Abstract WP170: Patients With Cerebral Microbleeds Are More Likely to Have Poor Collaterals at the Time of Mechanical Thrombectomy

Abstract

Introduction: Collateral cerebrovasculature is important for the survival of penumbral tissue in patients with large vessel occlusive (LVO) stroke, and adequate collateralization is associated with favorable outcomes. The presence of cerebral microbleeds (CMB) have been implicated in vessel wall remodeling and impaired vasoreactivity. We investigated whether the presence of deep and/or lobar CMB predicts poor collateral recruitment in patients with LVO stroke who underwent thrombectomy (MT). Methods: We performed a retrospective, single-center study of adults who presented with LVO stroke and had MT between January 2012-June 2020. CT angiography of head prior to MT was used to assess collateral status with poor status defined as ≤50% filling and good status as >50% filling. CMBs were assessed using T2*-weighted gradient echo or susceptibility weighted magnetic resonance images. Location of CMB was further characterized as deep vs. lobar using consensus definitions. Poor stroke outcome was defined as modified Rankin Scale (mRS) ≥3 at 90 days. Multivariable logistic regressions were performed to evaluate the associations between CMB, CMB types (lobar, deep, lobar+deep) and poor collaterals after adjusting for vascular risk factors. Results: Consecutive 151 subjects were included (mean age 68±15 years, 47% women), with a median NIHSS of 18. CMB were identified in 57 subjects (37.4%). Of these, 20.5% had lobar CMB, 5.3% had deep CMB and 11.9% had combined lobar & deep CMB. Patients with CMB had higher odds of having poor collaterals compared to those without CMB (OR 1.56, 95% CI 1.09-2.23, P=0.015). The association of CMB with poor collaterals was strongest in patients with combined lobar & deep CMB (OR 4.78, 95% CI 1.43-15.93, P=0.011), compared with those with exclusively deep CMB (OR 2.78, 95% CI 0.44-17.62, P=0.278) or lobar CMB (OR 0.52, 95% CI 0.19-1.38, P=0.188). Patients with poor collaterals were more likely to have poor outcome compared with those with good collaterals (48.4% vs. 31.0%, P=0.03). Conclusion: Cerebral microbleeds are independently associated with poor collateral recruitment when adjusted for traditional vascular risk factors. Further study is warranted to elucidate the differential effects of CMB types on collateral recruitment.