Clinical outcomes of delayed mechanical thrombectomy: Descriptive analysis and development of a screening tool.

BACKGROUND AND PURPOSE: Limited data guide the selection of patients with large vessel occlusion ischaemic stroke who may benefit from referral to a distant tertiary centre for mechanical thrombectomy (MT). We aimed to characterize this population, describe clinical outcomes and develop a screening system to identify patients most likely to benfit from delayed mechanical thrombectomy (MT). METHODS: We undertook a retrospective cohort analysis enrolling patients transferred from regional sites to one of two MT comprehensive stroke units with a time from non-contrast computed tomography (NCCT) of the brain to reperfusion of 4 h or more. We describe Alberta Stroke Programme Early Computed Tomography Score (ASPECTS), National Institute of Health Stroke Scale (NIHSS) and modified Rankin Scale (mRS) in our patients and compare these patients to those in extended-time-window trials. Lastly, we developed and validated a scoring model to help clinicians identify appropriate patients based on variables associated with poor outcomes. RESULTS: We included 563 patients, 46% of whom received thrombolysis; the median (interquartile range [IQR]) ASPECTS was 8 (7-10) and the median (IQR) NIHSS score was 16 (11-20). The median (IQR) symptom to mechanical reperfusion time was 390 (300-580) min. Eight patients (1%) had a symptomatic haemorrhage. We achieved good clinical outcome (defined as mRS score ≤2) in 299 patients (54%). Age, diabetes, NIHSS score and ASPECTS were used to create a weighted scoring system with a validated area under the curve of 0.83 (95% confidence interval 0.74-0.92). CONCLUSION: Our study shows, in highly selected patients, that delayed MT many hours after baseline NCCT is associated with good clinical outcomes. However, older patients with diabetes, high NIHSS score and low ASPECTS may not benefit from transfer to a hub centre many hours away for MT in this model of care.