Abstract TP82: Factors Associated With Good Collateral Flow in Acute Stroke Patients With Large Vessel Occlusion
Introduction: Retrograde collateral flow is critical to maintain tissue perfusion despite large vessel occlusion. However, the premorbid factors associated with good collateral flow remain unclear, with substantial discrepancies in the literature. Methods: Patients from the registries of 6 French stroke centres with the following criteria were included: (1) acute stroke with isolated M1 occlusion ( i.e , without tandem occlusion) referred for thrombectomy between May 2015 and March 2017; and (2) baseline brain MRI, including diffusion weighted imaging, MR-angiography and dynamic susceptibility-contrast perfusion-weighted imaging (PWI). A collateral flow map derived from PWI source data was automatically generated, replicating Kim et al’s previously validated method (Ann. Neurol., 2014). Collateral flow was dichotomized into good and poor. The association between good collateral flow and baseline clinical, biological and radiological variables was studied. Results: One hundred and sixteen patients were included, of which 66 (57%) had good collaterals. As expected, the latter patients had lower admission NIHSS (median: 15 vs . 18, P=0.005) and lower baseline DWI lesion volume (median: 7ml vs . 32ml, P<0.001) than patient with poor collaterals. Onset-to-imaging delay and M1 occlusion site (proximal vs . distal) were similar in both groups (123min vs . 118min, P=0.75; 70% vs. 68%, P=0.85, respectively). There was no significant (P>0.05) difference in gender, age, history of hypertension or diabetes, current smoking, baseline blood glucose and use of statins or antiplatelets between the good and poor collaterals groups. Conclusions: Despite the expected association between PWI-derived collateral flow and baseline clinical and radiological stroke severity in our sample of acute M1 occlusions, there was no association with premorbid factors previously found associated in some -but not all- studies. Our findings suggest that collateral flow is predominantly explained by genetic factors.