Purpose: We observed that in several cases of embolectomy treated acute stroke patients complicated by parenchymal hematoma (PH), there was marked susceptibility change (“blooming artifact”) on admission gradient echo MRI sequences surrounding the occlusive thrombus. We sought to determine if this imaging feature was associated with hemorrhagic risk. Methods: We identified 91 consecutive embolectomy treated patients who had admission susceptibility weighted MRI sequences (51 MPGR, 26 perfusion GRE, and 14 low B) and follow-up imaging. PH was defined as PH1 or PH2 per ECASS criteria. A “susceptibilty ratio score” (SRS) was determined by dividing the maximum width of the blooming artifact of the involved vessel (n=69 proximal MCA, 13 ICA/MCA, and 10 terminal ICA) by that of the corresponding segment of the contralateral uninvolved vessel. Patients were stratified according to SRS <2 (Group 1), 2-4 (Group 2), or >4 (Group 3). Univariate and multivariate analyses were performed to test the association between SRS scores and PH on post-treatment follow-up imaging. Results: Mean age was 67.3 ± 17.7, and median NIHSS score was 17 (IQR 14-20). The mean SRS was 2.5 ± 1.3. Using ECASS criteria, no hemorrhages were seen in 46 (50.5%); HI1 in 26 (28.6%); HI2 in 10 (11.0%); PH1 in 4 (4.4%); and PH2 in 5 (5.5%). The PH rate was 0% (0/22) for Group 1, 9.6% (5/52) for Group 2, and 24% (4/17) for Group 3 (p=0.04, two-tailed Fishers Exact Test). Adjusting for age, NIHSS score, TICI reperfusion, and IV tPA pre-treatment, ordinal SRS was an independent predictor of parenchymal hematoma (O.R. 4.13, 95%C.I.:1.2-13.8; p=0.02). There was an interaction between ordinal SRS and IV tPA (multivariate p=0.02 for interaction term), such that IV tPA prior to embolectomy resulted in a higher rate of PH in the SRS >4 group (33.3% vs 12.5%). Conclusion: An easy to calculate “susceptibility ratio score” appears to be associated with development of PH in acute stroke patients treated with mechanical thrombectomy, with low risk (SRS<2, 0% PH incidence) and high risk (SRS>4, 24% PH incidence) groups. For the high risk group, PH was more likely in those with IV tPA pre-treatment. We speculate that this imaging sign may be a biomarker for occlusive thrombus burden and/or the degree of baseline vessel wall injury surrounding the occlusive clot. Replication of this novel observation in independent cohorts is needed. Figure 1 A. Sample measurement for calculation of SRS from MPGR image. Figure 1 B. CT scan post mechanical thrombectomy demonstrates acute development of PH2 superior to the region of maximal susceptibility.