0.3).

Conclusion Although CTP selection increases the rate of favorable clinical outcomes in patients treated with mechanical thrombectomy, there may be no overall population benefit or harm in comparison to standard CT/CTA selection alone in the early interventional time window. Further randomized studies would be required to control for institutional and imaging selection bias to assess if any trends towards greater population benefit with CT/CTA selection may be realized, or conversely if futile thrombectomy can be curtailed.

Disclosures M. Darwish: None. P. Golnari: None. A. Muzaffar: None. A. Shaibani: None. M. Hurley: None. M. Potts: None. B. Jahromi: None. S. Ansari: None.