Abstract WP190: Comparison of Collateral Score to ASPECTS to Exclude "Very Large" Infarct Core Volumes in Anterior Circulation Large Vessel Occlusion Strokes at 0-24 Hours

Sidney Lin,Avtar Raince,Kory Byrns
DOI: https://doi.org/10.1161/str.55.suppl_1.wp190
IF: 10.17
2024-02-01
Stroke
Abstract:Purpose: ASPECTS 100mL) for reperfusion using endovascular thrombectomy (EVT). While specific, ASPECTS 100mL while maintaining similar specificity to ASPECTS. Methods: Retrospective analysis included consecutive stroke patients arriving within 24 hours of onset, with intracranial ICA and/or M1 occlusion on CTA, and had concurrent CT perfusion. RAPID AI software estimated ICV using the rCBF<30% threshold. CS was scored on CTA using the method of Tan et al on a 4-point scale for % collaterals compared to contralateral side: 0 none, 1 >0 but <50%, 2 >=50 but <100%, and 3 100%. ROC curve analysis was used to find the optimal CS cutpoint with high specificity for ICV>100mL. Sensitivity and specificity of that CS threshold vs ASPECTS<3 were compared using the N-1 chi square test. Results: 164 patients were included with median (IQR) age of 74 (66-84), NIHSS of 20 (12-25), and time of onset-to-CT of 4 hours (2-7.5). The LVOs were 17 ICA (10.4%), 108 M1 (65.9%), and 39 both (23.8%), with median ICV of 12mL (0-45). 16 patients had ICV>100mL (9.8%) with median of 136mL (105-172). Using ASPECTS<3 identified only 1 of these (ICV of 118mL at 18.5 hours), sensitivity of 6.3% (95% CI: 0.16-30.2%), with no false positives, specificity of 100% (95% CI: 97.6-100%). ROC analysis for CS showed AUC=0.818 (p<0.001), with maximal specificity at CS=0 (no visible collaterals). Using CS=0 identified 6 of these infarcts, sensitivity of 37.5% [95% CI: 15.2-64.6%], with 2 false positives, specificity of 98.7% [95% CI: 95.2-99.8%]. The 2 false positives were ICV of 97mL at 2 hours and 96mL at 3 hours. The improved sensitivity of CS=0 over ASPECTS<3 was significant (p=0.036) while the difference in specificity was not (p=0.144). Conclusion: In our cohort of LVOs, CS=0 was a more sensitive surrogate marker for ICV>100mL compared to ASPECTS<3 while maintaining near 100% specificity. If confirmed in future prospective studies, CS=0 has the potential to further streamline patient selection for EVT.
peripheral vascular disease,clinical neurology
What problem does this paper attempt to address?