E-255 Catheter blood vessel ratio for middle cerebral artery occlusions requiring mechanical thrombectomy: a propensity adjusted analysis. (23rd July 2022)

• Record Type: Journal Article
• Authors: Catapano, J; Naik, A; Pacault, M; Sigh, R; Stonnington, H; Rangel, I; Koester, S; Winkler, E; Desai, S; Jadhav, A; Albuquerque, F; Ducruet, A
• Is Part Of: Journal of neurointerventional surgery. Volume 14(2022)Supplement 1
• Page Start: A218
• Page End: A219
• Publication Date: 2022-07-23
• Subjects: Nervous system -- Surgery -- Periodicals; Cerebrovascular disease -- Surgery -- Periodicals; 617.48
• DOI: 10.1136/neurintsurg-2022-SNIS.366
• Languages: English
• ISSNs: 1759-8478

Abstract

Abstract : Background: Mechanical thrombectomy is considered a gold standard procedure for the management of large vessel occlusions in the acute stroke setting. There are currently no guidelines present to select optimal catheter diameter, and such decisions are based on surgeon preference and experience. In this study, we seek to determine whether catheter-blood vessel-ratio (CVR) is predictive of reperfusion and patient outcomes including post-procedure intracerebral hemorrhage (ICH) for patients undergoing mechanical thrombectomy. Methods: A retrospective analysis was performed of all patients with a proximal middle cerebral artery (M1) occlusion at a large comprehensive stroke center who underwent a mechanical thrombectomy from 1/1/2020 to 6/30/2021. Study included patients with an available pre-intervention axial CTA in which cross-sectional diameter was measured of the occluded M1 proximally. An aspiration catheter outside diameter and vessel cross-sectional diameter ratio (CVR) was calculated. Patients were grouped and compared based on a CVR threshold of 0.9. Additional data extracted for analysis included: demographics, occlusion characteristics, intraoperative and post-operative management, and in-hospital and discharge outcomes. Univariate statistics used Welch's two-sample t-test for continuous data and chi-squared test for frequency-based variables. Multivariate analysis used multivariate linear and Firth's logistic regression. A propensity-score adjustment was used consisting of age, gender, admission NIHSS, ASPECT score, prior anti-coagulation, or anti-platelet use, TPA usage, and systolic blood pressure. Results: During the 19-month study period, 60 patients met inclusion criteria with 25 patients (42%) found to have CVR≥0.9 (vs 35 (58%) with CVR< 0.9). There was no difference between cohorts in demographics or patient presentation on univariate analysis. Of the 25 patients with a CVR≥0.9, 8% (N=2) had >2 passes compared to 20% (N=7 of 35) in the CVR< 0.9 cohort (p=0.22). Puncture to revascularization time, admission NIHSS, and discharge NIHSS were comparable between both cohorts. On a propensity adjusted multivariable logistic regression analysis a CVR of <0.9 was associated with a post-operative ICH (OR 3.4, 95% CI 1.1–12.5, p=0.047). However, the association became non-significant when number of passes was included into the regression model. Greater than 2 passes were associated with a lower odd of acceptable post-operative reperfusion (TICI ≥ 2B, OR = 0.06, p = 0.009). Additionally, in predicting discharge NIHSS, CVR added an δAUC of 0.13 and for TICI ≥ 2B prediction, CVR increased the AUC by 0.04. Conclusion: Lower CVR is associated with an increase odd of ICH following mechanical thrombectomy for M1 occlusions. However, this association appears to be confounded by number of passes. Additionally, CVR does improve the predictive value of linear and logistic models to predict NIHSS at discharge and TICI score. Larger prospective trials are warranted to estimate the impact of CVR for mechanical thrombectomy. Disclosures: J. Catapano: None. A. Naik: None. M. Pacault: None. R. Sigh: None. H. Stonnington: None. I. Rangel: None. S. Koester: None. E. Winkler: None. S. Desai: None. A. Jadhav: None. F. Albuquerque: None. A. Ducruet: None.