Background: The benefit of mechanical thrombectomy (MT) in acute ischemic stroke (AIS) patients with M2 occlusion is uncertain. Observational studies suggest that perfusion imaging may help select optimal M2 candidates for MT. We aimed to study the prevalence and factors associated with core/perfusion mismatch, as assessed by both CT and MRI, in an unselected population of AIS with M2 occlusion. Methods: Retrospective observational study including AIS patients with M2 occlusion and baseline perfusion imaging (MRI or CT) performed within 24hrs from last seen well. To avoid selection bias, patients were selected from the prospective registries of 3 centers with systematic use of perfusion imaging, including both MT-treated and untreated patients. Core/perfusion mismatch was defined as mismatch ratio (Tmax>6s/ core volume) >1.8 and mismatch volume (Tmax>6s - core volume) >15mL. Results: Overall, 156 and 196 patients were included in the MRI and CT cohorts, respectively. Last-seen-well to imaging time was 5.6hrs (IQR 3.9-9.7) and 3.5hrs (1.5-9.8) in the MRI and CT cohorts, and 103/156 (66%) and 134/196 (68%) had proximal M2 occlusions (the remainder had distal M2 occlusions). Median ASPECT score was 7 (5-8) and 8 (7-9), and core/perfusion mismatch was present in 70/156 (45%) and 167/196 (85%) patients. In both cohorts, mismatch was less frequent in distal M2 occlusions and among patients with lower ASPECTS (Figure). Multivariable analysis showed higher ASPECT scores (OR=3.09, 95%CI=2.11-4.52, P <0.001 and OR=2.41, 95%CI 1.63-3.58, P <0.001 in the MRI and CT cohorts) and proximal M2 occlusion (with distal M2 as a reference: OR=3.94, 95%CI=1.68-9.26, P =0.002 and OR=32.78, 95%CI 8.36-128.47, P <0.001 in the MRI and CT cohorts) were independently associated with core/perfusion mismatch. Conclusion: Perfusion imaging may be useful in triaging M2 patients, particularly for those with ASPECT score <8 and distal M2 occlusions. This has implication for trials.