Background The use of endovascular thrombectomy (EVT) is increasing in patients with large vessel occlusion (LVO) strokes. A few areas of uncertainty remain, and multiple trials are under way to address some of these questions. To guide future trials and to address practice gaps, knowledge of the cases where physicians are hesitant to offer EVT will be helpful.

Methods An international web-based survey was sent to practitioners in 38 countries to assess their decision-making toward selecting the management approach for LVO strokes in real-world. Ten case scenarios were randomly presented to each respondent from a pool of 22 cases. Participants were asked to choose whether they would offer EVT in that scenario in their current practice settings. We identified and summarized the cases in which the decision to perform EVT was significantly lower than the overall median for EVT use. We investigated the factors that may have influenced the decision to withhold EVT and the level of evidence for EVT in these cases according to the American Stroke Association guidelines.

Results 607 physicians (mean age of 44 (SD 8.5) years, 83.5% men, 53.6% neurologists, 28.7% neuro-interventionists, 13.3% neurosurgeons, 4.7% other) participated. The median number of participants who scored each scenario was 276. In the overall cohort, the median responses in favour of EVT in the participants’ current practice setting were 78.7%.

Ten scenarios received a significantly lower response in favour of EVT than the cohort median. These cases had a median EVT use of 60.3%. Nine scenarios describe cases in which level 2B guideline recommendation for EVT exist while one scenario had a 1A level of evidence.

The scenarios with low EVT use describe patients with poor baseline functional status (EVT use 66.4%), patients with M2-occlusion (EVT use 61.3%), mild clinical deficits (EVT use 60.4%), baseline ASPECTS ≤ 4 (EVT use 57%), or with isolated intracranial carotid occlusion but patent MCA (EVT use 55.6%).

Conclusion Occlusion site, clinical symptoms severity, and low ASPECTS scores were determinant factors in the low responses in favour of EVT in our survey.

Disclosures N. Kashani: None. J. Ospel: None. A. Wilson: None. M. Hill: None. B. Menon: None. G. Saposnik: None. M. Goyal: None. M. Almekhlafi: None.